UWB radar antenna

An Ultrawide-band Microwave Radar Sensor for Nondestructive Evaluation of Pavement SubsurfaceJoongsuk Park and Cam Nguyen,Fellow,IEEEAbstract—A new ultrawide-band(UWB)microwave radar sensor operating from0.6to5.6GHz has been developed using microwave integrated circuits for pavement subsurface charac-terization.UWB antennas operating from0.5–10-GHz have been designed and tested for use in the sensor.A new simple,yet effec-tive,accurate procedure was also developed to compensate for the common amplitude deviations and nonlinear phase errors pro-duced by the inherent imperfection of the system.The developed compensation method is applicable to other systems and effec-tively reduces the potential masking of adjacent targets as well as facilitating and increasing the accuracy for target identification of the sensor.The sensor has been used to assess a pavement sample with less than0.1in of error in the pavement’s layer thickness. The developed system represents thefirst UWB stepped-frequency radar sensor completely realized using microwave integrated circuits over the frequency range of0.6–5.6GHz for subsurface sensing applications.Index Terms—Evaluation of civil infrastructures,microwave sensors,nondestructive evaluation,stepped-frequency radar, ultrawide-band(UWB)systems.I.I NTRODUCTIONM ICROW A VE sensing has been proven as a valuable technique for nondestructive characterization of sub-surfaces.Microwave SFR can provide both deep penetration andfine resolution simultaneously for subsurface sensing and is,thus,attractive for subsurface evaluation such as measuring layer thickness and detecting damages in pavement structures. The instantaneous bandwidth of this radar at each frequency is very narrow,resulting in high signal-to-noise ratio at the receiver.Additionally,nonlinear effects caused by the inherent imperfection of the transmitter and receiver can be corrected through appropriate digital signal processing.Moreover,very low sampling frequency for the analog-to-digital converter (ADC)can be used,hence facilitating the system design. Most of the reported microwave SFRs for subsurface appli-cations operate at low frequencies and have narrow bandwidths. For example,a SFR,developed for detection of water below the pavement surface,operates from600MHz to1.112GHz [1].Another SFR operating from490to780MHz was devel-oped for detection of buried objects[2].An HP network ana-Manuscript received September9,2003;revised September7,2004.This work was supported in part by the National Science Foundation and in part by the National Academy of Sciences.The associate editor coordinating the review of this paper and approving it for publication was Dr.Dwight Woolard.J.Park is with the Digital Media R&D Center,Samsung Electronics Com-pany,Ltd.,Suwon,443-742,Korea(e-mail:joongsuk.park@). Professor C.Nguyen is with the Sensing,Imaging,and Communications Sys-tems Laboratory,Department of Electrical Engineering,Texas A&M University, College Station,TX77843-3128USA(e-mail:cam@).Digital Object Identifier10.1109/JSEN.2005.851002Fig.1.Block diagram of the UWB SFR sensor.lyzer operating from0.5to6GHz was also used as a SFR todetect concrete cracks[3],which is bulky and very expensive.Recently,an integrated-circuit SFR has been reported for sub-surface sensing[4].In this paper,we report a new compact UWB microwave SFRsensor operating from0.6to5.6GHz for measuring subsur-face characteristics of pavement.It is an extension of the workpresented in[4].The sensor is implemented using a coherentsuper-heterodyne scheme and completely realized using MICs.It employs two UWB antennas operating from0.5GHz to morethan10GHz.Its performance has been demonstrated throughaccurate measurements of pavement subsurface.A new com-pensation technique was also developed and used for correctingthe amplitude and phase errors of the system.II.S YSTEM P RINCIPLESFR sensor transmits sequencesof sinusoidal signals of different frequencies toward a target,receives return signalsfrom the target,and processes them for extracting the targetcharacteristics.In each sequence,the frequency is shifted indiscrete values—each value is held constant for a period oftime and then changed to a next higher value[5].The received signals at step frequencies,reflected from thetarget,are down-converted into an immediate-frequency(IF)signal.This IF signal is then demodulated into in-phase ()andquadrature-phase()signals in base band.The/signals rep-resent both the amplitude and phase information of the targets. 1530-437X/$20.00©2005IEEEFig.2.Photograph of the transceiver.Upon converting these analog/signals into digital/sam-ples,digital signal processing is applied on the/samples to retrieve a representative complex vector at each frequency.Each setof samples is combined to form a complex vectorarray,,from which a time-domain synthetic pulse or range pro-file representing the target is extracted and processed to reveal the pavement ’s subsurface characteristics.III.S YSTEM D ESIGNFig.1shows the system block diagram of the newly devel-oped UWB (0.6–5.6GHz)microwave SFR sensor based on the coherent super-heterodyne architecture.This frequency range was chosen to allow suf ficient penetration while achieving good resolution.The sensor consists of a transceiver,two antennas,and a digital signal processing.The transceiver architecture can be either a homodyne or super heterodyne scheme.The super-heterodyne scheme is more complex,but enables easier correc-tion of the/errors.Hence,it was selected for the developed sensor.A.TransceiverThe temperature compensated crystal oscillator (TCXO)in the transceiver generates a signal of 10MHz,which is used as the LO signal for the quadrature detector and the IF signal for the up-converter.The up-converter converts the incoming 0.59–5.59-GHz LO signals from the synthesizer to 0.6–5.6-GHz signals to be transmitted toward the targets (through an UWB transmit antenna).Alternately,the down-converter converts the returned signals from the targets (through the receiver antenna)to an IF signal of 10MHz by mixing them with the coherent LO signals from the synthesizer.The IF signal is then converted into the baseband/signal in the quadrature detector by mixing it with the coherent LO signal from the TCXO.The/signals are finally digitized with the ADCs and processed in the digital signal processing blocks to extract the target information.The transceiver is separated into two parts for easy fabrica-tion,evaluation,and trouble-shooting.One is for low-frequencycircuits and the other is for high-frequency circuits.Both low-and high-frequency circuits were fabricated on 31-mil FR-4sub-strates.The high-frequency circuits include an up-converter,a cascaded RF ampli fier,two LO ampli fiers,a low-noise ampli-fier (LNA),and a down-converter.The up-converter modulates the IF signals into the RF signals with the aid of external LO signals.The cascaded ampli fier increases the power of the trans-mitting RF signals,and the two LO ampli fiers boost the external LO up to the required power level for pumping the up-con-verter and down-converter,respectively.The LNA reduces the total noise figure of the transceiver and increases the power of the received RF signals.The down-converter demodulates the received RF signals into a single frequency called the IF signal.The low-frequency circuits consist of a stable local os-cillator (STALO),attenuators,low-pass filters (LPFs),power di-viders,an IF ampli fier,an LO ampli fier,an/detector,and a two-channel video ampli fier.A temperature controlled crystal oscillator (TCXO)is used for the STALO.The attenuators limit the power of the LO and IF signals below the speci fications of the subsequent circuits.The LPFs reduce the high frequency harmonics included in the IF signal and the IF harmonics added in the baseband/signals.The power divider splits the output of the TCXO into two,one for the IF of the up-converter and the other for the LO of the quadrature detector.The LO ampli fier increases the LO power to pump the quadrature detector.The quadrature detector down-converts the single frequency input,which includes information on targets,into the baseband/signals.The two-channel video ampli fier increases the power of the baseband/signals to meet the input range of the ADC.Fig.2depicts the photograph of the developed transceiver.B.AntennaTEM horn antennas are attractive for UWB radars owing to their inherent characteristics of wide bandwidth,high direc-tivity,good phase linearity,and low distortion.Various types of TEM horn antennas have been developed [6]–[8].A TEM horn antenna,however,needs a balun at its input,prohibiting aPARK AND NGUYEN:ULTRAWIDE-BAND MICROW A VE RADAR SENSOR3Fig.3.Sketch of the UWB antenna.direct connection between antennas and the transceiver circuit. The use of balun also limits the antenna’s operating bandwidth. Moreover,direct coupling between the transmitting and re-ceiving antennas in a bi-static system is severe.Novel UWB antennas[9]are employed in our system to allow a direct integration with the transceiver while main-taining adequate isolation between the two antennas.Fig.3 shows a sketch of the antenna.It consists of a conductor on top of a grounded dielectric substrate and,hence,resembling a microstrip structure.The dielectric medium,and,hence,the height of the conductor above the ground plane,can be changed in any particular fashion,and the conductor’s profile depends on the contour of the dielectric substrate.The antenna can be directly connected to the connector without any balun and/or transition,making it simpler physically and higher-performance electrically.When two such antennas used for the transmitter and receiver are placed against each other,the common ground plane acts as a shield between these antennas,resulting in a high isolation between them.This unique feature is extremely attractive for radar applications where a certain degree of isola-tion is always needed between transmit and receive antennas. The operation of the antenna is based on the principle of wave propagation along a transmission line.In the uniform section of the microstrip line,where the spacing between the top conductor and the ground plane is very small compared to a wavelength, wave propagation is mostly confined within the dielectric be-tween the top conductor and the ground plane.However,as the separation between the conductor and ground plane gradually increases and approaches approximately a half-wavelength or more,the energy begins to radiate in the end-fire mode and,con-sequently,the wave is no longer guided between the conductor and ground plane.The entire structure effectively behaves as an antenna.The width of the antenna aperture primarily controls the radiation at low frequencies and,hence,sets the low-fre-quency limit for radiation,while the antenna length and the con-ductor and dielectric contours control the matching over the op-erating bandwidth.The antenna used in the sensor was designed to present at least10dB of return loss over a wide band of0.5–10GHz.The length of the antenna,which is primarily restricted by the lowest operating frequency,is set to16in.Styrofoam is used as the di-electric medium to support the antenna’s top conductor.Reflec-tions from the open end and the edges were significantly reduced by appending a resistive pad to the open end and absorber tothe Fig. 4.Antenna’s reflection coefficient in the(a)time domain and (b)frequency domain,where(I)indicates the antenna alone and(II)represents the antenna with a resistive pad and absorber.edges.The resistive pad,which is made of a metalfilm with thickness and resistivity of0.025in and250/,respectively, was tuned empirically to an optimal size ofputer simulations were performed using Ansoft’s HFSS program to theoretically verify the reflection coefficients and the farfield radiation patterns.The gain and3-dB beamwidth of theantenna -plane was within6–17dBi and25–45from0.6–5GHz,re-spectively.These results led to a lateral resolution of9.5in at a distance of12in.Fig.4shows the measured reflection coefficient in both the time and frequency domains.The reflection coefficient at the low frequency end,as seen in the frequency-domain plot,is im-proved significantly due to the incorporation of the resistive pad and absorber.1The resistive pad and absorber,however,reduce the gain of the antenna and degrade the system’s sensitivity and dynamic range.A better illustration of the impact of these acces-sories is shown in the time domain plots.An additional narrow peak,indicating deterioration of the input reflection coefficient, is observed at around3.5ns when the resistive pad and absorber were not incorporated.C.Digital Signal Processing and Systematic Error Correction The baseband analog/signals are digitized into digital/signals at ADCs.These digitized/signals need signal processing to be transformed into a synthetic pulse in the time 1The resistive pad and absorber absorb energy at the low frequencies,which cannot be radiated by afinite-size antenna.4IEEE SENSORS JOURNALdomain.Therefore,signal processing including/error com-pensation and inverse discrete Fourier transform (IDFT)was de-veloped using LabView.2In the absence of errors in the/channels,thephase of thebasebandand signals,expressed in terms of the targetrange and frequencystep ,is givenby(1)where is the speed of the electromagnetic wave in themedium,is the angular frequency step,and is the number of the frequency steps.The timedelay is equal to a two-way traveltimeof.A practical (nonideal)system produces common and differ-ential amplitude and phase errors intheand channels.The common errors are caused by the common signal path such as the antennas,ampli fiers,mixers,transmission lines,filters,etc.On the other hand,the differential errors are generated in the quadrature detector due to the difference between the two mixers and the phase imbalance of the 90coupler contained in the quadrature detector.The differential amplitude and phase errors generate a Hermitian image of the response in the re-sultant synthetic range pro file,resulting in a reduction of the sensor ’s unambiguous range by one half [9].In a super-hetero-dyne system,these errors are constant in the operating frequency range,as a single constant intermediate frequency is used for the quadrature detector.Consequently,measurement and com-pensation of these errors is simple.The differential amplitude and phase errors intheand channels at an intermediate fre-quency can be measured by using the methods presented in [5]or [10].Following these techniques,the differential amplitude and phase errors were measured as 1dB and 3,respectively,for the developed microwave SFR sensor.After these differen-tial errors are corrected and normalized,the complexvectors,,for a fixed angular frequency can be written in terms of therangeas(2)where is the common amplitude deviationand is the common phase error,which consists of a linear phaseerror and a nonlinear phaseerroras(3)The common linear phase error results in a constant shift of the response in the synthetic range pro file due to the fact that a fre-quency-dependent linear phase is transformed into a constant time delay through the inverse Fourier transform (IFT)[11].Therefore,it is not needed to correct for the common linear phase error.However,the nonlinear phase error causes shifting as well as imbalance of the response in the synthetic range pro-file.The common amplitude error affects the shape of the syn-thetic range pro file signi ficantly,as they tend to defocus the re-sponse in the pro file and increase the magnitudes of the side2Agraphical software system called “Virtual Instruments ”from National In-struments.Fig. 5.Phase of the complex vector I +jQ versus frequency.(a)Linear transformation of the trace of calculated phases to a linear phase line,( + )!.(b)Magni fied drawing of (a)showing the trace of calculated phases obtained by cumulating the phase differences 18;...;18;...;18.(c)Nonlinearity of the calculated phases in polar form,where C is the k th complex vector after compensating for the common amplitudedeviation.Fig.6.Amplitude deviations and nonlinear phase errors of the complex vectors due to the imperfection of the system.lobes.Therefore,these common nonlinear phase and amplitude errors need to be corrected.To this end,a new,simple,yet ef-fective,and accurate compensation technique for the common amplitude and nonlinear phase errors of the system has been developed.In the process of correction using the new compensation tech-nique,the complex vector is measured when a metal plate is moved along a track at a fixed frequency.The metal plate has a size of33ft in order to accommodate the sensor ’s lateral reso-lution.From (2),it is seen that these complex vectors will rotate circularly as the/channels are completely balanced during the movement of the metal plate at a fixed frequency.The mag-nitude of the rotating vector is then measured and stored.This procedure is repeated at each frequency step across the oper-ating frequency range.These measured magnitudes are used as reference data to compensate for the common amplitude error.After compensating for the common amplitude error,the common nonlinear phase error needs to be corrected.Fig.5PARK AND NGUYEN:ULTRAWIDE-BAND MICROW A VE RADAR SENSOR5Fig.7.Flowchart describing the compensation procedure for the amplitude deviations and nonlinear phase errors.depicts the calculated phases of the complexvector,,over a frequency range.Cumulating the phase dif-ference between two consecutive frequencysteps,,unwraps the calcu-lated phases and makes it easy to draw the trace of the calculated phases as shown in Fig.5(a)and (b).Fig.5(c)shows that the rate of rotation of the complex vector is not constant due to the non-linear phaseerror .After drawing an appropriate linear phase line as shown in Fig.5(a),the nonlinear phaseerrors is then determined by subtracting the linear phase line from the trace ofthe calculated phases.The nonlinear phaseerrorsat all the frequency steps for the metal plate are stored in memory and used as reference data to compensate for the nonlinear phase error of an actual target.Consequently,after corrected for the nonlinear phase error,the complex vector is obtainedas(4)Fig.6shows the amplitude deviations and nonlinear phase er-rors of the measured vectors for the metal plate in the frequency band of interest.In order to compensate for the common ampli-tude and nonlinear phase errors occurred in the complex vec-tors measured for the targets,the reference data obtained for the metal plate described earlier are applied to these measuredvectors.The stored reference data()for the common amplitude errors are normalized to the maximum value of the reference data and inversed.The inverted values are stored into memory as compensation factors for the common amplitude de-viation.These factors are multiplied with the new complex vec-tors collected from the targets.The stored referencedatafor the common nonlinear phase errors are subtracted from the ex-tracted phases of the new complex vectors collected from thetargets.The procedure of compensation for the common am-plitude deviations and nonlinear phase errors is illustrated in a flowchart shown in Fig.7.Fig.8shows the normalized/outputs of the quadrature de-tector before and after compensating for the common amplitude and nonlinear phase errors.The simulation result for a (fixed)point target is shown in Fig.9,which shows that the developed compensation method for the common errors not only reduces but also balances the side lobes of the synthetic range pro file.Reduction of the side lobes reduces the possibility of masking the responses from adjacent targets and,hence,facilitating their detection.Balancing the side lobes increases the possibility and accuracy in identifying the target.Upon compensating for the errors intheand channels,thedigitaland components are combined into a complex vectorfor each frequency step.Anarray consistingofcomplex vectors correspondingto frequency steps is then formedas(5)where.Adding zeros to the complex vectorarray generates a newarrayof ele-ments.This zero padding is needed to improve the accuracy of the range and the speed of IDFT using the fast Fourier transform(FFT).Applying -point IDFT on thearraythengives (6)6IEEE SENSORSJOURNALFig.8.Normalized I/Q(a)before and(b)after compensating for the amplitude deviations and nonlinear phaseerrors.Fig.9.Synthetic range profile obtained from a target,whose main peak indicates the target location(a)before and(b)after compensating for amplitude deviations and nonlinear phase errors.Simplifying and rearranging(6)gives the magnitude response of the synthetic range profileas(7)where is the frequency stepand is the speed of light inair.From(7),the rangeinformation of a target can be derived.It is noted that an appropriate window function is also used toreduce the side lobes of the synthetic range profile,which mightmask other profiles produced by multiple targets.This windowfunction can be selected based on individual target responses.IV.M EASUREMENTA pavement sample was constructed with two layers in awooden box of3636in The top layer is asphalt having a thick-ness of2.6–2.7in.The bottom layer is base.It has a thickness of4.1in and isfilled with limestone.The sensor’s antennas werepointed directly onto the sample through air.Fig.10illustrates the single reflected signals at the interfacesbetween the sample’s layers.Multiple reflections in thesampleFig.10.Sketch of the pavement sample in a wooden box together with theincident and reflected waves.E is the incident wave;E,E,and E arethe reflected waves at the interfaces between layers0and1,layers1and2,andlayer2and the wooden box,respectively.were ignored,as they are typically very small compare to thesingle reflected signal.It is assumed that the layers are homo-geneous and have negligible loss.These assumptions are notcorrect for practical pavement materials.However,as will beseen later,accurate measured results,for practical engineeringpurposes,were achieved for a practical pavement sample,thusjustifying the supposition.With reference to Fig.10,the singlereflectedwave is givenas(8)where represents the incidentwave;is the reflectioncoefficient at the interface between layers1and2;andandare the transmission coefficients from layers0to1and1to0,respectively.The relative dielectricconstant oflayer(,1,2)can be derived,assuming an incident wavefrom layer tolayer,as(9)where is the relative dielectric constant of layerandis the reflection coefficient at the interface between layersand.The reflection coefficient between the air and asphaltlayer can be expressedas(10)PARK AND NGUYEN:ULTRAWIDE-BAND MICROW A VE RADAR SENSOR7Fig.11.Synthetic range pro files obtained from a metal-plate target and the pavement sample.where is the amplitude of a wave re flected from a metal plate,which is approximately equal to the amplitude of the in-cidentwave.is the amplitude of the re flected wave at the interface between the air and asphalt layer.The relative di-electricconstantof the asphalt layer can be obtained from (9)and (10).The re flection coef ficient at the interface between the asphalt and base layers is obtained from (8)as(11)where represents the amplitude of the re flected wave at the asphalt ing (9)and (11),the relative dielectric constant of the base layer can be calculatedas(12)The thickness of layer can be derived from (7)in Section IIas(13)where is the range cell number between layers andlayer .It should be mentioned here that this procedure can also be applied for structures containing more than three layers.Fig.11shows the synthetic range pro files obtained from the pavement sample and the metal plate.Table I shows the mea-sured parameters of the pavement sample along with the ac-tual values.The measured thickness of each layer agrees well with the actual values.Note that the theoretical relative dielec-tric constants of the sample ’s asphalt and base materials are not known.Several values of dielectric constants for asphalt and base can be found from literature but these are not used for com-parison here since 1)the reported values vary over a wide range,2)they are at frequencies different from those used here,and 3)the reported asphalt and base materials are not the same as those in our pavement sample.It should be noted that the sensor can also be used to as-sess other samples that have more complicated layer structuresTABLE IC OMPARISON B ETWEEN A CTUAL AND M EASURED DATAwithin the constraints of penetration depth and resolutions pro-vided by the sensor.V .C ONCLUSIONA new integrated-circuit UWB microwave stepped-fre-quency radar sensor including antennas has been successfully developed and demonstrated for accurate measurement of pavement subsurface.A new correction procedure was also developed to compensate for the imperfection of the system.This simple compensation method reduces and balances the side lobes of the synthetic range pro file representing the target,which helps reduce the potential masking of adjacent targets and facilitate and increase accuracy for identi fication of the target responses.The sensor ’s operation has been veri fied by measuring layer thickness and relative dielectric constants of a pavement sample.The system represents the first 0.6–5.6-GHz portable microwave SFR developed for pavement subsurface sensing and is useful not only for nondestructive evaluation of pavements and other highway structures,but also for other subsurface sensing applications like detection of buried mines.A CKNOWLEDGMENTThe authors would like to thank T.Scullion and L.Gustavus of the Texas Transport Institute for providing the pavement sample.8IEEE SENSORS JOURNALR EFERENCES[1]R.C.Pippert,K.Soroushian,and R.G.Plumb,“Development ofa ground-penetrating radar to detect excess moisture in pavementsubgrade,”in Proc.2nd Government Workshop on GPR—Advanced Ground Penetrating Radar:Technologies and Applications,Oct.1993, pp.283–297.[2] ngman,S.P.Dimaio,B.E.Burns,and M.R.Inggs,“Develop-ment of a low cost SFCW ground penetrating radar,”in Proc.IEEE Geo-science and Remote Sensing Symp.,1996,pp.2020–2022.[3] D.Huston,J.O.Hu,K.Maser,W.Weedon,and C.Adam,“GIMAground penetrating radar system for monitoring concrete bridge decks,”J.Appl.Geophys.,vol.43,pp.139–146,2000.[4]J.Park and C.Nguyen,“An ultra-wideband microwave radar sensor forcharacterizing pavement subsurface,”in Proc.IEEE IMS,Jun.2003,pp.1443–1446.[5] D.R.Wehner,High Resolution Radar.Norwood,MA:Artech House,1995.[6] E.A.Theodorou,M.R.Gorman,P.R.Rigg,and F.N.Kong,“Broad-band pulse-optimized antenna,”Proc.Inst.Elect.Eng.H,vol.128,pp.124–130,Jun.1981.[7]J.D.Cermignani,R.G.Madonna,P.J.Scheno,and J.Anderson,“Mea-surement of the performance of a cavity backed exponentiallyflared TEM horn,”Proc.SPIE,vol.1631,pp.146–154,1992.[8]M.Kanda,“The effects of resistive loading of“TEM”horns,”IEEEput.,vol.EC-24,no.2,pp.245–255,May 1982.[9] C.Nguyen,J.Lee,and J.Park,“Novel ultra-wideband microstrip quasi-horn antenna,”Electron.Lett.,vol.37,no.12,pp.731–732,2001.[10] F.E.Churchill,G.W.Ogar,and B.J.Thompson,“The correction ofI and Q errors in a coherent processor,”IEEE Trans.Aerosp.Electron.Syst.,vol.AES-17,no.1,pp.131–137,Jan.1981.[11]J.G.Proakis and D.G.Manolakis,Digital Signal Processing.Engle-wood Cliffs,NJ:Prentice-Hall,1996.Joongsuk Park received the B.S.degree in electricalengineering from Yonsei University,Seoul,Korea,in1988and the Ph.D.degree in electrical engineeringfrom Texas A&M University(TAMU),College Sta-tion,in2003.From1988to1996,he was with LG ElectronicsCompany,Seoul,as a Senior Researcher,developingdigital televisions.From1998to2003,he wasa Research Assistant in electrical engineering atTAMU,developing microwave and millimeter-waveintegrated circuits and systems.In2004,he joined the Digital Media Research Center,Samsung Electronics,Suwon,Korea, as a Principal Engineer,where he is currently developing a wireless home networking system.His research interests include the design of microwave and millimeter-wave integrated circuits and antennas and wireless sensors and communicationsystems.Cam Nguyen(F’03)joined the Department of Elec-trical Engineering,Texas A&M University(TAMU),College Station,in December1990,where he nowholds the position of Texas Instruments EndowedProfessor.From2003to2004,he was ProgramDirector at the National Science Foundation,respon-sible for research programs in RF electronics andwireless technologies.From1979to1991,he heldvarious positions in industry.He was a MicrowaveEngineer with ITT Gilfillan Company,a Member ofTechnical Staff with Hughes Aircraft Company(now Raytheon),a Technical Specialist with Aeroject ElectroSystems Company, a Member of Professional Staff with Martin Marietta Company(now Lock-heed-Martin),and a Senior Staff Engineer and Program Manager at TRW(now Northrop Grumman).While in industry,he led numerous microwave and mil-limeter-wave activities and developed many microwave and millimeter-wave integrated circuits and systems up to220GHz for communications,radar,and remote sensing.His research group at TAMU focuses on CMOS RF ICs and systems,microwave and millimeter-wave ICs and systems,and UWB devices and systems for various engineering applications.Particularly,his research group has been at the forefront of developing UWB ICs and systems for radar and wireless communications and pioneered the development of microwave and millimeter-wave integrated circuit systems for sensing applications.He has published more than140papers,one book,and several book chapters.Dr.Nguyen is the Founding Editor-in-Chief of Subsurface Sensing Technolo-gies and Applications:An International Journal.。