关于DSP的外文文献

合集下载

DSP外文资料9

Procedia Engineering 29 (2012) 2864 – 28681877-7058 © 2011 Published by Elsevier Ltd.doi:10.1016/j.proeng.2012.01.405Available online at vailable online at 2012 International Workshop on Information and Electronics Engineering (IWIEE) The Study of Fluorescence Thermal Measurement Based on DSPDanping Jia *, Zhuo Yuan, Hongli San, Lu Gao School of Information Science and Engineering,Shenyang University of Technology,Shenyang110870,ChinaAbstractA FFT algorithm with the DSP processor based on TMS320C5402 and fluorescence thermal measurement based on DSP are proposed. Basing on the Fast Fourier transform (FFT), the fluorescence lifetime is obtained from the tangent function of the phase angle of the first non-zeroth items of the FFT result. This method has a series of advantages such as quick calculation, high accuracy and immunity to the base signal, so the stability, real time and speediness of fluorescence thermal measurement system can be advantaged. © 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Harbin University of Science and Technology Keywords:DSP;Fluorescence lifetime;FFT;TMS320C54021.IntroductionTemperature is a basic physical quantity which express the hot and cold degree of object and is one of the most important measurement parameters in measured science and industrial process control. All the processes in nature are closely related to temperature. With the level of industrial automation increased and continuous production scale expanded the need of temperature measurement becomes greater and greater. Due to the high temperature, the high voltage, the big electromagnetic interference, traditional electronic temperature sensor can’t be used in many conditions, the alternative method must be adopted [1]. The temperature sensor based on optical fiber with the advantages of anti-electromagnetic interference, high temperature resistance, corrosion resistance, small size and other characteristics will be an effective prepared scheme .It has a very wide application prospect in industrial production. Optical fiber with a wide spectrum that light can get through, low transmission loss is convenient, whatever is used directly or to spread in short distance. Moreover the diameter of optical fiber is small and can be* Corresponding author. E-mail address : winnerjia@.2865Danping Jia et al. / Procedia Engineering 29 (2012) 2864 – 2868used in a single 、bundle 、Y or array way. The structural arrangement of optical fiber is easy and has a small size [2]. Therefore, the temperature sensor based on optical fiber is applicable for all detection objects. It also can be used in special occasions where other temperature sensors are difficult to use such as seals, the high voltage, the strong magnetic fields, radiation, rigorous proof, waterproof, corrosion resistance, especially small space or especially small parts and so on[3]. The fluorescence thermal measurement based on DSP is proposed in the article.2.The principle of temperature measurement systemAccording to the differences of processing approach of fluorescence signal, fluorescence temperature sensor can be divided into three types ,namely fluorescence intensity, the ratio of fluorescence intensity, fluorescence lifetime. The sensor based on fluorescence lifetime is used in this design. The biggest advantage using fluorescence lifetime to measure temperature is a single temperature conversion value determined by the relationship between the fluorescence lifetime .It can’t be affected by other external conditions(such as the changes of the intensity of excitation light ,the efficiency of optical fiber transmission, the changes of coupling degree and so on),thus it has more obvious advantage on the principle than the method of temperature measurement （using the intensity of fluorescence peak or the method of temperature measurement using the ratio of intensity ） which uses light intensity as the temperature sensing signal[4,5]. At present the data processing methods of fluorescent afterglow curves are the least squares method 、the integrated area ratio method L-M method and FFT. FFT is not influenced by DC component. The fluorescence lifetime is obtained from the tangent function of the phase angle of the first non-zeroth items of the FFT result. This method has a series of advantages such as quick calculation, high accuracy and immunity to the base signal. Compared with several other traditional fitting methods, the measured error of FFT is small. In addition, the method of FFT is not influenced by the base signal. thus It doesn’t need to remove the base signal when we process signal, can reduce the time of measurement and improve the resolution of measurement.3.The analysis of Fast Fourier algorithmAccording to the exponential decay law of the fluorescent afterglow, the guiding ideology of fluorescence lifetime method based on FFT algorithm is proposed .when the excitation light stops, fluorescence will not immediately disappear and decline in the form of index. The time which fluorescence disappears is called fluorescence lifetime. It is expressed by τ. Fluorescence lifetime τ is a function of temperature T and has nothing to do with the light intensity. Fluorescence decay signal can be expressed [6]:()exp(/)k f t A k t B τ=−•Δ+ 0,1,,1k N =−L (1)Where ΔT is the interval of sampling time. The n term of the Fourier transform in (1) is:1,,2,1]2exp[])2(exp[1010−⋅⋅⋅=−++Δ−=∑∑−=−=N n k N n j B k N n j t A F N k N k n ππτ (2)2866 Danping Jia et al. / Procedia Engineering 29 (2012) 2864 – 2868The other items have no relationship with the DC component except 0 item. The fluorescence lifetime is calculated by using the first non-zero entries. The tangent function of the angular value is :1111Im tan Re F Q F ϕ== (3) From the formula, it can be concluded that the phase tangent Q 1 of the first non-zero item about Fourier transform is a single-valued function of the fluorescence lifetime τ, which has no relationship with the initial light intensity A and the base signal B , Fluorescence lifetime can be calculated as:()(){}11ln cos 2/sin 2//t Q N N Q τππΔ=−⎡⎤⎣⎦ (4)4.The system of fluorescence thermal measurementThe fluorescence processing system mainly consists of: Optical fiber fluorescent temperature sensing probe 、photo electric conversion part 、the part of the light pulse trigger 、weak signal amplification 、 fast A / D converter and DSP digital signal processor and so on. Fluorescent sensing probes are directly connected with optical fiber, which becomes a part of optical fiber. Actually it is made into an end of optical fiber where there is a special phosphor which can give out fluorescence after irradiation by light. when measuring ,optical fiber fluorescence sensing probe must be fixed with measured object to achieve a number of direct contacted temperature measurement of the measured objects. Optical fiber fluorescence sensing probe and optical fiber coupler are connected with the part of photo electric conversion through conductive fiber. So the isolation with high magnetic field and other harsh environment is achieved very well. It has high dielectric strength 、the ability of resistance to strong electric and magnetic fields. The weak signals which are Processed by the multi-level amplification transmit to A / D. Finally it is processed by digital signal processor .Thus it extends the distance of temperature measurement greatly. The measurement system block diagram of fluorescence afterglow is shown in figure 1. When the entire temperature measurement system works, at first, the control circuit of MCU issued a control signal which makes Light-emitting diode sent out a light pulse. It is transmitted to the spectral slice by conductive optical fiber, then it is transmitted to optical fiber fluorescence temperature sensor probe by Spectral slice. The phosphor which is at the top of optical fiber fluorescence temperature sensor probe is excited by the excitation light pulse ,then it will give out fluorescence whose margin decay with time changing .The fluorescence whose margin decay with time changing are mixed with the trigger light pulse signal and it is transmitted back through conductive fiber. Then the fluorescence whose margin decay with time changing are separated by spectral slice, at the same time it is transmitted to the part of photo electronic conversion. The weak current which is obtained by converting is processed through multi-level amplification and A / D converter circuit and is transmitted to DSP where it will get further processing. A large amount of data which is collected is mathematically analyzed and processed to calculate fluorescence lifetime τ, thus the corresponding temperature is calculated by further calculating. Excitation light is not visible. It can’t be transmitted through spectral slice and only be reflected. Fluorescent afterglow is visible and can be transmitted through spectral slice.2867Danping Jia et al. / Procedia Engineering 29 (2012) 2864 – 2868The filter has two effects: the first is to remove a small amount of excitation light and other stray light;the second is to filter out other peaks of the fluorescence spectra and only to keep light which is near theband of the useful peak. thus it is helpful to improve the accuracy of the system.The light path map of the system is shown in fig.2:Fig. 1. The measurement system block diagram of fluorescence afterglowFig. 2. The light path map of the systemThe rising and declining curve of the fluorescence signal which is excited is generated and it is described as the first-order exponential function. Time constant is a function of temperature, however, the leakage of the excitation light, background noise and other interference signals must be suppressed effectively in practical application. The above problems are well solved by using the technology of DSP to measure the decay time of fluorescent sensing material real time. In order to fully play the fast processing data capabilities of the DSP chip, we select Maxim's high-speed A/D MAX1420. MAX1420 requires a 3.3V DC voltage to supply, whose quantify median is 12 and the maximum conversion rate of it is 60MS / S. A / D converter circuit is shown in fig.3. τ is calculated through the output data of A/D conversion which is processed by using fast Fourier transform (FFT) with the TMS320C5402, moreover the temperature is calculated according to the value of τ.5.ConclusionThe method of time constant based on fluorescence lifetime is to achieve the measurement of temperature by measuring the time constant of fluorescence lifetime. The biggest advantage of the method is that the measured temperature only depends on the time constant of fluorescence material. It has nothing to do with other variables of the system. Compared with the method of light intensity, it hasOptical lens2868Danping Jia et al. / Procedia Engineering 29 (2012) 2864 – 2868more obvious advantages on the principle. DSP chip is used to process the digital signal. Digital signal processor (DSP) is a processing digital signal chip which is invented for the application of digital signal processing. Compared with general purpose processor, it has higher performance in the field of digital signal processing. It is superior to general purpose processor in the area of low power, the execution efficiency of the instruction, cost and so on. Maxim's high-speed A / D is used by A / D converter sampling, which keeps up with the processing speed of DSP fast. Thus the high speed and stability of the system of Fluorescence Thermal Measurement are improved greatly.Fig. 3. A / D converter circuitAcknowledgementsThis work is supported by the Natural Science Foundation of Liaoning Province under grants No.20092048 and Doctor Foundation of Shenyang University of Technology. The author wishes to thank all colleagues who previously provided technical support.References[1] CUI.Z.S. "the measurement and testing of temperature ".beijing: China Metrology Publishing House,1998.[2] LIAO.Y.B. 'Fiber Optics". Qinghua University Publishing House,2000: 54-55[3] Jia, D .P. "A study of dynamic coupling in multi-channel fluorescent temperature measurement ". Proceedings of ISIST. 2004, (1):713-717.[4]V.C .Fernicola,L. Rosso, R. Galleano, "Investigations on exponential lifetime measurements for fluorescence thermometry". Review of Scientific Instruments, 2000, 71 (7): 2938-2943.[5] Jia, D .P,LIN.L.L,YUAN.W.Q. "The study of stability of high-precision optical fluorescence temperature measurement ". Review of Measurement Science, 2008.29(2):129-133.[6]Sun.W.M,Zhang.J.Z,Yu.L. 'The fitting method of the fast Fourier transform based on fluorescence lifetime." Review of optics, 2004.24(6): 838-841.。

DSP外文资料13

Optik124 (2013) 5280–5284Contents lists available at ScienceDirectOptikj o u r n a l h o m e p a g e:w w w.e l s e v i e r.d e/i j l eoReal time fabric defect detection system on an embedded DSP platform Jagdish Lal Raheja a,Bandla Ajay b,Ankit Chaudhary c,∗a Machine Vision Lab,CEERI/CSIR,Pilani,RJ,Indiab Department of Electrical&Electronics Engineering,BITS,Pilani,RJ,Indiac Department of Electrical&Computer Engineering.,The University of Iowa,USAa r t i c l e i n f oArticle history:Received1March2012 Accepted18March2013Keywords:Fabric defectsTextureGray level co-occurrence matrix DSP kitEnergy computationSliding windowFDDS a b s t r a c tIn industrial fabric productions,automated real time systems are needed toﬁnd out the minor defects.It will save the cost by not transporting defected products and also would help in making company image of quality fabrics by sending out only undefected products.A real time fabric defect detection system (FDDS),implemented on an embedded DSP platform is presented here.Textural features of fabric image are extracted based on gray level co-occurrence matrix(GLCM).A sliding window technique is used for defect detection where window moves over the whole image computing a textural energy from the GLCM of the fabric image.The energy values are compared to a reference and the deviations beyond a threshold are reported as defects and also visually represented by a window.The implementation is carried out on a TI TMS320DM642platform and programmed using code composer studio software.The real time output of this implementation was shown on a monitor.© 2013 Elsevier GmbH. All rights reserved.1.IntroductionIdentifying defects in fabric is a major concern for fabric indus-tries as it is important for their brand images for quality products. Automated visual inspection system to detect possible defects in fabric provides a more reliable and consistent quality control process than human eye view.Many research works has been demonstrated in this area but a robust real time system for color fabrics is still needed.Efﬁcient algorithms are also needed to reduce the response time in real time.Karayiannis[1]wrote for the need of a real time system aboutﬁfteen years back,“In the best case,a human can detect not more that the60%of the present defects and he cannot deal with fabric wider than2meters and moving faster than30m/min”.However,implementing automated visual inspec-tion has to meet with some challenges.A number of techniques for describing image texture have been proposed in the research literature.A good review on these techniques is described in[2].There are a variety of defects occurring in fabrics,in terms of texture pattern violation,weaving defects,yarn misplacements etc. Tuceryan[3]deﬁnedﬁve major categories of features for texture analysis:statistical,geometrical,structural,model based and spec-tral features.Spectral features are based on the periodicity of the texture feature.The high degree of periodicity of basic texture prim-itives,such as yarns in the case of textile fabric,permits the usage ∗Corresponding author.E-mail addresses:jagdish@ceeri.ernet.in(J.L.Raheja),ajayb5632@ (B.Ajay),ankit-chaudhary@(A.Chaudhary).of spectral features for the detection of defects.Also,Chan[4]has been able to identify defects on simulated,real fabrics and then could classify these defects based on some extracted parameters. To perform real time critical operations for an industrial applica-tion,it requires specialized hardware in operations,to ensure faster decision making time.The wide variety of defects and textures encountered in fab-ric industries are difﬁcult to identify by a fully automated visual inspection system.This paper limits the variety of fabric defects to some simple ones as those described by[5,6].This paper is organized as follows.In the next section the background of the application and methods are discussed.A brief introduction of GLCM is shown in Section3and mathematical model for fabric detection is described in Section4.Hardware and software com-ponents used for our systems are detailed in Sections5and6 ter section7discusses obtained results followed by conclusions.2.BackgroundApplication of co-occurrence texture features for interpreting fabric defects are discussed in this section.Many research works could be found on real time fabric defect implementation using different methods for real time applications.Guang[7]presents a defect detection algorithm based on distance calculation among images where he claims accracy of92%.Han[8]use Gaborﬁlter with GA to detect fabric defects.Shu[9]detailed a method of detec-ting the fabric defects automatically based on multi-channel and multi-scale Gaborﬁltering.It is based on energy response from the0030-4026/$–see front matter© 2013 Elsevier GmbH. All rights reserved. /10.1016/j.ijleo.2013.03.038J.L.Raheja et al./Optik124 (2013) 5280–52845281convolution of Gaborﬁlter banks in different frequency and ori-entation domains.These sets of banks are to be convolved with the image to generate output as a set of images.Further,these images need to undergo smart segmentation processing to iden-tify defective regions.After this,these images should be under a decision making system which decides for defects in the image. The implementation using Gaborﬁlter banks is computationally very expensive.Theseﬁlters are16x16arrays,which need to be convolved in two-dimensional spaces with the image.Karayiannis[1]presented multiresolution decomposition based real time FDDS.He used AT&T32C DSP board for implementation while a pentium PC was used with PC bus to show output.Jia[10] presented an edge-based seedﬁlling algorithm for fabric detection. Different edge detections methods with connecting broken edges have been discussed in[11].Shi[12]used a segmentation method based on local contrast deviation for fabric defect detection.Real time region of interest cropping for faster processing is discussed in[13].Zhang[14,15]presented a multiple window gray ratio based algorithm for cord fabric detection.Although it work in real time but cord is a different kind of fabric and he used gray images toﬁnd the defect.Siew[16]has shown assessment of carpet wear using spatial gray level dependence matrix(SGLDM).Also methods similar to GLCM have been applied to wood inspection[17],surface defect detection[18],and fabric defect detection[19].The original investi-gation into SGLDM features was pioneered by Harlick[20].Texture features such as energy,entropy,contrast,homogeneity,and cor-relation are then derived from the co-occurrence matrix.However only six of such features have been used for the defect detection on wood and fabric defect detection has been shown with only two of these six features.Interestingly Conners[17]has used six features of co-occurrence matrix,to identify nine different kinds of surface defect in wood.Also,Tsai[19]has detailed fabric defect detection while using only two features and achieved a classiﬁcation rate as high as96%.Zhang[21]came with synthetic aperture radar to over-come disadvantages of brightness effect of light at the time of defect detection but results were not so clear.As far as real time implementation is concerned,Bariamis[22] has proposed an FPGA based architecture for real time image fea-ture extraction,which describes a FPGA based system to compute GLCM.Mak[23]used iterative tensor tracking using spatial his-togram[24]of gradient orientations on a GPU NVIDIA GTX260 O.C.;He showed good compuatation results as the image resolution increases.3.Gray level co-occurrence matrixThe GLCM method,also known as the spatial gray-level dependence method,has been widely used in texture analysis,as discussed in previous section.It is based on repeated occurrences of different gray level conﬁgurations in a texture.Automatic visual inspection techniques for textured images generally compute a set of textural features in the spatial domain or in the spectral domain. The co-occurrence probabilities provide a second order method for generating texture features.A brief presentation of the GLCM method follows but a more complete explanation is provided by Haralick[20,25].This GLCM matrix contains the conditional joint probabilities of all pair wise combinations of gray levels given two parameters: inter-pixel distance(say d)and inter-pixel orientation(sayÂ).They were measured counter clockwise from the horizontal axis.The probability measure is deﬁned by Barber[26]in terms of number of occurrences of gray levels within a region for a particular value of displacement and orientation.Different statistical information can be determined from each GLCM.However,statistics that are gray level shift invariant are important so that the classiﬁcation is not a function of tone.Eight such shift invariant statistics are presented in Haralick’s work[25].Energy(ENG),Entropy(ENT)and maximum probability(MAX)are linear(scale and shift)invariant statistics.4.Defect detection model using GLCMThe system architecture was designed on the principle of com-ponents[27].There are a number of different parameters that must be indicated in order to generate co-occurrence data:window size (n),orientation(Â),pixel separation distance(d)and the number of quantized gray levels(G).This paper proposes a defect detec-tion method where a moving window,slides over the whole image in steps computing GLCM at each position.Then energy feature is extracted from the GLCM.These energy values are compared to that of a reference and then any deviation beyond a threshold is identiﬁed as defect.For classiﬁcation studies,the window size is oftenﬁxed and assumes that sufﬁcient area of the texture is represented to capture an appropriate measurement.TypicallyÂis set to a value among 0,45,90and135degrees,since this is easiest to implement.To program the generation of co-occurrence data for arbitrary orien-tations is not reasonable.Short pixel separation distances have typi-cally achieved the best success[21].Various values of d can yield dif-ferent features.However,it is possible if there is any priori method for selecting its value based on the given image characteristics.Here for defect detection,parameters were chose as d=1,Â=0, G=256(for8bit image)and window size to50×50.Haralick[25] introduced several texture features on GLCM where we use energy feature to identify defects.The minimum window size should be such that it should include the at least one periodic pattern of the texture.The fabric should have a repeating pattern of the image in the window as.A larger window which includes three or more periodic patterns will identify the defects but it will fail to properly localize the defective region in the fabric.That is,the whole window region will be shown as defective in the event of an observable difference in energy.Although a larger window scans the image in less number of steps,it takes more time to perform extraction operation of the region of interest.The number of computations fora single image is as follows(1)Number of steps to scan the whole image=Rw×Cw(1) where R and C are the number of rows and columns in image and w is the window size.‘ ’denotes theﬂoor value.(2)Number of computations at each position of the window: (2.1)to extract the window region from the input image=w×w(2) (2.2)to compute GLCM=G×G(3)So the total number of steps required to identify defects in an image of size R×C isR×C×w×w×G×G=R×C×G×G(4) Table1shows the number of computations for an image of dimensions R=461,C=512and G=256(8bit gray scale).Hence,the larger windows slightly reduce the numbers of com-putations.The window size should be chosen to balance the two factors i.e.localizing the defect and reducing computations.Initially to identify important textural features,the deviations of energy along with other features were noted as a window moved over the image.The moving window type defect detection logic using energy, correlation homogeneity and contrast was also tested.Out of these5282J.L.Raheja et al./Optik 124 (2013) 5280–5284Fig.1.Images on the left are defective image samples,images on the right are graphs of the features (1)contrast,(2)homogeneity,(3)energy,and (4)correlation,extracted from the sample image as a 50×50window slides over the whole image.The x -axis denotes the count of movement of the window as it slides through the whole image.The window moves in steps of 50pixels.features,energy feature always gave prominent peaks in the defec-tive regions compared to the remaining textural features.This was tested for several defective images,which also showed observable peaks for energy feature.Hence this energy feature was used to identify texture defects in the image.Fig.1shows samples and their graph of their features.It concludes that energy feature contains information to identify the defects in the image.The defect detec-tion results of this logic,using energy feature are implemented in MATLAB.Fig.2shows few results of defect detection.5.HardwareThe FDDS was implemented on a Texas Instruments’TMS320DM642Evaluation Module.It has three on board video ports with 2decoders and 1encoder.The Input video was taken from a digital color camera in NTSC format.The gray level informa-tion was extracted from one of the channels.The direct memoryTable 1Total number of computations.Window sizeNo of steps required to slide through the whole image aTotal no of computations3025515e9509014e91002013e9aA count variable keeps track of this number in the program.access (DMA)feature of the DSP kit facilitates memory to memory transfer operations without disturbing the processor operation cycle.The image was loaded to the memory from SDRAM through this DMA feature.Image processing functions were performed and the ﬁnal output with markings of defects if any,is transferred to the output buffer.This code was tested for real time opera-tions by taking the input from a real fabric with somedefectsFig.2.Defect detection results of moving window implementation by using energy feature for texture defect detection.J.L.Raheja et al./Optik 124 (2013) 5280–52845283Fig.3.Flow of control in the program.as described in [5,6].Fig.3shows the algorithmic ﬂow of the proposed approach.6.SoftwareThis paper uses code composer studio IDE with additional libraries which come along with the Texas Instruments DSP kit to implement FDDS.This Texas Instruments Chip support Libraries (CSL)APIs [29]were used for conﬁguring and controlling the DSP on-chip peripherals.The proposed logic was implemented after testing it in MATLAB ®[28]environment running onMicrosoft Windows ®.The implementation program in code composer stu-dio was written in C language.The libraries in code composerFig.4.FDDS results on the monitor.which have implemented certain register level operations whichthe user can not do explicitly.It also has some library functions performing major chip level operations which are named under CSL.It also has predeﬁned video/image-processing libraries which reduce the program development time for the user.The software also provide supports to perform functions relating to video port handles,exchange etc.After the code has been developed,the program was loaded on to the DSP kit.The program includes some necessary information about location of memory segment labels,location of main,heap location etc.Once the program is loaded on to the chip it runs on the DSP kit performing the implemented program.More information about programming using Code Composer Studio can be found in the manuals [29,30]which come with the DSP kit.7.ResultsFDDS implemented on a Texas Instruments TMS320DM642Evaluation Module,process one frame at a time.Moving window analysis is performed on each frame and the defects are marked with white squares on the input fabric image.This image with defective regions marked as white squares is shown on a moni-tor connected to the DSP kit.The camera was positioned to capture 4cm 2area of the fabric in each frame.The response time of results on the monitor for the frames processed is 0.5s.In laboratory setup,the fabric was manually moved to get the next focus area.The tech-nique is still primitive in terms of setting up motors to keep the fabric rolling,moving the camera etc.,which are required for a full scale industrial use.The results of FDDS on the monitor are shown in Fig.4.8.ConclusionsThe implementation of an automated visual inspection system for fabric defect detection is a challenging task.Most of the defect detection algorithms implemented previously in this area require some modiﬁcations in the program to properly identify the defects in case a new texture is needed to process.This paper implements a real time FDDS on a Texas Instruments DSP kit with input taken from a piece of fabric.The properties to identify the defects were tested using MATLAB environment.The program is fully automated and do not need any manual interaction once started.The results show that energy feature would be best to use to get information to distinctly identify defects in fabric textures.The results shown in this paper clearly identify the parameter to detect defects in tex-tures.Also,the algorithm is simple in its decision making process and can be modiﬁed for extension.The response time is quite good and easily implemented in industrial applications.AcknowledgmentsAuthors would like to thank the Director of Central Electronic Engineering Research Institute/Council of Scientiﬁc and Industrial Research (CEERI/CSIR),Pilani,for providing research facilities,and for his active encouragement and support.References[1]Y.A.Karayiannis,R.Stojanovic,P.Mitropoulos,C.Koulamas,T.Stouraitis,S.Koubias,G.Papadopoulos,Defect detection and classiﬁcation on web tex-tile fabric using multiresolution decomposition and neural networks,in:Proceedings of 6th IEEE International Conference on Electronics,Circuits and Systems,5–8September,1999,pp.765–768.[2]P.M.Mahajan,S.R.Kolhe,P.M.Patil,A review of automatic fabric defect detec-tion techniques,p.Res.1(2)(2009)18–29.[3]M.Tuceryan,A.Jain,Texture analysis,in:C.H.Chen,L.F.Pau,P.S.P.Wang (Eds.),The Handbook of Pattern Recognition and Computer Vision,2nd ed.,World Scientiﬁc Publishing Co.,Singapore,1998,pp.207–248.5284J.L.Raheja et al./Optik124 (2013) 5280–5284[4]C.Chan,G.Pang,Fabric defect detection by Fourier analysis,in:ConferenceRecord of the1999IEEE Industry Applications Conference,Thirty-Fourth IAS Annual Meeting,vol.3,1999,pp.1743–1750.[5]M.Tunák,A.Linka,Simulation and Recognition of Common Fabric Defects,in:In Proceedings of13th International Conference on Structure and Structural Mechanics of Textiles,Liberec,Czech Republic,2006,pp.363–370.[6]J.Kula,M.Tunak,A.Linka,Real-time quality control of fabric based on multivari-ate control charts,in:Proceedings of7th International conference on TEXSCI, September6–8,CZ Republic,2010.[7]G.Zheng,J.Wang,Fabric defect detection method based on image distancedifference,in:Proceedings of8th International Conference on Electronic Mea-surement and Instruments,2007,pp.2822–2825.[8]H.Runping,Z.Lingmin,Fabric defect detection method based on Gaborﬁltermask,in:WRI Global Congress on Intelligent Systems,May19–21,vol.3,2009, pp.184–188.[9]Y.Shu,Z.Tan,Fabric defects automatic detection using Gaborﬁlters,in:Proceedings of the5th World Congress on Intelligent Control and Automation, Hangzhou,China,2004,pp.3378–3380.[10]X.Jia,Fabric defect detection based on open source computer vision libraryOpenCV,in:Proceedings of2nd International Conference on Signal Processing Systems,5–7July,2010,V1-342–V1-345.[11]A.Chaudhary,K.Vatwani,T.Agrawal,J.L.Raheja,A vision-based method toﬁndﬁngertips in a closed hand,rm.Process.Syst.8(September(3))(2012) 399–408.[12]M.Shi,R.Fu,S.Huang,Y.Guo,B.Xu,A method of fabric defect detection usinglocal contrast deviation,in:Proceedings of2nd International Congress on Image and Signal Processing,October17–19,2009,pp.1–5.[13]J.L.Raheja,K.Das,A.Chaudhary,Fingertip detection:a fast method with naturalhand,IJESC3(December(2))(2011)85–89.[14]W.Zhang,J.Zhang,Y.Hou,S.Geng,MWGR:a new method for real-timedetection of cord fabric defects,in:Proceeding of International Confer-ence on Advanced Mechatronic Systems,18–21September,2012,pp.458–461.[15]W.Zhang,Q.Zhao,L.Liao,Development of a real-time machine vision systemfor detecting defeats of cord fabrics,in:Proceeding of International Conference on Computer Application and System,22–24October,2010,pp.539–543.[16]L.H.Siew,R.M.Hodgson,E.J.Wood,Texture measures for carpet wear assess-ment,IEEE Trans.Pattern Anal.Mach.Intell.10(1)(1988)92–105.[17]R.W.Conners,C.A.Harlow,A theoretical comparison of texture algorithms,IEEETrans.Pattern Anal.Mach.Intell.PAMI–2(3)(1980)204–222.[18]J.Iivarinen,Surface defect detection with histogram-based texture features,Proc.SPIE4197(2000)140–145.[19]I.Tsai,C.Lin,J.I.Lin,Applying an artiﬁcial neural network to pattern recognitionin fabric defects,Text.Res.J.65(March)(1995)123–130.[20]R.M.Harlick,K.Shanmugam,I.Dinstein,Textural features for image classiﬁca-tion,IEEE Trans.Syst.Man Cybern.3(1973)610–621.[21]L.Zhang,X.H.He,D.S.Si,H.Zhang,A fabric defects detection method using SARimaging,in:Proceedings of6th International Conference on Wireless Com-munications Networking and Mobile Computing,23–25September,2010,pp.1–4.[22]D.G.Bariamis,D.K.Iakovidis,D.E.Maroulis,S.A.Karkanis,An FPGA-based archi-tecture for real time image feature extraction,in:Proceedings of the17th International Conference on Pattern Recognition,23–26August,2004,pp.801–804.[23]K.L.Mak,X.W.Tian,Iterative tensor tracking using GPU for textile fabric defectdetection,in:Proceedings of International Conference on Green Circuits and Systems,21–23June,2010,pp.375–380.[24]J.L.Raheja,M.B.L.Manasa,A.Chaudhary,S.Raheja,ABHIVYAKTI:hand ges-ture recognition using orientation histogram in different light conditions,in: Proceedings of the5th Indian International Conference on Artiﬁcial Intelli-gence,India,2011,pp.1687–1698.[25]R.M.Haralick,Statistical and structural approaches to textures,Proc.IEEE67(5)(1979)786–804.[26]D.G.Barber,E.F.LeDrew,SAR sea ice discrimination using texture statistics:amultivariate approach,Photogramm.Eng.Remote Sens.57(April(4))(1991) 385–395.[27]A.Chaudhary,B.K.Verma,J.L.Raheja,Product line development architecturalmodel,in:Proceedings of3rd IEEE International Conference on Computer Sci-ence and Information Technology,9–11July,2010,pp.749–753.[28]MATLAB,[29]TMS320DM642Evaluation,ModuleTechnical Reference.[30]TMS320manuals,。

DSP外文资料43

Comparison of impedance measurements in a DSP using ellipse-ﬁt and seven-parameter sine-ﬁt algorithmsPedro M.Ramos a,b,*,Fernando M.Janeiro a,c ,TomášRadil aaInstituto de Telecomunicações,Av.Rovisco Pais 1,1049-001Lisbon,PortugalbDEEC,Instituto Superior Técnico,UTL,Av.Rovisco Pais 1,1049-001Lisbon,Portugal cUniversidade de Évora,Departamento de Física,Rua Romão Ramalho,n °.59,7000-671Évora,Portugala r t i c l e i n f o Article history:Received 13March 2009Received in revised form 12May 2009Accepted 14May 2009Available online 23May 2009Keywords:Impedance measurements Digital signal processor DSP based instrument Sine-ﬁt algorithms Ellipse-ﬁt algorithmsExperimental uncertainty analysisa b s t r a c tIn this paper,two DSP implemented algorithms for impedance measurements are com-pared.Previously published results demonstrate the usefulness of sine-ﬁt algorithms and ellipse-ﬁt algorithms for impedance measurements where two channels are simulta-neously acquired with analog to digital converters.The comparison between the two implemented algorithms is done by analyzing the average execution time,memory requirements and experimental standard deviation of the estimated impedance parame-ters.For the ﬁrst time,the seven-parameter sine-ﬁt algorithm is adapted so that it does not need the construction and manipulation of its largest matrix thus requiring less overall memory.This improvement can be used to acquire and process more samples (leading to reduced experimental standard deviations of the estimated parameters)with the same memory size.Ó2009Elsevier Ltd.All rights reserved.1.IntroductionTraditional impedance measurement techniques are being improved [1]due to the development of highly efﬁ-cient analog to digital converters,the use of digital signal processing techniques and the rapid development of new,more powerful digital signal processors that can now per-form millions of operations per second.Impedance mea-surements are a relevant topic with many new developments published recently.Some works focus on impedance sensor measurements [2]while others make use of recently published signal processing algorithms to improve impedance estimation [3,4].In [5]a high imped-ance spectroscopy probe is analyzed and in [6]a novel method to estimate the individual parameters of the com-ponents from an impedance model was presented.A simple impedance measurement technique presented in [7]and improved in [8]minimizes the analog front-end circuitry to reduce the inﬂuence of its frequency depen-dence.This technique is based on the volt-ampere method where a reference impedance and an unknown impedance are placed in series and supplied with the same current from a sine generator.The voltage across each impedance is simultaneously acquired with a two-channel analog to digital converter and the impedance’s magnitude and phase are extracted using digital signal processing algorithms.Sine-ﬁt algorithms are an option to estimate the sine signal parameters from a set of acquired samples.Since the ratio between the sine frequency and the sampling rate is,most of the times,not accurately known,the algorithms must also estimate the sine frequency.This makes the problem a nonlinear regression that is solved using an iter-ative procedure which is called the four-parameter sine-ﬁt algorithm [9].For impedance measurements,this algo-rithm can be applied to each channel separately,however the overall experimental uncertainty can be reduced by0263-2241/$-see front matter Ó2009Elsevier Ltd.All rights reserved.doi:10.1016/j.measurement.2009.05.005*Corresponding author.Address:Instituto de Telecomunicações,Av.Rovisco Pais 1,1049-001Lisbon,Portugal.E-mail addresses:pedro.ramos@Lx.it.pt (P.M.Ramos),fmtj@uevora.pt (F.M.Janeiro),tomas.radil@lx.it.pt (T.Radil).Measurement 42(2009)1370–1379Contents lists available at ScienceDirectMeasurementj ou r na l h om e pa ge :w w w.e lse vi e r.c om /lo c at e /me as u r em en tforcing the common frequency in what is called the seven-parameter sine-ﬁt algorithm[10].Nevertheless,the itera-tive nature of the sine-ﬁt algorithms results in a large num-ber of operations that the processor must execute.Another issue that can affect the performance of sine-ﬁt algorithms implemented in DSP devices is the memory requirements. In fact,the seven-parameter algorithm as proposed in[10] must build a matrix with seven columns and2N rows,cor-responding to the total number of samples acquired in both channels.This large matrix and its manipulation to-gether with the limited memory available in the DSP re-stricts the number of samples that can be processed.In this paper,the algorithm proposed in[10]is shown to be applicable in a memory restricted system by bypassing the need to build the2NÂ7matrix.This modiﬁcation al-lows the DSP to process more samples to estimate the sine-wave parameters with the seven-parameter sine-ﬁt.Although the frequency must be accurately known for a correct estimation of the sine signals’amplitudes,phases and dc components by the sine-ﬁt algorithms,in imped-ance measurements an accurate frequency value is not needed.The time information,and thus the sine signals frequency,can be removed from the problem by making an XY plot of the two waves.The result is an ellipse whose parameters can be estimated by ellipse-ﬁt algorithms such as the one presented in[11]and then improved in[12].The sine signals’parameters can then be extracted from the el-lipse parameters as proposed in[13].This algorithm has been recently adapted and optimized for use in DSP based impedance measurements[14].In the modiﬁed ellipse-ﬁt method published in[14]there is no limit on the number of samples that can be used since only nine values must be stored and the samples themselves can be discarded after their contribution to the nine stored values is taken into account.These characteristics make the algorithm a prime candidate for efﬁcient implementation in a DSP based impedance measurement instrument.In this paper,the performance of the DSP implementation of the ellipse-ﬁt algorithm and the seven-parameter sine-ﬁt algorithm is compared through the evaluation of the memory requirements,speed of execution and experimental uncer-tainty of both algorithms.The system used for the comparison is based on a commercialﬂoating-point DSP kit,with few external electronics,for baseline assessment and require-ments deﬁnition.The performance tests were executed at fre-quency f=1kHz,for impedances with magnitudes from 100X to15k X and phases in the±90°range.The paper is divided into four sections including the introduction(Section1)and the conclusions(Section4). The description of the two algorithms is presented in Sec-tion2together with the measurement setup and the algo-rithms DSP implementation details.The performance comparison is performed in Section3where the experi-mental results are shown and analyzed.The conclusions are presented in Section4.2.Measurement methodThis section describes the measurement setup and method as well as the two signal processing algorithms used.The measurement method is based on measuring the voltages across two impedances placed in series and supplied with a current from a sine generator that sets the measurement frequency.This method is also com-monly called the series comparison method.The voltages are simultaneously acquired by two differential input acquisition channels and the samples are then transmitted to the DSP in which the signal processing algorithms are implemented.The two acquired sine signals are modeled byu1ðtÞ¼C1þD1cosð2p ftþ/1Þu2ðtÞ¼C2þD2cosð2p ftþ/2Þð1Þwhere C k is the dc component,D k is the sine amplitude and /k is the phase of channel k.The two algorithms compared in this paper,estimate the sine signals amplitude ratio D2/D1,their phase differ-ence D/=/2À/1and the corresponding dc components. In addition,the seven-parameter algorithm also estimates the common frequency f.2.1.Measurement circuitThe basic measurement circuit is presented in Fig.1. The use of different reference impedances is needed so that the amplitude of the acquired signals is as close as possible to the ADC input range,thus maximizing the use of the ADC dynamic range.In addition,the different reference impedances are also used to avoid ellipse degeneration [14].Two analog switches are used to select the appropri-ate reference impedance from aﬁxed set of impedances. Theﬁrst switch selects the reference impedance through which the currentﬂows and the second switch selects where the ADC channel is connected.This switch isneededFig.1.Basic measurement circuit.FG is the function generator(Agilent 33250A).The DSP controls the two switches to select the reference impedance(Z R)in order to maximize the ADC input voltages.The ADCs included in the acquisition channels are also controlled by the DSP and sample both channels simultaneously.Z is the impedance under measurement.P.M.Ramos et al./Measurement42(2009)1370–13791371to assure that theﬁrst switch’s impedance is not measured together with the reference impedance.The impedance of the second switch is not measured because it is in series with the voltage circuit of the four-terminal measurement setup of the reference impedance.This setup removes the error caused by the impedance of the switches and is a modiﬁcation of the four-wire impedance measurement, used especially for the measurement of low resistors by eliminating the inﬂuence of resistances of connecting wires and terminals.The switches were implemented using the Analog Devices ADG4078-Channel,high perfor-mance analog multiplexer.With channel1sampling the voltage across the refer-ence impedance and channel2sampling the voltage across the unknown impedance,the estimated impedance magni-tude isj Z j¼D2D1j Z R jð2Þwhere j Z R j is the magnitude of the reference impedance at the measurement frequency.The estimated impedance phase isuZ ¼D/þu ZRð3Þwhere u ZR is the reference impedance phase at the mea-surement frequency.The acquisition process consists in:(i)an initial acquisi-tion and quick estimation of the unknown impedance magnitude and phase;(ii)with this initial estimate,the DSP selects the reference impedance that best matches the unknown impedance magnitude and avoids ellipse degeneration;(iii)ﬁnal acquisition using the selected ref-erence impedance;(iv)estimation of the sine parameters using either the ellipse-ﬁt algorithm or the seven-parame-ter sine-ﬁt algorithm;(v)estimation of the impedance parameters using(2)and(3).Note that,for the seven-parameter sine-ﬁt,the step(ii)does not have to avoid ellipse degeneration.Nevertheless,to ensure comparable results between the two methods,the same procedure is done for both algorithms.2.2.Seven-parameter sine-ﬁtSine-ﬁt algorithms were standardized in[9]for the characterization of ADCs.Their basic objective is to deter-mine a set of parameters of theﬁtting model(analytical expression of a sine signal)that minimizes the sum of the squared errors between the model and the sampled data.Two sine-ﬁt algorithms are described in[9]:the three-parameter algorithm and the four-parameter algo-rithm.Theﬁrst estimates the phase,amplitude and dc component when the signal frequency is known and is a multiple linear regression method that requires no itera-tions.In the four-parameter algorithm,the frequency is also estimated which makes the model nonlinear and re-quires an iterative procedure in order to obtain the best parameters.The convergence of the algorithm depends on the initial estimates of the parameters.For two-channel systems with a common frequency,as is the case of imped-ance measurements,all the information from both records should be used to obtain a better estimate of the common frequency.This is obtained with the seven-parameter sine-ﬁt[10]which is also an iterative algorithm much like the four-parameter.The main difference is that it estimates, in each iteration,the two amplitudes,the two phases,the two dc components and the common frequency.In[15] the effectiveness of this method was demonstrated by the smaller Cramér Rao Lower Bound(CRLB)of the fre-quency and phase difference when compared with the application of two four-parameter sine-ﬁts.In the seven-parameter sine-ﬁt,the estimated parame-ter vector for each iteration isxðiÞ¼AðiÞ1BðiÞ1CðiÞ1D fðiÞAðiÞ2BðiÞ2CðiÞ2h i Tð4Þwhere A k and B k are the in-phase and quadrature ampli-tudes from which the sine amplitudes D k and phases/k are obtained[9].D f(i)is the frequency correction that up-dates the estimated common frequency.The iterative pro-cess ends when the relative frequency adjustment D f(i)/f(i) is below a preset threshold–currently set at5Â10À7. After the iterative part of the algorithm is completed and theﬁnal frequency is determined,the amplitudes A k and B k are determined with the three-parameter sine-ﬁt.The estimated parameter vector is obtained fromxðiÞ¼½½DðiÞ T DðiÞ À1½½DðiÞ T y ð5Þwhere y is the concatenated sample vectory¼½u1;1u1;2...u1;N u2;1u2;2...u2;N Tð6Þandð7Þwith(x=2p f(i))QðiÞk¼cosðx t k;1Þsinðx t k;1Þ1cosðx t k;2Þsinðx t k;2Þ1.........cosðx t k;NÞsinðx t k;NÞ1266664377775ð8ÞandrðiÞk¼a k;1a k;2...a k;N266664377775with a k;n¼À2p AðiÞk t k;n sinðx t k;nÞþ2p BðiÞk t k;n cosðx t k;nÞ:ð9ÞThe calculation of D(i)requires2NÂ7words(in DSP imple-mentations,each word usually corresponds to a32-bit long single precisionﬂoat[16]).[D(i)]T D(i)requires an addi-tional7Â7words while[D(i)]T y requires7words.Overall, the method requires17N+63words(the17N part corre-sponds to2N for the samples,14N for D(i)and1N for the FFT used in the interpolated discrete Fourier transform (IpDFT)[17]to estimate the initial frequency).Due to the restricted DSP internal memory and the space occupied by the program itself and other internal variables,the number of samples N is limited to1024in case of the em-ployed DSP.Note that,although external DSP memory is1372P.M.Ramos et al./Measurement42(2009)1370–1379available and can be used,this would considerably slow down the algorithm.The proposed optimization to remove much of the memory constraints associated with the num-ber of acquired samples is based on the direct calculation of[D(i)]T D(i)and[D(i)]T y further optimized considering simultaneous acquisition(i.e.,t1,n=t2,n=t n)½DðiÞ T DðiÞ¼E1;1E1;2E1;3E1;4000E1;2E2;2E2;3E2;4000E1;3E2;3N E3;4000E1;4E2;4E3;4E4;4E4;5E4;6E4;7000E4;5E1;1E1;2E1;3000E4;6E1;2E2;2E2;3000E4;7E1;3E2;3N26666666666643777777777775ð10ÞwithE1;1¼X Nn¼1cos2ðx t nÞ;E1;2¼X Nn¼1cosðx t nÞsinðx t nÞ;E1;3¼X Nn¼1cosðx t nÞ;E1;4¼X Nn¼1cosðx t nÞa1;n;E2;2¼X Nn¼1sin2ðx t nÞ;E2;3¼X Nn¼1sinðx t nÞ;E2;4¼X Nn¼1sinðx t nÞa1;n;E3;4¼X Nn¼1a1;n;E4;4¼X Nn¼1a21;nþa22;n;E4;5¼X Nn¼1cosðx t nÞa2;n;E4;6¼X Nn¼1sinðx t nÞa2;n;E4;7¼X Nn¼1a2;n:Notice that[D(i)]T D(i)is a symmetric matrix.Finally,for[D(i)]T y½DðiÞ T y¼P Nn¼1cosðx t nÞu1;nP Nn¼1sinðx t nÞu1;nP Nn¼1u1;nP Nn¼1u1;n a1;nþu2;n a2;nP Nn¼1cosðx t nÞu2;nP Nn¼1sinðx t nÞu2;nP Nn¼1u2;n266666666666666666666666666664377777777777777777777777777775:ð11ÞWith this method,the memory usage is reduced to3N+63 (the3N part corresponds to2N for the samples and1N for the FFT used in the IpDFT).For the initial A k and B k estimates,the three-parameter sine-ﬁt is applied to each channel using the frequency ob-tained with the IpDFT.These two algorithms are also implemented in the DSP.In Fig.2,96samples acquired with the proposed system at1kHz measurement frequency sampled at48kS/s are shown together with the reconstructed sine signals using the seven-parameter algorithm.In this case D2/D1’1.393and D/’À44.2°.2.3.Ellipse-ﬁtThe ellipse-ﬁt algorithm estimates the ellipse parame-ters that bestﬁt the XY pairs of voltages from the two channels.From the ellipse parameters,the sine amplitudes, dc components and the phase difference are determined. The ellipse-ﬁt algorithm is a non-iterative method based on Lagrange multipliersﬁrst published in[11]and im-proved in[12].The reﬁned implementation,ﬁrst published in[14],requires the construction of small matrices(3Â3) instead of the NÂ3large matrix used in[12],their multi-plication and inversion and also the determination of the eigenvectors of a3Â3matrix.Through algebraic manipu-lation the common frequency can be eliminated from(1) u1ÀC1D12þu2ÀC2D22À2ðu1ÀC1Þðu2ÀC2ÞD1D2ÂcosðD/ÞÀsin2ðD/Þ¼0:ð12ÞEq.(12)corresponds to a conicFðu1;u2Þ¼au21þbu1u2þcu22þdu1þeu2þg¼0;ð13Þwhich describes an ellipse,whenb2À4ac<0:ð14ÞBy introducing a scaling constant q,this condition can be set to4ac-b2=1.For conic(12)to correspond to an ellipse three condi-tions must be veriﬁed:D1–0;D2–0and cos2(D/)<1, i.e.,(12)is either a line segment(whenever at least one of the conditions is not true)or an ellipse.From the ellipse parameters,the sine parameters are obtained throughD1¼1ﬃﬃﬃﬃﬃﬃqap;D2¼1ﬃﬃﬃﬃﬃqcp;ð15ÞC1¼2cdÀbeb2À4ac;C2¼2aeÀdbb2À4acð16ÞP.M.Ramos et al./Measurement42(2009)1370–13791373andcosðD/Þ¼ÀsignðqÞÂb2ﬃﬃﬃﬃﬃacp:ð17ÞFor impedance measurements the actual sine amplitudes are of no interest since it is the amplitude ratio D2/D1that determines the impedance magnitude.This ratio is given byD2 D1¼ﬃﬃﬃacrð18Þtherefore the scaling constant q does not need to be deter-mined.Also,the sign of the scaling constant,required in (17),can be replaced by the sign of parameter a(due to scaling,parameters a and c can both be negative which corresponds also to q<0).The only parameter that needs further calculation is the sign of D/which is unavailable from(17)but can be re-trieved from the rotation direction of the ellipse–if the el-lipse is constructed clockwise D/>0and D/<0if it is constructed counterclockwise.The sign of the phase differ-ence between two consecutive sample pairsd n jn¼1;...;NÀ1¼sign arg u1;nþ1À1X Ni¼1u1;iþj u2;nþ1À1X Ni¼1u2;i!"# (Àarg u1;nÀ1X Ni¼1u1;iþj u2;nÀ1X Ni¼1u2;i!"#)ð19Þcan be used to determine whether the ellipse is being con-structed clockwise or counterclockwise.The ellipse center is estimated using all the acquired samples.However,due to the presence of noise,some of the consecutive samples may give the wrong rotation direction,therefore a voting system was implemented.The sign of the sum of the votes deﬁned in(19)determines the sign of D/signðD/Þ¼ÀsignX Nn¼1d n!:ð20ÞThis system is specially designed for noisy and low ampli-tude situations.For example,for f=1kHz,f S=96kS/s and 1024samples,this voting system has never failed in simu-lations for SNR>15dB.The ellipse-ﬁt requires2N+42words(2N for the sam-ples,four3Â3matrices and two3element vectors).With the current sine-ﬁt implementation,the memory needed by the ellipse-ﬁt is not a big improvement from the mem-ory needed by the seven-parameter sine-ﬁt.The main advantage is the total number of operations–not only for the matrix construction and manipulations but also due to the fact that the sine-ﬁt is iterative and the el-lipse-ﬁt is not.In Fig.3,96samples acquired at48kS/s with the pro-posed system at1kHz measurement frequency are shown together with the reconstructed ellipse obtained using the ellipse-ﬁt.This situation corresponds to the one depicted in Fig.2.Two periods are represented that correspond to two ellipse rotations.Due to low noise and nearly coherent sampling,the data points of the second period are over-lapped with the points from theﬁrst rotation.2.4.DSP implementationThe DSP is an Analog Devices ADSP-21369SHARC high performance32-bitﬂoating point processor operating at 332MHz performing2GFLOPS/664MMACS with2Mbit on-chip SRAM.The DSP is included in an evaluation kit with a multichannel Sigma-Delta ADC(AD1835-with24 bit and sampling rate of48kS/s or96kS/s)and2Mbit SPI FLASH memory.The basic block diagram of the DSP system interface to the circuit of Fig.1is shown in Fig.4.Outside the DSP kit, two instrumentation ampliﬁers are used since the kit ana-log inputs are referenced.The ellipse-ﬁt and the seven-parameter sine-ﬁt algo-rithms were implemented in C using the Analog Devices VisualDSP++to program the DSP kit.The implemented subroutines include:acquisition(communication with the AD1835),IpDFT,matrix operations(e.g.,determination of eigenvectors and eigenvalues)and simple operations for the estimation of the impedance magnitude and phase.A basic set of commands to interface the DSP with a PC through a fast serial connection(115200baud)was also implemented.The implemented commands enable the impedance measurements either with the ellipse-ﬁt or the seven-parameter sine-ﬁt.Also included is a command to retrieve the unprocessed acquired samples.This option enables some signal analysis at a lower level to assess the analog circuitry and the analog to digital conversion. The complete program is stored in the non volatileﬂashFig. 4.Block diagram of the DSP system that interfaces the basic measurement circuit.1374P.M.Ramos et al./Measurement42(2009)1370–1379memory and upon boot up or reset of the DSP kit,the pro-gram is loaded from theﬂash memory to the DSP and exe-cuted.It then enters a command mode waiting for instructions from the serial interface.In addition to executing the algorithms,the DSP also controls the selection of the reference impedance.3.Measurement resultsThe impedance measurement system was tested for both algorithms with the measurement of105different impedances at a frequency of1kHz with magnitudes from 100X up to15k X and phases in the±90°range.3.1.Reference impedanceIn the seven-parameter sine-ﬁt,the reference imped-ance selection usually only depends on the unknown impedance’s magnitude.However,since the ellipse-ﬁt algorithm cannot estimate the ellipse parameters when the signals are in phase or in opposition,i.e.,when cos2(D/)=1because the conic is not an ellipse,the selec-tion of the reference impedance must also be based on the unknown impedance’s phase.Another consequence of this is that the reference impedances must include some phases substantially different from0°.For the selected impedance range,a set of six reference impedances were used.The measured impedances together with the delimit-ing regions used to select the reference impedances are de-picted in Fig.5.The selection of the reference impedance is done as follows:(i)one period measurement of the two sine signals with a resistive reference impedance;(ii)pre-liminary estimation of the impedance magnitude and phase;if the estimated impedance magnitude is below 3.5k X then the selected reference impedance has1k X magnitude,otherwise a5k X magnitude reference imped-ance is used;(iii)if the estimated phase is between0°and 10°,a capacitive reference impedance is applied(R=100X in series with C=159nF for the lower magnitude and R=100X in series with C=31.8nF for the higher magni-tude);(iv)if the estimated phase is betweenÀ10°and 0°,an inductive reference impedance is used(R=100X in series with L=157mH for the lower magnitude and R=100X in series with L=796mH for the higher magni-tude);(v)otherwise a resistive reference impedance is cho-sen(R=1k X for the lower magnitude and R=5k X for the higher magnitude).3.2.Channel mismatch compensationTo compensate for the mismatch between the two acquisition channels,the same signal was applied to both channels and repeated acquisitions at different frequencies (from100Hz up to10kHz)were made.The seven-param-eter sine-ﬁt was then used to obtain the amplitude and phase calibration coefﬁcients.Their average and standard deviations are plotted in Fig.6.These calibration coefﬁ-cients are then used in each measurement to correct the estimated amplitude ratio and phase difference.3.3.Evaluation of measurement standard deviationsTo assess the algorithms’performance,repetitive mea-surements of the impedances were executed.Each imped-ance was measured5000times at a frequency of1kHz, with1024points per channel acquired at96kS/s.For each measured impedance,the average value of the impedance magnitude and phase and the corresponding experimental standard deviation values were registered.The experimen-tal standard deviation results obtained with the ellipse-ﬁt algorithm are shown in Figs.7and8.From the results as a function of the measurement impedance phase(Figs.7 and8on the right side)it is clear that the algorithm is independent on the impedance phase.Note that this inde-pendence for the ellipse-ﬁt is a result of the method used to select the reference impedance,i.e.,the ellipse-ﬁt meth-od is phase dependant because it does not work for phase differences near0°[14].Regarding the impedance magni-tude it is also clear that the slightly worse results occur when the measured impedance magnitude is near the lim-its of the measurement range.This is justiﬁed by the re-duced number of reference impedances and their values –i.e.,near the impedance measurement range,the mis-match of the reference impedance and the unknown impedance magnitude is larger.The maximum relative standard deviation of the impedance magnitude is0.016% while the maximum standard deviation of the estimated impedance phase is0.0089°.Even in the most mismatched measurement situation of the105impedances,the signal to noise ratio(SNR)is greater than54dB.This ensures that the ellipse-ﬁt is not biased(as shown in[14],the algorithm is slightly biased but only for SNR below30dB).In Figs.9and10the results obtained with the seven-parameter sine-ﬁt algorithm are shown for the same situ-ations as those depicted in Figs.7and8.For the seven-parameter sine-ﬁt algorithm,the worst results correspond to a relative standard deviation of the impedance magnitude of0.015%and0.0083°for the exper-imental standard deviation of the estimated impedance phase,occurring for the measured impedance withhighest P.M.Ramos et al./Measurement42(2009)1370–13791375。

dsp存储器英文文献及翻译（精）

dsp存储器英文文献及翻译（精）dsp存储器英文文献及翻译dsp存储器英文文献及翻译the ROM.ROM/RAM security option. This option is the most secure, because it protects both the on-chip ROM and the on-chip RAM. When the ROM/RAM security option is enabled, any instruction fetched from external memory is prohibited from accessing on-chip ROM or RAM and reads invalid data (0FFFFh). Only instructions fetched from on-chip ROM or on-chip RAM can access the on-chip memory. The ROM/RAM security option also internally forces the device into microcomputer mode (MP/MC bit forced to zero), preventing the ROMfrom being disabled.Table 1–4. On-Chip Memory Secruity HPI Access Restrictions 1.4.7 Program MemoryThe standard external program memory space on the ’54x devices addresses up to 64K 16-bit words. Software can configure their memory cells to reside inside or outside of the program address map. When the cells are mapped into program space, the device automatically accesses them when their addressesare within bounds. When the program-address generation (PAGEN) logic generates an address outside its bounds, the device automatically generates an external access. The advantages of operating from on-chip memory are as follows:Higher performance because no wait states are requiredLower cost than external memoryLower power than external memoryThe advantage of operating from off-chip memory is the abilityto access a larger address space.1.4.7.1 Relocatable Interrupt Vector TableThe reset, interrupt, and trap vectors are addressedin program space. These vectors are soft —meaning that the processor, when taking the trap, loads the program counter (PC) with the trap address and executes the code at the vector location. Four words are reserved at each vector location to accommodate a delayed branch instruction: either two 1-word instructions or one 2-word instruction, whichallows branching to the appropriate interrupt service routine with minimal overhead.At device reset, the reset, interrupt, and trap vectors are mapped to address FF80h in program space. However, these vectors can be remapped to the beginning of any 128-word page in program space after device reset. This is done by loading theinterrupt vector pointer (IPTR) bits in the PMST register with the appropriate 128-word page boundary address. After loading IPTR, any user interrupt or trap vector is mapped to the new 128-word page.NOTE: The hardware reset (RS) vector cannot be remapped, because the hardware reset loads the IPTR with 1s. Therefore, the reset vector is always fetched at location FF80h in program space. In addition, for the ’54x, 128 words are reserved in the on-chip ROM for device-testing purposes. Application code written to be implemented in on-chip ROM must reserve these 128 words at addresses FF00h–FF7Fh in program space.1.4.7.2 Extended Program Memory论文网http:///doc/bb11097156.html,Selected ’5 4x devices use a page-extended memory scheme in program space to allow access of up to 8M of program memory. This extended program memory is organized into a maximum of 128 pages (0–127), each 64K in length. Devices which implement theextended program memory scheme include the following additional features:Maximum of seven additional address lines (for a total of 23). An extra memory-mapped register [program counter extension register (XPC)]. Six new instructions for addressing extended program memory space:FB[D] —Far branch无耻悲鄙下流的网~学'网总是抄六.维^论^文.网FBACC[D] —Far branch to the location specified by the value in accumulator A or accumulator BFCALA[D] — Far call to the location specified by the value in accumulator A or accumulator BFCALL[D] —Far callFRET[D] —Far returnFRETE[D] —Far return with interrupts enabledTwo ’54x instructions are extendedREADA —Read program memory addressed by accumulator A and store in data memoryWRITA —Write data to program memory addressed by accumulator A for more information on these six new instructions and the two extended instructions, refer to the instruction set summary (Table 1–11) and to the TMS320C54x DSP Reference Set, Volume 2, Mnemonic Instruction Set, literature number SPRU172. And for more information on extended program memory, refer to the TMS320C54x DSP Reference Set, Volume 1, CPU and Peripherals, literature number SPRU131.Table 1–5 shows the extended program addressing options available in the ’54x familyTable 1–5. Extended Program Memory Addressing Options 1.4.8 Data MemoryThe data memory space on the ’54x device addresses 64K of 16-bit words.The device automatically accesses the on-chip RAM when addressing within its bounds. When an address is generated outside the RAM bounds, the device automatically generates an external access.The advantages of operating from on-chip memory are as follows: Higher performance because no wait states arerequired Higher performance because of better flow withinthe pipeline of the CALU Lower cost than external memory Lower power than external memoryIn addition to general-purpose data memory, the CPU maintains a set of memory-mapped registers in data memory for processor configuration and configuration/communication with the device peripherals. For detailed information on the implementation of the memory-mapped CPU and peripheral control registers, see the device-specific data sheets.1.5 On-Chip PeripheralsAll the ’54x devices have the same CPU structure; however, they have different on-chip peripherals connected to their CPUs. The on-chip peripheral options provided are: Software-programmable wait-state generator Programmable bank-switching Parallel I /O ports DMA controller Host-port interface (standard 8-bit, enhanced 8-bit, and 16-bit) Serial ports (standard, TDM, BSP, and McBSP) General-purpose I/O pins 16-bit timer with 4-bit prescaler Phase-locked loop (PLL) clock generator1.5.1 Software-Programmable Wait-State GeneratorsThe software-programmable wait-state generator can be used to extend external bus cycles to interface with slower off-chip memory and I/O devices. The software wait-state generator is incorporated without any external hardware. For off-chip memory access, a number of wait states can be specified for every 32K-word block of program and data memory space, and for one 64K-word block of I/O space within the software wait-state register (SWWSR). The software wait-state generator is programmable up to 7 or 14 wait states depending on the device. For more specific information on the software wait-state generation capability, see the device-specific data sheet.1.5.2 Programmable Bank-SwitchingProgrammable bank-switching can be used to insert one cycle automatically whencrossing memory-bank boundaries inside program memory or data memory space. One cycle can also be inserted when crossing from program-memory space to data-memory space (’54x) or from one program memory page to another program memory page on selected devices. This extra cycle allows memory devices to release the bus before other devices start driving the bus; thereby avoiding bus contention. The size of memory bank for the bank-switching is defined by the bank-switching control register (BSCR). For specific information on the bank-switching capabilities of a specific device, see the device-specific data sheet.1.5.3 Parallel I/O Ports论文网/doc/bb11097156.html,Each ’54x device has a total of 64K I/O ports. These ports can be addressed by the PORTR instruction or the PORTW instruction. The IS signal indicates a read/write operation through an I/O port. The devices can interface easily with external devices through the I/O ports while requiring minimal off-chip address-decoding circuits.1.5.6 Serial Ports无耻悲鄙下流的网~学'网总是抄六.维^论^文.网The ’54x devices provide high-speed, full-duplex serial ports that allow direct interface to other ’54x devices, codecs, and other devices in a system. There is a standard serial port, a time-division-multiplexed (TDM) serial port, a buffered serial port (BSP), and a multichannel buffered serial port (McBSP).T able 1–1 shows the availability of each of the serial port types in the ’54x family.1.5.6.1 Standard Serial PortThe general-purpose serial port utilizes two memory-mapped registers for data transfer: the data-transmit register (DXR) and the data-receive register (DRR). Both of these registers can be accessed in the same manner as any other memory location. The transmit and receive sections of the serial port each have associated clocks, 上一页[1] [2] [3] [4] [5]下一页。

DSP外文资料142

Available online at Behavioural Brain Research188(2008)271–280Research reportNeurotoxic effects of DSP-4on the central noradrenergicsystem in male zebraﬁnchesSusanna A.Waterman a,b,Cheryl F.Harding a,b,∗a Psychology Department,Hunter College,CUNY,New York,NY10016,United Statesb Biopsychology and Behavioral Neuroscience Doctoral Program,Graduate Center,CUNY,New York,NY10016,United StatesReceived25July2007;received in revised form29October2007;accepted11November2007Available online19November2007AbstractWhen administered systemically,the noradrenergic neurotoxin N-(2-chloroethyl)-N-ethyl-2-bromobenzylamine(DSP-4)appears to target the noradrenergic innervation originating in the locus coeruleus causing long-term decrements in noradrenergic function.In songbirds,DSP-4-treatment decreased female-directed singing by males and copulation solicitation responses of females to male songs.However,DSP-4treatment in songbirds did not lower measures of NE function in the brain to the same extent as it does in mammals.The current study had two goals:determining if two DSP-4treatments10days apart would cause signiﬁcant decrements in noradrenergic function in male zebraﬁnches and determining if,as in other species,the noradrenergic innervation of midbrain and cortical areas would be profoundly affected while hypothalamic areas were spared. Dopamine-␤-hydroxylase immunoreactivity(DBH-ir)was quantiﬁed in thirteen brain regions(ﬁve vocal control nuclei,one auditory nucleus,two hypothalamic nuclei,andﬁve additional areas that demonstrated high DBH labeling in controls).Within20days,DSP-4treatment profoundly reduced the number of DBH-ir cells in both the locus coeruleus and ventral subcoeruleus.Unlike a previous study,DBH labeling delineated four out ofﬁve vocal control nuclei and an auditory nucleus.As expected,DSP-4treatment signiﬁcantly decreased DBH labeling in all areas examined in the mesencephalon and telencephalon without signiﬁcantly affecting DBH-ir in hypothalamic areas.This double treatment regime appears to be much more effective in decreasing noradrenergic function in songbirds than the single treatment typically used.©2007Elsevier B.V.All rights reserved.Keywords:Songbird;V ocal control;Hypothalamus;Locus coeruleus;Norepinephrine;Dopaine-␤-hydroxylase;Immunoreactivity1.IntroductionNorepinephrine(NE)appears to mediate a heightened state of arousal and attention to changes in the environment and to species-speciﬁc stimuli.In response to a novel stimulus,NE neurons in the locus coeruleus(LoC)react as a group with increased activity[56].NE has been implicated in the con-trol of behavior and reproduction[13,16].The LoC may affect reproductive behavior directly through its effects on arousal and attention,e.g.,[47]or indirectly through its modulation of hypothalamic–pituitary–gonadal function[38,52,65].Research also suggests that NE plays a role in some aspects of learn-ing,e.g.,discrimination[31],olfactory learning[18–21,62],and temporal learning[1,36].∗Corresponding author at:Psychology Department,Hunter College,695Park Avenue,New York,NY10065,United States.Tel.:+12127725047;fax:+12126503647.E-mail address:harding@(C.F.Harding).Central NE-producing cell bodies form cell groups that are located in the brainstem.The LoC is the largest of these with the remaining cell groups forming the lateral tegmental system (LT).For the most part,the target areas of the LoC and LT do not overlap.The LoC sends highly collateralized noradrener-gicﬁbers to the cortex/telencephalon,the cerebellum,and some brain-stem sensory nuclei[56].In both rats and pigeons,LoC projections are the primary source of NE projections to the cor-tex/telencephalon[40,43,44].In general,projections of the LT are restricted to thalamic and hypothalamic brain regions[24].The noradrenergic neurotoxin N-(2-chloroethyl)-N-ethyl-2-bromobenzylamine(DSP-4)is widely used to lower brain NE to investigate the functions of the central noradrenergic system [7,11,12,55,64].When administered systemically in rats,DSP-4 crosses the blood–brain barrier and causes rapid,profound,long-term decreases in NE and markers of noradrenergic function in the cortex but leaves hypothalamic NE function relatively unaltered[39,58].These differential effects on NE function in the cortex and hypothalamus presumably result from DSP-4’s0166-4328/$–see front matter©2007Elsevier B.V.All rights reserved. doi:10.1016/j.bbr.2007.11.004272S.A.Waterman,C.F.Harding/Behavioural Brain Research188(2008)271–280high selectivity for noradrenergic axon terminals that originate in the LoC[33,34,37,39,57,58].Singing in zebraﬁnches(Taenopygia guttata)is controlled by a series of hormone-sensitive interconnected brain nuclei that are located primarily in the telencephalon and are known collectively as the vocal control system(VCS)[26,51].Only male zebraﬁnches sing.Barclay and Harding[9,10]found high levels of NE in the male VCS.Hormone treatments which rein-stated high levels of courtship singing in castrates signiﬁcantly increased levels and turnover of NE in vocal control nuclei (VCN)on the motor pathway controlling the muscles of the syrinx,the avian vocal organ.Data suggest that NE may also be involved in song learning.When we quantiﬁed NE func-tion across development in male zebraﬁnches,NE levels and turnover in brain areas controlling singing showed striking peaks during the sensory acquisition phase of song learning[35].Dur-ing sensory acquisition,juvenile males attend to and memorize the songs of adult tutors[15].The increase in NE function was not seen outside auditory and vocal control nuclei,suggesting that heightened NE function in these areas might be related to song learning.To determine whether NE played a signiﬁcant role in con-trolling singing behavior in adults,we used DSP-4in an attempt to lower NE levels in the telencephalic vocal control nuclei but leave NE levels in the hypothalamus relatively unaffected.When DSP-4was administered systemically,it produced signiﬁcant deﬁcits in courtship singing[11].While brain NE levels were not signiﬁcantly depleted by DSP-4treatment,the nonsigniﬁ-cant decreases in NE levels in three VCN on the motor pathway (interfacial nucleus of the nidopallium—NIf,robust nucleus of the arcopallium—RA,and dorsomedial portion of the intercol-licular nucleus—DM)were positively correlated with decreased frequencies of singing.Since this systemic DSP-4treatment did not signiﬁcantly lower telencephalic NE levels,in a second study we administered DSP-4centrally via the third ventricle in an attempt to cause greater depletions[12].Central adminis-tration of DSP-4also caused signiﬁcant deﬁcits in courtship singing and,in addition,signiﬁcantly depleted NE levels in three VCN(Area X,lateral magnocellular nucleus of the ante-rior nidopallium—lMAN,and NIf).However,NE levels in two hypothalamic nuclei,the magnocellular portion of the paraven-tricular nucleus(PVN)and the preoptic area(POA)were also signiﬁcantly lowered.In both studies[11,12],the effects on male sexual behavior appeared to be mediated by decreased attentive-ness to females.Males took a longer time before beginning to court,but once they began,their behavior appeared normal.The current study had two goals.We wanted to determine if we could signiﬁcantly decrease the central NE innervation if birds were given two systemic injections of DSP-4,because a single systemic dose failed to signiﬁcantly deplete central NE levels in our previous study[11].We also wanted to ver-ify that the hypothalamic NE innervation would be spared as it was previously when we administered DSP-4systemically. To make our results more comparable to work done in rats and other species,e.g.,[3,26,64],we quantiﬁed the effects of DSP-4 treatment in male zebraﬁnch brains using standard immuno-cytochemical procedures to visualize dopamine-␤-hydroxylase (DBH)-containing neurons.DBH is the enzyme that catalyzes the conversion of dopamine to NE,and as such is a good marker of noradrenergic cells.2.Materials and methods2.1.SubjectsAdult zebraﬁnches were obtained from Canary Bird Farm(Englishtown, NJ).Males and females were treated with ivermectin to reduce parasite infes-tations and housed in separate aviaries until needed.Bird rooms were kept on a14-h light:10-h dark cycle with the temperature controlled(24±2◦C)and the humidity kept over50%to maintain optimal breeding conditions.Birds were fed a vitamin-supplemented(8in1,Pet Products)commercialﬁnch seed mix,grit,water and cuttlebone ad libitum,supplemented with fresh greens and oranges.During the experiment,males were housed in individual cages(56cm)3 and stimulus females were group housed until needed.All animal care,experi-mental procedures,and euthanasia were carried out in accordance with protocols approved by the Institutional Animal Care and Use Committee of Hunter College of the City University of New York.2.2.DSP-4injectionsThe two drugs used in this study were zimelidine dihydrochloride,a sero-tonergic re-uptake blocker,and DSP-4(generous gifts of Trevor Archer,Astra Pharmaceuticals,Ltd.).Males were randomly assigned to either experimental or control groups.All males received200␮g,i.p.zimelidine dihydrochloride to protect serotonergic neurons[58].After30min,experimental males(n=8) received500␮g DSP-4i.p.and controls(n=8)received saline vehicle alone i.p. Administration of DSP-4occurred within10min of drug preparation to prevent the formation of the non-toxic aziridinium derivative[42].These doses were based on the typical rat dosages which are expressed as mg/kg,e.g.,50mg/kg DSP-4[45,66].Since the average weight of ourﬁnches was10g,our doses were close to the typical rat dosage.This is the dose we have used withﬁnches in our two previous studies.To maximize noradrenergic depletion,drug treatments were repeated10days later.One hour after the second drug administration,each male was housed with a stimulus female.Our hypothesis was that the two groups would interact differently with the females and this would maximize differences in the availability of DBH between groups.We expected the control males to court the females more frequently,increasing their hormone levels.Increased gonadal steroid levels should stimulate their noradrenergic neurons and,given our prior results,increase NE levels[9,10].We hypothesized that a treatment that increased NE levels should also increase DBH-ir.Ten days later,males were anesthetized,perfused,and their brains processed for immunocytochemistry and cresyl violet staining.2.3.PerfusionMales were mildly anesthetized with a combination of xylazine:ketamine (5␮g each/g body weight)in0.5ml saline injected into the pectoral mus-cle.Metofane(Pitman-Moore)was administered as necessary to keep birds deeply anesthetized during the procedure.Males were transcardially perfused with0.2M phosphate buffer(PB)followed by4%paraformaldehyde in0.1M PB.Brains were quickly removed and post-ﬁxed in4%paraformaldehyde for 2h.The post-ﬁxing solution was replaced with cryoprotectant(300g sucrose, 10g polyvinyl-pyrrolidone,30ml ethylene glycol,2g sodium azide,volume adjusted to1000ml with0.1M PB)and stored at5◦C.The brains stayed in the cryoprotectant until they sank[14,22].2.4.ImmunocytochemistryBrains were mounted onto frozen cryostat chucks with double-distilled water.Frozen brains were equilibrated in an American Optical Cryo-cut II microtome or an IEC Minotome Plus at−15◦C for an hour,and then sliced into40-␮m-thick coronal sections.Three series of sections were collected into glass vials containing cryoprotectant.Sections were either stored at−70◦C untilS.A.Waterman,C.F.Harding/Behavioural Brain Research188(2008)271–280273needed or stored at5◦C for immediate immunocytochemical processing.One vial was used for immunocytochemistry.Another vial containing unprocessed tissue was rinsed,sorted,mounted onto gelatin-coated slides and stained with cresyl violet.This set of slides served as a cytoarchitectural reference for the DBH-immunolabeled slides.Sections were processed to optimize DBH staining inﬁnch brains.Unless otherwise stated,all rinses were performed three times for5min each in Buffer A (0.05M tris-buffered saline)while sections were gently agitated.To control for variations in staining intensity,tissues from experimental and control birds were always processed in the same assay.Free-ﬂoating sections were treated accord-ing to Bamshad et al.[8]with the following modiﬁcations.After pretreatment with0.1%NaBH4for10min in Tris–NaCl,sections were incubated in the fol-lowing solutions:(1)1%H2O2in Tris–NaCl for10min;(2)Tris–Triton rinse for10min;(3)anti-DBH serum(Eugene Tech International,Inc,New York,NY; DBH TE103),1:500in Tris–Triton for24h at5◦C;(4)Tris–Triton rinse.The next steps followed the directions of the Vectastain ABC Elite kit(Vector Labs, Burlingame,CA);(5)incubation with biotinylated anti-rabbit IgG at a dilution of1:600in Tris–Triton for45min at room temperature;(6)two Tris–Triton rinses,then two Tris–NaCl rinses;(7)incubation with ABC reagent diluted in Tris–NaCl for40min;(8)rinsed and treated with DAB peroxidase substrate (SigmaFast3,3 -diaminobenzidine tablet sets)in Tris–NaCl for5min.After the ﬁnal rinse with Tris–NaCl,sections were sorted and mounted onto gelatin-coated slides.Slides were dehydrated in a graded alcohol series,cleared with xylene and coverslipped with Permount.Adrenergic neurons also contain DBH;however,in rats and birds,adren-ergic neurons are restricted to the medulla[50,60].Since tissues from the medulla were not included in this study,all DBH-positive cells are presumably noradrenergic.2.4.1.Speciﬁcity checkThere were two checks for the speciﬁcity of DBH-antiserum inﬁnch brains. In the positive control,the antiserum was preadsorbed with an excess of DBH (Sigma)and was used in an assay.In the negative control,incubation with the anti-DBH serum was omitted from the assay.Both speciﬁcity checks resulted in DBH-immunonegative tissue.2.5.Image processingSections were examined using a Nikon Optiphot microscope.Images were captured by a Dage MTI CCD72video camera interfaced via a Perceptics Pixel pipeline framegrabber to a MAC2FX computer.Quantiﬁcation of the density of DBH-ir cell bodies and/orﬁbers in a given brain area was performed using NIH Image version1.55.Images were sharpened to focus the digitized image. The density slice feature was used for thresholding and adjusted so that only immunopositive cell bodies and/orﬁbers were quantiﬁed.For all measurements, all images which contained a given brain area were captured and measured and the mean per coronal section was then calculated.For each brain area,two calculations were performed.Theﬁrst measured the area of DBH-ir within the outlined brain area.The brain area of interest in each image was outlined with the freehand selection tool.This outline was saved also for later calculation of percent area labeled.The atlas of Stokes et al.[61]was used as a reference,but areas are designated using the updated avian nomenclature[53].The program was calibrated by measuring lines of known length so that Image could convert the area of labeled pixels in the area of interest into square microns.The second calculation divided the area of DBH-ir in the deﬁned regions by the size of the brain area measured to determine the percentage of each brain area that was labeled(i.e.brain area with thresholding enabled divided by the same brain area without thresholding enabled).Percent area calcula-tions corrected for the possibility that DSP-4treatment altered the size of the nucleus in addition to affecting the area of DBH innervation.This measure-ment also allowed meaningful comparisons between brain areas of different sizes.2.6.Figure preparationSections were examined using a Nikon Eclipse E400microscope and dig-itized using SPOT version3.5.5for Windows(Diagnostic Instruments,Inc.).Images were autobalanced using PhotoShop6.0and labels added in CorelDraw 9.0.2.7.Data analysesDBH-ir cells were counted on a Nikon Optiphot-2microscope(10×power) equipped with a reticle.These data were collected from all sections that con-tained DBH-ir cells.Since only one third of brain sections were processed for DBH,the resulting cell counts were multiplied by three to estimate total DBH-ir cells.All data were analyzed by unpaired Student’s t-tests(GraphPad Prism version4.03for Windows,GraphPad Software,San Diego,California,USA, ).If the variances of the two groups differed signiﬁcantly, the data were square-root or log transformed prior to analysis.All statisti-cal tests were one-tailed,because we predicted DSP-4would decrease DBH labeling.The LoC was divided into dorsal and ventral sections because these regions may project to different targets[63],and we wanted to determine if DSP-4 had differential effects within the LoC.Previous research in rats[63]based this division on the appearance of the DBH-labeled neurons.Inﬁnches,DBH-labeled cells were categorized on the basis of their position relative to the fasciculus longitudinalis medialis(FLM).At the level of the fourth ventricle,cells located above FLM were classiﬁed as dorsal LoC,and cells located lateral to FLM were classiﬁed as ventral LoC.Any DBH-ir cells below FLM were classiﬁed as ventral subcoeruleus(SCv).3.Results3.1.Noradrenergic cellsIn control birds,the mean estimated total DBH-ir cell num-ber in the LoC was667±30(±S.E.M.)and in the SCv was 616±47.In the LoC,DSP-4treatment signiﬁcantly decreased the total number of DBH-ir cells compared to saline controls [t(14)=8.993,p=0.0001;see Figs.1and3A].A signiﬁcant loss of DBH-ir cells occurred in both dorsal LoC[t(14)=6.221, p=0.0001]and in ventral LoC[t(14)=2.169,p=0.0239,see Fig.1].DBH-ir cells in the SCv cell groups were also sig-niﬁcantly decreased in the drug-treated group[t(14)=7.765, Fig.1.Mean(±S.E.M.)estimated DBH-ir cells in locus coeruleus(LoC),ven-tral subcoeruleus(SCv),dorsal LoC,and ventral LoC in saline-treated(SAL)and DSP-4-treated males.Signiﬁcant decrease in cell numbers compared to saline group,*p<0.05,***p<0.0001.274S.A.Waterman,C.F .Harding /Behavioural Brain Research 188(2008)271–280p =0.0001;see Fig.1].The reductions in cell numbers were quite striking.DSP-4treatment decreased DBH-ir cells in the LoC by 60%,the dorsal LoC by 60%,the ventral LoC by 58%,and the SCv by 80%.DBH-ir cells in the LoC and SCv were arranged differently.Cells in the dorsal LoC were usually tightly clustered while cells in ventral LoC were diffuse,and cells in SCv were even more diffuse.In terms of their morphology,LoC cells were either round or angular while those in the SCv were predominately elongated.In ﬁnches,there were clear differences in the intensity of DBH labeling of cell bodies within the LoC and SCv of the saline-treated and drug-treated groups.In controls,some cells were intensely stained while others were very lightly stained.In the drug-treated group,no lightly stained cells were detected in the LoC,and the DBH-ir cells that remained were not tightly clustered.3.2.DBH-ir labeling across brain areasDrug treatment signiﬁcantly reduced the areas of DBH-ir cells and ﬁbers in the LoC [t (12)=2.593,p =0.0118],and DBH-ir ﬁbers in cerebellum (Cb)[t (12)=7.76,p <0.0001],ventral tegmental area (VTA)[t (11)=2.373,p =0.0185],optic tectum (TeO)[t (12)=2.638,p =0.0108]and septum (S)[t (11)=2.566,p =0.0131;see Figs.2A and 3A–D].DSP-4treatment did not signiﬁcantly affect the mean area covered by DBH-ir ﬁbers in the two hypothalamic nuclei,PVN and POA (Figs.2A and 3E and F ).In this study,as in our previous research,POA was deﬁned as the area under the septopalliomesencephalic tract (TSM).Other authors have included a much larger area,including this area and its posterior extension located above PVN and directly underthe anterior commissure as the medial preoptic nucleus based on aromatase labeling [6].In comparison to Mello et al.[48],POA as deﬁned here included only the anterior extent of POM,i.e.,the area under TSM.DSP-4treatment signiﬁcantly decreased DBH-ir label-ing and the mean area covered by DBH-ir ﬁbers in two of the four VCN examined,{high vocal center (HVC)[t (12)=2.64,p =0.0108],robust nucleus of the arcopallium (RA)[t (11)=4.375,p =0.0006],and one auditory nucleus (AN)—dorsal part of the lateral mesencephalic nucleus (MLd)[t (9)=3.27,p <0.005];see Figs.2B,4C–E and 5A and B }.Data from two other VCN just missed being signiﬁcantly affected by DSP-4treatment {nucleus uvaeformis (Uva)[t (11)=1.773,p =0.0520]and lateral magnocellular nucleus of the anterior nidopallium (lMAN)[t (9)=1.783,p =0.0541];see Figs.2B and 4A and B }.3.3.Percent of each brain area labeled by DBHDSP-4treatment signiﬁcantly decreased the percent area of DBH-ir labeling in the LoC [t (12)=3.925,p =0.001],Cb [t (12)=7.509,p <0.0001],VTA [t (11)=2.745,p =0.0095],TeO [t (12)=4.328,p =0.0005],S [t (11)=2.636,p =0.0116],(see Fig.2C).The percent area covered by DBH-ir ﬁbers was not signiﬁcantly affected by drug treatment in the two hypothala-mic nuclei examined;PVN and POA (Fig.2C).In the VCN and AN,DSP-4treatment signiﬁcantly decreased the percent area of DBH-ir labeling in all nuclei:Uva [t (11)=2.005,p =0.0351],HVC [t (12)=2.757,p =0.0087],RA [t (11)=4.289,p =0.0006],lMAN [t (9)=1.847,p =0.0489],MLd [t (9)=3.115,p =0.0062],(see Fig.2D).Fig.2.Mean area (±S.E.M.)covered by DBH-ir labeling per coronal section of (A)areas outside the vocal control system and (B)vocal control and auditory nuclei.Mean (±S.E.M.)percent of each brain area covered by DBH-ir labeling per coronal section in (C)areas outside the vocal control system and (D)in vocal control and auditory nuclei.Note broken Y axes in A and D.Scale is not linear in A.Signiﬁcant change compared to saline-treated males,*p <0.05,**p <0.01,***p <0.001.See text for abbreviations.S.A.Waterman,C.F.Harding/Behavioural Brain Research188(2008)271–280275Fig.3.Photomicrographs of DBH ir in areas outside of the vocal control system in male zebraﬁnch brains.In each pair,the brain on the left is from a saline-treated male,that on the right from a DSP-4-treated male.(A)LoC,(B)VTA,(C)S,(D)Cb,(E)POA and(F)PVN.See text for abbreviations.4.Discussion4.1.Typical effects of DSP-4treatmentThe actions of DSP-4have been most thoroughly documented in rats.In this species,DSP-4treatment causes a rapid,dose-dependent reduction in NE levels and NE-ir which is maximal by 8–16h after treatment[3,26].NE depletion is most pronounced in brain areas innervated by the LoC,including the hippocam-pus,cortex,and olfactory bulb.NE levels in these areas are extremely low following treatment with50mg DSP-4/kg body weight,the dose of DSP-4commonly used in rats.Brain areas like the hypothalamus and basal forebrain that are innervated by the LT are much less affected.[3,26].The neural distribution of DBH-irﬁbers is affected in a similar fashion following DSP-4 treatment,except that signiﬁcant decreases in DBH-ir are not seen until4days after treatment with profound decrements by day5[26].In rats,NE cell bodies in the LoC are not affected until several months after treatment[26,30].Typically,there is recovery of NE function over time,with surviving LoC neurons showing strong regenerative responses[30].4.2.Effects of DSP-4treatment on noradrenergic cells and projections inﬁnchesNE levels Inﬁnch brains are approximately ten times higher than those found in comparable brain areas in rats[9].This may account for the inability of a single systemic treatment with the standard rat dose of DSP-4to signiﬁcantly deplete telen-cephalic NE levels[11].Our current results document that giving birds two systemic injections of DSP-4spaced10days apart is much more effective in decreasing noradrenergic function. This study also conﬁrms,that when administered systemically, DSP-4does not signiﬁcantly affect noradrenergic function in the hypothalamus.The double DSP-4treatment profoundly decreased the num-ber of DBH-ir cell bodies inﬁnches-by60%in the LoC and 80%in the SCv.This dramatic cell loss occurred within20days of theﬁrst drug treatment.While control birds had both lightly-and intensely-labeled cells in the LoC and SCv,the DBH-ir cells remaining in DSP-4-treated males were all darkly stained, suggesting that cells showing low levels of DBH-ir are more vulnerable to the neurotoxin.As might be expected from the fact that DSP-4-treated males had fewer DBH-ir cell bodies,they also had fewer DBH-irﬁbers in most brain areas compared to the saline-treated controls.The differences in DBHﬁber staining between the two groups were obvious and in some brain regions could be seen without the aid of a microscope.In general,the distribution of DBH-ir ﬁbers in the saline-treatedﬁnches formed a clear network of ﬁbers,though the intensity of staining varied from lightly to densely stained.DSP-4-treated males did not have thisﬁbrous network,and their remaining DBH-irﬁbers appeared turgid and fragmented.Drug treatment reduced DBH-irﬁbers in two of the four VCN (HVC and RA)and signiﬁcantly reduced the percentage tis-sue labeled in all four VCN(HVC,RA,Uva,and lMAN).The VTA has a strong noradrenergic innervation[41],and DSP-4 treatment signiﬁcantly reducedﬁber staining in this area.NE is noted for its modulation of sensory inputs[23],and so it276S.A.Waterman,C.F.Harding/Behavioural Brain Research188(2008)271–280Fig.4.Photomicrographs of DBH-ir labeling in vocal control and auditory nuclei in male zebraﬁnch brains.In each pair,the brain on the left is from a saline-treated male,that on the right from a DSP-4-treated male.(A)Uva,(B)lMAN and X,(C)HVC,(D)RA and(E)TeO and MLd.See text for abbreviations.Note that in the saline-treated controls,the DBH-ir labeling inside the vocal control nuclei is denser than that in surrounding tissue,except in Area X.There are two issues here. First,there are more aggregations of dense punctate labeling andﬁbers within the vocal control nuclei compared to surrounding tissue.Second,the background of the VCN is darker than that of the surrounding tissue.This is caused by smaller,lighter puncta of DBH ir,making the vocal control areas in the DSP-4-treated birds darker than the surrounding tissue.DSP-4treatment signiﬁcantly lowered,but did not eliminate DBH-ir.The VCN in DSP-4-treated birds stand out because they still have more DBH-ir than the surround.For the most part,tissue from DSP-4-treated birds lacks the denseﬁbers and puncta seen in controls.This can be seen in higher magniﬁcation in Fig.5.The arrows in HVC and the asterisks in lMAN in thisﬁgure correspond to those in these areas in Fig.5.S.A.Waterman,C.F.Harding/Behavioural Brain Research188(2008)271–280277Fig.5.Higher magniﬁcation of DBH-ir labeling in the sections of HVC and lMAN shown in Fig.4.In each pair,the brain on the left is from a saline-treated male,that on the right from a DSP-4-treated male.The arrows in HVC and the asterisks in lMAN in thisﬁgure correspond to those in these areas in Fig.4.Note how the pattern and density of DBH-ir labeling over lMAN in the control bird is different from that outside the nucleus.This is particularly clear in comparison to the area below lMAN.If you refer back to Fig.4,you can see that the arrow in HVC in the control bird points to an area of dense labeling on the ventral border of the nucleus.It is clear at higher magniﬁcation that no labeling below the arrow matches the aggregations of high density DBH-ir labeling seen within HVC.was not surprising that DSP-4treatment signiﬁcantly decreased DBH labeling in the primary visual(TeO)and auditory pro-cessing areas(MLd).Noradrenergic modulation of these brain areas may be related to birds’attentiveness to conspeciﬁcs.DSP-4treatment has been shown to decrease the responsiveness of maleﬁnches to females placed in their home cages[11,12]and the responsiveness of female canaries to the songs of conspe-ciﬁc males[2].DSP-4treatment also decreased DBH ir in the Cb and S.In rats,both areas are innervated by and their func-tion modulated by the LoC[24].However,ﬁber labeling in the two hypothalamic nuclei,PVN and POA,was not signiﬁcantly affected by drug treatment.4.3.Most VCN receive stronger noradrenergic innervation than surrounding tissueIn control birds,DBH labeling clearly delineated all VCN examined from surrounding tissue except for Area X.In most birds,there was a denser network of darkly stained DBH-ir inside MLd,Uva,HVC,RA and lMAN than there was in the surround-ing tissue.This striking labeling of the vocal control nuclei was not observed by Mello et al.,except in DM[48].Four method-ological differences might account for discrepancies in DBH labeling between the two studies.First,the concentration of anti-body used in our study was double that used by Mello et al.[48]. However,if the staining differences were attributable only to the stronger antibody concentration,then one would expect an over-all increase in staining intensity across all brain areas,which was not the case.Second,Mello et al.performed their assay on slide-mounted tissue,not free-ﬂoating sections.The free-ﬂoating sections used in our study provided two surfaces and increased the opportunity for the antibody to bind to available antigens.Third,Mello et ed sagittal sections to examine many brain areas.Our study used only coronal sections.The orientation ofﬁbers(cross-sections versus tracts)may also have contributed to the observed differences in bothﬁber density and innervation.Finally,all males in our study were housed with females to maximize neurochemical differences between treat-ment groups.These males were given the opportunity to court and mate with females for10days prior to sacriﬁce.Since our prior studies suggested that high levels of singing behavior were correlated with high levels of NE in the motor pathway control-ling singing[11],we expected that housing with females would increase levels of DBH in the saline-treated males,maximizing differences between treatment groups.Although we previously found that levels of NE in NIf were higher than those in two other telencephalic vocal control nuclei, HVC and RA[10,11],NIf was not strongly labeled by DBH-ir in these brains,so we did not quantify labeling in this vocal control nucleus.4.4.Mismatch between DBH-ir labeling and other noradrenergic markers in Area XDespite the fact that we consistently found moderate to high levels of NE in Area X in previous studies[9–12,35],like Mello et al.,[48]we found few if any DBH-irﬁbers in Area X and the surrounding medial striatum.Area X is very large and highly vis-ible making it easy to accurately microdissect.Although DBH labeling was very limited or absent within and around Area X, signiﬁcant levels of␣2-adrenergic receptors were found in Area X in both zebraﬁnches[54]and starlings[5].A recent study[17] documented the existence of projections from both the LoC and SCv to Area X inﬁnches.The mismatch between the existence of these projections,high NE levels,and signiﬁcant levels of␣2-adrenergic receptors yet the lack of signiﬁcant levels of DBH-ir ﬁbers remains to be explained.parison of DSP-4’s effects across speciesIn saline-treatedﬁnches,the estimate of total LoC DBH-ir cell numbers(667±30,mean±S.E.M.)were much lower than the1300reported in Japanese quail and in chickens[4,49] and the2900reported in albino rats[63].Our estimates were based on counting every DBH-ir cell found in every third section through the LoC.While counting cells in every section is clearly。

【最新推荐】基于DSP的IIR滤波器设计外文文献

学科分类号本科毕业设计题目(中文)：基于DSP的IIR滤波器设计(英文)：The Design of IIR Filter Basedon DSP Chip姓名学号院(系)专业、年级指导教师二〇年月目录摘要 (1)Abstract. (2)1 绪论 (2)1.1 认识数字信号处理和IIR数字滤波器 (3)1.2 数字滤波器的实现方法 (4)1.3 主要研究内容 (6)2 滤波器原理基础 (6)2.1 IIR数字滤波器的优缺点 (7)2.2 IIR数字滤波器的设计方法和原理 (9)2.2.1 脉冲响应不变法 (12)2.2.2 双线性变换法 (14)2.3 IIR滤波器的基本结构 (17)3 IIR滤波器的设计过程及DSP的实现 (21)3.1 IIR滤波器的设计过程 (21)3.2 DSP系统的设计流程 (22)3.3 IIR数字滤波器在DSP上的实现 (22)参考文献 (27)附录 (28)致谢 (31)外文文献译文......................................................................................... 1-3 外文文献原文基于DSP的IIR滤波器设计摘要：数字信号处理(Digital Signal Processing，DSP)是一门涉及许多学科而又广泛应用于众多领域的新兴学科。

早在20世纪60年代，数字信号处理(即信号的数字化及数字处理)理论已经被被提出，到20世纪70年代，DSP理论和算法基础才被人提出。

不久之后，1982年世界上第一枚DSP芯片诞生了。

这枚DSP芯片在当时运算速度很快，尤其是在编码解码和语音合成方面得到广泛应用。

随着科学技术的飞速发展，数字化硬件技术得到长足的发展，这就带动了数字信号处理的飞速发展，也使得它得到了很多的实际应用，由此奠定了DSP这一词的地位。

之后，DSP芯片的科研不断推陈出新，每一代的DSP芯片都向着使运算速度更快、精度更高的目标发展，应用于通信、语音、医疗、仪器仪表和家用电器等人类生产生活的各个领域。

DSP外文资料80

Behavioural Brain Research119(2001)131–142Research reportCardiovascular and behavioural responses to psychological stress in spontaneously hypertensive rats:effect of treatment with DSP-4 Maarten van den Buuse a,b,*,Gavin Lambert a,Mark Fluttert c,Nina Eikelis aa Baker Medical Research Institute,Prahran,Vic.,Australiab Beha6ioural Neuroscience Laboratory,Mental Health Research Institute,Locked Bag11,Park6ille,Vic.3052,Australiac Leiden/Amsterdam Centre for Drug Research,Leiden Uni6ersity,Leiden,The NetherlandsReceived1May2000;received in revised form25July2000;accepted21August2000AbstractWe used a model of psychological stress combining exposure to an open-ﬁeld novel environment,radio-telemetric measurement of blood pressure and heart rate,and behavioural tracking analysis of behavioural parameters.All rats showed signiﬁcant increases in blood pressure and heart rate for the duration of open-ﬁeld exposure,with spontaneously hypertensive rats(SHR) showing markedly greater pressor responses and tachycardia when compared to either Wistar–Kyoto(WKY)or Sprague-Dawley rats(SD rats).Behavioural responses in the open-ﬁeld were unrelated to the magnitude of cardiovascular responses.Open-ﬁeld exposure on4consecutive days induced similar pressor responses and tachycardia on each day.By contrast,behavioural responses were reduced from the second day of open-ﬁeld exposure.Treatment of SHR and WKY rats with DSP-4,to deplete central noradrenaline levels,did not affect cardiovascular responses in SHR,whereas WKY rats showed a trend towards inhibition. However,in WKY rats,but not SHR,DSP-4treatment caused a marked reduction in behavioural activity in the open-ﬁeld.In conclusion,these data show that:(1)SHR display marked cardiovascular hyperreactivity to psychological open-ﬁeld stress when compared to two normotensive rat strains;(2)unlike behavioural responses,cardiovascular stress responses do not habituate upon repeated stress exposure;and(3)noradrenergic projections from the locus coeruleus do not appear to play a major role in cardiovascular stress responses in SHR or WKY rats,although they may be involved in behavioural responses in WKY rats.©2001Elsevier Science B.V.All rights reserved.Keywords:Behaviour;Stress;Blood pressure;Heart rate;Spontaneously hypertensive rats;Noradrenaline;Locus coeruleus;DSP-4/locate/bbr1.IntroductionBlood pressure and heart rate are not stable during the day butﬂuctuate in response to environmental stimuli.Over the circadian cycle,cardiovascular parameters vary,with blood pressure and heart rate being highest in rats during the dark period,when these animals display the highest level of behavioural activity, and lowest during the light period,when the animals are resting[46,47].Thus,in these studies cardiovascular and gross behavioural parameters appeared to change in parallel.Indeed,it has been suggested that cardiovas-cular circadian variability is merely the result of changes in motility and wakefulness of the subjects[9]. Also acutely,cardiovascular parameters are con-stantly adjusted to metabolic need in response to envi-ronmental stimuli.Several studies have addressed the effect of environmental stress on blood pressure and heart rate in experimental animals.The majority of these studies in rats have employed stressors such as footshock[27],restraint[24]or noise stress[14,21]. These stressors could be classiﬁed as‘physical stress’and the relationship of such stimuli to more naturalistic environmental factors in rats or to stress in humans can*Corresponding author.Current address:The University of Mel-bourne,Behavioural Neuroscience Laboratory,Mental Health Re-search Institute,155Oak Street,Parkville,Vic3052,Australia.Tel.:+61-3-93892967;fax:+61-3-93875061.E-mail address:m.vandenbuuse@.au(M.van denBuuse).0166-4328/01/$-see front matter©2001Elsevier Science B.V.All rights reserved. PII:S0166-4328(00)00349-1M.6an den Buuse et al./Beha6ioural Brain Research119(2001)131–142 132be questioned[40,45].Some rat studies have attempted a more naturalistic approach,using dominant/submis-sive behaviours[38]or simple novelty[31]as a stimulus. The use of radio-telemetry recording of heart rate[6] has facilitated such studies by removing the need for external catheters or tethers which could inﬂuence be-haviour of the animals.Few studies have,however, combined measurement of blood pressure as well as heart rate by telemetry with detailed behavioural obser-vations,making it difﬁcult to interpret the results.We recently developed a model of novelty stress in rats, using telemetry to measure blood pressure,heart rate and gross locomotor activity and automated video tracking analysis to measure behavioural parameters [48].When placed in an open-ﬁeld,rats display marked rises in blood pressure and heart rate,which are depen-dent on increased vasomotor and cardiac sympathetic tone,respectively.The rats also display exploratory locomotor activity,mostly in the outer segments of the open-ﬁeld,and occasional rearing and grooming.Be-havioural activity was found to habituate rapidly dur-ing each session and over repeated daily exposures to the open-ﬁeld,in contrast to the pressor responses and tachycardia which did not decrease over at least4days [48].In addition,treatment with the anxiolytic di-azepam completely prevented the cardiovascular effects while causing only minor changes in behavioural activ-ity in the open-ﬁeld[48].Combined analysis of cardio-vascular and behavioural parameters thus allowed us to show that the pressor response and tachycardia were not simply the result of an increase in behavioural activity but were temporally and pharmacologically separable.Such combined observations may prove use-ful in drug development studies and on the central mechanisms involved in autonomic responses to stress. In the present study,we further investigate cardiovas-cular and behavioural responses to novelty stress in the open-ﬁeld telemetry model.Weﬁrst extended our initial observations in normotensive Wistar rats to other rat strains,including spontaneously hypertensive rats (SHR).In studies on the role of stress in the develop-ment of hypertension,these genetically hypertensive rats have been shown by a number of authors to display exaggerated cardiovascular responses to stress [7,16,26].However,it has been suggested by others that this apparent cardiovascular hyperreactivity is merely a result of behavioural hyperactivity of these rats and unrelated to the development of their hypertension [22,41].In addition,there is considerable debate as to which control strain is most appropriate to compare SHR with[23,27].Ourﬁrst aim was therefore to com-pare blood pressure,heart rate and behavioural re-sponses in the open-ﬁeld model between SHR and two normotensive strains,Wistar–Kyoto(WKY)rats and Sprague-Dawley(SD)rats.We compared these novelty stress-induced cardiovascular changes over several con-secutive days with those occurring over the circadian cycle.The central mechanism behind cardiovascular re-sponses to stress has not been fully elucidated.A num-ber of studies have suggested a role of the amygdala, but other brain regions may be involved as well,sug-gesting a global modulatory action to prepare the sub-ject at several levels to deal with impending danger [15,35,45].The nucleus locus coeruleus is the region of origin of widespread noradrenergic innervation of corti-cal and limbic regions[1]and has been suggested to change the signal to noise ratio of activity of forebrain neurons during alerting and stress responses[2,37]. Cellular activity and neurotransmitter turnover in the locus coeruleus have been shown to increase in situa-tions of severe stress[2,21,39].Administration of yohimbine in humans,which activates locus coeruleus neurones,causes anxiety and panic attacks,particularly in patients with post-traumatic stress disorder[4].In contrast,treatments which potentially inhibit central noradrenergic activity,such as clonidine,reduce stress responses,including cardiovascular effects[3,43,44].On the other hand,electrical or chemical stimulation of the locus coeruleus decreases blood pressure[30,42]and reduced locus coeruleus activity has been described in SHR[12,13,32].Taken together,these data predict that lesions of the locus coeruleus will result in altered cardiovascular responses to stress,although the direc-tion of the effect is unclear.Moreover,the effect of locus coeruleus lesions on behavioural activation to stress,which may indirectly inﬂuence cardiovascular stress responses,remains to be determined in parallel to the cardiovascular effects.The second aim of this study was therefore to investigate the effect of lesions of the locus coeruleus noradrenaline system on blood pres-sure,heart rate and behavioural responses to novelty stress.We treated SHR and controls with DSP-4(N-(2-chloroethyl)-N-ethyl-2-bromobenzylamine)which has been shown to induce selective destruction of locus coeruleus neurons,while sparing medullary noradrener-gic cell groups[19,20,36].2.Methods2.1.Animals and surgeryMale SHR,WKY rats and SD rats of13weeks of age were obtained from the Baker Medical Research Institute breeding stock.Before experiments com-menced,the rats were housed3–4per plastic cage under standard animal house conditions with ad libi-tum access to pellet food and tap water.After surgery for telemetry probe implantation,rats were housed singly under similar conditions,except for the SHR and WKY rats in the DSP-4study which were placed onM.6an den Buuse et al./Beha6ioural Brain Research119(2001)131–142133telemetry receivers in a self-contained rat isolator unit (Iffa Credo,L’Arbresle,France).Light in the animal room or isolator were on at0700and off at1900h.All surgical techniques,treatments and experimental proto-cols were performed in accordance with the Australian Code of Practise for the Care and Use of Animals for Scientiﬁc Purposes(1990)set out by the National Health and Medical Research Council of Australia. All rats were instrumented with a telemetry transmit-ter(TA11PA-C40,Data Sciences,St.Paul)as described in detail previously[46,47].Brieﬂy,the rats were anes-thetized with a mixture of30mg/kg of pentobarbital sodium(Nembutal,Boehringer Ingelheim,Artarmon, NSW,Australia),40mg/kg of methohexitone sodium (Brietal,Lilly,West Ryde,NSW,Australia)and0.5 mg/kg of atropine sulfate(Apex Laboratories,St. Marys,NSW,Australia).Through a midline incision the abdominal aorta was exposed and clamped off with surgical clamps.The tip of the probe catheter was inserted rostrally through a small hole in the abdominal wall andﬁxed in position with a drop of tissue glue. The body of the transmitter was positioned in the abdominal cavity and sutured to the inside of the muscle wall.After the incision was suture closed,the rats received5mg/kg of Carprofen to relieve postoper-ative discomfort.The animals were kept for7–10days in the animal house where they were closely monitored for adequate postoperative recovery.Starting3days before the ex-periment,data for circadian variation of blood pres-sure,heart rate and behavioural activity were obtained while the rats were in their home cage in the animal house.2.2.Experimental design and proceduresThe experimental procedure consisted principally of subjecting the rats to the mild psychological stress of an unfamiliar and unpredictable environment,i.e.place-ment in an open-ﬁeld.On the day of the experiments, resting blood pressure and heart rate values and activity scores were recorded while the rats were still in their home cage but already in the laboratory.Baseline car-diovascular values reported here are the mean of15 min,starting20min before the individual rats were transferred gently into the open-ﬁeld for15min.Be-cause most of the rats were asleep during these mea-surements,baseline activity scores in the home cage were usually zero and are not reported.The open-ﬁeld was a90cm diameter circular arena surrounded by a30cm high wall.A distracter object (black cylinder,10cm diameter,10cm high)was put in the centre of the open-ﬁeld to encourage the rats to enter the‘inner circle’(see below).Two60W bulbs provided lighting.Eight RL1010telemetry receivers (Data Sciences)were placed under theﬂoor of the open-ﬁeld to register the transmitter signal and were connected to an RMX10multiplexer(Data Sciences) which selects the receiver with the strongest signal. Behaviour of the rats in the open-ﬁeld was recorded on video with a commercial VHS video camera.All experiments were done during the light period of the rats between0900h and1400h.After testing the rats were transferred back to their individual home cages and to the animal house.Two groups of experiments were done.Theﬁrst group consisted of eight SHR,WKY rats and SD rats which were tested in the open-ﬁeld on4consecutive days to assess interstrain differences and habituation. The second group consisted of eight SHR and WKY rats which were tested on two occasions,1week apart. All rats in this latter group received an intraperitoneal injection of DSP-4(N-2-chloroethyl-N-ethyl-2-bro-mobenzylamine HCl,Research Biochemicals,Natick) to deplete central noradrenaline levels[36,49]on the afternoon of the day of theﬁrst control open-ﬁeld test. One,3days and7days after treatment,resting blood pressure and heart rate were measured in these rats. While1day after injection there was a marked fall in resting blood pressure,this effect was partially(3days) and completely(7days)recovered before the second test.This time-course is compatible with earlierﬁndings on rapid,but transient,noradrenaline depletion in sym-pathetic nerve endings[36].After the second open-ﬁeld test,all rats were decapitated and their brains dissected to obtain frontal cortex and hypothalamus samples[49] for analysis of catecholamine levels.We obtained a similar number of untreated control SHR and WKY rats for control data on catecholamine levels.2.3.High performance liquid chromatography analysis On the day of catecholamine analysis,brain samples were thawed and weighed prior to being homogenised on ice in0.5ml0.4M perchloric acid containing0.01% EDTA and a known amount of the internal standard 3,4-dihydroxybenzylamine.The homogenate was then rapidly centrifuged and the supernatant collected for subsequent neurochemical analysis.Catecholamines were extracted from the perchloric acid supernatant with alumina adsorption,separated by high perfor-mance liquid chromatography and the amounts quantiﬁed by electrochemical detection according to previously described methods[29].The chromato-graphic system consisted of a Model480High Precision Pump,Model Gina autosampler,Model STH585 column oven,Chromeleon3.03Chromatography Data System(Gynkotek,Germering,Germany),Model 5100A coulometric detector equipped with a Model 5021conditioning cell and a Model5011analytical cell (Environmental Sciences Associates,MA,USA)and a 25cm Altex Ultrasphere column(ODS4.6mm×25M.6an den Buuse et al./Beha6ioural Brain Research119(2001)131–142 134cm,5m m particle size,Beckman Instruments,CA). Analysis was performed at24°C with the operating potentials set at+0.35V for the guard cell and−0.35 and+0.29V for detectors1and2,respectively.All measurements were made using the oxidising potential applied at detector2and compounds in brain extracts were identiﬁed by their retention behaviour compared to that of authentic standard solutions.2.4.Data analysisData for blood pressure,heart rate and gross be-havioural activity were obtained with the Data Sciences data acquisition system.Circadian variation was deter-mined over48h by sampling every10min and averag-ing these data over1and12-h periods in an Excel spreadsheet.Resting values just before the open-ﬁeld test were the average of15min.In the open-ﬁeld, parameters were obtained every20s for the duration of 15min and later averaged per minute in Excel. Video tapes of behaviour were analyzed using the Ethovision automated tracking program(Noldus Infor-mation Technology,The Netherlands).For analysis,an ‘inner circle’of 30cm was deﬁned.Parameters ob-tained were distance moved in total arena and inner circle,velocity of movement,and time spent in inner circle(expressed as percentage of the total15min session).Behavioural data were taken in1min bins and averaged over the total15-min period.Data were expressed as mean values9standard error of the mean(SEM).Differences between groups were analyzed with one-or two-way analysis of variance (ANOVA)using the Sigmastat1.0statistics software package(Jandel Scientiﬁc)and considered to be signiﬁ-cant when P B0.05.3.ResultsAs expected,SHR had signiﬁcantly higher resting blood pressure when compared to WKY rats and SD rats,whereas WKY rats had signiﬁcantly higher blood pressure values compared to SD rats.This difference was found both during the circadian cycle(Fig.1)and in the home cage before the open-ﬁeld test(Table1). Resting heart rate was lower in SHR and WKY rats when compared to SD rats(Fig.1and Table1).During the circadian cycle,blood pressure,heart rate and locomotor activity values were higher during the dark period(1900–0700h)when compared to the light period(0700–1900h),a pattern which was not different between strains(Fig.1).Baseline blood pressure and heart rate before open-ﬁeld testing did not differ signiﬁcantly over the4con-secutive days in any of the strains(Table1).Exposure to the open-ﬁeld caused a marked rise in blood pressure and heart rate in all strains(Fig.2).Both the pressor response and tachycardia were signiﬁcantly greater in SHR than those in WKY rats and SD rats which did not differ signiﬁcantly from each other.Upon repeated exposure to the open-ﬁeld,pressor responses and tachy-cardia did not signiﬁcantly differ from day to day in any of the strains(Fig.3).By contrast,video tracking analysis showed that indices of behavioural activity Fig.1.Circadian rhythms of blood pressure(top panel),heart rate (middle panel)and behavioural activity(lower panel)in SHR,WKY and SD-rats as measured by telemetry in the home cage.Data are mean9SEM of seven SHR,seven WKY and eight SD-rats per group,collected during the light period(0700–1900h,white bars) and the dark period(1900–0700h,dark bars)over2consecutive days.For blood pressure data,ANOVA showed a signiﬁcant overall effect of strain(F(2,41)=47.7,P B0.001)and of time of day(F(1,41)= 168,P B0.001).For heart rate data,there was a signiﬁcant overall effect of strain(F(2,41)=50.0,P B0.001),of time of day(F(1,41)=936, P B0.001)and an interaction between these factors which was of borderline signiﬁcance(F(2,41)=3.3,P=0.045).For behavioural ac-tivity,there was a signiﬁcant effect of strain(F(2,41)=4.3,P=0.020) and of time of day(F(1,41)=573,P B0.001).M.6an den Buuse et al./Beha6ioural Brain Research119(2001)131–142135Table1Resting blood pressure and heart rate of SHR,WKY and SD-rats as measured by telemetry in the home cage before exposure to the open-ﬁeld aMean blood pressure Heart rate(mm Hg)(beats/min)SHR14095bDay129196b14394b,cDay23129932791214796b,cDay330498Day414096b,cWKYDay129797b12194Day212093298910bDay312493b30397b3019312293bDay4SD-rats3519810294Day110194Day235291110194cDay3349911337959894cDay4a Data are mean9SEM of eight SD rats,seven WKY rats and seven SHR.ANOVA on blood pressure data indicated a signiﬁcant overall effect of strain(F(2,19)=36.0,P B0.001)and a signiﬁcant difference over the4days between SHR and WKY rats,between SHR and SD rats,and between WKY rats and SD rats.Analysis of heart rate data indicated a signiﬁcant overall effect of strain(F(2,19)= 23.0,P B0.001)and a signiﬁcant difference over the4days between SHR and SD rats,and between WKY rats and SD rats.b P B0.05for difference with SD rats on the same day.c P B0.05for difference with WKY rats on the same day.while absolute circadian changes in blood pressure did not signiﬁcantly differ between the strains,when ex-pressed as percentage of baseline,circadian blood pres-sure variation was signiﬁcantly greater in SD rats than in either SHR or WKY rats.In contrast,absolute and percentage changes in blood pressure after novelty stress in SHR were nearly twice as great as those in WKY rats and SD rats.Circadian changes in heart rate,were greatest in WKY rats,both as absolute and percentage values.In sharp contrast,stress-induced in-creases in heart rate in SHR were more than twice those found in WKY rats and SD rats,which did not differ(Table3).Treatment of SHR and WKY rats with DSP-4,to induce central noradrenaline depletion,caused an im-mediate fall in blood pressure which lasted2–3days. Fig.2.Changes in blood pressure and heart rate in SHR,WKY and SD-rats during15min exposure to open-ﬁeld novelty stress.Data are mean change9within-animal SEM of eight rats per group.For blood pressure,ANOVA showed an overall signiﬁcant effect of strain (F(2,19)=27.4,P B0.001),of time in open-ﬁeld(F(14,266)=2.4,P= 0.004)and a strain×time interaction(F(28,266)=2.0,P=0.003).Fur-ther analysis showed that SHR data were signiﬁcantly higher than those in WKY rats and SD rats,which did not differ.For heart rate, ANOVA showed an overall signiﬁcant strain effect(F(2,19)=27.6, P B0.001),time effect(F(14,266)=4.7,P B0.001)and strain×time interaction(F(28,266)=4.8,P B0.001).Further analysis showed that SHR data were signiﬁcantly higher than those from WKY rats and SD rats,which did not differ.were reduced from the second day of open-ﬁeld expo-sure(Table2).During theﬁrst and subsequent exposures to the open-ﬁeld there were distinct differences in behavioural parameters between SHR and both WKY rats and SD rats,but also between WKY rats and SD rats.For total distance,data from SHR were signiﬁcantly higher than from either WKY rats or SD rats which did not differ. For all strains overall,data on day1were signiﬁcantly higher than on either days3or4.For distance moved in the inner circle,data from SD rats were signiﬁcantly lower than those from either WKY rats or SHR which did not differ.For percentage time spent in the inner circle,there was a gradual rise in values in SHR over repeated testing as opposed to a gradual reduction in SD rats and no change in WKY rats.Overall,data from SD rats were signiﬁcantly lower than those from either WKY rats or SHR which did not differ.For velocity of movement,WKY rats showed signiﬁcantly lower values than either SD rats or SHR which did not differ(Table2).Table3shows a comparison of the changes in blood pressure and heart rate over the circadian cycle with those found during novelty stress.Circadian variation of blood pressure and heart rate was much smaller than that found during open-ﬁeld exposure.In addition,M.6an den Buuse et al./Beha6ioural Brain Research119(2001)131–142 136Fig.3.Average15-min changes in blood pressure and heart rate in SHR,WKY and SD-rats exposed to the open-ﬁeld on4consecutive days.ANOVA on blood pressure data showed a signiﬁcant effect of strain only(F(2,19)=36.3,P B0.001).Overall,SHR data differed signiﬁcantly from those of both WKY and SD-rats,which did not differ from each other.SHR had signiﬁcantly greater blood pressure responses when compared to SD-rats on all days and when compared to WKY on days1,3,and4.A similar analysis for heart rate data again showed an overall strain effect only(F(2,19)=25.5,P B0.001) and signiﬁcantly greater responses in SHR when compared to either WKY or SD-rats that did not differ.SHR had signiﬁcantly greater tachycardias when compared to SD-rats on all days and when compared to WKY on days1and4.tion in resting blood pressure in both strains,and signiﬁcant reductions of both heart rate and be-havioural activity in the dark period,but not the light period after DSP-4treatment(Fig.4).These changes were present in both SHR and WKY rats.Exposure to the open-ﬁeld caused the same pattern of increases in blood pressure and heart rate before and after treat-ment with DSP-4although the time-course of the pres-sor response and tachycardia was slightly,but signiﬁcantly different in WKY rats(Fig.5).Fifteen minute average increases in blood pressure in SHR were2791mmHg before treatment versus2691 mmHg after treatment.In WKY rats these values were 1991versus1591mmHg,respectively.Fifteen minute average increases in heart rate in SHR were 13598beats/min before treatment versus14996 beats/min after treatment.In WKY rats,these values were11795and8998beats/min,respectively.Video tracking analysis of behaviour showed a marked reduc-tion of behavioural activity in WKY rats,but not SHR (Table4).Indeed,after being placed in the open-ﬁeld and showing some initial exploratory locomotor activ-ity,the WKY rats would often remain in the same position in the open-ﬁeld for much of the duration of the session.Analysis of noradrenaline levels in the brain of SHR and WKY rats showed a marked reduction in frontal cortex,but not hypothalamus in both strains.Do-pamine and adrenaline levels were not affected by DSP-4treatment(Fig.6).4.DiscussionThis study produced a number of mainﬁndings. Firstly,we observed that SHR showed markedly en-hanced increases in blood pressure and heart rate in response to novelty stress when compared to WKY rats as well as SD rats,which did not differ.In the open-ﬁeld,SHR had highest scores for distance moved and velocity of movements,followed by SD rats and then WKY rats.In all strains,cardiovascular stress re-sponses were not reduced upon repeated exposure to the open-ﬁeld,while differential habituation of be-havioural responses was seen.As expected,baseline blood pressure was increased in SHR compared to both normotensive strains throughout the circadian cycle, with highest values for all parameters occurring in all strains during the dark period as opposed to the light period.These circadian changes in blood pressure and heart rate were much smaller than those during novelty stress.Moreover,while the circadian changes were sim-ilar in magnitude in all strains,the SHR stood out from the other two strains with respect to stress-induced responses.We furthermore observed that marked deple-tion of noradrenaline in the brain after treatment withBy7days after treatment,resting blood pressure and heart rate,obtained just before the open-ﬁeld test,were not signiﬁcantly different when compared to values obtained in the same rats before treatment.In this case, blood pressure before and1week after DSP-4treat-ment was17392versus16693mmHg in SHR and 14095and13595mmHg in WKY rats,respectively. Heart rate before and1week after DSP-4treatment was323912and30699beats/min in SHR and3149 15and307914beats/min in WKY rats.Body weight of the rats before and1week after DSP-4treatment was29897versus31394g in SHR and337912 versus338910g in WKY.Analysis of circadian rhythms revealed a slight reduc-M.6an den Buuse et al./Beha6ioural Brain Research119(2001)131–142137DSP-4did not alter cardiovascular stress responses signiﬁcantly in SHR and WKY rats or behavioural responses in SHR.The already relatively low be-havioural responses in WKY were reduced even further after DSP-4treatment.Spontaneous behavioural activ-ity in the home cage was reduced in both SHR and WKY rats in the dark period after treatment.4.1.Strain differencesGreater blood pressure and heart rate responses to stress in SHR when compared to WKY rats have been described before[7,14,16,24,26,27].However,there are several important differences between these early stud-ies and the present one.Unlike many of the previous studies,we used novelty stress which can be considered a more‘naturalistic’stress than the more physical stres-sors like footshock or restraint.Our results furthermore show that the apparently greater stress responses in SHR were not merely the result of relatively smaller responses in WKY rats as a similar difference was found between SHR and SD rats.The markedly greater cardiovascular changes in SHR were furthermore spe-ciﬁc for stress,inasmuch as there were no major strain differences in the magnitude of circadian cardiovascular variation.Finally,the much greater stress-induced car-diovascular responses in SHR were not a non-speciﬁc result of behavioural hyperactivity in this strain.Al-though our behavioural analysis showed that SHR travelled the greatest distance in the open-ﬁeld,fol-lowed by SD rats and then WKY rats,cardiovascular responses in the openﬁeld in the latter two strains were essentially similar with the SHR being markedly differ-ent from either WKY rats or SD rats.In addition, habituation was found in behavioural,but not cardio-vascular responses after repeated exposure to the stress. It has now been established that behavioural hyper-activity of SHR as observed in some stress paradigms is unrelated to cardiovascular reactivity and can be sepa-rated by selective breeding[22].If we had performed theﬁrst part of our study with only SHR and WKY rats,we would not have been able to draw any conclu-sions about the speciﬁcity of apparently enhanced car-diovascular stress reactivity in SHR as a marked difference in behavioural responses is similarly found between the strains.A number of studies have suggested that SHR differ from WKY rats in the duration of cardiovascular changes to stress and by showing reduced habituation [28,33].It was concluded that in SHR the prolonged stress responses may contribute to the development of hypertension in this strain.We exposed our rats to the open-ﬁeld for15min on4consecutive days and found little habituation of cardiovascular responses within and between sessions in either SHR,WKY rats or SD rats.By contrast,behavioural responses were differen-tially reduced over the course of the4days and also during each session.The difference between our resultsTable2Behaviour of SHR,WKY rats and SD rats during exposure to the open-ﬁeld on4consecutive days as measured by video tracking analysis aDistance inner(cm)%time innerDistance total(cm)Velocity(cm/s)SHRDay18779135481930209413.490.65210926Day23694144591513.590.613.490.9Day34695128191345339331252911446793113.290.8Day44295WKY rats2571924 6.790.6Day158971194914Day22700924131091453967.390.83896 6.890.7Day391891322959252151920 6.390.7929910Day45396SD rats400193912.191.0Day115935999912.091.3263493632597893Day22473943437910Day389312.892.0 230593129197Day410.691.1793a Data are mean9SEM of eight SD rats,seven WKY rats and seven SHR.For total distance,2-way ANOVA for repeated measures using strain and day as factors showed a signiﬁcant effect of strain(F(2,19)=17.5,P B0.001)and of day(F(3,57)=5.0,P=0.004).Data from SHR were signiﬁcantly higher than from either WKY rats or SD rats which did not differ.Data on day1were signiﬁcantly higher than on either day3or day4.For distance moved in the inner circle,ANOVA showed a signiﬁcant effect of strain only(F(2,19)=15.5,P B0.001).Data from SD rats were signiﬁcantly lower than those from either WKY rats or SHR which did not differ.For percentage time spent in the inner circle,ANOVA showed a signiﬁcant effect of strain(F(2,19)=20.0,P B0.001)and a signiﬁcant strain×day interaction(F(6,57)=3.6,P=0.004),reﬂecting the gradual rise in values in SHR over repeated testing as opposed to a gradual reduction in SD rats and no change in WKY rats.Overall,data from SD rats were signiﬁcantly lower than those from either WKY rats or SHR which did not differ.For velocity of movement,ANOVA showed a signiﬁcant effect of strain(F(2,19)=14.6,P B0.0001)with WKY rats data being signiﬁcantly lower than either SD rats or SHR which did not differ.。

DSP外文资料91

Denervation of the locus coeruleus projections by treatment withthe selective neurotoxin DSP-4[N (2-chloroethyl)-N -ethyl-2-bromobenzylamine]reduces dopamine release potential in thenucleus accumbens shell in conscious ratsRiina Ha ¨idkind a ,Toomas Kivastik b ,Marika Eller a ,Ivo Kolts c ,Lars Oreland d ,Jaanus Harro a,*aDepartment of Psychology,Centre of Behavioural and Health Sciences,University of Tartu,EE-50410T artu,EstoniabDepartment of Pharmacology,University of T artu,EE-50410Tartu,Estonia cDepartment of Anatomy,University of T artu,EE-50410Tartu,EstoniadDepartment of Neuroscience,Pharmacology,University of Uppsala,Biomedical Centre,Uppsala,SwedenReceived 14June 2002;received in revised form 5July 2002;accepted 18July 2002AbstractPretreatment with DSP-4,a neurotoxin highly selective for the locus coeruleus (LC)noradrenergic projections,2weeks before in vivo microdialysis in conscious rats had no effect on baseline extracellular dopamine (DA)levels in the nucleus accumbens shell,but reduced dose-dependently the dopamine response to depolarisation induced by 50mM KCl.DA metabolism in the frontal cortex,as measured ex vivo,was increased in animals treated with a low (10mg/kg)but not with a high dose (50mg/kg)of DSP-4,possibly indicating an increased sensitivity to stress in these animals and thus suggesting differential regulation of DA in the forebrain by the LC lesions.The reduced DA release potential in the nucleus accumbens after DSP-4treatment suggests that weakening of the LC input to DA nerve cells contributes to motivational deﬁcits.q 2002Elsevier Science Ireland Ltd.All rights reserved.Keywords :Nucleus accumbens;Frontal cortex;Dopamine;DSP-4[N (2-chloroethyl)-N -ethyl-2-bromobenzylamine];Locus coeruleus;In vivo microdialysisDopaminergic (DA-ergic)neurotransmission,particu-larly DA release in the nucleus accumbens,has been impli-cated in the mechanism of action of addictive drugs [1,4],and reduction in DA-ergic neurotransmission in the accum-bens may serve as the basis for anhedonia and loss of energy in depression [6].It is well established that there is a nora-drenergic (NA-ergic)stimulatory input from the locus coer-uleus (LC)to the mesencephalic DA-ergic cell groups (see ref.[6]for references),which provide DA-ergic innervation to forebrain regions.There is some evidence that lesions of the LC projections can be a reason for a reduced DA release in the accumbens.Thus,in anaesthetised rats,a 28%reduc-tion in basal DA overﬂow was observed using in vivo microdialysis in the nucleus accumbens after pretreatment with DSP-4,a selective NA-ergic neurotoxin [13].Thisstudy also observed a reduction in extracellular DA in the striatum after DSP-4treatment.On the other hand,we have previously been unable to see any change in either basal or KCl-induced DA release in the frontal cortex after DSP-4pretreatment [12].In a recent study in conscious rats no decrease in the extracellular DA in the nucleus accumbens core was found after DSP-4treatment [10].The aim of the present investigation was to study whether partial or nearly complete denervation of the LC by DSP-4treatment inﬂu-ences DA overﬂow in the nucleus accumbens shell in conscious animals.Male Wistar rats (300–400g,Kuopio,Finland)were housed individually under 12h light/dark cycle (lights on at 07:00h)with food and water available ad libitum.DSP-4[N (2-chloroethyl)-N -ethyl-2-bromobenzylamine](Astra,Sweden)was administered as a single dose of 10or 50mg/kg (expressed as for hydrochloride)intraperitoneally (IP)at the start of individual housing.Each dose was weighed separately,dissolved in distilled water and imme-Neuroscience Letters 332(2002)79–820304-3940/02/$-see front matter q 2002Elsevier Science Ireland Ltd.All rights reserved.PII:S0304-3940(02)00817-0/locate/neulet*Corresponding author.Tel.:1372-7-375-911;fax:1372-7-375-900.E-mail address:jharro@psych.ut.ee (J.Harro).diately injected.Control animals received an injection of distilled water.After the administration of DSP-4the animals ’body weight was monitored daily.Surgery was carried out 1week after the toxin treatment.The animals were anaesthetised with chloral hydrate (350mg/kg IP)and mounted in a Kopf stereotactic frame.Guide cannulas for the microdialysis probes (MAB 6;AgnThos AB,Sweden)were implanted above the left nucleus accumbens shell according to the following coordinates relative to bregma:AP:11.7,ML:11.2,and DV:22.8(the ﬁnal DV coordi-nate after probe insertion 28.2;according to Paxinos and Watson [15]).Three stainless steel screws and dental acrylic was used to ﬁx the cannula to the scull.After the surgery the animals were given 1week to recover.During this period the rats were handled and weighed daily.Microdialysis was conducted in awake freely moving rats.In the morning of the experiment day (at about 08:00h)the animals were transported to a separate experiment room.The animals remained in their individual home cages throughout the experiment,though the metal cage covers were replaced with plastic ones having an open ceil-ing.Microdialysis probes with 2mm active polyethersul-phone membrane,cut-off 15kD (MAB 6,AgnTho ’s AB,Sweden)were connected to a syringe pump (World Preci-sion Instruments,USA)and a CMA/142microsampler (CMA,Sweden)via a two-channel swivel.Probes were inserted in the morning of the day of experiment and perfused with arti ﬁcial cerebrospinal ﬂuid (147mM NaCl,3.0mM KCl,1.3mM CaCl 2,1.0mM MgCl 2,1mM Na 2H PO 4,and 0.2mM NaH 2PO 4;pH 7.3–7.4)at a constant rate 1m l/min.After the probe insertion the perfusate was discarded during the ﬁrst 120min.This was followed by collection of 15dialysate fractions (each for 20min)into the vials pre ﬁlled with 10m l of 0.3M perchloric acid.After the tenth sample the system was switched to the perfusion solu-tion containing 50mM KCl and left so for 40min.Upon completion of the experiment the animals were deeply anaesthetised with chloral hydrate (350mg/kg IP)and decapitated;the brains were removed,immediately frozen in ice cold acetone,and kept at 2808C.The brains were sectioned in a cryostatic microtome (Microm GmbH,Germany)after the whole frontal cortex had been taken for separate analysis;the probe placements were determined according to the atlas by Paxinos and Watson [15]and data of animals with probe placements outside the nucleus accumbens shell were excluded from the analysis.The quantity of DA in the samples was determined by high-performance liquid chromatography with electroche-mical detection.The chromatography system consisted of Hewlett Packard series 1100pump and autosampler,a LiChrospher 100RP-18column (250£3mm,5m m),an ESA 5021conditioning cell (1300mV),an ESA 5011analytical cell (ﬁrst electrode 2300mV,second electrode 1400mV)and an ESA Coulochem II controller unit.The mobile phase composition was:0.05M sodium citrate buffer,pH 5.5,0.02mM EDTA,2mM sodium octylsupho-nate,0.3mM triethylamine,12%methanol.The content of monoamines and their metabolites in brain tissue was deter-mined as previously described [9].After exclusion of animals with incorrect placement of microdialysis probes or incomplete biochemical data due to sampler failures,the number of animals in analysis was 18(control n ¼5,DSP-410mg/kg n ¼8,DSP-450mg/kg n ¼5).Repeated measures analysis of variance (ANOVA)was used,followed by a post-hoc Fisher ’s analysis per time point.As shown in Fig.1,no signi ﬁcant effect of LC denerva-tion by DSP-4pretreatment was observed on basal extra-cellular DA in the nucleus accumbens shell in conscious rats.Nevertheless,local depolarisation elicited by perfusion with 50mM KCl was affected by this treatment,as revealed by signi ﬁcant two-way ANOVA sample £treatment inter-action (F ð2;28Þ¼3:51,P ,0:0001).Post-hoc Fisher ’s protected least signi ﬁcant difference tests revealed that theR.Ha ¨idkind et al./Neuroscience Letters 332(2002)79–8280Fig.1.Effect of DSP-4on DA levels in the nucleus accumbensshell in conscious rats.The values (mean ^SEM)are expressed as fmol/25m l of sample.Solid bar indicates infusion of 50mM KCl.For statistical evaluation of the data,see text.Table 1Effect of DSP-4on monoamine levels in the rat frontal cortex aVehicleDSP-4(10mg/kg)DSP-4(50mg/kg)NA 2.75^0.54 2.38^0.240.72^0.18b,c DA 0.53^0.120.59^0.110.52^0.09HVA 0.20^0.020.28^0.03d,e 0.20^0.02DOPAC 0.12^0.020.19^0.020.14^0.025-HT 2.45^0.55 2.47^0.22 2.57^0.265-HIAA2.91^0.372.98^0.433.02^0.26aThe values (mean ^SEM)are expressed as pmol/mg wet weight tissue.bSigni ﬁcant difference from the Vehicle group,P ,0.001;post-hoc Fischer ’s PLSD test after signi ﬁcant ANOVA.cSigni ﬁcant difference from the DSP-410mg/kg group,P ,0.005;post-hoc Fischer ’s PLSD test after signi ﬁcant ANOVA.dSigni ﬁcant difference from the Vehicle group,P ,0.05;post-hoc Fischer ’s PLSD test after signi ﬁcant ANOVA.eSigni ﬁcant difference from the DSP-450mg/kg group,P ,0.05;post-hoc Fischer ’s PLSD test after signi ﬁcant ANOVA.peak extracellular DA was signiﬁcantly(P¼0:011) reduced after the higher(50mg/kg)dose of DSP-4,whereas the effect of the lower dose(10mg/kg)missed the conven-tional level of signiﬁcance(P¼0:076).DSP-4treatment reduced NA levels in the frontal cortex(Fð2;13Þ¼10:8, P,0:005).However,only the higher dose reduced NA levels signiﬁcantly(Table1).DSP-4pretreatment had no effect on cortical levels of5-hydroxytryptamine(5-HT)and DA,but there was a signiﬁcant effect on homovanillic acid levels(Fð2;13Þ¼4:71,P,0:05).Thus,the lower but not the higher dose of DSP-4increased HVA levels in the fron-tal cortex.DSP-4is a noradrenergic neurotoxin highly speciﬁc to the nerve terminals originating from the LC[3,16],while it has no direct effect on DA-or5-HT-ergic neurons[11].DSP-4does not affect tissue DA levels but both DA utilisation and3,4-dihydroxyphenylacetic acid production rates have been found to be lower after DSP-4treatment[2,5].There is also striatal DA D2receptor upregulation as well as behavioural supersen-sitivity to the effects of amphetamine after DSP-4treatment [8].In contrast to theﬁndings of Lategan et al.[13]in anaes-thetised rats,a recent study in conscious animals did notﬁnd any reduction of extracellular DA in the nucleus accumbens core after DSP-4treatment[10].Similarly,we did notﬁnd any signiﬁcant reduction of extracellular DA in the nucleus accumbens shell in conscious rats.As determined by DA over-ﬂow,induced by local depolarisation,there is,however,a signiﬁcantly lower DA release potential in LC-denervated animals,which may translate into lower DA release in the nucleus accumbens in response to stimuli activating the LC. Chemical denervation of the LC projections by DSP-4treat-ment reduced the effectiveness of positive reinforcers[14]. Thus it is conceivable that weakening of the LC input to DA nerve cells,which results in diminished DA output in the accumbens,contributes to motivational deﬁcits,occurring, e.g.in depression[6].On the other hand,as low basal extra-cellular DA in the nucleus accumbens shell has been also associated with peak drug-seeking behaviour[4],the possibi-lity that functional weakness of the LC contributes to drug addiction(which is frequently associated with depression) should be examined.Unexpectedly,there was evidence for an increased DA metabolism in the frontal cortex after the lower dose of the neurotoxin.Increases in frontal cortical HVA have been suggested to be strongly indicative as a reaction of stress [17,18],which has previously also been observed in this laboratory[9].We have previously found that DSP-4treat-ment has a biphasic effect in rats submitted to forced swim-ming:10mg/kg of DSP-4increased immobility but50mg/ kg rather reduced this measure[7].It is thus tempting to speculate that partial LC denervation(10mg/kg DSP-4) renders animals more susceptible to stress.Nevertheless, such an interpretation as well as the idea that the effect of LC lesions on DA-ergic neurotransmission is different for the mesolimbic and the mesocortical pathways would require speciﬁcally designed experiments which control for stress and compare both DA release and metabolism in these respective projection areas in parallel.This work was supported by grants from the Estonian Science Foundation(No4531),the Estonian Ministry of Education(No0814),and the Swedish Academy of Sciences.Prof.Svante Ross(Astra)kindly provided DSP-4.We are grateful to Prof.Tomas Ho¨kfelt for support in setting up the microdialysis lab and Ms Margit Kirk for skilful technical assistance.[1]Di Chiara,G.and Imperato,A.,Drugs abused by humanspreferentially increase synaptic dopamine concentrations in the mesolimbic system of freely moving rats,Proc.A,85(1988)5274–5278.[2]Eller,M.and Harro,J.,Partial denervation of the locus coer-uleus projections by treatment with the selective neuro-toxin DSP-4potentiates the long-term effect of parachloroamphetamine on5-hydroxytryptamine metabo-lism in the rat,Neurosci.Lett.,322(2002)53–56.[3]Fritschy,J.-M.and Grzanna,R.,Immunohistochemical analy-sis of the neurotoxic effects of DSP-4identiﬁes two popula-tions of noradrenergic axon terminals,Neuroscience,30 (1989)181–197.[4]Gerrits,M.A.,Petromilli,P.,Westenberg,H.G.,Di Chiara,G.and van Ree,J.M.,Decrease in basal dopamine levels in the nucleus accumbens shell during daily drug-seeking beha-viour in rats,Brain Res.,924(2002)141–150.[5]Hallman,H.and Jonsson,G.,Pharmacological modiﬁca-tions of the neurotropic action of the noradrenaline neuro-toxin DSP-4on central noradrenaline neurons,Eur.J.Pharmacol.,103(1984)269–278.[6]Harro,J.and Oreland,L.,Depression as a spreading adjust-ment disorder of monoaminergic neurons:a case for primary implication of the locus coeruleus,Brain Res.Rev.,38(2001)79–128.[7]Harro,J.,Pa¨hkla,R.,Modiri,A-M.,Harro,M.,Kask,A.andOreland,L.,Dose-dependent effects of noradrenergic denervation by DSP-4treatment on forced swimming and b-adrenoceptor binding in the rat,J.Neural Transm.,106 (1999)619–629.[8]Harro,J.,Meriku¨la,A.,Lepiku,M.,Modiri,A.-R.,Rinken,A.and Oreland,L.,Lesioning of locus coeruleus projections by DSP-4neurotoxin treatment:effect on amphetamine-induced hyperlocomotion and dopamine D2receptor binding in rats,Pharmacol.Toxicol.,86(2000)197–202.[9]Harro,J.,To˜nissaar,M.,Eller,M.,Kask,A.and Oreland,L.,Chronic variable stress and partial5-HT denervation by parachloroamphetamine treatment in the rat:effects on behavior and monoamine neurochemistry,Brain Res.,899 (2001)227–239.[10]Hatip-al-Khatib,I.,Mishima,K.,Iwasaki,K.and Fujiwara,M.,Microdialysates of amines and metabolites from core nucleus accumbens of freely moving rats are altered by dizocilpine,Brain Res.,902(2001)108–118.[11]Jonsson,G.,Hallman,H.,Ponzio,F.and Ross,S.,DSP-4[N(2-chloroethyl)-N-ethyl-2-bromobenzylamine]:a useful denervation tool for central and peripheral noradrenaline neurons,Eur.J.Pharmacol.,72(1981)173–188.[12]Kask,A.,Harro,J.,Tuomainen,P.,Ra¨go,L.and Ma¨nnisto¨,P.T.,Overﬂow of noradrenaline and dopamine in frontal cortex after[N-2(chloroethyl)-N-ethyl-2-bromobenzyla-mine](DSP-4)treatment:in vivo microdialysis study inR.Ha¨idkind et al./Neuroscience Letters332(2002)79–8281anaesthetised rats,Naunyn-Schmiedeberg’s Arch.Pharma-col.,355(1997)267–272.[13]Lategan,A.J.,Marien,M.R.and Colpaert,F.C.,Suppressionof nigrostriatal and mesolimbic dopamine release in vivo following noradrenaline depletion by DSP-4:a microdialy-sis study,Life Sci.,50(1992)995–999.[14]Morley,M.J.,Shah,K.,Bradshaw,C.M.and Szabadi,E.,DSP-4and Herrnstein’s equation:further evidence for a role of noradrenaline in the maintenance of operant beha-viour by positive reinforcement,Psychopharmacology,96 (1988)551–556.[15]Paxinos,G.and Watson,C.,The Rat Brain Stereotaxic Coor-dinates,Academic Press Inc,San Diego,1986.[16]Ross,S.B.,Long-term effect of N-2-chloroethyl-N-ethyl-2-bromobenzylamine hydrochloride on noradrenergic neurons in the rat brain and heart,Br.J.Pharmacol.,58 (1976)521–527.[17]Roth,R.H.,Tam,S.,Ida,Y.,Yang,J.and Deutch,A.Y.,Stressand the mesocorticolimbic dopamine systems,In P.W.Kali-vas and C.B.Nemeroff(Eds.),The Mesocorticolimbic Dopa-mine System,The New York Academy of Sciences,New York,1988,pp.138–147.[18]Thierry, A.M.,Tassin,J.P.,Blanc,G.and Glowinski,J.,Selective activation of mesocortical DA system by stress, Nature,263(1976)242–243.R.Ha¨idkind et al./Neuroscience Letters332(2002)79–82 82。

DSP外文资料72

Systematic error cancellations and fault detection of resolver angular sensors using a DSP based systemSantanu Sarma *,A.VenkateswaraluControl Systems Group,ISRO Satellite Centre,Vimanapura PO,Airport Road,Bangalore 560017,Indiaa r t i c l e i n f o Article history:Received 30March 2007Accepted 3September 2009Keywords:Angular position and speed sensing Resolver-to-digital (R/D)converter Phase-locked loops Phase delay estimation Magnitude estimationSynchronous amplitude demodulator Adaptive algorithma b s t r a c tResolver sensor based angular position and speed sensing are extensively used in safety critical servo applications that demands accurate as well as high-resolution position and speed information for feed-back control.In this paper,a novel scheme for position and speed sensing along with fault detection and identiﬁcations of a resolver sensor with systematic errors like magnitude imbalance,imperfect quad-rature,and inductive harmonics is presented.The proposed scheme of resolver-to-digital (R/D)conver-sion mitigates the errors in position and speed estimate due to these common resolver imperfections and provides fault indicators such as good resolver signal,degradation of signal,and loss of signal for fault tolerant operation and diagnosis of malfunctions in the sensor system for safety critical systems.The pro-posed method incorporates software generation of the resolver carrier using a digital ﬁlter for synchro-nous demodulation without unintended time delay of the processed outputs,in such a way that there is substantial saving in hardware,for instance,carrier oscillator and associated digital and analog circuits for amplitude demodulators.The R/D converter incorporates an adaptive phase-locked loop (APLL)that accurately estimates the angular position and speed for a large range of operation along with superior tracking performance under dynamic conditions.Also,it provides the estimate of the magnitudes of the resolver outputs,estimate of the imperfect quadrature,and indication of harmonic distortion in the sensing angle,which can be used to directly access the quality of the resolver sensor -puter simulations and experimental results demonstrate an accurate R/D converter with adaptive capa-bilities to mitigate all the major systematic disturbances with reduced hardware complexity.Ó2009Elsevier Ltd.All rights reserved.1.IntroductionResolvers are very robust and cost-effective angular position sensors [1]that are extensively used in many safety critical appli-cations like aircrafts,satellite antennas,robots,machine tools,and radars-to mention few.They have also been used in electrically as-sisted power steering in mid and small size commercial cars [2],and fuel saving starter alternators [3].Most of these applications require smooth torque control of dc motor that demands accurate,high-resolution position and/or speed feedback information from the resolver sensor for commutation and control purpose.How-ever,due to the imperfections and non-ideal characteristics of the resolver sensor,accurate position and speed sensing with the popular methods [3–12]becomes a challenge.Above all,when there is no margin of failure that can be tolerated in such safety critical application,the need and motivations for new methods of fault detection and identiﬁcation become unavoidable.In additionto that there is also the necessity for cost effective realization,ﬂexibility of operation,reduction of weight and size,and wide range of operation.In this paper,a scheme is presented that mitigates the position error incurred due to the common resolver imperfections like amplitude imbalance,imperfect quadrature,and resolver inductive harmonics.Besides,it provides fault indicators such as good resol-ver signal,degradation of signal,and loss of signal for fault tolerant operation of safety critical systems.Also,it incorporates an adap-tive phase-locked loop (APLL)to accurately estimate the angular position and speed for a large range even in dynamic conditions.Moreover,the scheme provides the estimate of the magnitudes of the resolver outputs,the phase delay between envelops due to imperfect quadrature,and the measure of harmonic distortion in the sensing angle.In addition to that the scheme presents a novel,delay less demodulation of the resolver outputs that greatly re-duces the hardware.In fact,reduction in hardware like oscillator,peak detectors and other associated digital and analog circuitries causes a drastic reduction in weight,size and cost.The paper is organized in seven sections.While the basic oper-ation and errors in the sensor are discussed in Section 2,the0957-4158/$-see front matter Ó2009Elsevier Ltd.All rights reserved.doi:10.1016/j.mechatronics.2009.09.002*Corresponding author.Tel.:+918025082343/805082343;fax:+91805082306.E-mail addresses:santanu@.in ,sarma_santanu@ (S.Sarma).Mechatronics 19(2009)1303–1312Contents lists available at ScienceDirectMechatronicsj o ur na l h om e pa ge :w w w.e ls e v ie r.c o m/lo c at e/me c h at r on icsproposed R/D scheme is presented in Section3.Section4presents the fault detection and identiﬁcation methods,which is followed by simulation results in Section5.The experimental result for the proposed scheme is corroborated in Section6followed by the concluding remarks in Section7.2.Resolver operation and errorsResolvers resemble small motors and have magnetically cou-pled rotor and stator windings[1–2].The analog output of the re-solver contains the angular position information that is obtained in digital form using a resolver-to-digital(R/D)converter.Many pop-ular methods[3–12]exist when an ideal resolver is considered with sinusoidal excitation voltagev ref¼V sinðx c tÞ;ð1Þand stator signals generated asV Cðt;hÞ¼g v v ref cosðhÞ¼A cosðhÞsinðx c tÞ;ð2ÞV Sðt;hÞ¼g v v ref sinðhÞ¼A sinðhÞsinðx c tÞ;ð3Þwhere A is a constant that depends on the construction of the re-solver,h the angular position of the rotor with respect to one pair of stator winding,and x c is the frequency of the excitation or carrier signal to the rotor.Unfortunately,practical resolvers are imperfect and the resolver output signal deviates considerably from the assumed ideal.This results in severe degradation in the position accuracy performance[13]and error at converter output when ideal methods[3–12]are used.It is therefore important that the effect of such systematic disturbances,partic-ularly,amplitude imbalance,imperfect quadrature,and resolver inductive harmonics be considered in the design and develop-ment of the R/D converter for angular position and speed estima-tion.These systematic errors are therefore described in the following subsections.Amplitude imbalance refers to occurrence of amplitude differ-ence between the two resolver signals such that the resolver signal can be represented asV Cðt;hÞ¼A cosðhÞsinðx c tÞ;V Sðt;hÞ¼ð1þmÞA sinðhÞsinðx c tÞ;ð4Þwhere m represents the amount of amplitude imbalance.On the other hand,imperfect quadrature refers to a spatial misalignment between the two resolver phases.That is,the case when the two inductance proﬁles of the two phases are not exactly p/2radians out of phase with one another.In this situation the resolver signal can be expressed asV Cðt;hÞ¼A cosðhÞsinðx c tÞ;V Sðt;hÞ¼A sinðhþ/Þsinðx c tÞ;ð5Þwhere u is the amount of quadrature error.In fact,it is impossible to construct a resolver with inductance proﬁle that is perfect sinu-soidal and co-sinusoidal functions of position.More generally,the inductance will contain the harmonics which creates resolver out-put signal of the form:V Cðt;hÞ¼A K0þX1n¼1K n cosðn hÞ()sinðx c tÞ;V Sðt;hÞ¼A K0þX1n¼1K n sinðn hÞ()sinðx c tÞ:ð6ÞWith all the above three errors,the generalized resolver signal out-put becomes V Cðt;hÞ¼A K0þX1n¼1K n cosðn hÞ()sinðx c tÞ;V Sðt;hÞ¼ð1þmÞA K0þX1n¼1K n sin½nðhþ/Þ()sinðx c tÞ:ð7ÞIn[13],the percentage of contribution due to different errors is de-rived.It is clear from[13]that the R/D conversion process is very sensitive to the non-ideal characteristics.For example,an ampli-tude imbalance of0.62%is all that required to create a1/2LSB posi-tion error out of10bits resolution.Additionally,when a DC inductive component term(K0)is present,0.22%of the fundamental component is sufﬁcient to produce1/2LSB position errors.Roughly, the maximum limit values for offset errors due to mechanical mis-alignment and the amplitude differences of resolver is about4%, whereas the total harmonic distortions(THDs)due to harmonics, which is less-distinctive,is less than1%[14].Considering this,the offset,amplitude,and phase errors become the dominant factors and the harmonic distortion being not signiﬁcant,results in the re-solver output signals as:V Cðt;hÞ¼A f K0þK1cosðhÞg sinðx c tÞ;V Sðt;hÞ¼ð1þmÞA f K0þK1sinðhþ/Þg sinðx c tÞ;ð8ÞIn most of the cases,K0is very small or zero by design,such that the resolver output signals can be considered to beV Cðt;hÞ¼m1cosðhÞsinðx c tÞ;V Sðt;hÞ¼m2sinðhþ/Þsinðx c tÞ;ð9Þwhere m1=AK1and m2=(1+m)AK1.It has been shown in[13]that the errors incurred due to these imperfections are quite signiﬁcant on the measurement accuracy. However,the solution suggested for correction of amplitude imbal-ance in[13]is based on calibration and number of experiments that are tiresome and time consuming.The method suggested for correcting quadrature imbalance,further,results in amplitude imbalance,a sign change and a constant phase offset in the de-coded angle,which is not acceptable in most practical applications. Moreover,the effect of inductance harmonics on the position error was reduced only under certain conditions by adjusting the ampli-tude imbalance.But,as mention in[13],it is impossible to apply the suggested technique in[13]that would hold for all harmonics. More recently,a new method has been proposed in[14]that over-comes some of the limitations of[13]for non-ideal resolver.In this method,the resolver or encoder signal are represented as a Fourier series,and the complex coefﬁcients are obtained by evaluating the cross-correlation of the line signals over certainﬁxed number of periods of the line signals.This demands an evaluation of an inte-gral of the squared error over number of periods of the resolver output signal.As a result,this method is not valid for online adap-tation for fast point-to-point positioning with small distances for resolver line signals with less than one period[14].The proposed method,however,overcomes the limitations of[13,14]and elimi-nates the position error incurred due to the common resolver imperfections like amplitude imbalance,imperfect quadrature, and resolver inductive harmonics in addition to other important features as described in the next section.3.Proposed scheme of R/D converter for angular position and speed sensingThe proposed scheme of R/D converter block diagram for angu-lar position and speed sensing along with fault indicators is shown in Fig.1.The problem considered in this paper is to estimate h;_h;/, m1,m2along with generation of fault indicators such as good resol-ver signal(GRS),degradation of signal(DOS),and loss of signal1304S.Sarma,A.Venkateswaralu/Mechatronics19(2009)1303–1312(LOS)for fault tolerant operation of safety critical systems.Besides,the proposed scheme also provides the indication of the Total Har-monic Distortion (THD)in position sensing due to harmonics and manufacturing defects,which can greatly help in veriﬁcation of the sensor quality and speciﬁcations.The proposed scheme consists of a sine–cosine generator and associated peak detection logic for synchronous demodulation,an adaptive PLL (APLL)for angular position,speed and magnitude esti-mation,and an adaptive quadrature phase detector to estimate imperfect quadrature phase delay.The sine–cosine generator gen-erates the sinusoidal carrier at the required frequency x c and amplitude V that can be controlled as per the requirement by sim-ply modifying their values in the software.The resolver carrier v ref is applied to its carrier terminals to produce the quadrature signals V C (t ,h )and V S (t ,h )corresponding to the shaft rotation.The quadra-ture signals that are amplitude modulated are synchronously demodulated by sampling at the peak of the sinusoid carrier as in [8].One of the demodulated envelop that contains the angular information is then used by the adaptive PLL to estimate the angu-lar position and speed of the rotating shaft.The other envelop that is having a phase shift with respect to the former due to the imper-fect quadrature is used in a adaptive quadrature phase detector to give an estimate of that phase shift.Moreover,the quadrature car-riers needed for its operations are obtained from the APLL.Theamplitude estimates ^m1;^m 2and the known carrier amplitude V are used in the fault detection logic to generate the above fault indi-cators to know the health and status of the R/D conversion scheme.The proposed R/D converter provides several signiﬁcant im-proved features as compared to some of the competitive methodsas tabulated in Table 1.Some of the major features of the proposed method are as follows:The proposed scheme performs software synchronous demodu-lations such there is considerable saving in hardware without introducing any additional time delay to affect the dynamics of the servo loop.It provides direct estimate of the position,speed,and magni-tudes of the resolver outputs,the phase delay between envelops due to imperfect quadrature,and the measure of harmonic dis-tortion in the sensing angle,which can greatly help in veriﬁca-tion of the speciﬁcations and sensor quality.It provides fault indicators such for fault tolerant operation of safety critical systems.These indicators can be used to derive conclusion such as the resolver is operating correctly,possibility of short or open circuit in the winding,and failure of the resolver.The method adapts online to time-varying systematic distur-bances and errors.It has a wide of operational range and good tracking performance.Required computation power is low and the scheme can be implemented on a low-cost microcontroller or DSP with improved ﬂexibility.Saving in size,weight,and cost.It is clear from Table 1that unlike other methods the proposed scheme provides number of advantages that are very attractive.Having being realized mostly in software,the scheme provides great ﬂexibility,repeatability of results,and ease of modiﬁcations.Table 1Comparison of features of competitive methods.R/Dconversion method Saving on amplitude demodulators Saving on hardware oscillators Compensatesamplitude imbalance Compensates quadrature phase imperfection Compensatesharmonic distortions Faultindicators Both x &h available Murray [3]No No No No No Yes Yes Yim [4]No No No No No No No Benammar [6]No No No No No No No Sarma [8]Yes Yes No No No No No Schwesig [9]No No NoNoNo No Yes Ishizuka [10]No No Yes (off-line)Yes (off-line)No No No Hanselman [13]No No Yes (off-line)No Yes No No Bünte [14]No No Yes No Yes No Yes ProposedYesYesYesYesYesYesYesFig.1.Propose scheme of S.Sarma,A.Venkateswaralu /Mechatronics 19(2009)1303–13121305The following subsections described brieﬂy these software blocks involved in the proposed scheme.3.1.Resolver carrier generation and synchronous demodulation The carrier signal to the resolver is generated by using a digital sine–cosine generator [15],which is basically a second order digi-tal ﬁlter with its pole on the unit circle [8].There are many possible representations of this sine cosine generator;however,the single multiplier structure with three associated addition is used for com-putation efﬁcient software implementation.This digital ﬁlter pro-duces the sinusoids given by [15,8]s 1½n ¼a Ãsin ðx c =x F n Þ¼a Ãsin ðX c n Þ;s 2½n ¼b Ãcos ðx c =x F n Þ¼b Ãcos ðX c n Þð10Þwhere a ,b are constants,x F the sampling frequency,X c the digital equivalent of the carrier frequency x c ,and n are the time index.The digital sine–cosine generators obtained when b =a tan (x c /2)and a =-b tan (x c /2)are expressed,respectively,as [8,15]:s 1½n þ1 s 2½n þ1¼cos ðX c Þcos ðX c Þþ1cos ðX c ÞÀ1cos ðX c Þs 1½n s 2½n:ð11ÞThe sinusoidal carrier is multiplied with V in software to get the re-quired carrier amplitude and applied as resolver carrier using a dig-ital-to-analog converter (DAC)of appropriate resolution.The outputs of the resolver can be demodulated by using any of the synchronous demodulation techniques as in [16–18].However,a simple,hardware-efﬁcient way to achieve this is by sampling simultaneously the quadrature outputs of the resolver at the posi-tive peak of the carrier that avoids any extra delay in the demodu-lated sine cosine envelops [8].This results in the synchronous amplitude demodulation of the resolver outputs in (5)as,x 1ðt ;h Þ¼m 1cos ðh Þ;x 2ðt ;h Þ¼m 2sin ðh þ/Þ:ð12ÞThe detection of the carrier peak plays an important role in the accuracy of the measurement and the software approach gives the ﬂexibility to set it appropriately.It also provides the ability to con-trol the number of samples per shaft rotations to improve the envel-op smoothness,change in the carrier amplitude and frequency to improve the estimation accuracy and performance.3.2.Angular position,speed,and magnitude estimationThe estimation of the angular position,speed,and magnitude of the resolver demodulated envelop is performed by an adaptive PLL that uses a modiﬁed phase detector.The block diagram of the adaptive PLL with the modiﬁed phase detector is shown in Fig.2.Unlike conventional and Costas PLL [19,5],it can be observed fromFig.2that the phase detector of this PLL uses the difference of the reference and generated signal to produce the phase error signal.Let us considering the input signal to this PLL to be a sinusoid,x 1ðt Þ¼m 1cos ðh ðt ÞÞ;ð13Þwhere the phase is related to the frequency by h (t )=x Ãt +a .The magnitude m 1,the frequency x as well as the initial phase h (0)=a are unknown parameters.The adaptive PLL produces an estimate of the input sinusoid as,^x 1ðt Þ¼^m1cos ð^h ðt ÞÞ;ð14Þwith ^h ¼^xt þ^a .The equations describing this adaptive PLL of Fig.2become:_^m 1ðt Þ¼2g m ½x 1ðt ÞÀ^x 1ðt Þ cos ð^h ðt ÞÞ;_^x ðt Þ¼À2g x ½x 1ðt ÞÀ^x 1ðt Þ sin ð^h ðt ÞÞ;^h ðt Þ¼K p ^xðt ÞþZ t^x ðs Þd s ;ð15Þwhere K p ,g x ,g m are design parameters.At convergence,the adap-tive PLL provides the estimate of the frequency and the amplitudewith their nominal values ^x Ã¼x ;^m Ã1¼m 1.Also,the third param-eter namely the initial phase is indirectly determined through thetime history of ^xðt Þ,when the signal ^x 1ðt Þis equal to x 1(t )with the parameters reaching their nominal value.The proposed adaptive PLL is based on the gradient descent algorithm that minimizes the square of the error e ðt Þ¼x 1ðt ÞÀ^x 1ðt Þ.The phase error information for the proposed adaptive PLL can be related with conventional multiplier phase detector from the expression [17],x x ¼e ðt ÞÂsin ð^h ðt ÞÞ%m 12sin ð^h ðt ÞÀh ðt ÞÞ:ð16ÞHowever,unlike conventional multiplier phase detector that incor-porates an additional low pass ﬁlter,the high frequency compo-nents can be considered here to cancel each other without any additional ﬁlter due to the low-pass property of the whole loop and the incorporation external magnitude loop.Proceeding similar to (16),we have magnitude error signal as,x m ¼e ðt ÞÂcos ð^h ðt ÞÞ%1½m 1cos ðh ðt ÞÀ^h ðt ÞÞÀ^m1ðt Þ :ð17ÞSubstituting (16),(17)in (15)and deﬁning the phase and frequencyestimation error as ~h ðt Þ¼^h ðt ÞÀh ðt Þ;and ~xðt Þ¼^x ðt ÞÀx ðt Þ,we get the error dynamic representation of the adaptive PLL as [17],1306S.Sarma,A.Venkateswaralu /Mechatronics 19(2009)1303–1312_^m1¼g mðm1cosð~hÞÀ^m1Þ; _~h¼~xÀgxm1K p sinð~hÞ;_~x¼Àgx m1sinð~hÞÀa;ð18Þwhere a=d x/dt.The stability and convergence of the proposed adaptive PLL has been studied in[17]by deﬁne a candidate Lyapunov function as,Vð^m1;~x;~hÞ¼^m21=2þ~x2=2þg x m1ð1Àcosð~hÞÞ;ð19Þwhich is positive deﬁnite when g x m1>0for all h.One can obtain the stability conditions of(15)using Lyapunov’s direct method such that the derivative of the candidate Lyapunov function given by, _Vð^m1;~x;~hÞ¼^m1_^m1þ~x_~xþðg x m1sin~hÞ_~h¼Àg m^m1ð^m1Àm1cos~hÞÀa xÀg2x m21^m21K p sin2~h:ð20Þis negative deﬁnite.The negative deﬁniteness of(20)i.e. _Vð^m1;~x;~hÞ<0can be satisﬁed for the following conditions:g m ^m1>0;^m1!m1;K p>0;a~x!0;8~h:ð21ÞThe condition in(21)results in asymptotic stability of(18)such that the magnitude,phase,and frequency parameters converge to the equilibrium points,^mÃ1¼m1;^xÃ¼x;^hÃ¼hÆ2n p8n¼0;Æ1;Æ2;::ð22Þfor a=0.The analysis of the system described by(18)can easily be carried out using phase plane plots as the phase and frequency dynamics are decoupled from the magnitude estimate.The phase plane portraits of the error dynamics of the phase and frequency estimate shows spiral sinks,at(±2n p,0),where n is0,±1,±2...[17].In other words,for any phase error within the capture range of the APLL,the error in frequency and phase converges to zero when n=0.Further,an improved version of the proposed adaptive PLL that provides faster convergence and wide frequency acquisi-tion range can also be obtained by adapting the gain g x with the magnitude estimate^m l[21].In that scenario,the gain g x is divided by^m l,for which(18)becomes:_^m1¼g mðm1cosð~hÞÀ^m1Þ;_~h¼~xÀðgx=^m1Þm1K p sinð~hÞ;_~x¼Àðgx =^m1Þm1sinð~hÞÀa:ð23ÞThe stability and the convergence of(23)can also be shown using a similar candidate Lyapunov function as discussed in[17].It is also important to study the effect of design parameters on the perfor-mance of the proposed scheme to get useful insights.We consider the characteristic equation of linearized system of(18)as, s2+g x m1K p s+g x m1=0,to approximately relate the convergence time with the design parameters g x,g m,and K p.The system in(18)is stable wheng x m1K p>0;g x m1>0;ð24Þand its settling time is given byT APLL/1=g x m1K p:ð25ÞThe conditions in(24)is also are in agreement with the conditions in(21).In[20,21]the convergence time of the frequency estimation loop has been related with the design parameters for two limiting cases as:T x%~xð0Þ2=ðg x m1Þ2K p when g x m1K p(~xð0Þ%K P ln½~xð0ÞÀln~x1 when g x m1K p)~xð0Þð26Þwhere the frequency estimator error~x1is a small value deﬁned when convergence is achieved.It can be inferred from(26)thatthe convergence time decreases with decreasing value of~xð0Þand increasing values of g x,m1,and K p when g x m1K p very smallcompared to~xð0Þand does not depend on the values of g x when g x m1K p very large compared to~xð0Þ.When g x m1K p is close to ~xð0Þ,the convergence time may be determined by the combination of the above two limiting cases.The performance of the scheme also depends on the noise rejec-tion capability of the adaptive PLL.The noise rejection capability ofthe frequency estimation loop can be understood from the varianceof the frequency estimate given by[20,21]r2APLL¼ðg x r21Þ=ðm1K pÞ;ð27Þwhere r21is the variance of the additive Gaussian noise at the input. It is evident from(27)that a large value of K p will provide a better frequency estimate but may result in a large convergence time as it is proportional in(26).3.3.Adaptive quadrature phase detector(AQPD)The estimation of the phase shift due to imperfect quadrature isperformed by the proposed AQPD.The AQPD provides an onlineestimate of the quadrature imperfection error using two synchro-nous demodulators[18].The basic idea of this approach can beunderstood from the expressionx2ðtÞ¼m2cosðhðtÞÆ/ðtÞÞ¼m2½cosð/ðtÞÞcosðhðtÞÞÇsinð/ðtÞÞsinðhðtÞÞ :ð28ÞThis expression can be interpreted as summation of two amplitude modulated signal with carrier cos(h(t))and sin(h(t))for theﬁrst and second part,ing two synchronous demodulator [18]with appropriate carrier as in Fig.3,an estimate of the message signals m2cos(u(t))and m2sin(u(t))can be derived.The sine and cosine carriers generated by the adaptive PLL are used in the adap-tive quadrature phase detector to derive^m2cosð^/ðtÞÞ,^m2sinð^/ðtÞÞusing the delayed signal x2(t).The estimates of two parts of(28)ob-tained from the two demodulators are added to produce the esti-mate of input signal^x2ðtÞ.The error d(t)between this generated signal and the input signal is minimized to achieve converge.It can be shown that when^x2ðtÞconverges to x2(t),the AQPD produces a phasor representation of the magnitude m2.The phasor compo-nent of the delayed signal magnitude is multiplied with the respec-tive carriers to reconstruct the input signal.The expression for the reconstructed signal and the difference signal are given by:^x2ðtÞ¼^mðtÞsinðhðtÞþ^/ðtÞÞ;¼^m QðtÞcosðhðtÞÞþ^m IðtÞsinðhðtÞÞ;ð29ÞdðtÞ¼x2ðtÞÀ^xðtÞ¼~m I cosðhðtÞÞþ~m Q sinðhðtÞÞ;ð30Þwhere~mI¼m IÀ^m I;~m Q¼m QÀ^m Q:ð31Þ1()e t±××sin(())tθ2()x t()d t××ˆ()x t cos(())tθˆQmˆIm××+( )2()e t2ˆ()m t2ˆ()x t2ˆˆcos()mφ2ˆˆsin()mφQg2Ig2Fig.3.Adaptive quadrature phase detector.S.Sarma,A.Venkateswaralu/Mechatronics19(2009)1303–13121307。

DSP外文资料138

(62%),but the remaining38%of digestion occurred in the course of the subse-quent intermittent encounters.The complex formation was maintained after the DNA break.This is suggestive of an intrinsic behaviour in which the DNA and protein molecules are continuously held together by switching their binding po-sitions with short-range interactions.It also provides a clue to understanding an efﬁcient and effective reaction by the DNA-binding proteins in bacterial cells. Membrane Physical Chemistry III3546-Pos Board B274Rapid Determination of Geometry and Elastic Constants of Lipid Nano-tubesPavel Bashkirov1,Anna Shnyrova2,Ksenia Chekashkina1,Eva Rodriguez Hortelano2,Petr Kuzmin1,Vadim Frolov2,3.1Laboratory of Bioelectrochemistry,Frumkin Institute of Physical Chemistry and Electrochemistry of RAS,Moscow,Russian Federation,2Biophysics Unit(CSIC,UPV/EHU)and Department of Biochemistry and Molecular Biology,University of the Basque Country,Leioa,Spain,3IKERBASQUE, Basque Foundation for Science,Bilbao,Spain.Membrane nanotubes(NT)are cylinders made of a lipid bilayer.They are widely used to study the effects of curvature on the mechanics of lipid bilayers and proteo-lipid interactions.Fast and accurate determination of the size (length and radius)and the elastic parameters(membrane tension and bending rigidity)of the tube is imperative to quantify its interaction with proteins.Here we describe how measurements of electrical conductivity of the NT interior un-der different voltage protocols allows for simultaneous real time measurements of its geometry and mechanical properties.We present the correlative analysis of conductance andﬂuorescence microscopy measurements performed on a single microns-long nanotube.For submicron nanotubes we discuss the preci-sion of the method(which approaches0.5nm for the measurements of the NT radius)and demonstrate that bending moduli obtained here for the nanocon-ﬁned NTs of different lipid compositions correspond to those obtained from the bulk measurements.3547-Pos Board B275Interaction of Digitonin and Cholate with Complex MembranesHelen Y.Fan,Dar’ya S.Redka,Heiko Heerklotz.University of Toronto,Toronto,ON,Canada.The non-ionic detergent digitonin,extracted from purple foxglove,is widely used in the solubilization and reconstitution of membrane proteins.Its physical-chemical properties are not well characterized,however,and literature on its aggregation behavior is limited.In contrast,the bile salt sodium cholate which also is commonly used in the solubilization of membrane proteins,has been studied extensively.In order to understand the role of digitonin-lipid in-teractions in the reconstitution of G protein-coupled receptors,we have studied the interactions between a particular digitonin-cholate mixture and a mixed membrane composed of phosphatidylcholine,phosphatidylserine and choles-terol.Isothermal titration calorimetry was used,along with time-resolvedﬂuo-rescence leakage assays and light scattering,to study the self-assembly of the mixed-surfactant system and its interactions with lipid membranes.The mech-anism by which the digitonin-cholate surfactant mixture aids in the reconstitu-tion of membrane proteins will be discussed.The insights gained from this work will facilitate the selection of detergents in future studies on the solubili-zation of membrane proteins.3548-Pos Board B276Membrane Leakage and Antimicrobial Action of Polymers and Surfac-tantsSara G.Hovakeemian1,Runhui Liu2,Samuel H.Gellman2,Heiko Heerklotz1.1Pharmaceutical Sciences,University of Toronto,Toronto,ON,Canada,2Chemistry,University of Wisconsin,Madison,WI,USA.Surfactants and nylon-type polymers have been found to induce leakage of lipid membranes as detected by dye efﬂux from liposomes.They also show inhibi-tory or cytotoxic activity against living cells.We aim at better understanding the quantitative correlation between these two activities.We hypothesize that this correlation depends crucially on the mechanism of membrane leakage. The graded mechanism involves partial efﬂux of entrapped dye from all vesi-cles,suggesting that it is based on frequent yet very small and short-lived de-fects distributed over all liposomes.Partial efﬂux by the"all-or-none" mechanism means that some liposomes leaked out all entrapped dye whereas others remained fully intact.This scenario implies the existence of distinct pores that develop in some of the liposomes and remain there during the incu-bation time of the experiment.The experiments are done using theﬂuorescence lifetime-based leakage assay.3549-Pos Board B277Inﬂuence of Cholesterol Microstructures on Fluctuation Spikes in Nystatin Channel Currents in Phospholipid/Cholesterol BilayersCarl S.Helrich,Dennis R.Chavez.Physics,Goshen College,Goshen,IN,USA.Nystatin(NYS)is an antifungal agent that preferentially forms ion channels in membranes containing ergosterol(ERG)or cholesterol(CHOL).In phospho-lipid bilayers ERG or CHOL segregate into ordered(L o)and disordered(L d) domains.We prepared POPC bilayers containing CHOL mol fraction0.18 %c ERG%0.50separating two chambers containing KCl solutions.The con-centration gradient between chambers was435/150mM KCl.We allowed theCHOL/POPC bilayer to settle15min then added19or38microM NYS to the reference(Cis)chamber,imposed50mV across the bilayer,and stirred the Cis chamber at4Hz.The NYS molecules adhered to the bilayer forming channels on the boundaries of the L o domains.[1]The resultant bilayer current exhibited prominent spikes of very short duration.A plot of the frequency of these spikes against c ERG revealed prominent spikes at c ERG=0.19,0.25and0.40.These correspond well to the dips in dehydroergosterolﬂuorescence observed by Chong[2],which he understood in terms of a superlattice.We conclude that ﬂuctuations in NYS channels on the perimeter of the L o CHOL domains depend strongly on L o domain structure.1.Helrich,C.S.,J.A.Schmucker,and D.J.Woodbury2006.Evidence that Nystatin Channels Form at the Boundaries,Not the Interiors of Lipid Domains. Biophys.J.91:1116-1127.2.Chong,P.L-G.1994.Evidence for regular distribution of sterols in liquid crystalline phosphatidylcholine bilayers,A 91:10069-10073.3550-Pos Board B278Optimizing Drug Release:Bilayer to Inverted Hexagonal Phase Transition of Cationic XTC2and Anionic DSPS Lipid System is Inﬂuenced by pH, Temperature,and Salt ConcentrationSiyun(Linda)Wang1,Ismail M.Hafez2,Jason Wang1,Mo Ashtari3,D.Peter Tieleman3,Pieter R.Cullis2,Jenifer L.Thewalt1.1Simon Fraser University,Burnaby,BC,Canada,2University of British Columbia,Vancouver,BC,Canada,3University of Calgary,Calgary,AB, Canada.The effectiveness of a macromolecular drug delivered in lipid nanoparticles(LNP)depends upon the biophysical properties of the delivery vehicle.Recent research has shown that the design of the cationic lipid component of LNPs im-proves the intracellular delivery of therapeutic siRNA[1].Even in these opti-mized LNPs,only1%of the siRNA taken up by the cell via endocytosis is actually released into the cell cytosol.We proposed a mechanism of endosome disruption that relies on the formation of non-bilayer phases in the presence of anionic endosomal lipid and synthetic cationic lipids.A model system using prototypical anionic lipid1,2-distearoyl(d70)-sn-glycero-3-[phospho-L-serine] (DSPS-d70)in1:1molar ratio to the cationic lipid DLin-KC2-DMA(XTC2) (pKa~6.7)was characterized by2H and31P NMR spectroscopy.Through spec-tral analysis,we determined that at physiological pH(~7.4)the XTC2/DSPS system exhibits a stable gel phase for temperatures below45o C while an isotropic signal emerges at higher temperatures-no inverted hexagonal(H II) phase is observed.At low pH(~4.75),the XTC2/DSPS system is principally in a bilayer gel phase at low temperatures with a non-bilayer H II phase predom-inating at higher temperatures.The transition from gel to H II phase is dependent on salt concentration and is most evident in the range of15-25o C for0.25M [Naþ],20-30o C for0.5M[Naþ]and35-45o C for1M[Naþ].Through depaking the spectra,order parameter proﬁles S CD have been obtained and compared for DSPS-d70chains in bilayer and H II phases.These will be useful for computa-tional simulation and eventually to design in vivo animal model experiments.[1]Semple,S.C.,et al.,Rational design of cationic lipids for siRNA delivery. Nat Biotechnol,2010.28(2):p.172-6.3551-Pos Board B279Modelling of the Interaction between Cationic Lipid Dlin-Kc2-Dma (XTC2)and Anionic Lipid Distearoylphosphatidylserine(DSPS) Mohammad Ashtari1,D.Peter Tieleman1,Linda Wang2,Jenifer Thewalt2, Peter R.Cullis3.1University of Calgary,Calgary,AB,Canada,2University of Simon Fraser, Vancouver,BC,Canada,3University of British Columbia,Vancouver,BC, Canada.Short pieces of double stranded small interfering RNA(siRNA)could be used as a potential drug to cure cancer by binding to cancer gene’s messenger RNA. However,the siRNA is not stable in the blood stream,and tends not to penetrate target cell membrane.So,it must be encapsulated in a lipid nanoparticle.LNPs have been shown to be strong candidates for drug delivery[1].Lipid700a Wednesday,February19,2014。

DSP外文资料135

Ž.European Journal of Pharmacology3211997149–155Potentiation by DSP-4of EEG slowing and memory impairment in basalforebrain-lesioned ratsKohji Abe),Masahito Horiuchi,Kohji YoshimuraDepartment of Pharmacology,DeÕelopment Research Laboratories,Shionogi and Co.Ltd.,Toyonaka,Osaka561,JapanReceived20May1996;revised26November1996;accepted29November1996AbstractŽ.The effects of cholinergic and noradrenergic depletion,alone and in combination,on spatial memory and electroencephalogram EEG activity were investigated.Basal forebrain-lesioned rats exhibited a significant decrease in cortical choline acetyltransferase activity and spatial memory impairment.In the cortical EEG,the basal forebrain lesion induced EEG slowing such as an increase in delta powerŽŽ.activity and a decrease in beta power activity.Noradrenergic depletion following a treatment with DSP-4N-2-chloroethyl-N-ethyl-2-.bromobenzylamine had no effect on cortical choline acetyltransferase activity and spatial memory,but it aggravated the cognitive impairment induced by the basal forebrain lesion.DSP-4itself increased delta power activity in non-lesioned rats,whereas DSP-4 potentiated the EEG slowing induced by the basal forebrain lesions.Systemic administration of tetrahydroaminoacridine at1or3mg r kg, i.p.,ameliorated the memory deficits and EEG slowing induced by the basal forebrain lesion.However,the drug could not attenuate the EEG slowing and memory impairment in rats that had received a combination of DSP-4and basal forebrain lesion.These results suggest that noradrenergic depletion aggravated the EEG slowing and the spatial memory impairment induced by cholinergic dysfunction and may decrease the efficacy of an anticholinesterase agent in reversing the cortical cholinergic hypofunction.ŽŽ..Ž.Keywords:DSP-4N-2-chloroethyl-N-ethyl-2-bromobenzylamine;Noradrenaline;EEG electroencephalogram;Memory;Basal forebrain lesion; Tetrahydroaminoacridine1.IntroductionIt has been proposed that the cholinergic system plays a pivotal role in learning and memory,and in the regulationŽ.Žof the cortical electroencephalogram EEG Deutsch,.1971;Bartus et al.,1985;Winkler et al.,1995.Neocorti-cal EEG and learning behavior have been shown to depend on the proper functioning of cholinergic neurons in theŽbasal forebrain Ray and Jackson,1991;Riekkinen et al., .1990.The memory impairment and EEG slowing found in patients with Alzheimer’s disease may be attributable to degeneration of the cholinergic neurons in the basal fore-Žbrain Davies and Maloney,1977;Gordon and Sim,1967; Gustafson et al.,1987;Hollander et al.,1986;Coben et al.,.1990;Schreiter-Gasser et al.,1993.In addition to cholin-ergic dysfunction,a variety of other neurotransmitter sys-tems such as noradrenergic and serotonergic systems haveCorresponding author.Shionogi Laboratories,Shiga520-34,Japan.Ž.Ž.Tel.:81-74888-3281;Fax:81-74888-6508.also been shown to be degenerated in patients withŽAlzheimer’s disease Palmer et al.,1987;Rossor and. Iversen,1986;Reinikainen et al.,1990;Whitehouse,1985.In particular,the noradrenergic system has been shown to be deficient in Alzheimer’s disease patients,as evi-denced by cell loss in the locus coeruleus,and by reduced noradrenaline content and dopamine-b-hydroxylase activ-Ž.ity Bondareff et al.,1981;Cross et al.,1981.The nora-drenergic systems also make a significant contribution toŽthe regulation of the neocortical EEG Buzsaki et al.,.1988b;Morgan and Pfeil,1979.Furthermore,noradrener-gic systems have been reported to influence spatial mem-ory function although the mechanisms remain to be deter-Ž. mined Decker and McGaugh,1989;Spangler et al.,1990.These previous reports suggest the EEG slowing seen in Alzheimer’s disease patients may be due to concurrent aggravation of both the cholinergic and the noradrenergicŽ. innervation of the forebrain Soininen et al.,1992.Thus, the cortical cholinergic innervation seems to play an im-portant role in mediating memory and EEG activity,but the relationship is not likely to be that simple,with non-0014-2999r97r$17.00Copyright q1997Elsevier Science B.V.All rights reserved.Ž.PII S0014-29999600934-X()K.Abe et al.r European Journal of Pharmacology3211997149–155 150cholinergic neurons potentially contributing to the EEG and memory function.In this study,we investigated the effects of combining a basal forebrain lesion with chronic noradrenergic depletionŽŽinduced by the systemic administration of DSP-4N-2-..chloroethyl-N-ethyl-2-bromobenzylamine,a noradrener-gic neurotoxin,in order to compare the disruptive effects of basal forebrain lesion on memory and EEG with the effects of noradrenergic depletion.Furthermore,we exam-ined the effects of tetrahydroaminoacridine,an anti-cholinesterase agent,on memory impairment and EEG slowing induced by the combination of DSP-4and basal forebrain lesion.2.Materials and methods2.1.AnimalsTen-week-old male rats of the Wistar strain were ob-tained from Japan SLC.The initial free-feeding weights were280–300g.The rats were housed in groups of threeŽ.per cage at a constant temperature23"28C and main-tained under a12-h ligh r dark cycle with lights on at7:00 a.m.Experiments were conducted during the light portion of the day.2.2.Surgery and lesioningŽRats were anesthetized with sodium pentobarbital40 .mg r kg,i.p.and fixed on a stereotaxic apparatus with the incisor bar set3.3mm below the interaural line.In all surgical procedures,the stereotaxic coordinates were ac-Ž. cording to the atlas of Paxinos and Watson1986.Each rat was surgically implanted with EEG electrodes.BipolarŽ. stainless steel electrodes0.1mm in diameter with an interpolar distance of0.5mm were implanted in the left Žamygdala 2.1mm posterior to the bregma and4mm.Žlateral to midline and hippocampus 3.8mm posterior to.bregma and 1.8mm lateral to midline.Bipolar silver Ž.electrodes1mm were placed1mm apart on the dura mater in the parietal cortex.An integrated circuit socket was attached to the skull with dental cement,and insulated leads were routed from this plug to the electrodes.EEG electrodes connected to female pins were fixed on the skullŽ.with dental cement Polyset;Yata Chemical.The rats were allowed to recover for a minimum of one week before being connected to a flexible tether and slip-ring, and accustomed to the soundproof recording chambers.AŽstainless steel guide cannula length s40mm;outer diam-.eter s0.5mm was stereotaxically placed2mm above the basal forebrain region to allow the infusion of ibotenic acid.The cannula was kept in position by dental cement. Bilateral neurotoxic lesions of the basal forebrain wereŽ. achieved by the injection of ibotenic acid Sigma through the cannula on each side with an interval of2days.AŽ. stainless steel injection needle outer diameter0.25mm connected via polyethylene tubing to a10-m l micro-syringe was inserted into the basal forebrain.The coordinates of the basal forebrain were:2.6mm posterior to the bregma, 2.8mm lateral to the midline and8mm ventral from the skull surface.Ibotenic acid was dissolved in0.1M phos-Ž.phate buffer pH7.2at a concentration of10m g r m l,and then infused in a volume of1m l over a period of3min. The injection needles were left in place for a further5min to ensure that the drug had diffused away from the needle tip.Sham-operated rats received0.1M phosphate buffer in the same way.At the end of the experiments,the place-ment and size of the lesions were checked.2.3.Water maze taskA circular pool178cm in diameter was used.The tank was filled with water at approximately238C.A transparent platform,10cm in diameter,was located at a constant position in the center of one quadrant inside the tank,the upper surface of which was3cm below the surface of the water.During each trial,we recorded the time it took ratsŽ.to escape onto the platform within300s latency.If the rat failed to find the platform within300s,it was placed on it for30s,and the training session was terminated and a maximum score of300s was assigned.The behavioral trace,the speed and the distance of swimming in the water maze were monitored with a TV video camera and ana-Žlyzed using a computer system AXIS60,Neuroscience, .Tokyo,Japan.The latency to locate the platform was recorded in this experiment.Behavioral tests were per-formed in rats21days after the basal forebrain lesion. Each rat was tested with one trial r day for5days.2.4.EEG recordingsThe EEG values of freely moving rats were wire-relay recorded between9:00and16:00h using a polygraph Ž.system Nihon Kohden.The animals were allowed to adapt for at least48h before the test began.The EEG was recorded before and21–28days after basal forebrain lesioning.The behavioral states were selected waking-im-mobility for EEG spectral components monitored with a video camera.On each day of the experiment,the EEGŽwas recorded with an analog tape recorder XR-310, .TEAC for each rat.Spectral analysis of the EEG wasŽperformed by spectral analysis software DSS98SV,Kano-.pus.Six consecutive10-s artifact-free epochs were se-lected from each tape-recorded EEG sample.The samples of EEG were filtered to pass frequencies below50Hz withŽ.an analog filter ASIP-0260;Kanopus,and digitized at aŽsampling rate of500Hz,using an AD converter ADX98E;.Ž. Kanopus connected to the computer PC-H98,NEC.The digitized EEG was converted to power spectral quantitiesŽby fast Fourier transformation.Relative power power in.frequency band r total power was calculated for the delta()K.Abe et al.r European Journal of Pharmacology3211997149–155151s1–3.9Hz,theta s4–6.9Hz,alpha s7–12.9Hz,and beta s13–25Hz bandwidths.2.5.Choline acetyltransferase actiÕityRats were decapitated21days after basal forebrain lesion,and the brain was removed rapidly and dissected on ice into the frontal cortex,parietal cortex,occipital cortex and the hippocampus.The tissues were stored at y808C until choline acetyltransferase assay.Choline acetyltrans-ferase activity was measured according to the radioenzy-Ž.w14xmatic method of Fonnum1975,using C acetyl coen-Ž.zyme A Amersham,and expressed as nmol acetyl-Žcholine r h r mg protein.Protein determination Bradford, .1976was performed on all samples using a commercially Ž.available kit Bio-Rad,Richmond,CA,USA.2.6.Noradrenaline leÕelOn the14th day after the administration of DSP-4,the rats were decapitated and their brains were removed rapidly and dissected on ice into the frontal cortex,parietal cortex, occipital cortex and the hippocampus.Each tissue sample was examined with high-performance liquid chromatogra-Ž.Žphy HPLC with an electrochemical detector Eicom, .Kyoto,Japan for analysis of monoamine levels.TheŽ. working electrode was a WE-3G graphite electrode Eicom set at a detector potential of q0.65V against an Ag r AgCl reference electrode.Monoamines and metabolites wereŽseparated on an Eicompack MA-5ODS column 4.6=250 .mm.The mobile phase consisted of Na HPO containing240.1mM EDTA,0.5mM sodium1-octanesulfonate,andŽ.20%v r v methanol at pH3.8.The flow rate was set at 1.0ml r min.Samples were homogenized on ice in a0.2M HClO r104M EDTA solution containing a fixed amount of the inter-nal standard isoproterenol.The samples were then cen-Ž.trifuged at48C for10min12000=g,and20m l of the supernatant was injected into the HPLC system.Nora-drenaline levels are shown as ng r mg protein.2.7.DrugsŽŽ.DSP-4N-2-chloroethyl-N-ethyl-2-bromobenzyla-.Žmine;Sigma and THA tetrahydroaminoacridine hydro-.chloride;Nakarai Tesque,Kyoto,Japan were dissolved in saline.Both drugs were administered intraperitoneally in a volume of1ml r kg.Rats were treated with tetrahydro-aminoacridine30min before each training trial in the water maze.In EEG recordings,this drug was injected24 h after the last behavioral test.2.8.StatisticsData for the water maze task are expressed as medians and were analyzed using the Kruskal-Wallis test,followed by Mann-Whitney’s U-test.Biochemical data are ex-pressed as means"S.E.and were analyzed using a one-Ž.way analysis of variance ANOVA and the Dunnett test. Data for EEG were analyzed by Tukey’s test.3.Results3.1.Water maze testingThe basal forebrain lesion produced ataxia for3–20 days after surgery.These behavioral changes were amelio-rated after10days.On the first day of testing,the escape latency in the sham-operated group did not differ signifi-cantly from that in the basal forebrain-lesioned group. However,the escape latencies in the basal forebrain-le-sioned group shortened gradually with repeated testing, while those in the sham-operated group shortened rapidly. The swimming speed during the five trials did not differ among groups.The basal forebrain-lesioned rats showed a significant increase in escape latency on the3rd,4th and 5th trials of test sessions when compared to the sham-oper-Ž.ated group Fig.1.DSP-4at50mg r kg,given i.p.14 days before testing,had no significant effect on the latency of water maze performance in the normal rats,and the swimming speed during the five trial blocks did not differ among experimental groups.However,the basal forebrain-lesioned rats treated with DSP-4at50mg r kg had significantly increased escape latencies on the3rd,4th and5th trials compared with the basal forebrain q vehicle-treated group,indicating that DSP-4potentiated the mem-Ž. ory impairment in basal forebrain-lesioned rats Fig.1.Tetrahydroaminoacridine at1and3mg r kg dose depen-dently improved the memory disruption induced by the basal forebrain lesion.Tetrahydroaminoacridine at1 mg r kg significantly decreased the escape latencies in the 4th trials,while3mg r kg produced a significantdecreaseFig. 1.Effects of DSP-4plus basal forebrain lesion on acquisition performance in the water maze task.The values are expressed as medi-ans.n s7–8per group.BF s basal forebrain-lesioned group.†P-0.05,††P-0.01vs.Sham q Vehicle,)P-0.05,))P-0.01vs.BF qa aaŽVehicle,P-0.05,P-0.01vs.Sham q DSP-4Mann-Whitney’s .U-test.()K.Abe et al.r European Journal of Pharmacology 3211997149–155152Ž.Fig.2.Effects of tetrahydroaminoacridine THA on memory impairment induced by the basal forebrain lesion,as assessed in the water maze.The Ž.values are expressed as medians.THA 1or 3mg r kg was injected i.p.30min before the training trials.n s 7–8per group.†P -0.05vs.))Ž.BF q Vehicle,P -0.01vs.Sham q Vehicle Mann-Whitney’s U -test .in the escape latencies in the 4th and 5th trials when compared to those of the vehicle-treated basal forebrain-le-Ž.sioned group Fig.2.However,tetrahydroaminoacridine at 1or 3mg r kg had no effect on the memory impairment induced by the combination of DSP-4and basal forebrain Ž.lesion Fig.3.No behavioral change was measured in the sham rats after injection of tetrahydroaminoacridine at 1Ž.and 3mg r kg data not shown .3.2.EEG recordingThe basal forebrain-lesioned rats showed a significant increase in delta power activity and a decrease in beta power activity,but no significant effects on alpha or theta activity.No recovery of this EEG slowing occurredduringŽ.Fig.3.Effects of tetrahydroaminoacridine THA on memory impairment induced by the combination of DSP-4and basal forebrain lesion,as Žassessed in the water maze.The values are expressed as medians.THA 1.or 3mg r kg was injected i.p.30min before the training trials.n s 7–8)))Žper group.P -0.05,P -0.01vs.Sham q Vehicle Mann-Whitney’s .U -test.Fig.4.Effects of DSP-4plus basal forebrain lesion on cortical EEG Ž.power activity in rats.The values %are expressed as means "S.E.n s 7–8per group.)P -0.05,))P -0.01vs.Sham q Vehicle,†P -0.05,†P -0.01vs.BF q Vehicle,a P -0.05,aa P -0.01vs.Sham q Ž.DSP-4Tukey’s test .the course of the experiment,49days.DSP-4increased only the delta power activity of awake immobility.Fur-thermore,DSP-4significantly potentiated the EEG slow-ing,i.e.,it was followed by an increase in delta power activity and a decrease in beta power activity induced by Ž.the basal forebrain lesion Fig.4.Tetrahydroaminoacri-dine at 1mg r kg failed to improve these EEG changes in basal forebrain-lesioned rats compared to their post-lesion baseline values.In contrast,tetrahydroaminoacridine at 3mg r kg significantly normalized the increase in delta power activity and decrease in beta power activity in basal fore-Ž.brain-lesioned rats Fig.5.However,tetrahydroaminoacri-dine at 1and 3mg r kg had no effect on the cortical EEG slowing induced by the combination of DSP-4and basal Ž.forebrain lesion Fig.6.3.3.Choline acetyltransferase acti ÕityThe basal forebrain lesion caused significant reductions in choline acetyltransferase activity in the cortex but not in the hippocampus or striatum,as shown in Table 1.LesionsŽ.Fig.5.Effects of tetrahydroaminoacridine THA on cortical EEG slow-Ž.ing induced by the basal forebrain lesion in rats.The values %are expressed as means "S.E.n s 7–8per group.)P -0.05,))P -0.01††Žvs.Sham q Vehicle,P -0.05,P -0.01vs.BF q Vehicle Tukey’s .test .()K.Abe et al.r European Journal of Pharmacology3211997149–155153Ž.Fig.6.Effects of tetrahydroaminoacridine THA on cortical EEG slow-ing induced by the combination of DSP-4and basal forebrain lesion in Ž.rats.The values%are expressed as means"S.E.n s7–8per group. ))Ž.P-0.01vs.Sham q Vehicle Tukey’s test.of the basal forebrain decreased the choline acetyltrans-ferase activity by an average of30%in the frontal cortex, 30%in the parietal cortex and19%in the occipital cortex,Ž.as compared to sham-operated animals Table1.DSP-4 treatment did not affect the choline acetyltransferase activ-ity in the non-lesioned rats and did not potentiate the decrease in choline acetyltransferase activity in the basal forebrain-lesioned rats.3.4.Noradrenaline concentrationNoradrenaline levels are shown in Table 2.Nora-drenaline content was decreased following basal forebrain lesion only in the occipital cortex.The intraperitoneal injection of DSP-4at50mg r kg significantly reduced noradrenaline levels in all areas sampled.The parietal cortex noradrenaline levels dropped from0.39ng to0.23Ž.ng r mg tissue56%reduction,and levels in the hip-pocampal tissue dropped from0.51ng r mg tissue to0.18Ž.ng r mg tissue65%reduction,compared to sham-oper-ated animals.The rats that received a combination of DSP-4and basal forebrain lesion showed a decrease in noradrenaline similar to that seen in the DSP-4-treated group.4.DiscussionIbotenic acid infusion into the basal forebrain produced a marked reduction in cortical choline acetyltransferase activity.We also confirmed that basal forebrain lesioning induced a memory deficit in the rats,using the water maze task.Further,lesioning the basal forebrain increased corti-cal delta wave and decreased beta wave activity.This EEG slowing induced by the basal forebrain lesion did not recover spontaneously with the present experimental schedule.The cholinergic deficit induced by basal fore-brain lesioning has been previously reported to produce aŽmemory impairment and EEG slowing Ray and Jackson,.1991;Riekkinen et al.,1990.These findings and those of the present study suggest the importance of cholinergic neuron loss in the memory deficit and EEG slowing induced by basal forebrain lesion.The present study indicated that noradrenaline deple-tion,achieved with DSP-4,aggravated the spatial memory deficits induced by basal forebrain lesion.DSP-4is widelyŽused as a selective noradrenergic neurotoxin Jonson et al., .1981.Systemic administration of50mg r kg DSP-4has been reported to produce a rapid,marked and selective reduction of noradrenaline levels in various brain regionsŽand in peripheral neurons Chrobak et al.,1985;Fritschy et .al.,1989.In the present study,DSP-4at50mg r kg had no significant effect on spatial memory in the water mazeTable1Ž.Brain regional choline acetyltransferase ChAT activities induced by the combination of DSP-4and basal forebrain lesionBrain region Sham q Vehicle Sham q DSP-4BF q Vehicle BF q DSP-4a,b a,bFrontal cortex48.9"1.247.8"2.134.5"2.535.5"1.8a,b a,bParietal cortex50.5"2.347.3"3.335.3"0.832.6"2.1a,b a,bOccipital cortex36.0"1.639.4"1.529.2"1.429.6"0.5Hippocampus40.7"8.447.5"4.349.3"1.245.2"3.5Ž.ChAT activities nmol ACh r h r mg protein were determined2weeks after basal forebrain lesioning.Each value represents the mean"S.E.Eight ratsa bŽ.were used in each group.BF s basal forebrain-lesioned group.P-0.01vs.Sham q Vehicle,P-0.01vs.Sham q DSP-4Dunnett’s test.Table2Brain regional noradrenaline contents induced by the combination of DSP-4and basal forebrain lesionBrain region TreatmentSham q Vehicle Sham q DSP-4BF q Vehicle BF q DSP-4a a,bFrontal cortex0.38"0.010.28"0.030.33"0.010.30"0.02a,b a,bParietal cortex0.39"0.010.24"0.020.37"0.010.23"0.04a,b a a,bOccipital cortex0.37"0.010.15"0.020.31"0.010.17"0.04a,b a,b Hippocampus0.51"0.020.18"0.040.52"0.040.15"0.10Ž.Noradrenaline ng r mg tissue levels were determined2weeks after DSP-4administration.Each value represents the mean"S.E.Eight rats were used ina bŽ.each group.BF s basal forebrain-lesioned group.P-0.01vs.Sham q Vehicle,P-0.01vs.BF q Vehicle Dunnett’s test.()K.Abe et al.r European Journal of Pharmacology3211997149–155 15414days after treatment,when noradrenaline levels de-creased about40%in the parietal cortex.However,the treatment aggravated the memory deficits induced by the basal forebrain lesion.Noradrenergic depletion induced by DSP-4has been reported to fail to alter the effects of scopolamine on the acquisition of reference memory in the Žwater maze Decker and McGaugh,1989;Spangler et al., .1990.These previous data suggest that such interactions between cholinergic and noradrenergic dysfunctions are not involved in reference memory performance.Indeed,Ž.Ohno et al.1993recently demonstrated that noradrener-gic depletion induced by DSP-4potentiated the impair-ment of scopolamine-induced working memory.Thus,these findings suggest that an interaction between cholinergic and noradrenergic dysfunctions is evident only when a working memory procedure is used rather than a reference memory procedure.Our present findings thus indicate that this water maze task may test the working memory proce-dure involved in attention function rather than reference memory procedure,for the task was performed with one trial per day for5days,which is a short period.The firing rates of the neurons in the locus coeruleus have been related to the level of vigilance,and it is possible that the noradrenergic input informs the nucleus basalis cholinergic cells of the appearance of behaviorally important stimuliŽoriginating from the external milieu Aston-Jones and.Bloom,1981;Aston-Jones et al.,1986.Therefore,our data suggest that the central cholinergic,but not the nora-drenergic,system plays an essential role in spatial cogni-tion processes,also indicating that the regulation of mem-ory is at least partially controlled by noradrenergic neu-rons.A most important finding in the present study was that DSP-4increased delta power activity in normal rats,and potentiated the increase in delta power and decrease in beta power activity induced by the basal forebrain lesion.Ž.Riekkinen Jr.et al.1990a,b have demonstrated that thea-adrenoceptor antagonists increase fast wave activity, 2and that an a-adrenoceptor agonist increases slow wave 2activity during periods of immobility and mobility.In the present study,the noradrenergic impairment induced by DSP-4was accompanied by an increase in cortical delta power activity but it had no effect on spatial memory in the water maze task.These data indicate that EEG slow-ing,such as an increase in delta power activity induced by noradrenergic impairment,may not be correlated with spatial cognitive disruption.Neocortical choline acetyl-transferase activity has been reported to correlate with delta power activity induced by nucleus basalis lesion Ž. Riekkinen et al.,1990;Ray and Jackson,1991.Our data indicate that DSP-4has no effect on choline acetyltrans-ferase activity in normal rats and does not potentiate the reduction of choline acetyltransferase activity induced by basal forebrain lesions.Therefore,the increase in delta power activity induced by DSP-4may indirectly affect the cholinergic impairment associated with cognitive function.Interestingly,DSP-4potentiated not only the increase in delta power activity but also the decrease in beta power activity,as it aggravated the spatial cognitive disruption induced by the basal forebrain lesion.Previous findings have underscored the involvement of a cholinergic deficit in age-related EEG slowing and that memory deficits are associated with cholinergic deficits.In addition,in agedŽ. rats a change was observed in the slow wave delta,theta activity,but a large decrease was observed in fast wave Ž.Ž.alpha,beta activity Buzsaki et al.,1988a.Aged rats have also been reported to show a deficit of workingŽ. memory in an8-arm radial maze Blanco et al.,1994. These data suggest that high frequency activity rather than low frequency activity may be more important for the memory and learning involved in working memory.We also examined the effect of tetrahydroaminoacridine on memory and EEG activity in the present study.This drug could reverse the memory deficit and EEG slowing induced by basal forebrain lesion,but it could not amelio-rate the memory deficit or EEG slowing produced by the combination of DSP-4and basal forebrain lesion.Riekki-Ž.nen et al.1991have shown that tetrahydroaminoacridine at7.5mg r kg partially reverses the increases in delta and theta amplitudes induced by nucleus basalis lesions.In addition,tetrahydroaminoacridine could not reverse the decrease in beta amplitude induced by nucleus basalis lesioning.Our present data show that tetrahydroaminoacri-dine at3mg r kg significantly improved the EEG slowing induced by basal forebrain lesion,although it could not attenuate the memory deficit and EEG slowing induced by the combination of DSP-4and basal forebrain lesion.Ž.Hartounian et al.1986also reported that dorsal noradren-ergic bundle lesions decreased the efficacy of physostig-mine in reversing passive avoidance deficits in rats with basal forebrain lesions.These findings taken together sug-gest that the presence of noradrenaline within the neocor-tex is sufficient to ameliorate learning deficits and restore EEG slowing following damage to the basal forebrain. However,the efficacy of tetrahydroaminoacridine in re-versing the cortical cholinergic hypofunction induced by noradrenergic depletion may be limited.Furthermore,these data suggest that cognitive impairment accompanied the EEG slowing induced by cholinergic and noradrenergic deficits.The mechanism and generality throughout the cholinergic system of this interaction cannot be determinedŽ.from the present study.Ray and Cole1985stated that it should be noted that changes in fast EEG activity are associated with cognitive and emotional processes.Further studies are required to clarify the correlation to high frequency activity and cognitive functions.In conclusion,in the present study we observed that noradrenergic deficits aggravate the EEG slowing and the memory impairment induced by cholinergic dysfunction, and we suggest that the noradrenergic deficits may de-crease the efficacy of the anticholinesterase agent tetrahy-droaminoacridine in reversing the cortical cholinergic hy-。

DSP的发展毕业论文外文翻译

英文文献DSPs Back to the Future[13]W. Patrick HaysPublished online：23 June 2004Springer Science + Business Media B.V. 2004AbstractOur complex world is characterized by representation, transmission, and storage of information-and information is mostly processed in digital form. With the advent of DSPs (digital signal processors), engineers are able to implement complex algorithms with relative ease. Today we find DSPs all around us-in cars, digital cameras, MP3 and DVD players, modems, and so forth. Their widespread use and deployment in complex systems has triggered a revolution in DSP architectures, which in turn has enabled engineers to implement algorithms of ever-increasing complexity. A DSP programmer today must be proficient in not only digital signal processing but also computer architecture and software engineering.Keywords DSP；Motor control systemThe Intel 2920 included on-chip (D/A) digital/analog and A/D (analog/digital) converters but lacked a hardware multiplier and soon faded from the market.An NEC project resulted in the NEC µPD7720—one of the most successful DSPs of all time.The Bell Labs DSP-1 and NEC µPD7720 were announced at ISSCC80.DSP-1,achieved 5-MHz clock speed，executing 1.25-MB multiply-accumulates per second at four clock cycles each—enough to allow Touch-Tone receiver filters to execute in realtime.The once formidable performance demands of the Touch-Tone receiver are now ludicrously easy, but new applications in turn arose throughout the last 20 years to put new demands on DSP technology (see figure 1).According to Will Strauss,president and principal analyst at Forward Concepts, “DSP shipments were up a healthy 24 percent in 2003, and we are forecasting a bit higher growth for 2004, at 25 percent. Longer term,we forecast a 22.6 percent compound growth rate through 2007.”So the game has been: Boost DSP performance, run the algorithm atan acceptable cost, and open up a new commercial market.It is perhaps too glib to project this trend indefinitely into the future,In fact, savvy analysts have periodically predicted the demise of the DSP.Will performance requirements outstripthe ability of programmable DSP architectures to keep up, thus demanding a new approach?or if DSPs are to maintain their historical growth curve，what kinds of tools and architectures are needed? Ultimately，these questions will be answered by creative architects，mar-ket competition，and application demands. The goal of this article is to illuminate current and future trends by reviewing how technology and application pressures have shaped DSP architecture inthe past. WHAT IS A DSP？At the outset，it is important to distinguish between digital signal processing and digital signal processors. The techniques and applications of digital signal process-ing, as compared with analog signal processing，are well established and are more important commercially than ever. Throughout this article,DSP refers to the VLSI (very large-scale integration) processor component. Therefore，what special demands in digital signal processing make a DSP different from another programmable processor？In other words,what makes a DSP a DSP?The Realtime Requirement. The essential application characteristic driving DSP architecture is the requirement to process realtime signals.Realtime means that the signal represents physical or “real” events. DSPs are designed to process realtime signals and must therefore be able to process the samples at the rate they are generated and arrive. Adding significant delay,or latency,to the output can be objectionable. While high realtime rates often demand that DSPs be “fast” ,fast and realtime are different concepts.For example，simulations of VLSI designs must be fast—the faster the better but the application doesn’t fail if the simulator completes a little slower. Conversely，a realtime application need not be fast—for example,a hospital room heart monitor doesn’t need to be fas t (30-Hz sample rate) but does need to be realtime; it would be disastrous if the processing of a sample took so long that after a few hours，the monitor was displaying five-minute-old data. Not all digital signal processing applications require realtime processing.Many applications are performed offline. For instance，encoding high-fidelity audio for mastering CD-ROMs uses sophisticated digital signal pro-cessing algorithms，but the work isn’t done in realtime.Consequently，a DSP isn’t required—any old processor fast enough for the engineer to get home for dinner will do. To summarize, the most important distinguish-ing characteristic of DSPs is that they process realtime signals—the signals can be fast or slow, but they must be realtime.Programmability. Do DSPs need to be programmable? No:it’s quite feasible to process digital signals without a programmable architecture.In this article, however, DSP refers to programmable DSP—more specifically，to user-programmable DSPs, because my bias is that that’s where the most interesting architectural issues lie. Often,the most demanding applications have required nonpro-grammable architectures. For instance,first-generation programmable DSPs could execute a single channel of the 32-Kbps ADPCM/DLQ,(adaptive differential pulse code modulation/dynamic locking quantizer) codec,whereas a special custom-integrated circuit that was not program-mable but deeply pipelined could run eight channels in the same technology.The reason for this is that programmability comes at a cost：Every single operation in a programmable chip—no matter how simple—requires fetch-decode-execute. That’s a lot of silicon area and power devoted to,say,shifting left by two bits. Nonprogrammable architectures succeed when the shift-left-by-two-bits function is a small building block，allowing other building blocks to operate simultaneously. It’s easy to imagine many building blocks working simultaneously to achieve a 10x performance advantage in nonprogrammable logic.The problem with specialized DSP hardware is that you have to develop a new chip for each application. As development costs increase，the break-even point isconstantly shifting in favor of using a programmable architecture.More Power. Higher clock speed permits more instructions to be executed during a fixed time interval. In 1980,the Bell Labs team struggled to run DSP-1 at 5 MHz; today in 130-nm technology,clock speeds greater than 500 MHz can be attained. The advantage of more instructions in a fixed time period can be used to achieve one or more of the following.1.At a fixed data rate more complicated algorithms can be programmed.2.At a fixed data rate, more channels of the same algo-rithm can be programmed.3.At a higher data rate,algorithms of similar complexity can be programmed.An example of the first case is G.729A,a CELP (coded-excited linear predictor) speech codec. It allows good quality at low data rates.The algorithm requires about 30 times more computations per sample than G.711 PCM. Examples of number 2 are V oIP (voice over IP) applica-tions where four channels are supported for SoHo (small office/home office) products，and up to 256 or more channels for CO (central office) products. Channel den-sity is the key metric for V oIP processing. An example of the third case is the MPEG-2 video compression algorithm applied to decode DVDs at dif-ferent picture resolutions.The computational power is directly proportional to the video resolution.Stretching MPEG-2 from NTSC (National Television System Commit-tee) resolution to high definition requires not only a six-fold increase in processing power but new blue laser DVD technology for faster readout of the data from the disk.In addition, advancing VLSI permits the program-mable architecture to reduce power and/or cost for a fixed algorithm at a fixed data rate.Advancing technology conspires in many ways to move the boundary in favor of programmable DSPs. Applications that require highly spe-cialized design today become programs for inexpensive DSPs tomorrow;costly power-hungry DSPs today become the jelly beans of tomorrow. The past 25 years has seen the ascendancy of the user-programmable DSP as the dominant architectural approach to implementing digital signal processing applications.DSP architecture is driven by a number of specialized application characteristics. Let’s look at a few of these before returning to the architectureal influence of the all-important realtime constrain. The DSP program must sustain processing at the realtime rate under all circumstances.and.in fact，the programmer must somehow know that this has been accomplished so that the application doesn’t fail in th e field. In other words the DSP program must deterministically allocate realtime. Sources of indeterminacy，common in desktop CPUs，can be catastrophic for the DSP programmer. For example.page faults and cache misses can cause hundreds of cycles to be missed by the CPU as it is idled while the operation is satisfied.If you must sample a value every microsecond, then the page fault or cache miss could cause the window to be missed.As a result, DSPs need either fixed memories or caches that can be locked after the program is booted.Other less critical examples of indeterminacy include branch prediction and data-dependent termination of functions such as “divide” . Although nice for the average case, the DSP program must also allow for the worst case.Deterministic allocation of realtime not only must be achieved，but tradition-ally DSPs have made it straightforward to achieve. In newer DSPs, realtime allocation is indeed know-able at compile time, but very careful profiling and iterative programming are oftenrequired to achieve the desired outcome. ILLUSTRATION: THE TI TMS320C54XXTo bring the discus-sion down to earth, let’s illustrate with a real DSP. Targeted at the cellphone, TI’s TMS320C54xx was introduced in 1994; in a sense, it is the fruition of TI’s 16-bit DSP product line, which started with the introduction of the TMS32010 in 1983 and moved through the ’C1x, ’C2x, ’C2xx, and ’C5x generations to the ’C54xx. Although strict compat-ibility wasn’t maintained, the follow-on architectures were close enough for TI to migrate its growing customer base with each new product generation much as Intel has done with the x86 family. Earlier TI DSPs sacrificed much performance to improve ease of use. Numerous other shortcomings such as the lack of accumulator guard bits were also rectified over the years. Table 1 shows how the TMS320C54xx has addressed each of the DSP features discussed here. For later comparison the TMS320C62xx is also listed. THE 16-BIT DSP RUNS OUT OF GASAnother form of “accumulation” other than the mu ltiply-accumulate arithmetic operation was taking place in the TI DSP product line: by the mid-1990s, the TI architecture had grown to more than 130 instructions. New specialized instructions are one way of improv-ing performance—the way early DSPs used to meet cost goals. It became difficult to pack new instructions into the TMS320C54xx’s burdened instruction set. Clock speed can be increased over time but does not take full advantage of advancing technology if the CISC instruc-tion growth continues. Somehow the DSP architecture needed to find a way to use the extra transistors of later-generation IC technology to increase performance. Deeper pipelining gives little benefit because the deeper pipeline must benefit all critical paths and CISC instruc-tions have many complex critical paths. An alternative strategy, VLIW (very long instruction word) parallelism, boosts performance by executing multiple instructions in parallel. VLIW is relatively ineffective on CISC instruction sets because it’s difficult to id entify instructions that are commonly executed in parallel. It is also important to note that compilers have had little success with complex 16-bit DSP instruction sets.Yet as higher clock speeds and larger local memories per-mit larger programs, the demand for good DSP compilers becomes paramount. Consequently, the 16-bit DSP is out of gas: It’s too complicated to scale performance with Moore’s law and too complicated to support good compi-lation While the ’C54xx was running 160 MHz using full custom 0.15µm circuit design, the StrongARM RISC broke 600 MHz in 0.18µm.Faced with this crisis, in 1997 TI introduced the all-new 32-bit VelociTI instruction set with its TMS320C62xx architecture. The TMS320C62xx has had enormous publicity as an eight-issue VLIW architecture (thus, the real instruction length is 8 x 32 or 256 bits, and it is pos-sible to execute eight 32-bit instructions in parallel on the chip). Less remarked, but equally important, is that each instruction is a relatively simple 32-bit RISC-like instruc-tion. In fact, it’s ironic that RISCs have included multiply-accumulate instructions since the mid-1990s, but TI—the company that has shipped more multiply-accumulates than any vendor—chose to “out-RISC the RISCs” by requiring a multiply followed by a n add instruction to implement the common DSP kernel.I call the new RISC-like DSP instruction sets,”RISC-DSP” . For an illustration of RISC-DSP. let’s return to the FIR filter program. We saw that the instruction count of the inner loop is nine times betterin the DSP case than for the conventional RISC.Keep in mind, though，that clock speeds today are 100 times that of the 1980s when DSPs started down the CISC path. As a result.the RISC can execute the FIR almost 10 times faster than a 1980s DSP but one-tenth the speed of an optimized DSP architecture with the same clock speed-assuming of course that the in 1980 In 1980 the dial needed to be turned all the way to “DSP” in order to meet the minimal performance goals；today the architect can choose different po ints on the spectrum with less performance. At today’s clock speeds，RISC-DSP perfor-mance will be sufficient for many applications and have other advantages as well.Sources and destinations from a general-purpose register file are easily encoded in a 32-bit RISC-DSP instruction，making compilers more successful. Decoupling data loads from execution permits higher clock speed because data can be preloaded into a general-purpose register file. For each special feature, careful study of the potential number of instructions saved,critical path impact, interrupt overhead,and,of course, compila-tion is required.To summarize, 32-bit RISC-DSP instruction sets have moved DSPs onto the historic RISC technology learning curve.APPLICATIONS OF RISC + DSPWe’ve seen that the need for good tools and continued performance scaling have forced DSP architects to break with the complex 16-bit instruction sets of the past.RISC-DSP. However，really comes to fruition in applications combining both “RISC tasks” and”DSP tasks” . Applica-tions of this type are proliferating commensurate with DSP applications on packet networks. An important example is the 3G wireless handset of the near future，with video communications and speech recognition.Table 3 lists the key tasks，classifying them aeither conventional RISC tasks or conventional DSP tasksWe see that a single RISC-DSP at about 200 MHz has suf-ficient performance for all tasks.9 The important advan-tage achievable in this application is that separate RISC and DSP chips—or separate RISC and DSP cores—aren’t required. Significant architectural efficiency is gained because data doesn’t need to be communicated between two different subsystems.This efficiency translates into hardware and performance advantages and therefore reduced cost and power. Because this is a handheld con-sumer device，opportunities to save power and cost are critically important. The key barrier to merging DSP applications and RISCapplications on a common processor is the need for deterministic response time.The RISC processor often supports an operating system complicating the realtime problem. Packetized networks，however, have relaxed the realtime constraint somewhat：The samples arrive in packets; therefore,the realtime response rate is the (somewhat irregular) packet arrival rate. For example,80,000 packets per second are transmitted over 1 GigE (Gigabit Ethernet). MontaVista’s sponsored Linux preemptive kernel guarantees a worst-case kernel preemption latencyunder 1 millisecond on several CPUs.VxWorks guaran-tees an interrupt response of a few microseconds on a 500-MHz Pentium.So DSP applications can often now be run under major real-time operating systems.The Siemens Tricore deserves recognition as one of the first RISC-DSPs:“Tri”signifies that microproces-sor，DSP，and microcontrol functions are combined in a common processor. Intel and Analog Devices have recently collaborated to build the Intel MSA (Micro Signal Architecture). The first produc,the ADSP-21535 (Blackfin)，appears tobe targeted at the 3G cellphone.StarCore and the Philips Trimedia are two additional high-perfor-mance RISC-DSP architectures，each with VLIW ing from the RISC side,all vendors are taking digital signal processing requirements into account:ARM with “E” extensions，Hitachi with SH-DSP now into its third generation，IBM’s PowerPC with Book E. MIPS has recently announced CoreExtend.Jonah Probell has demonstrated that DSP extensions with CoreExtend can achieve 3x speed-up on audio applications.10 INTO THE FUTUREVLIW architectures applied to RISC-DSP instruction sets offer an important path for increasing performance，but the silicon cost of these architectures is not neg-ligible.Although the eight functional units in the TI TMS320C62xx data path have capability well beyond the ’C54xx，when applied to a typical case like the direct-form FIR，the eight-issue ’C62xx architecture uses 256 instruction bits to accomplish about what the ’C54xx can do in 16 bits. The extra silicon cost also extends to data-path elements and to the 15 ports on the register file required to sustain the eight functional units.As a result,TI’s VelociTI products are positioned for high-perfor-mance applications that,at the same time are not price-and power-sensitive.Another turn of the technology crank will be needed before VLIW architectures crowd out older 16-bit DSPs altogether.The H.264 codec is an example of an application requiring VLIW DSP.UB Video has developed H.264 decoder software for the 600-MHz TI TMS320DM642. This device uses the ’C64xx core along with specialized audio and video interfacing. It is capable of 4，800 MMACs (million multiply-accumulates per second) in eight-bit precision.The UB Video software supports decode at SDTV (standard-definition TV) resolutions.It’s impor-tant for H.264 decoder ICs to support the vast number of MPEG-2-encoded DVDs,as well as other codecs and future evolution in the ITU-T/ISO standard itself．The view of TI’s Eric Braddom,worldwide manager for DSP video imaging，that “programmability is essential at this stage,” is understandable from a technical as well as busi-ness perspective.11Meanwhile,there is still room for architectural inno-vation within the framework of VLIW using RISC-DSP instruction sets,The promise of H.264 in the DVD market is its potential for high-definition (1080i,720p) decoding. Currently,HD resolution in H.264 is beyond the 600-MHz ’DM642.TI has announced it will apply its forthcoming 1-GHz DSP.Other high-end VLIW DSPs such as Philips Trimedia,with five-instruction-issue aren’t waiting and are expected to attack the HD problem soon.Other competi-tors will reduce programmability,resorting to specialized hardware for MPEG-2 and H.264 alone．To see how far DSP architectures are from “matu-rity”，it’s eye-opening to look at Appendix C，“Survey of Architectures”，in Hennessey and Patterson’s text．12 The authors compare five RISC architectures.After a decade of research on compiler performance using well-established benchmarks,we find that RISCs are more alike than dif-ferent.DSPs are a long way from “Appendix C status，” but now that DSP applications are mainstream and both RISC vendors and DSP vendors are converging on RISC-DSP, the increasing cost of software development—along with the emergence of good benchmarks such as BDTI-mark2000 from Berkeley Design Technology (BDTI) and EEMBC (Embedded Microprocessor Benchmark Consor-tium) with which tomeasure design progress—will drive DSP architectures to become more similar, not unlike the path RISC followed a generation ago. Whenever a programmable .DSP architecture can meet an applications cost and power goals,it will be the pre-ferred solution. But what about the applications that just can’t fit? In the desktop market. application and system software seem to lag VLSI capabilities;in digital signal processing,the demands of “”faster/cheaper/lower-power” have always pushed DSP VLSI. Nick Tredennick. believes that the “leading-edge wedge … of zero-cost, zero-power, and zero-delay segments of the embedded systems mar-ket，”will drive DSPs to dynamic logic design附录B中文翻译DSP的发展W. PATRICK HAYS在线出版： 2004年6月23日施普林格科学+商业媒体b.v. 2004摘要现在的世界是以信息的传播、存储和处理为代表的数字时代。

Digital-Signal-Processing数字信号处理大学毕业论文英文文献翻译及原文

毕业设计（论文）外文文献翻译文献、资料中文题目：数字信号处理文献、资料英文题目：Digital Signal Processing 文献、资料来源：文献、资料发表（出版）日期：院（部）：专业：班级：姓名：学号：指导教师：翻译日期： 2017.02.14数字信号处理一、导论数字信号处理（DSP）是由一系列的数字或符号来表示这些信号的处理的过程的。

数字信号处理与模拟信号处理属于信号处理领域。

DSP包括子域的音频和语音信号处理，雷达和声纳信号处理，传感器阵列处理，谱估计，统计信号处理，数字图像处理，通信信号处理，生物医学信号处理，地震数据处理等。

由于DSP的目标通常是对连续的真实世界的模拟信号进行测量或滤波，第一步通常是通过使用一个模拟到数字的转换器将信号从模拟信号转化到数字信号。

通常，所需的输出信号却是一个模拟输出信号，因此这就需要一个数字到模拟的转换器。

即使这个过程比模拟处理更复杂的和而且具有离散值，由于数字信号处理的错误检测和校正不易受噪声影响，它的稳定性使得它优于许多模拟信号处理的应用（虽然不是全部）。

DSP算法一直是运行在标准的计算机，被称为数字信号处理器（DSP）的专用处理器或在专用硬件如特殊应用集成电路（ASIC）。

目前有用于数字信号处理的附加技术包括更强大的通用微处理器，现场可编程门阵列（FPGA），数字信号控制器（大多为工业应用，如电机控制）和流处理器和其他相关技术。

在数字信号处理过程中，工程师通常研究数字信号的以下领域：时间域（一维信号），空间域（多维信号），频率域，域和小波域的自相关。

他们选择在哪个领域过程中的一个信号，做一个明智的猜测（或通过尝试不同的可能性）作为该域的最佳代表的信号的本质特征。

从测量装置对样品序列产生一个时间或空间域表示，而离散傅立叶变换产生的频谱的频率域信息。

自相关的定义是互相关的信号本身在不同时间间隔的时间或空间的相关情况。

二、信号采样随着计算机的应用越来越多地使用，数字信号处理的需要也增加了。

dsp的英语作文

dsp的英语作文Certainly, here is an English essay on the topic of DSP (Digital Signal Processing):Digital signal processing, or DSP, is a fundamental field of study in the realm of electrical engineering and computer science. It involves the analysis, manipulation, and interpretation of digital signals, which are representations of real-world physical phenomena, such as sound, images, or sensor data, in a digital format. DSP has become an essential component of a wide range of technologies, from audio and video processing to telecommunications, medical imaging, and industrial automation.At the core of DSP is the concept of converting analog signals, which are continuous-time and continuous-amplitude, into digital signals, which are discrete-time and discrete-amplitude. This process, known as analog-to-digital conversion (ADC), is typically performed using specialized hardware, such as analog-to-digital converters (ADCs). Once the analog signal has been digitized, a variety of mathematical algorithms and techniques can be applied to manipulate, analyze, and extract meaningful information from the digital data.One of the primary applications of DSP is in the field of audio processing. Digital audio signals can be processed to enhance, filter, or modify the sound in a variety of ways. For example, DSP algorithms can be used to reduce noise, apply equalization, or apply special effects like reverb or echo. This has led to significant advancements in the quality and capabilities of audio equipment, from consumer electronics like smartphones and music players to professional-grade audio systems used in recording studios and live performances.Another important application of DSP is in the field of image and video processing. Digital cameras and video cameras capture images and video as digital data, which can then be processed using DSP techniques. These techniques can be used to enhance image quality, apply filters, perform image recognition and analysis, or even create special visual effects. DSP is also essential in the compression and transmission of digital video, enabling the efficient storage and streaming of high-quality video content.In the field of telecommunications, DSP plays a crucial role in the transmission and reception of digital signals. DSP algorithms are used to modulate and demodulate signals, perform error correction, and implement various communication protocols. This has enabled the development of advanced communication technologies, such as cellular networks, high-speed internet, and digital television.Beyond these traditional applications, DSP is also finding increasing use in a wide range of emerging technologies, such as medical imaging, industrial automation, and renewable energy systems. In medical imaging, for example, DSP techniques are used to process data from various imaging modalities, such as MRI, CT scans, and ultrasound, to produce high-quality diagnostic images. In industrial automation, DSP is used to control and monitor various processes, such as motor control, sensor data analysis, and quality control.The importance of DSP cannot be overstated, as it has become a fundamental component of countless modern technologies. As technology continues to evolve, the role of DSP is likely to become even more prevalent, with new and innovative applications constantly emerging. Understanding the principles and techniques of DSP is essential for anyone interested in the fields of electrical engineering, computer science, or technology in general.。

DSP外文资料143

of high affinity and one of low affinity. Only the high affinity site corresponds to the noradrenaline reuptake site (Backstrom et al., 1989; Backstrom and Marcusson, 1990). [3H]Mazindol binds with high affinity to dopamine as well as noradrenaline reuptake sites (Javitch et al., 1984, 1985), thus necessitating the use of a selective dopamine uptake inhibitor when quantitating noradrenaline reuptake sites. Nisoxetine is a potent and selective inhibitor of noradrenaline uptake with little or no affinity for a range of neurotransmitter receptors (Wong and Bymaster, 1976; Wong et al., 1982; Richelson and Nelson, 1984;
002%3908(95)00134-4
Neuropharmocology, Vol. 35, No. 1, pp. 63-70, 1996 Copyright 0 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0028-3908/96 $15.00 + 0.00

毕业论文外文翻译-DSP的广度和深度

毕业设计(论文)外文资料翻译专业：自动化姓名：学号：外文题目：The Breadth and Depth of DSP外文出处：The Scientist and Engineer's Guideto DSP1 DSP的广度和深度数字信号处理是最强大的技术，将塑造二十一世纪的科学与工程之一。

革命性的变化已经在广泛的领域：通信，医疗成像，雷达和声纳，高保真音乐再现，石油勘探，仅举几例。

上述各领域已建立了深厚的DSP技术，用自己的算法，数学，和专门技术。

这种呼吸和深度的结合，使得它不可能为任何一个人掌握所有已开发的DSP技术。

DSP教育包含两个任务：学习一般适用于作为一个整体领域的概念，并学习您感兴趣的特定领域的专门技术。

本章开始描述DSP已在几个不同领域的戏剧性效果的数字信号处理的世界，我们的旅程。

革命已经开始。

1.1 DSP的根源独特的数据类型，它使用的信号，数字信号处理是区别于其他计算机科学领域。

在大多数情况下，这些信号源于感觉来自现实世界的数据：地震的震动，视觉图像，声波等DSP是数学，算法，并用来操纵这些信号的技术后，他们已被转换成数字形式。

这包括了各种目标，如：加强视觉图像识别和语音生成，存储和传输的数据压缩，等假设我们重视计算机模拟 - 数字转换器，并用它来获得一个现实世界的数据块。

DSP回答了这个问题：下一步怎么办？DSP的根是在20世纪60年代和70年代数字计算机时首次面世。

电脑是昂贵的，在这个时代，DSP是有限的，只有少数关键应用。

努力开拓，在四个关键领域：雷达和声纳，国家安全风险是石油勘探，可以大量资金;太空探索，其中的数据是不可替代的;和医疗成像，可节省生活。

20世纪80年代和90年代的个人电脑革命，引起新的应用DSP的爆炸。

而不是由军方和政府的需求动机，DSP的突然被带动的商业市场。

任何人士如认为他们可以使资金在迅速扩大的领域突然一个DSP供应商。

DSP的市民等产品达到：移动电话机，光盘播放器，电子语音邮件。

DSP外文资料40

Desensitization of 5-HT 1A autoreceptors induced by neonatal DSP-4treatmentJoanna Da ˛browska *,Przemyslaw Nowak,Ryszard BrusDepartment of Pharmacology,Medical University of Silesia,38Jordana St.,41-808Zabrze,Poland Received 24October 2005;received in revised form 7February 2006;accepted 21March 2006Abstract To examine the effect of noradrenergic lesion on the reactivity of central 5-HT 1A receptors,DSP-4(50mg/kg)was administered neonatally 30min after zimelidine (10mg/kg)administration.5-HT 1A autoreceptors are involved in the regulation of serotonin (5-HT)synthesis.In HPLC assay R -(+)-8-OH-DPAT (0.03mg/kg)significantly decreased 5-HT synthesis rate in striatum,hypothalamus and frontal cortex of control,whilst nonsignificantly in DSP-4-lesioned adult rats (10—12weeks old).To determine which type of receptor ,pre-or postsynaptically located,is involved in the attenuated response to 5-HT 1A receptors’agonist,behavioral tests were conducted.R -(+)-8-OH-DPAT (0.015mg/kg)caused hyperphagia of control rats,but did not change feeding of DSP-4treated rats.R -(+)-8-OH-DPAT (0.1mg/kg)induced hypothermia and b 5-HT 1A syndrome Q in both control and DSP-4-lesioned animals.The nature of this phenomenon is attributable to the presynaptic adaptive mechanism and suggests the desensitization of 5-HT 1A autoreceptors of rats with neonatal lesion of the central noradrenergic system.D 2006Elsevier B.V .and ECNP .All rights reserved.1.Introduction5-HT 1A receptors are located presynaptically on the soma and dendrites of the 5-HT neurons of the mesencephalic raphe nuclei,mainly dorsal (DRN)and median raphe nuclei (MRN)as inhibitory autoreceptors and postsynaptically to the 5-HT neurons in the forebrain areas:hippocampus,lateral septum and cortex (Hamon,2000).It is well established that activation of the 5-HT 1A receptors causes neuronal hyperpolarisation,which is an effective inhibitorycontrol of serotonergic neurons at the presynaptic level(Barnes and Sharp,1999;Pineyro and Blier ,1999).Stimula-tion of the somatodendritic 5-HT 1A autoreceptors decreases 5-HT cells firing rate,synthesis and turnover of 5-HT as well as 5-HT release within raphe nuclei and subsequently within serotonergic projection areas (Blier et al.,1998).Apart from these effects measurable by biochemical methods,there are behavioral responses mediated by stimulation of presynaptic 5-HT 1A receptors;hyperphagia in rats (Ebenezer ,1992;Simansky,1996)and anxiolytic effect in rats (Jolas et al.,1995)and gerbils (File et al.,1996).On the other hand,hypothermia (Hadrava et al.,1996;O’Connell et al.,1992),stereotypic behavior in rats (O’Connell and Curzon,1996)as well as antidepressive effect in mice (Luscombe et al.,1993;0924-977X/$-see front matter D 2006Elsevier B.V .and ECNP .All rights reserved.doi:10.1016/j.euroneuro.2006.03.007*Corresponding author .Tel.:+48322713691;fax:+48322722683.E-mail address:joanna.dabrowska@dilnet.wroc.pl (J.Da ˛browska).KEYWORDS5-HT 1A receptor;DSP-4;Noradrenaline;Serotonin;R -(+)-8-OH-DPAT;WAY 100635European Neuropsychopharmacology (2007)17,129—137w w w.e l s e v i e r.c o m /l o c a t e /e u r o n e u r oMatsuda et al.,1995)and rats(Lucki et al.,1994)elicited by 8-OH-DPAT are attributable to the postsynaptic mechanisms. However,hypothermic response is equivocal and depends on species differences since in mouse lesion of5-HT neurons abolished the hypothermic response to5-HT1A agonist(Bill et al.,1991).The regions with high and moderate abundance of5-HT1A receptors receive direct noradrenergic innervation from the locus coeruleus(LC)which provides ascending projections to the frontal cortex,hippocampus,striatum and amygdala.Efferents from the lateral tegmentum(LT) provide innervation to a lesser extent mainly to the hypothalamus and extended amygdala nuclei(Cryan et al.,2002).In raphe nuclei where5-HT1A receptors act as inhibitory autoreceptors,DRN receives noradrenergic pro-jection from the LC(Peyron et al.,1996),while MRN from the LT and to some extent also from the LC(Hopwood and Stamford,2001).The central noradrenergic system is involved in the regulation of serotonergic system activity.Firing rate of5-HT neurons(Haddjeri et al.,1996)as well as5-HT release in DRN and MRN(Adell and Artigas,1999;Pudovkina et al., 2002,2003),medial prefrontal cortex(Amargos-Bosch et al.,2003)and hippocampus(Mongeau et al.,1997)depend on the noradrenergic input.Due to direct innervation of brain structures enriched in5-HT1A receptors by noradren-ergic terminals we examined whether the reactivity of these receptors could be modulated by neurotoxin DSP-4under conditions of nearly total depletion of noradrenergic neurons.An alkylating agent,DSP-4(Ross,1976)causes selec-tive destruction of noradrenergic projections(distal axons and terminals)ascending from the LC(Jaim-Etcheverry, 1998;Lapiz et al.,2001).Systemic administration of DSP-4produces a long-term depletion of brain NA in rats, without affecting either dopaminergic or adrenergic neurons.The loss of5-HT can be avoided when animals are pretreated with a selective5-HT reuptake inhibitor (SSRI)(Cryan et al.,2002;Jaim-Etcheverry,1998).In our experiment we used a model of neonatal noradrenergic lesion(Jonsson et al.,1982)described in detail elsewhere (Da˛browska et al.,2004;Labus et al.,2004),induced by subcutaneous injections of DSP-4(50mg/kg)30min after zimelidine(10mg/kg)in rats on the1st and3rd day of life.It is well known that monoamine neurons respond to lesion with a wide range of compensatory adaptations aimed at preserving their functional integrity.Those may include increased synthesis and release of neurotransmitters from fibers(Hall et al.,1999;Lapiz et al.,2001),axonal sprouting (Heal et al.,1993)as well as adaptive changes of receptors’reactivity.After DSP-4treatment up-regulation of A1-adrenoceptors,A2-adrenoceptors and B-receptors in frontal cortex has been reported(Dooley et al.,1983a,b;Harro et al.,1999a)in parallel to biochemical supersensitivity to A1-, A2-and B-adrenoceptor agonists(Harro et al.,1995;Yasuda et al.,1986).Several previous studies have been focused on influence of DSP-4lesion(performed in adult animals)on5-HT1A receptors’reactivity.Although DSP-4treatment did not alter antidepressant-like effect of8-OH-DPAT in the Porsolt test in mice(Luscombe et al.,1993)male rats’sexual behavior induced by8-OH-DPAT was partially blocked by DSP-4pretreatment(Fernandez-Guasti and Rodriguez-Manzo,1997).Additionally,DSP-4lesion abolished8-OH-DPAT induced mydriasis in mice(Prow et al.,1996)as well as analgesic effect of this compound in rats(Archer et al., 1987).Results from autoradiographic studies were provided by Laporte et al.(1995)and H7idkind et al.(2003).Although partial noradrenergic depletion had no effect on[3H]8-OH-DPAT binding to5-HT1A receptors in the hippocampus 6weeks after DSP-4(10mg/kg)administration in the latter study,Laporte et al.reported significant increase in the specific binding of[3H]8-OH-DPAT to5-HT1A receptors in the dorsal horn of the spinal cord15days after DSP-4injection (50mg/kg).In our experiments we have focused on both pre and postsynaptically located5-HT1A receptors’reactivity under the conditions of central noradrenergic lesion.Hence we have assessed effects mediated by presynaptic(5-HT syn-thesis rate,hyperphagia)as well as postsynaptic5-HT1A receptors(hypothermia,b5-HT1A syndrome Q).Furthermore, the novel approach of our study is the evaluation of5-HT1A receptors’reactivity after neonatal noradrenergic lesion. 2.Experimental procedures2.1.AnimalsNoradrenergic lesion was performed in male Wistar newborn rats, weighing6—12g.Biochemical and behavioral experiments were conducted in lesioned and respective control,weighing280—300g, adult rats(10—12weeks old at the beginning of experiments). Between experiments the rats were housed in groups of6in cages in a temperature-controlled room with free access to food and water and maintained under a12h:12h light/dark cycle(light on7:00). Each group of the behavioral experiment consisted of8animals, whilst of6(5-HT synthesis rate)and12(the assay of biogenic amines)in the biochemical study.Local Bioethical Committee approved the experiment(permission#15/04,issued on 28.04.2004).2.2.DrugsNSD1015(m-hydroxybenzylhydrazine dihydrochloride),DSP-4(N-(2-chloroethyl)-N-ethyl-2-bromobenzylamine),R-(+)-8-OH-DPAT hydrobromide(R-(+)-8-hydroxy-2-(di-n-propylamino)tetralin),WAY 100635maleate(N-[2-(4-[2-methoxyphenyl]-1-piperazinyl)ethyl]-N-2-pyridynylcyclohexanecarboxamide)and zimelidine dihy-drochloride([Z]-3-[4-bromophenyl]-N,N-dimethyl-3-[3-pyridynyl]-2-propen-1-amine)were purchased from Sigma Chemicals,St.Louis, MO,USA.All substances were dissolved in0.9%NaCl solution (saline).Saline,R-(+)-8-OH-DPAT,WAY100635,DSP-4and zimelidine were injected subcutaneously(sc),whilst NSD1015was adminis-tered intraperitonally(ip).All compounds were injected in volume 1ml/kg,aside from NSD1015administered in2ml/kg.In the case of DSP-4each dose was weighed separately and administered imme-diately after dissolving in saline solution.Doses of above compounds are expressed as bases.2.3.Neonatal noradrenergic lesionNewborn male Wistar rats were divided into2groups.The first group was injected on the1st and3rd day of life with DSP-4(50mg/ kg)and zimelidine(10mg/kg).Zimelidine was used30min before DSP-4administration to prevent5-HT depletion.Control(non-J.Da˛browska et al.130lesioned)animals received zimelidine(10mg/kg)injection30min before saline injection(1ml/kg)on the1st and3rd day of life. 2.4.Biogenic amines assayRegional brain noradrenaline(NA),serotonin(5-HT),4-hydroxy-3-methoxyphenylglicol(MOPEG)and5-hydroxyindole-3-acetic acid (5-HIAA)levels in frontal cortex(FC),hippocampus(CA),hypothal-amus(HP)and striatum(STR)were determined by HPLC with electrochemical detection(Magnusson et al.,1980;Nowak et al., 2003;Ossowska et al.,2005).Adult control and DSP-4lesioned rats were sacrificed by decapitation and their FC,CA,HP and STR were immediately dissected on an ice-chilled plate and weighed.The tissues were stored in deep freeze(À708C)pending assay.The tissue samples were homogenized in ice-cold0.1M perchloric acid(HClO4) containing0.05mM ascorbic acid.After centrifugation(15,000Âg, 20min),the supernatants were filtered through0.2A m cellulose membranes(Titan-MSF Microspin Filters,Lida Manufacturing Corp.) and were injected into the HPLC system.Chromatography conditions: pump-model302HPLC with manometric model802C(Gilson, France);precolumn-Hypersil BDS C18,10Â4mm,3A m(Thermo-Quest,GB);column-Hypersil BDS C18,250Â4,6mm,3A m(Thermo-Quest,GB);injector-model2175rheodyne inert injector20A l loop (USA);flow rate:0.7ml/min;mobile phase:75mM NaH2PO4Â2H2O (Avocado Res.Chem.),1.7mM1-octanosulfonic acid(Avocado Res. Chem.),5A M EDTA(Avocado Res.Chem.),100A l triethylamine/ 1l(Sigma),9.5%acetonitrile(Lab-Scan),pH3.0with phosphoric acid(Fluka)for amines assay,pH3.5for5-HTP assay.Detector conditions:detector-model141electrochemical detector(Gilson, France)with applied potential:+750mV;output voltage:1V; gain:10nA,filter settings5s.Levels of NA,5-HT,MOPEG and 5-HIAA were expressed as ng/g of wet tissue.2.5.5-HT synthesis rate5-HT synthesis rate of control and DSP-4lesioned rats after aromatic amino acids decarboxylase inhibitor(NSD1015)injection was determined by HPLC/ED technique.Control and DSP-4lesioned rats were divided into4subgroups.5-HT synthesis rate was estimated after administration of saline;selective,full and potent5-HT1A agonist,R-(+)-8-OH-DPAT and selective5-HT1A antagonist WAY 100635.Saline,R-(+)-8-OH-DPAT(0.03mg/kg)and WAY100635(0.05mg/kg)were administered15min before NSD1015(100mg/ kg).To establish if the observed result derived exclusively via5-HT1A receptor activation,WAY100635(Forster et al.,1995)was administered15min before R-(+)-8-OH-DPAT injection.30min after NSD1015administration animals were sacrificed by decapitation and their FC,CA,HP and STR were immediately dissected on an ice-chilled plate and weighed.Following procedures were conducted as above.Levels of5-hydroxytryptophan(5-HTP)were expressed as pg/g of wet tissue.5-HT synthesis rates(pmol/g/h)were calculated according to Carlsson et al.(1972).2.6.Hyperphagic responseThe lesioned and respective control rats were housed separately in cages with free access to food and water for24h.Immediately before drug or vehicle injection food pellets were replaced by strictly weighed portions of food(10g).30min after drug or vehicle administration,the remaining food pellets were removed and weighed.The experiment was performed after saline,R-(+)-8-OH-DP AT(0.015mg/kg)or WA Y 100635(0.03mg/kg)injection.T o establish if the observed effect derived from activation of5-HT1A receptors,WAY100635was administered20min before R-(+)-8-OH-DP AT(Ebenezer,1992).ponents of the5-HT1A syndromeOn the day of experiment,control and DSP-4lesioned rats were placed separately in transparent plastic boxes,15min before subcutaneous injection of R-(+)-8-OH-DPAT(0.1mg/kg)or saline. Components of the b5-HT1A syndrome Q as reciprocal forepaw treading and flat body posture were scored over five30-s periods at3-min intervals starting3min after R-(+)-8-OH-DPAT or saline injection(5observation sessions).The appearance of forepaw treading and flat body posture was noted,and their intensity was scored using the following scale:0,absent;1,equivocal;2,present; 3,intense.The maximal cumulative score was15per rat.WAY 100635(0.15mg/kg)was used20min before R-(+)-8-OH-DPAT administration(Testa et al.,1999).2.8.Body temperature measurementsBody temperature was recorded at40,20and0min before and at20,40,60,80,100,120,140,160and180min after drug orTable1NA,5-HT,MOPEG and5-HIAA levels(ng/g of wet tissue)and turnover ratios of NA and5-HT in different parts of the rats’brain after DSP-4(50mg/kg)neonatal treatmentNA MOPEG MOPEG/NA5-HT5-HIAA5-HIAA/5-HT Frontal cortexControl309.55F27.39249.55F23.63 1.63F0.27359.95F29.75230.10F27.070.59F0.03 DSP-417.34F6.05**275.85F43.7130.38F13.8**373.04F13.78214.02F48.020.53F0.11 HippocampusControl483.04F32.12262.05F8.89 1.06F0.1296.85F20.49340.36F17.76 1.06F0.14 DSP-420.07F4.72**249.09F27.8923.70F8.35**285.71F16.08373.77F17.43 1.21F0.17 StriatumControl125.37F7.81108.26F12.66 1.63F0.14239.81F11.04269.19F18.18 1.04F0.21 DSP-444.18F9.01*87.30F20.85 3.77F1.29*282.69F16.25271.83F29.240.89F0.07 HypothalamusControl2638.9F43.82311.28F16.560.22F0.01266.56F37.62336.90F39.48 1.17F0.19 DSP-42365.41F276.33319.97F9.270.27F0.03332.47F30.77283.38F57.430.79F0.16 DSP-4was administered subcutaneously on the1st and3rd day of life30min after zimelidine(10mg/kg)injection.All data are expressed as means F S.E.M.,n=12;**p b0.0005,*p b0.05using Student’s t-test compared to control,non-lesioned rats.Desensitization of5-HT1A autoreceptors induced by neonatal DSP-4treatment131vehicle administration.The thermistor probe (Yellow Springs Instrument Co.Inc.)was inserted into the rectum,2.5cm from the anal orifice (Yoshitake and Kehr ,2004).Body temperature measurements in non-lesioned and DSP-4treated rats were recorded after R -(+)-8-OH-DPAT (0.1mg/kg)injection.WAY 100635(0.15mg/kg)was administered 20min before R -(+)-8-OH-DPAT injection.2.9.Data analysisMonoamines and their metabolites tissue levels (ng/g of wet tissue),their metabolic ratios,the 5-HTsynthesis rates (pmol/g/h),the weight of consumed food pellets and body temperature were expressed as means F standard errors (S.E.M.).In the assay of 5-HT synthesis rate and behavioral experiments pretreatment (DSP-4)and treatment (5-HT 1A ligands)effects were examined by the analysis of variance (two-way ANOVA)followed by Newman-Keul’s for post hoc analysis.In the assay of biogenic amines the differences between control and DSP-4lesioned rats were examined using Student’s t -test.In each experiment p b 0.05was accepted as a statistically significant effect.The number of scores of b 5-HT 1A syndrome Q components:flat body posture and reciprocal forepaw treading was expressed as median;treatment effects were assessed by Kruskal-Wallis analysis of variance.All statistics were done using Statistica 6.0software (StatSoft).3.Results3.1.Regional brain NA,5-HT,MOPEG and 5-HIAA levels after DSP-4lesionDSP-4treatment decreased NA concentration in the FC to 5.6%(p b 0.0005),to 4.2%(p b 0.0005)in the CA and to 35.2%(p b 0.005)in STR in comparison to control,non-lesioned rats.In HP only nonsignificant reduction was observed (89.6%of control rats content).MOPEG levels were not affected by DSP-4-pretreatment in all examined brain regions,consequently MOPEG/NA metabolic ratios were significantly increased in FC (p b 0.0005as compared to non-lesioned animals),CA (p b 0.0005)and STR (p b 0.05)of lesioned rats.In HP MOPEG/NA metabolic ratio was unchanged by noradrenergic lesion.5-HT levels were not affected by DSP-4pretreatment in the CA,FC,although in STR and HP a nonsignificant tendency to increase 5-HT content was observed.5-HIAA concentrations and conse-quently 5-HIAA/5-HT metabolic ratios remained unaffected in all examined brain regions (T able 1).3.2.5-HT synthesis rate in different parts of the brainThere was no significant difference between 5-HT synthesis rate of control and DSP-4pretreated animals aftersalineFigure 15-HT synthesis rates (pmol/g/h expressed as means F S.E.M.,n =6)of control and DSP-4(50mg/kg)treated rats after saline,R -(+)8-OH-DPAT (0.03mg/kg)and/or WAY 100635(0.05mg/kg)injection in frontal cortex (A),striatum (B),hypothalamus (C)and hippocampus (D).WAY 100635was administered 15min before R -(+)-8-OH-DPAT injection.5-HT synthesis rate was estimated by chromatographic assay of L-5-HTP level after NSD 1015(100mg/kg)injection;*p b 0.013/7,**p b 0.0051/3;4/8by analysis of variance (ANOVA two-way)followed by Newman-Keul’s test for post hocanalysis.Figure 2Effect of DSP-4lesion on hyperphagic response to R -(+)-8-OH-DPAT (0.015mg/kg)and/or WAY 100635(0.03mg/kg)administration.WAY 100635was administered 20min before R -(+)-8-OH-DPAT injection;*p b 0.0051/3;3/7;3/4using ANOVA with Newman-Keul’s test as post hoc.Results are expressed as means F S.E.M,n =8.J.Da ˛browska et al.132injection in all brain regions tested.In control,non-lesioned rats,R -(+)-8-OH-DPAT (0.03mg/kg)caused a significant reduction of 5-HT synthesis rate by 46%in STR (p b 0.005in comparison to saline-treated control rats),by nearly 42%in FC (p b 0.005)and by 20%in HP (p b 0.005).Nevertheless,in CA there was only a nonsignificant tendency to decrease 5-HT synthesis rate.R -(+)-8-OH-DPAT (0.03mg/kg)failed to significantly decrease 5-HT synthesis rate in all examined brain regions of DSP-4lesioned rats.WAY 100635(0.05mg/kg),when injected alone,did not affect 5-HT synthesis rate in all brain regions tested in control and DSP-4lesioned rats.Pretreatment with WAY 10063515min before R -(+)-8-OH-DPAT injection completely abolished the inhibitory effect of R -(+)-8-OH-DPAT on the 5-HT synthesis rate of control rats in FC (p b 0.01in comparison to the control group after R -(+)-8-OH-DPAT injection only),STR (p b 0.01)and HP (p b 0.01).Unex-pectedly,pretreatment with WAY 100635and R -(+)-8-OH-DPAT administration caused significant elevation of 5-HT synthesis rate of DSP-4lesioned animals in the FC (p b 0.005in comparison to DSP-4lesioned animals re-ceived R -(+)-8-OH-DPAT injection only),in STR (p b 0.005)and in the HP (p b 0.005).Similar ,albeit nonsignificant increase was observed in CA (Fig.1).3.3.Hyperphagic responseR -(+)-8-OH-DPAT (0.015mg/kg)administered immediately before the feeding session caused a significant increase in food intake by 257%during the first 30min in control rats (p b 0.005versus saline-treated group).In DSP-4treated rats R -(+)-8-OH-DPAT did not produce any effect on food intake;post hoc Newman-Keul’s test revealed a significant differ-ence versus the control,non-lesioned group after R -(+)-8-OH-DPAT administration (p b 0.005).Pretreatment with WAY 100635(0.03mg/kg)completely abolished hyperphagic response to R -(+)-8-OH-DPAT administration of control rats (Fig.2).ponents of the 5-HT 1A syndromeControl,non-lesioned and DSP-4treated animals after saline injection did not exhibit the behavioral components of5-HT 1A syndrome.The subcutaneous injection of R -(+)-8-OH-DPAT at a dose of 0.1mg/kg induced within 1—2min reciprocal forepaw treading and flat body posture in control,non-lesioned rats as well as in DSP-4lesioned rats.Kruskal-Wallis analysis of variance did not reveal any difference in number of scores of flat body posture and reciprocal forepaw treading between control and DSP-4treated rats.WAY 100635(0.15mg/kg)used 15min before R -(+)-8-OH-DPAT completely abolished b 5-HT 1A syndrome Q in both groups of animals.3.5.Body temperature measurementsR -(+)-8-OH-DPAT (0.1mg/kg)induced hypothermia in con-trol,non-lesioned and DSP-4treated rats.Maximal drop in body temperature (À2.40F 0.368C for control and À2.20F 0.418C for DSP-4rats)was observed 40min after R -(+)-8-OH-DPAT injection in both groups.The body tem-perature had in both groups of animals returned to basal values 120min after agonist administration.WAY 100635(0.15mg/kg)was able to completely antagonize the hypothermic effect of R -(+)-8-OH-DPAT in control and lesioned rats (Fig.3).4.DiscussionThe main finding of this paper is that neonatal noradren-ergic lesion has attenuated biochemical as well as behavioral responses to R -(+)-8-OH-DPAT mediated by pre-synaptic 5-HT 1A receptors,whereas the effects of post-synaptic 5-HT 1A heteroreceptors’stimulation remained unaffected.In our experiment the neonatal DSP-4treatment caused the persistent depletion of cortical,hippocampal and striatal NA levels,in accordance with the fundamental paper by Jonsson et al.(1982)reporting permanent depletion of NA in the cerebral cortex after neonatal DSP-4lesion.In our study hypothalamic NA content was unaffected.Since hypothalamus is known to receive main noradrenergic projection from lateral tegmentum (Cryan et al.,2002)it supports previous findings that DSP-4selectively destroys noradrenergic projections from LC,alsoconfirmedFigure 3Body temperature of control,non-lesioned and DSP-4pretreated rats after R -(+)-8-OH-DPAT (0.1mg/kg)administration.WAY 100635(0.15mg/kg)was administered 20min before R -(+)-8-OH-DPAT injection.Results are expressed as means F S.E.M.,n =8.Desensitization of 5-HT 1A autoreceptors induced by neonatal DSP-4treatment 133by immunocytochemical studies(Schuerger and Balaban, 1999).On the other hand MOPEG levels were not reduced by DSP-4in all tested brain regions,also observed by H7idkind et al.(2003),which may suggest the increased NA turnover, especially in the cortex and hippocampus(Harro et al., 1999b).DSP-4lesion had no effect on the5-HT level in all examined brain regions in saline treated animals,because DSP-4is a specific neurotoxin to noradrenergic neurons when animals are pretreated with selective serotonin reuptake inhibitor(SSRI)(Jaim-Etcheverry,1998;Jonsson et al.,1982).Activation of5-HT1A autoreceptors by a prototypical agonist,8-OH-DPAT(Arvidsson et al.,1981),reveals an inhibitory action on the serotonergic cell firing in raphe nuclei and decreases synthesis of5-HT(Blier et al.,1998; Pineyro and Blier,1999).In the chromatographic assay of5-HT synthesis rate the dose of0.03mg/kg R-(+)-8-OH-DPAT was used to act exclusively via5-HT1A autoreceptors.R-(+)-8-OH-DPAT is an active enantiomer of racemic8-OH-DPAT (Cornfield et al.,1991)and such a relatively low dose turned out to be sufficient to inhibit5-HT synthesis in FC,HP and STR by42%,20%and46%,respectively.WAY100635was able to antagonize the effect of R-(+)-8-OH-DPAT in all tested brain areas of control rats.Interestingly,in brain regions innervated by DRN seroto-nergic projections(FC and STR)above dose of R-(+)-8-OH-DPATwas able to reduce synthesis rate in control rats by over 40%in both these areas,whilst by20%in HP innervated by MRN serotonergic projection and nonsignificantly in CA innervated by both MRN and DRN(McQuade and Sharp, 1997).It confirms the regional differences between both raphe areas,since it was established that DRN5-HT neurons are more sensitive to5-HT1A agonists(Casanovas and Artigas,1996;Hajos et al.,1995).Unexpectedly,R-(+)-8-OH-DPAT failed to significantly inhibit5-HT synthesis rate in all examined brain structures of DSP-4-treated rats.It should be emphasized that the attenuation of the inhibitory effect of R-(+)-8-OH-DPAT on 5-HT synthesis was also observed in HP of lesioned rats where NA content was not affected by DSP-4treatment. Those results may suggest that the possible change in receptors’response to R-(+)-8-OH-DPAT does not concern 5-HT1A receptors located postsynaptically in the FC and CA extensively affected by noradrenergic lesion.Thus,the attenuated biochemical response to R-(+)-8-OH-DPAT in FC and STR might suggest involvement of5-HT1A autoreceptors located on DRN of lesioned rats and similar effects observed in HP as well as in CA(albeit nonsignificant) cannot exclude a parallel phenomenon in MRN.However, the effect of R-(+)-8-OH-DPAT in FC,HP and STR of lesioned rats was further reinforced by WAY100635administration, hence the observed5-HT synthesis elevation may rather indicate some nonspecific action of R-(+)-8-OH-DPAT. Antagonism toward a2-adrenoceptors(Crist and Surpre-nant,1987;Winter,1988)seems plausible,especially in terms of a2-adrenoceptors’up-regulation reported after DSP-4treatment(Dooley et al.,1983a,b).However,it is still hardly believable because of the low dose of R-(+)-8-OH-DPAT used.WAY100635did not modify the5-HT synthesis rate in all tested brain areas of both control and lesioned rats. Although Haddjeri et al.(2004)observed a significant enhancement of the firing activity of DRN neurons after WAY100635(0.1mg/kg)administration in DSP-4treated rats,the absence of this effect in our experiment is probably related to the low dose used.Due to the tonic activation of5-HT1A autoreceptors in the DRN(Haddjeri et al.,2004),the same basal5-HT synthesis rates in both control and lesioned saline-treated rats argue against any alteration in5-HT1A receptors’sensitivity induced by DSP-4 lesion.However,it is not excluded if the5-HT1A receptors are not tonically activated under physiological conditions (Adell et al.,2002;Johnson et al.,2002),which is in good agreement with papers reporting unchanged firing rate of serotonergic neurons in DRN(Richer et al.,2002)as well as not altered basal efflux of5-HT in DRN(Bortolozzi et al., 2004)and forebrain regions(He et al.,2001)of5-HT1A knockout mice.It is well known that5-HT1A receptor agonists produce hyperphagia in satiated rats via stimulation of5-HT1A autoreceptors in raphe nuclei(Ebenezer,1992).In our experiment R-(+)-8-OH-DPAT(0.015mg/kg)produced a significant increase in food intake in control rats,without eliciting stereotypic behavior derived from stimulation of postsynaptic5-HT1A receptors and this effect was abolished by pretreatment with WAY100635.The lack of hyperphagic response to R-(+)-8-OH-DPAT injection in DSP-4rats may be relevant to desensitization of5-HT1A autoreceptors after neonatal noradrenergic lesion.Contrary to the hyperphagic response,hypothermia (Yoshitake and Kehr,2004)as well as b5-HT1A syndrome Q (Testa et al.,1999)induced by selective5-HT1A agonist injection are believed to be mediated by postsynaptic5-HT1A receptors in rats(Lin and Chuang,2002;O’Connell and Curzon,1996).In our experiment R-(+)-8-OH-DPAT (0.1mg/kg)induced hypothermia and b5-HT1A syndrome Q to a similar extent in both control and lesioned groups of rats;the effects were completely antagonized by WAY 100635(0.15mg/kg).These data lead to the proposal that reactivity of postsynaptically located5-HT1A receptors remains unchanged after DSP-4lesion.Interactions between the noradrenergic and serotonergic system have been particularly described for MRN and DRN (Adell and Artigas,1999;Bortolozzi and Artigas,2003).The presence of NA and NA synthesizing enzymes(Hopwood and Stamford,2001)as well as a relatively high level of NA transporter was reported in both nuclei(Ordway et al., 1997).Abundance of a1-adrenoceptors,especially a1B-subtype,presence of a2A-and a2C-adrenoceptors was confirmed by autoradiographic,immunohistochemical(Tal-ley et al.,1996)and in situ hybridization studies(Scheinin et al.,1994).Distribution of cells expressing a1B-adrenoceptors indicates their localization on the DRN5-HT neurons(Day et al.,1997),on the other hand in situ hybridization suggests that the majority of a2-adrenoceptors are located on noradrenergic terminals as autoreceptors controlling the release of NA(Scheinin et al.,1994).Furthermore,there is a strong evidence that noradrenergic fibers exert tonic excitatory stimulation of spontaneous firing rate of5-HT neurons within the raphe nuclei in vivo.Stimulation of cell body a1-adrenoceptors results in increased serotonergic cell firing(Vandermaelen and Aghajanian,1983),whereas the firing is suppressed by stimulation of a2-adrenoceptors located on noradrenergic terminals(Svensson et al.,J.Da˛browska et al.134。

DSP外文资料127

Fusion Engineering and Design 71 (2004) 77–82
Real-time DSP-based shape determination and plasma position control in the ISTTOK tokamak
B. Carvalho∗, H. Fernandes, C. Silva, D. Borba, C.A.F. Varandas
0920-3796/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.fusengdes.2004.04.014
78
B. Carvalho et al. / Fusion Engineering and Design 71 (2004) 77–82
2. Shape reproduction methods
Shape reproduction on axisymmetric tokamak devices is done traditionally by solving the homogeneous Grad–Shafranov equation in the vacuum region around the plasma for the poloidal ﬂux function Ψ(R, Z) = 2πRAϕ (Aϕ is the azimuthal component of the mag-
• expansion in terms of current ﬁlaments (CF); • expansion in toroidal harmonics (TH).
Due to the limited space inside the vessel, the magnetic ﬁeld probes are located rather close to the vessel and copper shell, and therefore, are strongly affected by eddy currents in these passive conductors. This effect was veriﬁed by comparing the measured signals in magnetic probes in response to currents applied to the poloidal ﬁeld coils (PF) against a free-space Green’s function numerical simulation. In the following section we analyze how this affects the plasma shape reproduction from external magnetic measurements.

DSP外文资料155

Neuroscience Letters 503 (2011) 20–22Contents lists available at ScienceDirectNeuroscienceLettersj o u r n a l h o m e p a g e :w w w.e l s e v i e r.c o m /l o c a t e /n e u l etAntinociceptive effects of neurotropin in a rat model of central neuropathic pain:DSP-4induced noradrenergic lesionTakashi Kudo,Tetsuya Kushikata,Mihoko Kudo,Tsuyoshi Kudo,Kazuyoshi Hirota ∗Department of Anesthesiology,Hirosaki University Graduate School of Medicine,Hirosaki 036-8562,Japana r t i c l ei n f oArticle history:Received 5June 2011Received in revised form 20July 2011Accepted 30July 2011Keywords:Central painNoradrenergic neuronsSerotonergic neurons Neurotropina b s t r a c tNeurotropin is a nonprotein extract isolated from inﬂamed skin of rabbits inoculated with vaccinia virus,and used for treatment of neuropathic pain.In the present study,we have determined whether neu-rotropin could exert antinociceptive action using the central neuropathic pain model that we recently established.Rats were randomly allocated to 3groups:Sham group (n =20),DSP-4[N-(-2-chloroethyl)-N-ethyl-2-bromobenzylamine]group (50mg/kg ip,n =18),and DSP-4+5,7-DHT [5,7-dihydroxytryptamine]group (ip DSP-450mg/kg +icv 5,7-DHT 200␮g,n =18).In Sham,DSP-4and DSP-4+5,7-DHT groups,the effects of ip neurotropin (100NU/Kg)on hot-plate latency in rats with no lesion,noradrenergic neu-ron depletion and both noradrenergic and serotonergic neuronal depletion were studied,respectively.Rats in each group were subdivided equally to 2subgroups:saline and neurotropin.After completion of the hot-plate tests,each rat was decapitated,the cerebral cortex was dissected from its internal struc-ture for measurement of norepinephrine contents.Hot-plate latency signiﬁcantly decreased by ∼40%10days after ip DSP-4or after ip DSP-4and 5,7-DHT.Norepinephrine contents in DSP-4treated rats (55.6±6.3ng/ng tissue)and DSP-4+5,7-DHT treated rats (35.3±6.3ng/ng tissue)were signiﬁcantly lower than those in intact rats (131.6±5.7ng/ng tissue,p <0.01).Neurotropin signiﬁcantly increased the area under the curve (AUC)of the hot-plate latency in the DSP-4and DSP-4+5,7-DHT groups but not in the Sham group.There was a signiﬁcant correlation between AUC and norepinephrine contents in saline subgroup (p <0.01,r =0.597)but not in neurotropin subgroup in DSP-4group.Neurotropin exerted an antinociceptive effect in DSP-4induced central neuropathic pain.The present data suggest neuronal pathways other than descending inhibitory noradrenergic and serotonergic systems may be involved in neurotropin mediated antinociception.© 2011 Elsevier Ireland Ltd. All rights reserved.Neuropathic pain is caused by damage of the peripheral and/or cen-tral nervous system [16,18].Neuropathic pain is often chronic and unresponsive to opioids but sometimes responsive to anticonvul-sants and antidepressants.Neuropathic pain models are classiﬁed as central and peripheral.Whilst there are several established peripheral neuropathic pain models,central pain models are based only on spinal cord injury.Brain noradrenergic neurons originate from seven noradrener-gic cell groups,designated as A1–A7in rodents [2].One originates from the A6and A4cell groups that are located in the locus coeruleus (LC)complex.The LC complex innervates several cere-bral loci such as the cerebral cortex,hippocampus,hypothalamus,parabrachial nucleus,nucleus tractus solitarius,and nucleus raphe dorsalis that produce anti-nociceptive effects on pain transmis-sion.The LC complex also has descending projections to the spinal cord.N-(-2-chloroethyl)-N-ethyl-2-bromobenzylamine (DSP-4)is∗Corresponding author.Tel.:+81172395113;fax:+81172395112.E-mail address:hirotak@cc.hirosaki-u.ac.jp (K.Hirota).a competitive inhibitor of norepinephrine uptake and selectively degenerates LC noradrenergic neurons originating in the LC whilst non-cerulean innervated noradrenergic axons are unaffected [3].We recently conﬁrmed that depletion of LC noradrenergic neu-rons by DSP-4produces central neuropathic pain in rats [6].In this recent study DSP-4signiﬁcantly decreased the response latency in the hot-plate but not tail-ﬂick test.These data indicate that DSP-4may facilitate central neuropathic pain as a response in hot-plate and tail-ﬂick implies supra-spinal process and spinal reﬂex,respec-tively [11].Neurotropin is a non-protein extract isolated from inﬂamed skin of rabbits inoculated with vaccinia virus,and is used for the treatment of neuropathic pain.Although several studies [8,10,14]sought to solve the mechanism of the antinociceptive action of neurotropin,the mechanism remains unclear.Most reports sug-gest that the antinociceptive effects of neurotropin are mediated via activation of descending pain inhibitory systems such as the noradrenergic and serotonergic systems projecting from supra-spinal sites to the spinal dorsal horn [8,10,14].Thus,in the present study,we have determined whether neurotropin could exert0304-3940/$–see front matter © 2011 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.neulet.2011.07.056T.Kudo et al./Neuroscience Letters503 (2011) 20–2221antinociceptive actions in DSP-4-induced supra-spinal pain model with and without depletion of serotonergic neurons by intracerebroventricular(icv)injection of5,7-dihydroxytryptamine (5,7-DHT)[9].With approval by the institutional ethical committee on animal research of Hirosaki University Graduate School of Medicine,56 male Sprague-Dawley rats weighing300–400g were used.Rats were randomly allocated to3groups:Sham(n=20),DSP-4 (n=18),and DSP-4+5,7-DHT groups(n=18).In Sham,DSP-4and DSP-4+5,7-DHT groups,the effects of ip neurotropin(100NU/Kg) on hot-plate latency in rats with no lesion,noradrenergic neuron depletion and both noradrenergic and serotonergic neurons deple-tion were studied,respectively.Rats in each group were equally subdivided to2subgroups:saline and neurotropin subgroups.An 8mm length of stainless-steel guide cannula(OD=0.5mm,AG-8,EICOM,Kyoto,Japan)for icv administration was implanted unilaterally into the lateral cerebroventricle under pentobarbital anesthesia(50mg/kg ip).One week later,hot-plate latency was measured to obtain a control.Each rat was placed individually on a hot plate maintained at50◦C(MK-350D;Muromachi Kikai Co., Ltd.,Tokyo)and the reaction time was measured starting from time the rat was placed on the plate until the rat either demonstrated hind paw licking or shaking or jumping.The measurement was per-formed every10min after ip neurotropin.Five successive values were averaged and used.The cut-off latency was40s to prevent tis-sue damage.These animals received ip saline,ip DSP-4(50mg/kg) or both ip DSP-4(50mg/kg)and icv5,7-DHT(200␮g).Ten days after the injection,hot-plate latency was measured again.After completion of the behavioral tests,each rat was decap-itated,the brain quickly removed and immersed in ice-cold Krebs-Ringer bicarbonate buffer solution(KRBS)oxygenated with 95%O2and5%CO2:NaCl133,KCl4.8,KH2PO41.2,MgSO41.2,CaCl2 1.5,glucose11.1,HEPES10(in mM,pH7.4).The cerebral cortex whose noradrenergic neurons almost fully originate from the locus coeruleus[5]was dissected from their internal structures,weighed, and then immediately sonicated in10×brain tissue weight of KRBS. The supernatant was collected by centrifugation of the sonicated material at15,000×g for30min and stored at−70◦C until cate-cholamine extraction was performed.Norepinephrine levels were determined directly by high-performance liquid chromatography with electrochemical detection(ESA Coulochem Model5100A, Tokyo,Japan).The lower detection limit was15pg/ml,and the intra-assay coefﬁcient of variation was3.3%.All data are presented as mean±SEM.The relationship between norepinephrine content in the cerebral cortex and area under the curve(AUC)of time-latency in the hot-plate test of the DSP-4group was examined using Pearson’s correlation coefﬁcient and a least squares linear regression line wasﬁtted using GraphPad Prism (1.03).Where appropriate,statistical analysis was by Student-t or one way ANOVA followed by the Bonferroni’s multiple comparison test.A p<0.05was considered signiﬁcant.In Sham group,hot-plate latency did not signiﬁcantly change during experiment.There was no difference in the latency of each measurement point between saline and neurotropin subgroups.In DSP-4group,hot-plate latency signiﬁcantly decreased by ∼40%10days after ip DSP-4.In addition,norepinephrine contents in DSP-4group(55.6±6.3ng/ng tissue)were signiﬁcantly lower than those in the Sham group(131.6±5.7ng/ng tissue,p<0.01). The latency at50min in neurotropin subgroup was signiﬁcantly higher than that in saline subgroup(F=4.309,p<0.05)although intra-group comparison did not show any signiﬁcant differences between time0and any time points in the neurotropin subgroup (Fig.1A).In addition,the AUC of the time-latency(984±72s min, p<0.05)in the hot-plate test in the neurotropin subgroup was also signiﬁcantly larger than that(779±74s min)in the saline sub-group.Moreover,there was a signiﬁcant correlation betweenAUC Fig.1.Effects of neurotropin in DSP-4treated rats.(A)Time course of hot-plate latency with saline or neurotropin,all data are mean±SEM(n=9each),Pre:before ip DSP-4,0(min):10days after DSP-4,**p<0.01vs.0min,#p<0.05vs.saline.(B) Signiﬁcant correlation between the AUC of the hot-plate latency and cerebrocortical norepinephrine content in the saline subgroup(p<0.01,r=0.597)but not in the neurotropin subgroup(p=0.246,r=0.288).and norepinephrine contents in the saline(p<0.01,r=0.597)but not the neurotropin subgroup(p=0.578,r=0.266)(Fig.1B).In DSP-4+5,7-DHT group,hot-plate latency signiﬁcantly decreased by∼40%10days after ip DSP-4and icv5,7-DHT. In addition,norepinephrine contents in DSP-4+5,7-DHT group (35.3±6.3ng/ng tissue)were signiﬁcantly lower than those in Sham group(131.6±5.7ng/ng tissue,p<0.01).Neurotropin signif-icantly increased latency at10,20and30min compared to that at 0min(F=3.283,p<0.05,<0.01and<0.01,respectively).Moreover, neurotropin signiﬁcantly increased the latency compared to saline (Fig.2),and the AUC(1326±121,p<0.05)in the neurotropin subgroup was signiﬁcantly larger than that(922±95)in the saline subgroup.In addition,AUC in the neurotropin subgroup was also signiﬁcantly larger than that in neurotropin subgroups of DSP-4 group(p<0.05).However,signiﬁcant correlation between AUC and norepinephrine contents was not observed in both subgroups of this group(saline:p=0.688,r=0.156;neurotropin:p=0.629, r=0.188).Neurotropin did not modify hot-plate latency in intact rats and this is in agreement with previous reports[8,12,14].Ohara and colleagues[8]reported that neurotropin exerted antinocicep-tive effects in rats with environmental temperature stress(SART stressed rats)but did not in rats with normal condition.Saleh et al.[12]showed antinociceptive actions of neurotropin in rats with mononeuropathic pain produced by sciatic nerve ligation but no effects were reported in intact rats.Suzuki et al.[14]found similar22T.Kudo et al./Neuroscience Letters 503 (2011) 20–22Fig.2.Time course of hot-plate latency with saline or neurotropin in DSP-4+5,7-DHT treated rats.All data are mean ±SEM (n =9each),Pre:before ip DSP-4,0(min):10days after DSP-4and 5,7-DHT,**p <0.01vs.0min,#p <0.05vs.saline.results in mice with and without L5spinal nerve ligation.In their report [14],injection of neurotropin in to the nerve-injured paw produced no effects in the L5spinal nerve ligation model.Moreover,Okazaki and colleagues [10]showed that icv,but not intrathecal,administration of neurotropin markedly attenuated allodynia in L5spinal nerve ligated rats.These ﬁndings suggest that the antinoci-ceptive actions of neurotropin may be mediated via supraspinal and not peripheral sites.As nerve injury activates the descending inhibitory noradren-ergic systems [7,17],an increase in potency and efﬁcacy of ␣2-agonists is observed after nerve injury.Thus,noradrenergic sys-tems may be involved in the mechanisms of neuropathic pain.Indeed,the present study shows a signiﬁcant correlation between cerebrocortical noradrenergic contents and AUC in the hot-plate latency test.However,in the present study,antinociceptive action of neu-rotropin was observed even in rats with cerulean noradrenergic neuronal depletion.Therefore,not only noradrenergic but also other neuronal systems are likely to contribute to the antinoci-ceptive action of neurotropin.Okazaki and colleagues [10]showed that the antinociceptive actions of neurotropin could be antago-nized by intrathecal ketanserin (5-HT2receptor antagonist)and MDL72222(5-HT3receptor antagonist)in L5spinal nerve ligation model.Thus,serotonergic systems may also be involved in the antinociceptive effects of neurotropin.In the present study,even when both noradrenergic and serotonergic systems were depleted,an antinociceptive action of neurotropin remains.These ﬁndings suggest that neuronal pathways other than descending inhibitory noradrenergic and serotonergic systems are likely to be involved.Hata and colleagues [4]found that bicuculline a GABA A recep-tor antagonist antagonized antinociceptive effect of neurotropin in SART stressed mice when both agents were co-administered intracisternally whilst Ohara et al.[8]showed that subcutaneous bicuculline could not reverse antinociceptive acion of ip neu-rotropin in SART stressed mice.Okazaki and colleagues [10]also reported that antiallodynic effect of iv neurotropine was inhib-ited by intrathecal administration of bicuculline (GABA A receptor antagonist)or CGP35348(GABA B receptor antagonist)in rats with L5-spinal nerve ligation.Therefore,antinociceptive action of neu-rotropin might also be mediated via GABAergic systems.Antinociceptive effects of neurotropin in the DSP-4+5,7-DHT group was larger than that in DSP-4group.The descending sero-tonergic system shows some plasticity in nerve injury,and thepotentiation of descending facilitatory 5-HT pathways may con-tribute to the development of neuropathic pain [1].Indeed,Suzuki and colleagues [13]reported that ondansetron,a 5-HT3receptor antagonist,inhibited allodynia in rats with L5spinal nerve liga-tion.Wang and colleagues [15]also reported that ketanserin,a 5-HT2A receptor antagonist,inhibited thermal hyperalgesia in rats with L5nerve ligation.Thus,in the present study,5,7-DHT may deplete descending facilitatory serotonergic system and conse-quently potentiated antinociceptive action of neurotropin which caused increase in hot-plate latency and bringing its peak effect earlier in DSP-4+5,7-DHT treated rats than those in DSP-4treated rats.In the DSP-4+5,7-DHT group,there was no signiﬁcant correla-tion between cerebrocortical noradrenergic content and AUC for hot-plate latency in both saline and neurotropin subgroups.This ﬁnding suggests that not only noradrenergic but also serotonergic pathway should also affect pain intensity.Neurotropin exerted an antinociceptive effect in a central neu-ropathic pain model produced by DSP-4induced lesioning of noradrenergic neurons in the rat.As neurotropin was antinocicep-tive when both noradrenergic and serotonergic neuronal systems were depleted,other neuronal pathways are likely to be involved.Further studies will be required to elucidate the mechanism of the antinociceptive effects of neurotropin.References[1]A.V.Apkarian,M.N.Baliki,P.Y.Geha,Towards a theory of chronic pain,Prog.Neurobiol.87(2009)81–97.[2]A.Dahlstroem,K.Fuxe,Evidence for the existence of monoamine-containingneurons in the central nervous system I.Demonstration of monoamines in the cell bodies of brain stem neurons,Acta Physiol.Scand.Suppl.232(1964)1–55.[3]J.M.Fritschy,R.Grzanna,Selective effects of DSP-4on locus coeruleus axons:arethere pharmacologically different types of noradrenergic axons in the central nervous system?Prog.Brain Res.88(1991)257–268.[4]T.Hata,T.Kita,E.Itoh,R.Oyama,A.Kawabata,Mechanism of the analgesiceffect of neurotropin,Jpn.J.Pharmacol.48(1988)165–173.[5]K.Hirota,T.Kushikata,Central noradrenergic neurones and the mechanism ofgeneral anaesthesia,Br.J.Anaesth.87(2001)811–813.[6]T.Kudo,T.Kushikata,M.Kudo,T.Kudo,K.Hirota,A central neuropathic painmodel by DSP-4induced lesion of noradrenergic neurons:preliminary report,Neurosci.Lett.481(2010)102–104.[7]W.Ma,J.C.Eisenach,Chronic constriction injury of sciatic nerve induces the up-regulation of descending inhibitory noradrenergic innervation to the lumbar dorsal horn of mice,Brain Res.970(2003)110–118.[8]H.Ohara,M.Kawamura,A.Namimatsu,T.Miura,R.Yoneda,T.Hata,Mechanismof hyperalgesia in SART stressed (repeated cold stress)mice:antinociceptive effect of neurotropin,Jpn.J.Pharmacol.57(1991)243–250.[9]S.Ohashi,M.Matsumoto,H.Togashi,K.Ueno,M.Yoshioka,The serotoner-gic modulation of synaptic plasticity in the rat hippocampo-medial prefrontal cortex pathway,Neurosci.Lett.342(2003)179–182.[10]R.Okazaki,H.Namba,H.Yoshida,H.Okai,T.Miura,M.Kawamura,The antial-lodynic effect of Neurotropin is mediated via activation of descending pain inhibitory systems in rats with spinal nerve ligation,Anesth.Analg.107(2008)1064–1069.[11]M.H.Ossipov,C.J.Kovelowski,M.L.Nichols,V.J.Hruby,F.Porreca,Characteri-zation of supraspinal antinociceptive actions of opioid delta agonists in the rat,Pain 62(1995)287–293.[12]M.R.Saleh,H.Muneshige,Y.Ikuta,Effects of neurotropin on hyperalgesia andallodynia in mononeuropathic rats,Life Sci.63(1998)1931–1938.[13]R.Suzuki,W.Rahman,S.P.Hunt,A.H.Dickenson,Descending facilitatory controlof mechanically evoked responses is enhanced in deep dorsal horn neurones following peripheral nerve injury,Brain Res.1019(2004)68–76.[14]T.Suzuki,Y.H.Li,T.Mashimo,The antiallodynic and antihyperalgesic effectsof neurotropin in mice with spinal nerve ligation,Anesth.Analg.101(2005)793–799.[15]D.Wang,Y.Gao,H.Ji,Y.Hong,Topical systemic administrations of ketanserinattenuate hypersensitivity and expression of CGRP in rats with spinal nerve ligation,Eur.J.Pharmacol.627(2010)124–130.[16]C.J.Woolf,R.J.Mannion,Neuropathic pain:aetiology symptoms,mechanisms,and management,Lancet 353(1999)1959–1964.[17]X.J.Xu,M.J.Puke,Z.Wiesenfeld-Hallin,The depressive effect of intrathecalclonidine on the spinal ﬂexor reﬂex is enhanced after sciatic nerve section in rats,Pain 51(1992)145–151.[18]M.Zimmermann,Pathobiology of neuropathic pain,Eur.J.Pharmacol.429(2001)23–37.。

关于DSP的外文文献

Procedia Engineering 29 (2012) 3763 – 37671877-7058 © 2011 Published by Elsevier Ltd.doi:10.1016/j.proeng.2012.01.567Available online at vailable online at 2012 International Workshop on Information and Electronics Engineering (IWIEE) Design of ECG Signal Acquisition System Based on DSPKening Wang, Shengqian Ma*,Jing Feng, Weizhao Zhang,Manhong Fan,Dan ZhaoCollege of Physics and Electronic Engineering, Northwest Normal University, Lanzhou 730070 P.R. ChinaAbstractA real-time ECG signal acquisition system with the DSP chip TMS320VC5509A as its core is introduced in this paper. It introduces the design of ECG signal acquisition circuit in detail. ECG signal is analog filtered and amplified and then converted from analog to digital field. Finally, it is displayed in the LCD after digital low-pass filtering in DSP. The collected ECG data can truly reflect the condition of human heart’s health. The system has advantages of small volume, little power consumption, low cost and real-time processing © 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Harbin University of Science and Technology.Keyword : DSP; ECG signal;Filter;Data collectionl;a1.IntroductionIn recent years, cardiovascular disease is still one of the main diseases that threats to human life. According to incomplete statistics, about 2000 people die of the cardiovascular disease around the world every year. Therefore, the prevention and diagnosis of cardiovascular disease becomes one of the primary issues that the medical profession faces today. The medical profession can already make early prediction and diagnose of the related lesions through the study of ECG pattern on time. Therefore, acquisiting and analyzing ECG signal accurately has important meaning. At present, the ECG data acquisition and analysis programs are mostly done by using MCU and ARM. However, the operation ability of SCM is limited and its storage space is too small, so it can’t process signals of long time acquisition and complex [1], and although ARM is powerful enough, DSP devices have more advantages in digital filtering and * Corresponding author. Tel.: +86-139********.E-mail address : s.q.ma@.3764Kening Wang et al. / Procedia Engineering 29 (2012) 3763 – 3767further signal operations analysis of ECG signal [2,7].2.Overall design2.1 Basic characteristics of ECG signalThe ECG signal is a bipolar low-frequency weak signal. According to the standard of the ECG institution of the USA, normal ECG signal frequencies vary in the range of 0.05-100Hz, mainly concentrate in the range of 0.05~ 35Hz and their amplitude ranges from 10 μ V to 4mV whose typical value is 1mV.2.2 System block diagramAccording to the basic characteristics of ECG signal and powerful signal processing ability of DSP [2] [3], this design uses a DSP chip TMS320VC5509A as its control core whose highest working frequency is 200M, SRAM128K*16, RAM128K*16, ROM32K*16. It is one of the 16-bit fixed-point digital signal processors with a abundance of on-chip peripherals [3].System schematic diagram is shown in Fig1(a):3.Dsign of ECG signal acquisition circuitThe ECG signal acquisition system includes input protection circuit, preamplifier, driver circuit of right leg, band-pass filter circuit, 50Hz notch filter, primary amplifier circuit and level uplift circuit.Fig. 1. (a) Schematic diagram of system; (b)The input protection circuitWhen doing electrocardiogram ,the first thing to consider is to secure human’s safe, and at the same time we must pay attention to how to reduce cardiac pacemaker pulse interference effects or damage on ECG testing equipments. The input protection circuit is designed in considering of these two issues. From Fig1(b), we can see ECG signal can pass without affection. When substantial interference enters, they will be limited to ± 0.7V by diode. The voltage follower made by OPA2604 is a buffer level [4], which actually is an impedance converter, so it plays an isolated role in the system.3.2 Preamplifier and driver circuit of right legWhen choosing an amplifier, we need to consider the aspects of gain, frequency response, input impedance, common mode rejection ratio, noise and drift and etc [1]. This system selects the amplifier AD620 [5] from AD company as the core components of ECG signal preamplifier.3765Kening Wang et al. / Procedia Engineering 29 (2012) 3763 – 3767Fig. 2. (a)The preamplifier and driver circuit of right leg; (b) Band-pass filter circuit Fig2(a)shows the design of the preamplifier and driver circuit of right leg for the system.The gain of amplifier circuit can be calculated by formula 3-1. is 5.1 K Ω, and the gain G is about 10.(1) 3.3 Band-pass filter circuitBand-pass filter circuit adopts feedback structure based second-order active RC filter [6]. The system achieves a 0.05~100Hz transmission bands by designing the series of a second-order active high-pass filter(HPF) and a second-order active low-pass filter(LPF) whose cut-off frequencies are 0.05Hz and 100Hz respectively with dual operational amplifier OPA2604, 0.05~100Hz band-pass filter circuit schematic diagram is shown in Fig2(b)It is implemented by using a dual operational amplifier OPA2604. For high-pass filter, the parameters are 8968R R K ==Ω, 212247C C F μ==, and the cut-off frequency isIf the Gain k is 1, then the quality factor Q is (3)For low-pass filter, the parameters are 1011 1.6R R K ==Ω,23241C C F μ==and the cut-off frequency is3.4 50Hz notch filterIn the process of ECG signal extraction, a band-stop filter is commonly used to restrain the 50Hz power frequency interference produced by outside or system itself, the band-stop filter is also called notch filter [7]. The double T-RC active notch filter is used in this system. It is an improved, Q value adjustable band-stop double T active filter circuit. The circuit uses an amplifier to make a positive feedback in order to reduce the bandwidth which is the difference between it and the previous double-T notch filter. So the amplitude near both sides of the stop-band center frequency increases, quality factor Q can be regulated If the Gain k is 1, then the quality factor Q is (5) by potentiometer Rs1. The schematic diagram of 50Hz notch filter is shown in Fig4(a).The transfer function of the circuit is G R 49.41Gk G R Ω=+0.05()p f Hz =≈10.53Q K ==−100()p f H z =≈10.53Q K ==−3766 Kening Wang et al. / Procedia Engineering 29 (2012) 3763 – 3767(6) Fig. 4.(a)50Hz notch filter schematic diagram;(b)The main amplifier and voltage lifting circuitIn the actual circuit, R is 6.8 and C is 0.47 6.8R K =Ω,0.47C F μ=,the center frequency is:(7) In the preamplifier circuits, we amplified ECG signal about 10 times, and then the main amplifier circuit should amplify it about 100 times to meet the requirements of post processing. The schematic diagram of main amplifier circuit is shown in Fig4.(b).In this design, the input range of the A/D converter we use is 0 ~ 3.3 V. In order to meet the A / D conversion range, we design a level lifting circuit.If we take 171R M =Ω,210.1Rs K =Ω,The magnification is (8) 4.A / D conversionThe TMS320VC5509A provide 10-bit ADC input interface of double-channel. This is a continuous approximation type ADC and its external reference voltage is 3.3 V. The ADC can be controlled by setting up four corresponding registers, ADCCtl 、ADCDate 、ADCC 、lkDiv and ADCClkCtl . In this system we use channel 1. According to the sampling theory, the sampling frequency must be more than or equal to twice as much as the signal frequency in order to obtain the digital signal without distortion. The maximum frequency of ECG signals is 100 Hz, so the sampling frequency must be more than 200 Hz. The highest sampling frequency of this system is 21.5 KHz, so the system finally set sampling frequency as 360 Hz, which can meet the sampling theory [9,10]. The system uses the ADC in TMS320VC5509A which reduces system cost, and the stability is relatively better. In the A/D conversion of ECG signal, we start sampling in ADC through setting ADC Start in ADC register, wait for completing data transformation by the enquiring the ADCBusy, obtain sampling data by read data from the ADCData register , the data of low-10-bit in ADCData is converted, so the high-six-bit should be resetted, the procedures are described below ： Data_in=ADCData; // Obtain data Data_in&=0x03ff; // Data conversion5.Digital filtering to Electrocardiological signal andTotal program flowIn order to get more cleaner ECG signal, the system should adds an analog filter after the digital filter. a 0 ~ 100 HzFIR low-pass filter is designed to filter out-of-band noise by using frequency sampling method[8], whose stop-band attenuation can reach to 80 dB. If a 0.05 ~ 100 Hz low pass filter is designed, the sampling density is too high, and calculation amount is too big, so it can't meet the real-time requirment.The software system uses C programming language and mixed language to program. The overall process of program is shown in Fig5(a):()222222()1()()411o i V s s R C H s V s s R C K +==+−+0150()2f Hz RC π=≈1721100R G R s =+≈3767Kening Wang et al. / Procedia Engineering 29 (2012) 3763 – 3767Fig5 .(a)The system flow chart6. System test resultsThe system uses Q15 to express integer and Q0 to express decimal [3]. The oscilloscope observation ofECG output signal is shown in Fig6(a).Real time ECG signal of LCD display is shown in Fig6(b)Fig.6.(a)ECG waveform of output;(b)ECG signal of LCD displayThis design adopts TMS320VC5509A with high cost performance as its core, which Completes ECGsignal acquisition, completes the filtering and real-time display, achieves a ECG acquisition system of portable and low power consumption. Therefore, it has strong practical values.AcknowledgementsThe author would like to acknowledge the support of the National Science Foundation of China (NSFC) under grant NO.61162067 and NO.20927114， and also the high school graduate tutor scientific research project of Gansu province under grant NO.1101-03.References[1] LIU hong-xia,WEI dong-fang Atmega .SCM ECG signal detection and processing system. Journal of Yunnan University(Natural science edition)2009,31(S2):248～251.[2] Bahoura,M.,Hassani,M.,”DSP implementation of Wavelet transfom for rael time ECG wave forms detection and heart rata analysis ”,Comput Mothods Programs Biomed,52:1,pp.35-44.,1997.[3] WANG chun-mei.TMS320C55xDSP principle and application .BeiJing:Publishing House of electronics industry,2008.[4] OPA2604—Dual FET-Input, Low Distortion OPERATIONAL AMPLIFIER. Burr-Brown Corporation, 1991.[5] AD620-Low Cost，Low Power Instrumentation Amplifier. Analog Device Inc.1999.[6] YANG zhi-min，MA yi-de,ZHANG xin-guo．Modern circuit theory and design．BeiJing:Tsinghua university press.[7] Cai kun,LU-yaosheng.Based on the median filter for ECG baseline correction rmation of Medical Equipment,2004,19(2):5-7.[8] ICL7660—Switched-Capacitor Voltage Converters.Maxim Integrated Products, 1994.[9] DOU ming-tao. Data acquisition system based on the design of TMS320VC5402 Electronic measurement technology.2004(4):37-38.[10] ZHONG yin-bao,XIAO ying-bin,SHI chang-ning, Full ECG monitoring system design and Implementation Informationof Medical Equipment.,2006,27(8):46.[11] XUE nian-xi. MATLAB in the application of digital signal processing . Beijing,tsinghua university press,2003.。

1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。

关于DSP的外文文献

DSP外文资料9

DSP外文资料13

DSP外文资料43

dsp存储器英文文献及翻译（精）

DSP外文资料142

【最新推荐】基于DSP的IIR滤波器设计外文文献

DSP外文资料80

DSP外文资料91

DSP外文资料72

DSP外文资料138

DSP外文资料135

DSP的发展 毕业论文外文翻译

Digital-Signal-Processing数字信号处理大学毕业论文英文文献翻译及原文

dsp的英语作文

DSP外文资料143

毕业论文外文翻译-DSP的广度和深度

DSP外文资料40

DSP外文资料127

DSP外文资料155

关于DSP的外文文献

DSP的发展毕业论文外文翻译