2009-PIER-A HETERODYNE SIX-PORT FMCW RADAR SENSOR
【精品】翻译综合
一个抑制肿瘤的连续模型-------艾丽斯H伯杰，阿尔弗雷德G. Knudson 与皮埃尔保罗潘多尔菲今年，也就是2011 年，标志着视网膜母细胞瘤的统计分析的第四十周年，首次提供了证据表明，肿瘤的发生，可以由两个突变发起。
这项工作提供了“二次打击”的假说，为解释隐性抑癌基因（TSGs）在显性遗传的癌症易感性综合征中的作用奠定了基础。
然而，四十年后，我们已经知道，即使是部分失活的肿瘤抑制基因也可以致使肿瘤的发生。
在这里，我们分析这方面的证据，并提出了一个关于肿瘤抑制基因功能的连续模型来全方位的解释肿瘤抑制基因在癌症过程中的突变。
虽然在1900 年之前癌症的遗传倾向已经被人认知，但是，是在19 世纪曾一度被忽视的孟德尔的遗传规律被重新发现之后，癌症的遗传倾向才更趋于合理化。
到那时，人们也知道，肿瘤细胞中的染色体模式是不正常的。
接下来对癌症遗传学的理解做出贡献的人是波威利，他提出，一些染色体可能刺激细胞分裂，其他的一些染色体 a 可能会抑制细胞分裂，但他的想法长期被忽视。
现在我们知道，这两种类型的基因，都是存在的。
在这次研究中，我们总结了后一种类型基因的研究历史，抑癌基因（TSGs），以及能够支持完全和部分失活的肿瘤抑制基因在癌症的发病中的作用的证据。
我们将抑制肿瘤的连续模型与经典的“二次打击”假说相结合，用来说明肿瘤抑制基因微妙的剂量效应，同时我们也讨论的“二次打击”假说的例外，如“专性的单倍剂量不足”，指出部分损失的抑癌基因比完全损失的更具致癌性。
这个连续模型突出了微妙的调控肿瘤抑制基因表达或活动的重要性，如微RNA（miRNA）的监管和调控。
最后，我们讨论了这种模式在┲⒌恼锒虾椭瘟乒 讨械挠跋臁！岸 未蚧鳌奔偎?第一个能够表明基因的异常可以导致癌症的发生的证据源自1960 年费城慢性粒细胞白血病细胞的染色体的发现。
后来，在1973 年，人们发现这个染色体是是第9 号和第22 号染色体异位的结果，并在1977 年，在急性早幼粒细胞白血病患者中第15 号和第17 号染色体易位被识别出来。
TheNewcastle-OttawaScale（NOS）forAssessingthe…
Development: Identifying Items
• Identify ‘high’ quality choices with a ‘star’
• A maximum of one ‘star’ for each item within the ‘Selection’ and ‘Exposure/Outcome’ categories; maximum of two ‘stars’ for ‘Comparability’
2. Representativeness of the cases a) consecutive or obviously representative series of cases ♦ b) potential for selection biases or not stated
3. Selection of Controls a) community controls ♦ b) hospital controls c) no description
Selection
1. Is the case definition adequate? a) yes, with independent validation ♦ b) yes, eg record linkage or based on self reports c) no description
Bias and Confounding
Globally networked risks and how to respond
BOX 1
Risk, systemic risk and hyper-risk
According to the standard ISO 31000 (2009; /iso/ catalogue_detail?csnumber543170), risk is defined as ‘‘effect of uncertainty on objectives’’. It is often quantified as the probability of occurrence of an (adverse) event, times its (negative) impact (damage), but it should be kept in mind that risks might also create positive impacts, such as opportunities for some stakeholders.
Stat 6-dependent myeloid derived suppressor cells physical injury nitric oxide endotoxin.
O RIGINAL A RTICLESStat6-Dependent Induction of Myeloid Derived Suppressor Cells After Physical Injury Regulates Nitric Oxide Response to Endotoxin Veronica Munera,MD,*Petar J.Popovic,MD,PhD,*Jodie Bryk,BS,†John Pribis,BS,*David Caba,MD,MS,MPH,*Benjamin M.Matta,BS,*Mazen Zenati,MD,MPH,PhD,*and Juan B.Ochoa,MD,FACS*Objective:To delineate the role of T-helper2(Th2)cytokines in the induction of trauma induced myeloid suppressor cells(TIMSC)and the regulation of nitric oxide production.Background:Trauma induces myeloid cells that express CD11bϩ/Gr1ϩand arginase1and exhibit an immune suppressing activity.This article explores the mechanisms that induce TIMSC and the effects on nitric oxide production in response to endotoxin.Methods:TIMSC were studied in response to Th2cytokines and a subse-quent challenge to endotoxin.The role of Th2cytokines was studied in STAT6Ϫ/Ϫmice.Accumulation of TIMSC in spleens was studied using flow cytometry and immunhistochemistry.Plasma was recovered to measure accumulation of nitric oxide metabolites.Results:TIMSC accumulated in the spleen of injured mice and were particularly sensitive to IL-4and IL-13with large inductions of arginase activity.Significant blunting in both the accumulation of TIMSC in the spleen and induction of arginase1was observed in STAT6Ϫ/Ϫmice after physical injury.Accumulation of nitric oxide metabolites to endotoxin was observed in STAT6Ϫ/Ϫmice.Conclusion:This study shows that induction of CD11bϩ/Gr1ϩcells after physical injury play an essential role in the regulation of nitric oxide production after a septic challenge.The accumulation and induction of arginase1in TIMSC is Th2cytokine dependent.To our knowledge,the role of TIMSC in the regulation of nitric oxide is a novelfinding.This observa-tion adds to the possibility that TIMSC could play an important role in immunosuppression observed after physical injury.(Ann Surg2010;251:120–126)P hysical injury,either by trauma or through surgical intervention, is associated with immune dysfunction,leading to increased risk of infectious complications.Infections after physical injury increase morbidity,length of hospital stay,costs,and mortality.1–3Recent studies have proposed that alterations in innate and adaptive immune responses may contribute to increased susceptibility to infection after physical injury.4,5In particular,T lymphocyte dysfunction,as typically observed after physical injury,is one important mechanism of immune dysfunction.4–6Arginine,a nonessential amino acid,plays an important role in the multiple metabolic pathways related to immune competence, and is essential for T lymphocyte function.7–9Arginine is also the sole substrate for the production of nitric oxide(NO)by nitric oxide synthases.10,11Several investigators including us have shown that physical injury leads to a significant decrease in arginine availabil-ity,potentially compromising both T lymphocyte function and NO production.12,13It is therefore not surprising that a decrease in NO production is observed in patients after physical injury,even those that develop sepsis,a condition otherwise associated with excessive NO production.14,15Similarfindings have been reported in a murine trauma model,12,16.The mechanisms that lead to decreased NO production in response to infection after trauma are unknown.Recently,our group reported that physical injury was associ-ated with an induction of CD11bϩ/Gr1ϩimmature myeloid cells that migrate to the spleen.These cells express high levels of arginase 1,an enzyme that metabolizes arginine into urea and ornithine, effectively depleting arginine from the surrounding environ-ment.6Through arginine depletion,CD11bϩ/Gr1ϩcells may exert a suppressive effect on T lymphocytes.It is for this reason we describe these particular cells as trauma induced myeloid suppressor cells(TIMSC).The mechanism of TIMSC up-regulation after physical in-jury,and subsequent arginase1induction is unknown.In vitro, T-helper2(TH2)cytokines are known to induce arginase1expres-sion in some cell lines.7,8Reports suggest that TH2cytokines, specifically IL-4and IL-13which signal through signal transducer and activator of transcription-6(STAT-6),may play an important role in immune suppression after injury.STAT-6is thought to play an important role in IL-4and13signal transduction.17–20Hence,we hypothesized that TH2cytokines were involved in the recruitment and stimulation of arginase1expressing TIMSC and that,through arginine depletion,NO production would be compromised.This article successfully demonstrates the essential role of TH2cytokines in the induction of arginase1expression in TIMSC,the migration of TIMSC to the spleen,and the subsequent regulation of NO production.MATERIALS AND METHODSThe current experimental protocol was approved by the Uni-versity of Pittsburgh Institutional Animal Care and Use Committee and Division of Laboratory Animal Research.ReagentsIL-4and IL-13were purchased from R&D Systems Inc. (Minneapolis,MN)and Escherichia coli endotoxin(Lipopolysac-charide͓LPS͔)from Sigma-Aldrich Inc(St.Louis,MO).Cell culture media was obtained from BioWhittaker RPMI-1640(Lonza, Walkersville,MD).Media was supplemented with10%fetal bovine serum(Hyclone,Logan,UT),L-Glutamine(2000M,Gibco-In-vitrogen),penicillin G(100units/mL,Invitrogen,Carlsbad,CA), and streptomycin(100g/mL,Invitrogen).All other reagents were obtained from Sigma-Aldrich Inc.(St.Louis,MO)unless otherwise stated.Mouse Animal ModelSix to8-week-old male BALBc(wild-type͓WT͔)and STAT6Ϫ/Ϫmice were obtained from Jackson Laboratories(BarFrom the*Department of Surgery,University of Pittsburgh School of Medi-cine,Pittsburgh,PA;and†University of Pittsburgh School of Medicine,Pittsburgh,PA.Supported by RO1NIH-NIGMS GM-065914.Reprints:Juan B.Ochoa,MD,FACS,F1264PUH-UPMC,200Lothrop St,Pittsburg,PA15213.E-mail:ochoajb@.Copyright©2009by Lippincott Williams&WilkinsISSN:0003-4932/10/25101-0120DOI:10.1097/SLA.0b013e3181bfda1cAnnals of Surgery•Volume251,Number1,January2010 120|Harbor,MN).The animals were kept under12hours light/dark cycles at a temperature of20°C to22°C in a pathogen-free facility. Food and water were available ad libitum.The mice were allowed an acclimation period of2weeks.Mice were randomized into2groups:(1)Controls,receiving anesthesia alone;(2)Trauma,undergoing physical injury.In the trauma group,after the administration of anesthesia(ketamine430 mg/kg and xylazine34mg/kg͓Phoenix Pharmaceutical͔),a midline laparotomy incision was made under sterile conditions.Intra-ab-dominal contents were teased for2minutes,taking care not to create injury to the viscera.The incision was closed in2layers,and animals were maintained under a heat lamp until fully recovered. Blood samples and spleens were collected.To collect samples,all animals were once again anesthetized at varying time points accord-ing to the procedure performed.Isolation of CD11b؉Cells From SpleensA single-cell suspension was prepared from the spleens ofcontrol and trauma animals.Erythrocytes were lysed using MQH2O,and splenocytes were washed in MACS buffer(1ϫPBS supple-mented with2mM EDTA and0.5%BSA).CD11bϩcells were isolated using corresponding MACS magnetic microbeads(Miltenyi Biotec).Cells were either cultured or frozen as pellets.CD11b؉Cell CultureIsolated CD11bϩcells were plated1ϫ106cells per well in 24well-plates in regular media for all procedures.Cells were stimulated with IL-4(0.3–30ng/mL)and IL-13(1–5ng/mL)for24, 48,and72hours.CD11b؉Cell Protein Extract PreparationTotal cell protein extracts were prepared by lysing washed isolated CD11bϩcell pellets in10L of lysis buffer/106cells (Triton X-1000.5%,HEPES50mmol/L͓pH7.55,Gibco͔,NaCl 150mmol/L,sodium orthovanadate1mmol/L,trypsin-chymotrysin inhibitor100g/mL,leupeptin50g/mL͓Roche͔,aprotonin50g/mL͓Roche͔,PMSF2mM)containing50g/mL aprotonin,50g/mL leupeptin(Roche),100g/mL trypsin-chymotrypsin inhib-itor,and2mM PMSF(Sigma-Aldrich).Cell pellets were incubated on ice for7minutes with red blood cell lysing buffer and then centrifuged at3000g(12,000RPM)for1minute at4°C.Total protein concentration was determined by the Bradford method.21 Prepared protein lysates were aliquoted and stored atϪ80°C until used for arginase activity assay or Western blot analysis.Arginase Activity AssayArginase activity in protein lysates of splenocyte subsets was measured from the conversion of L-arginine to L-ornithine according to the technique described by Kornarska and Tomaszewski.22Avail-able arginase was activated by the addition of10mM MnCl2(25L)to cell protein lysate(25L)and incubation at55°C for20 minutes.Carbonate buffer(150L,100mM,pH10)was then added along with100mM L-arginine(50L)to initiate the reaction, and incubated at37°C.Arginase activity was stopped after10 minutes by adding glacial acetic acid(750L).Ninhydrin solution (250L,2.5g of ninhydrin,40mL of6M phosphoric acid,and60 mL of glacial acetic acid)was added,and the samples and standards were boiled at90°C to100°C for1hour.Standards were created by using known amounts of L-ornithine from8to250nmol,and all reagents were added to standards as a control.Samples were cooled and colorimetric reaction measured with a spectrophotometer at515 nm(Spectramax340;Molecular Devices).Arginase assay was linear with time and is presented as nanomoles of ornithine gener-ated per minute per milligram of protein.Western Blot Analysis of Arginase1Total protein from isolated CD11bϩcells was separated by 12.5%SDS-acrylamide gel for45minutes at200V and electro-phoretically transferred to a nitrocellulose membrane(Invitrogen Life Technologies).The protein extract from the mouse liver,which is constitutively rich in arginase1,served as a positive control.The nitrocellulose membrane was blocked with5%nonfat dry milk in TBST(2.42g of Tris,8g of NaCl,and0.5mL of Tween20in1L of MQH2O)at4°C,overnight.The membrane was incubated with chicken IgG1antimouse ARG1primary antibody(Ab,donated by Dr.S.Morris,Jr.,University of Pittsburgh,Pittsburgh,PA),diluted at1/50,000in1%nonfat dried milk,1%BSA in TBST,for1hour at room temperature.23The membrane was washed and the second-ary peroxidase-conjugated rabbit antichicken IgG Ab(Jackson Im-munoResearch Laboratories)was diluted1/5,000and applied for1 hour at room temperature.As an internal control for total protein concentration,the membrane was stripped and washed,and goat antimouse-actin Ab(Santa Cruz Biotechnology)was diluted1/500 and applied for1hour,at room temperature.After washing,the membrane was incubated with goat antidonkey IgG Ab(Santa Cruz Biotechnology),diluted1/5000,and applied for1hour at room temperature.For both arginase1and-actin,immunoreactive pro-tein was visualized using enhanced luminol reagent and oxidizing reagent(Pierce).Prestained Kaleidoscope molecular mass marker (Bio-Rad)was used to determine the molecular weight of immuno-reactive bands.Immunohistochemical StainingImmunohistochemistry was performed on spleens harvested from control or24hours after trauma(surgery).Spleens were frozen in the presence of Tissue-Tek pound(Sakura Finetek; Torrance,CA)and stored until used atϪ70°C.Cryostat sections4m in width were used for immunohistochemical evaluation using rat IgG2b,k antimouse Ly-6G(GR1)monoclonal antibody(1/50, eBioscience)and goat polyclonal antimouse CD11b antibody(1/50, BD Pharmingen).Endogenous Biotin was blocked using Avidin Biotin Blocking Kit(Vector Laboratories)incubation for the avidin at room temperature15minutes followed by PBS rinse then a15 minutes room temperature incubation of biotin.Nonspecific binding was blocked with10%goat serum(Vector Laboratories)for15 minutes at room temperature.The sections were then incubated in a humidified chamber at room temperature for2hours with the primary CD11b(1:50).After3washes in Tween-20/PBS buffer for 5minutes,the sections were incubated for30minutes with biotin-ylated goat-anti Rat secondary antibody(1:100BD Pharmingen). After3washes in Tween-20/PBS buffer for5minutes,the sections were incubated for30minutes with HRP conjugated ABC(Vector Laboratories).The color reaction was developed for5minutes using AEC chromogen(Scytek laboratories).Alternatively,sections were incubated1.5hours with the primary antibody to Ly6G(GR-1,1:50) followed by quenching in0.3%H2O2solution for15minutes and blocking by avidin/biotin kit(Vector Laboratories).After washing in Tween-20/PBS buffer,the sections were incubated with the second-ary biotinylated antigoat antibody(Jackson Laboratories)at dilution 1:200for30minutes.After3washes in Tween-20/PBS buffer for5 minutes,the sections were incubated for30minutes with HRP conjugated ABC(Vector Laboratories).The color reaction was developed for5minutes using AEC chromogen(Scytek laborato-ries).After thefinal wash in distilled water,crystal mount is applied to tissue sections which were then air-dried,cleared,and cover-slipped using Permount(Fisher Scientific).Control slides included an irrelevant isotype-matched antibody in place of the primary antibody:purified rat IgG2b,isotype standard(1/500,BD Bio-sciences-Pharmingen),goat IgG1(1/500,BD Biosciences Pharmin-Annals of Surgery•Volume251,Number1,January2010Induction of Myeloid Derived Suppressor ©2009Lippincott Williams&Wilkins |121gen),or PBS in place of the primary antibody to evaluate nonspecific staining.Control sections generally demonstrated negligible levels of endogenous background.Flow Cytometry AnalysisHarvested cells were washed in immunofluorescence assaybuffer(1ϫPBS supplemented with0.1%BSA and0.1%NaN3)andstained with appropriately diluted antibodies directly conjugated with FITC or PE according to the standard procedure,and followed byfixation in1%paraformaldehyde.Antibodies used for staining were the following:FITC-labeled antimouse CD11b and PE-labeled GR1(BD Pharmingen).All staining procedures were conducted on ice.Fluorescence was measured using an Epics XLflow cytometer (Beckman Coulter).Plasma Nitric Oxide AssayNitrite,the metabolic end product of NO metabolism,in the supernatants was measured using Griess reaction after nitrate reduc-tase,using Nitrate/Nitrite Colorimetric Assay Kit(Cayman Chem-ical).Plasma samples were ultra-filtered using a10kDa molecular weight cut-offfilter.Filtered plasma(30L)in duplicate was added to aflat bottom96well-plate.Standard curve prepared using serial nitrate dilutions.Enzyme cofactor(10L)was added to both samples and standards,followed by nitrate reductase(10L)and incubated at room temperature for3hours.After1%sulfanilamide (50L)was added followed by0.1%naphthlethylenediamine di-hydrochloride(50L)in2.5%phosphoric acid,mixed and incu-bated at room temperature in the dark for10minutes.Absorbance was read at540nm on a Beckman DU640spectrophotometer. Nitrite concentration was calculated from the standard curve pre-pared from serial sodium nitrate dilutions.Statistical AnalysisFor a single comparison of2groups,student’s t test was used when normality was met.Analysis of variance(1-way)was per-formed to evaluate the significant differences between experimental groups.A post hoc multiple comparison procedure was used when the omnibus test was significant.We selected a Scheffe test to compute the new critical value for an F test conducted when comparing any2groups from the larger analysis of variance.The differences were considered statistically significant at a P value of less than0.05.Data is presented as meanϮSEM in the text and in all graphs.STATA was used for the analysis.RESULTSTrauma Induced Myeloid Derived Suppressor Cells are Sensitized by Physical Injury to Respond toT-Helper2CytokinesTo determine whether TIMSC were susceptible to stimulation by TH2cytokines,TIMSC were isolated from WT mice and cultured for48hours in the presence of increasing concentrations of IL-4. Naive CD11bϩcells(cells harvested from control animals not subjected to physical injury)were also cultured under the same conditions for comparison.Arginase activity was2.4-fold higher in TIMSC than in naive CD11bϩcells and further increased in both types of cells proportional to the increase in IL-4.The increase with varying doses of IL-4was significantly higher(2-to6-fold)in TIMSC than in naive CD11bϩcells(PϽ0.05).Average arginase activity in maximally stimulated naive CD11bϩand TIMSC was 2825.9Ϯ286.8nmol/min/mg and4589.9Ϯ92.0nmol/min/mg, respectively.Similar effects were observed after IL-13stimulation (data not shown).These experiments therefore demonstratethat FIGURE1.IL-4induction of arginase in TIMSC.CD11bϩcells(TIMSC)were isolated from WT mice spleens24hours after trauma.Naive CD11bϩcells from nontraumatized WT mice spleens were used as controls.A,Arginase activity in TIMSC was dependent on the concentration of IL-4provided.Trauma induced myeloid suppressor cells showed a2-to6-fold increase in arginase activity above that of control naive CD11bϩcells(PϽ0.05).B,shows the Western blot for arginase1expression24, 48,and72hours after in vitro stimulation with IL-4(10ng/mL).Equal amounts of protein from TIMSC and control naiveCD11bϩcells were blotted and detected using arginase1antibody with liver protein lysate used as a positive control.Argi-nase1is a35-to38-kDa large protein.-actin expression was used for semiquantative control.Arginase1protein expression was3.5and3.7times higher for48and72hour cell culture experiments in TIMSC compared with control naive CD11bϩcells.Munera et al Annals of Surgery•Volume251,Number1,January2010 122|©2009Lippincott Williams&WilkinsCD11b ϩcells respond to T-helper 2cytokines with an induction of arginase activity,and that TIMSC appear to be more sensitive to this stimulus than naive CD11b ϩcells (Fig.1A).Selective induction of arginase 1protein expression in TIMSC was investigated by Western Blot analysis,using -actin as a control.Arginase 1protein in nonstimulated naive CD11b ϩcells was not detected,implying that this enzyme is not constitutively expressed.TIMSC stimulated with IL-4showed a progressive in-crease in arginase 1protein expression over time.Arginase 1protein expression was induced by IL-4both in naive CD11b ϩcells and TIMSC.However,there was increased arginase 1protein expression in TIMSC for the same concentration of IL-4when compared with that observed in naive CD11b ϩcells.Densitometry analysis showed that by 48and 72hours after IL-4stimulation,the relative expres-sion of arginase 1protein was respectively 3.5and 3.7-fold higher in TIMSC than in naive CD11b ϩcells (Fig.1B).These results mir-rored the increased arginase activity observed in TIMSC.STAT6Signaling is an Important Factor in the Induction of Arginase After TraumaTo further determine the role of T H 2cytokines in the induc-tion of arginase 1,we evaluated the effects of blocking the signal transducer and activator of transcription 6(STAT6)pathway using STAT6Ϫ/Ϫmice.STAT6is essential for the transduction of the T H 2cytokines signal.Thus,we hypothesized that the absence of a T H 2cytokine signal in STAT6Ϫ/Ϫmice would result in blunted arginase 1induction after physical injury.Naive CD11b ϩand TIMSC cells were isolated from the spleens of WT and STAT6Ϫ/Ϫanimals 24and 48hours after trauma,and arginase activity was measured.Arginase activity in TIMSC increased appropriately across time in WT mice,from 13.8Ϯ1.7nmol/min/mg in control to 69.5Ϯ2.2nmol/min/mg at 24hours and 165.1Ϯ8.3nmol/min/mg at 48hours;showing a 5-and 12-fold increase over control values,respectively (P Ͻ0.01).Arginase activity was significantly blunted in STAT6Ϫ/Ϫmice (P Ͻ0.01),increasing to 11.9Ϯ1.8nmol/min/mg in control compared with 13.7Ϯ1.6nmol/min/mg at 24hours and 42.2Ϯ3.9nmol/min/mg at 48hours after trauma (Fig.2A).A modest increase in arginase activity 48hours after trauma was observed in STAT6Ϫ/Ϫmice,though this was still 4-fold lower than in WT mice (P Ͻ0.01).Western blot 48hours after trauma,confirmed significant blunting of arginase 1protein expression in STAT6Ϫ/Ϫmice (Fig.2B).These observations demonstrate that T-helper 2cytokines play a central,but not exclusive role,in the induction of arginase 1in TIMSC.T-Helper 2Cytokines are Partially Responsible for TIMSC Accumulation in the SpleenArginase 1expression in splenic tissues after physical injury is the result of both the induction of arginase 1expression and the influx of large numbers of TIMSC into the spleens.We wanted to determine whether,in addition to induction of arginase 1expression,T H 2cytokines were also essential for the accumulation of TIMSC in the spleen.To test this hypothesis,spleens were isolated from WT and STAT6Ϫ/Ϫanimals 48hours after trauma.Splenic cells were harvested and then analyzed by flow cytometry for TIMSC pheno-typic markers.As expected,significant accumulation of TIMSC in the spleen of WT mice was evident 48hours after trauma,increasing from a mean 2.7%Ϯ0.6%of total splenocytes in controls to 7.6%Ϯ0.6%after trauma (P Ͻ0.01).The accumulation of TIMSC in STAT6Ϫ/Ϫwas significantly lower (P Ͻ0.01).The accumulation of TIMSC in splenic tissues of STAT6Ϫ/Ϫmice was not statisti-cally significant from baseline (2.4%Ϯ0.45%–4.3%Ϯ0.6%,P ϭ0.19,Figs.3A,B).Accumulation of TIMSC in Marginal Zones of the Spleens is T-Helper 2Cytokine DependentWe have previously reported that TIMSC preferentially ac-cumulate in the marginal zones of the spleen in proximity to T lymphocytes,within hours after physical injury.Immunohistochem-istry was performed to determine whether T H 2cytokines affected the distribution of TIMSC in the different zones of the spleen in addition to inducing TIMSC migration to the spleen.Figure 3C shows representative slides that demonstrate a decrease in the accumulation of TIMSC cells in STAT6Ϫ/Ϫmice compared with WT animals.T-Helper 2Cytokine Dependent Induction of Arginase 1Plays an Important Role in Blunting Accumulation of NO Metabolites After TraumaAfter determining that T H 2cytokines induce arginase 1ex-pression in TIMSC and promote their migration to the spleen,we next wanted to determine if these events would subsequently limit NO production.LPS endotoxin (5mg/kg)was injected intraperito-neally to induce NO production 24hours after physical injury.Control mice received an intraperitoneally injection of LPS in the absence of physical injury.Plasma was harvested within 16hours and NO metabolites were measured.Control mice exhibited a large accumulation of circulating NO metabolites after injection of LPS.Similar to previous observa-tions reported in humans,mice subjected to physical injury exhibited a greater than 80%reduction in the accumulation of NO metabolites in response to LPS when compared with nontraumatized controls (434.8Ϯ25.7M in control to 80.8Ϯ23.1M after trauma,P Ͻ0.01).To determine whether the presence of TIMSC andincreasedFIGURE 2.Arginase induction after trauma in WT versus STAT6Ϫ/ϪTIMSC.A,CD11b ϩcells (TIMSC)from WT and STAT6Ϫ/Ϫmice were collected at different time points (0,24and 48hours)after trauma and arginase activity was measured.Arginase activity after trauma was significantly blunted in the absence of STAT6(P Ͻ0.01).B,Western blot for arginase 1protein expression in TIMSC collected 48hours after trauma from WT and STAT6Ϫ/Ϫmice was per-formed.Equal amounts of protein were blotted and tested with arginase 1antibody.-actin expression was used for semiquantative control.STAT6Ϫ/ϪTIMSC showed confirma-tory blunting of arginase 1expression after trauma com-pared with WT TIMSC.Annals of Surgery •Volume 251,Number 1,January 2010Induction of Myeloid Derived Suppressor©2009Lippincott Williams &Wilkins |123arginase 1expression played a regulatory role in the accumulation of NO metabolites after trauma,STAT6Ϫ/Ϫmice were injected with LPS.Not surprisingly,the accumulation of NO metabolites after phys-ical injury was restored in STAT6Ϫ/Ϫmice,producing 283.9Ϯ40.1M in response to LPS,3.5-fold higher than in WT mice subjected to trauma (P ϭ0.01,Fig.4).The accumulation of NO metabolites in STAT 6Ϫ/Ϫsubjected to trauma and then given LPS were similar to those observed in WT mice given LPS only in the absence of physical injury (P ϭns).This data supports the hypothesis that TIMSC may play an essential regulatory and systemic role in the production of NO in response to LPS.DISCUSSIONThis article is important,in that we demonstrate a role for trauma induced myeloid suppressor cells in regulating inducible NO synthase activity in response to endotoxin.This observation also helps provide a potential mechanism of how immunonutrition works.In 1991,we reported for the first time,that NO metabolites accumulated in plasma in septic patients and suggested that NO could play an important role in the pathophysiology of sepsis.In contrast,in the same article,we also reported that NO metabolites in traumatized patients failed to accumulate in plasma even when these patients became septic.15In addition,others demonstratedthatFIGURE 3.TIMSC (CD11b ϩ/Gr-1ϩ)cells accumulation in the spleen 48hours after trauma was blunted in STAT6Ϫ/Ϫmice.Migration of TIMSC to the spleen was evaluated with flow cytometry and immunohistochemistry.A,B,Splenocytes were har-vested 48hours after trauma and stained with anti-Gr-1-PE and anti-CD11b-FITC conjugated antibodies for two-color flow cytometry analysis.The percentage of splenic CD11b ϩ/Gr-1ϩcells after trauma was 2.8-fold higher in wild type (WT)control versus WT traumatized mice,but only 1.7-fold higher in traumatized STAT6Ϫ/Ϫmice.Hence,there was increased accumula-tion of TIMSC after trauma that was significantly blunted in the absence of STAT6(P Ͻ0.01).C,Immunohistochemistry was used to evaluate the amount and distribution of Gr-1ϩcells in spleen.Frozen sections of spleens were stained with Gr-1anti-body.Representative slides are shown.A dramatic increase in Gr-1ϩcells was observed in the spleens of WT traumatized mice.These cells were circumscribed within the marginal zones and periarteriolar lymphatic sheaths.In STAT6Ϫ/Ϫmice,this accumulation was diminished.Munera et al Annals of Surgery •Volume 251,Number 1,January 2010124|©2009Lippincott Williams &Wilkinsarginine plasma levels decreased in trauma,but not in sepsis.24The evidence presented in this article is the first to demonstrate that the accumulation of TIMSC is responsible for the biologic observations reported almost 20years ago.Immune dysfunction after trauma or major surgery appears to be a significant cause of increased morbidity and mortality,leading to increased susceptibility to infections and organ failure.Arginine plays a key function in the immune system and is necessary for both T lymphocyte function and the generation of NO.In 2006,our group hypothesized that a state of arginine deficiency observed in trauma could compromise immune func-tion.13Indeed arginine deficiency has been linked to immunosup-pression in several disease states including renal cell carcinoma,leprosy,tuberculosis,sickle cell anemia,and after major surgical interventions.All of these diseases are also associated with in-creased circulating arginase 1activity and the presence myeloid suppressive cell activity.25–27Not surprisingly,the supplementation of arginine at supraphysiologic doses in traumatized patients and those undergoing elective surgery is associated with restoration of T cell function and an increase in circulating NO metabolites.28The main goal of this article was to identify the signals responsible for the up-regulation of arginase 1and the accumulation of TIMSC after physical injury.Activation of the inflammatory response after physical injury leads to the release of multiple cytokines and other substances,such as IL-6,IL-10,TGF-,IL-4,IL-13,catecholamines,and prostaglandins.29–32Of these,T H 2cy-tokines appear to play a major role in the induction of arginase 1in myeloid cells.33–35Our results show that T H 2cytokines induce arginase activity in TIMSC and play a role in their accumulation in the spleen after trauma.Although not presented here,preliminary data generated through gene microarrays shows a progressive in-duction of STAT6mRNA,as well as IL-4and IL-13receptor mRNA early after trauma that may explain increased sensitivity of TIMSC to T H 2cytokines (unpublished observations).The observation that arginase 1expression was significantly blunted after trauma in STAT6Ϫ/Ϫmice suggests that arginase 1could possibly regulate NO production.In vitro,this hypothesis has been confirmed.We found that the accumulation of NO metabolites to a septic stimulus was restored by blocking arginase 1induction by T H 2cytokines using a STAT6Ϫ/Ϫrodent trauma model.To our knowledge,this is the first evidence that arginase 1express-ing TIMSC regulate production of NO in response septic stimulus after trauma.It is important to note that while our group has specifically described TIMSC in trauma,similar cells have been identified under other conditions of acute and chronic inflammation and in cancer,and are collectively called myeloid derived suppressor cells (MDSC).36–38There is considerable heterogeneity in the cells,although they do share some basic characteristics.Most identified cells are immature in origin and express arginase 1.38,39Some of the described MDSC have been shown to migrate to the spleen.Inter-estingly,in a breast cancer rodent model,it was shown that STAT6Ϫ/Ϫmice exhibit decreased numbers of MDSC after tumor removal.40In the same study,STAT6Ϫ/Ϫmice exhibited increased production of NO and decrease mortality compared with WT mice,likely due to a better immunologic response.In conclusion,by blocking the STAT6pathway necessary for T H 2cytokine signaling,induction of arginase 1after physical injury was blunted.Also,T H 2cytokines were found to play an important role in the accumulation of TIMSC in splenic tissues.Through one or both of these mechanisms,TIMSC may help to regulate the inflammatory NO response to a septic stimulus.ACKNOWLEDGMENTSThe authors thank Dr.Sidney Morris (University of Pitts-burgh,Pittsburgh,PA)and Dr.M.Isabel T.D.Correia (Federal University of Minas Gerais,Belo Horizonte,Brazil)for constructive comments during our work;and again to Dr.Sidney Morris for providing antimouse-Arg 1Ab.We also thank Lisa R.Chedwick HT A.S.C.P.(University of Pittsburgh,Pittsburgh,PA)for performing immunohistochemistry;Derek Donaldson and Dr.Hongmei Shen from the Starzl Transplant Institute for their assistance with flow cytometry.REFERENCES1.Dziedzic T,Slowik A,Szczudlik A.Nosocomial infections and immunity:lesson from brain-injured patients.Crit Care .2004;8:266–270.2.O’Brien JM Jr,Ali NA,Aberegg SK.Sepsis.Am J Med .2007;120:1012–1022.3.Schneider DF,Glenn CH,Faunce DE.Innate lymphocyte subsets and their immunoregulatory roles in burn injury and sepsis.J Burn Care Res .2007;28:365–379.4.Daniel T,Thobe BM,Chaudry IH,et al.Regulation of the postburn wound inflammatory response by gammadelta T-cells.Shock .2007;28:278–283.ler AC,Rashid RM,Elamin EM.The “T”in trauma:the helper T-cell response and the role of immunomodulation in trauma and burn patients.J Trauma .2007;63:1407–1417.6.Makarenkova VP,Bansal V,Matta BM,et al.CD11b ϩ/Gr-1ϩmyeloid suppressor cells cause T cell dysfunction after traumatic stress.J Immunol .2006;176:2085–2094.7.Bronte V,Serafini P,Mazzoni A,et al.L-arginine metabolism in myeloid cells controls T-lymphocyte functions.Trends Immunol .2003;24:302–306.8.Corraliza I,Moncada S.Increased expression of arginase II in patients with different forms of arthritis.Implications of the regulation of nitric oxide.J Rheumatol .2002;29:2261–2265.9.Rodriguez PC,Quiceno DG,Zabaleta J,et al.Arginase I production in the tumor microenvironment by mature myeloid cells inhibits T-cell receptor expression and antigen-specific T-cell responses.Cancer Res .2004;64:5839–5849.10.Bryk J,Ochoa JB,Correia MI,et al.Effect of Citrulline and Glutamine onNitric Oxide Production in RAW 264.7Cells in an Arginine-Depleted Environment.JPEN J Parenter Enteral Nutr .2008;32:377–383.11.Nathan CF,Hibbs JB Jr.Role of nitric oxide synthesis in macrophageantimicrobial activity.Curr Opin Immunol .1991;3:65–70.12.Ochoa JB,Bernard AC,Mistry SK,et al.Trauma increases extrahepaticarginase activity.Surgery .2000;127:419–426.FIGURE 4.Nitric oxide (NO)metabolite accumulation in plasma after trauma in WT and STAT6Ϫ/Ϫmice.Mice were injected with LPS (5mg/kg)to stimulate NO production 24hours after physical injury.Plasma was harvested 16hours later and NO metabolites measured by Griess assay.As ex-pected,NO metabolites were significantly blunted aftertrauma in WT mice (P Ͻ0.01).Plasma levels of NO metabo-lites in STAT6Ϫ/Ϫmice were 3.5-fold higher than in WT mice (P ϭ0.01).Annals of Surgery •Volume 251,Number 1,January 2010Induction of Myeloid Derived Suppressor©2009Lippincott Williams &Wilkins |125。
Error propagated image halftoning with time-varyin
专利名称：Error propagated image halftoning withtime-varying phase shift发明人：Feigenblatt, Ronald I.,Powell, Carl G.申请号：EP89121390.2申请日：19891118公开号：EP0378780A1公开日：19900725专利内容由知识产权出版社提供专利附图：摘要：The present invention achieves a method of displaying an image which builds upon the error propagation method for mosaic color displays. That method propagates error between elements diagonally for a mosaic color display having diagonal rows consisting of monochromatic elements. In the method of the present invention, called "pel interleaving", the "error" propagated into the first element in the diagonal row changes with each new image or frame processed. More specifically the errorpropagated into each diagonal's first element increases incrementally with each frame processed until it exceeds the maximum element intensity value, in which case it is started anew by subtracting the maximum value. The incremental increasing of this initial error, inthe binary display case, leads to the spatial drift of "on" elements along the diagonals. If all preload values are equally likely, the time integrated ensemble of the displays approaches the exact contone image as the number of displayed images increases. Thus, if the processing is fast, so that the eye integrates a number of displayed images for the same input image, the display perceived using the present image approaches the actual contone of the input image.申请人：International Business Machines Corporation地址：Old Orchard Road Armonk, N.Y. 10504 US国籍：US代理机构：Jost, Ottokarl, Dipl.-Ing.更多信息请下载全文后查看。
氯仿-甲醇提取贝氏柯克斯体组分Q热疫苗的免疫保护性评价
2009诺贝尔生理及医学奖
生老病死，这或许是人类生命最为简洁的概括，但其中却蕴藏了无数的奥秘。获得2009年诺贝尔生理学或医学奖的三位美国科学家，凭借“发现端粒和端粒酶是如何保护染色体的”这一成果，揭开了人类衰老和罹患癌症等严重疾病的奥秘。 在生物的细胞核中，有一种易被碱性染料染色的线状物质，它们被称为“染色体”。正常人的体细胞有23对染色体，它们对人类生命具有重要意义，例如众所周知，决定男女性别的就是一对染色体。在染色体的末端部分有一个像帽子一样的特殊结构，这就是端粒。而端粒酶的作用则是帮助合成端粒，使得端粒的长度等结构得以稳定。 “染色体携有遗传信息。端粒是细胞内染色体末端的‘保护帽’，它能够保护染色体，而端粒酶在端粒受损时能够恢复其长度。”获奖者之一的伊丽莎白·布莱克本介绍说：“伴随着人的成长，端粒逐渐受到‘磨损’。于是我们会问，这是否很重要？而我们逐渐发现，这对人类而言确实很重要。” 卡罗林斯卡医学院发布的新闻公报说，这三位科学家的发现“解释了端粒如何保护染色体的末端以及端粒酶如何合成端粒”。借助他们的开创性工作，人们知道，端粒不仅与染，端粒变短，细胞就老化。相反，如果端粒酶活性很高，端粒的长度就能得到保持，细胞的老化就被延缓。 不过需要指出的是，近年来陆续有研究发现，端粒和染色体等虽然与细胞老化有关，进而影响衰老，但并非唯一的因素，“生命衰老是一个非常复杂的进程，它有许多不同的影响因素，端粒仅仅是其中之一”。 “这是有关人类衰老、癌症和干细胞等研究的谜题拼图中重要的一片，”新闻公报说，“他们的发现使我们对细胞的理解增加了新的维度，清楚地显示了疾病的机理，并将促使我们开发出潜在的新疗法。”
From Data Mining to Knowledge Discovery in Databases
s Data mining and knowledge discovery in databases have been attracting a significant amount of research, industry, and media atten-tion of late. What is all the excitement about?This article provides an overview of this emerging field, clarifying how data mining and knowledge discovery in databases are related both to each other and to related fields, such as machine learning, statistics, and databases. The article mentions particular real-world applications, specific data-mining techniques, challenges in-volved in real-world applications of knowledge discovery, and current and future research direc-tions in the field.A cross a wide variety of fields, data arebeing collected and accumulated at adramatic pace. There is an urgent need for a new generation of computational theo-ries and tools to assist humans in extracting useful information (knowledge) from the rapidly growing volumes of digital data. These theories and tools are the subject of the emerging field of knowledge discovery in databases (KDD).At an abstract level, the KDD field is con-cerned with the development of methods and techniques for making sense of data. The basic problem addressed by the KDD process is one of mapping low-level data (which are typically too voluminous to understand and digest easi-ly) into other forms that might be more com-pact (for example, a short report), more ab-stract (for example, a descriptive approximation or model of the process that generated the data), or more useful (for exam-ple, a predictive model for estimating the val-ue of future cases). At the core of the process is the application of specific data-mining meth-ods for pattern discovery and extraction.1This article begins by discussing the histori-cal context of KDD and data mining and theirintersection with other related fields. A briefsummary of recent KDD real-world applica-tions is provided. Definitions of KDD and da-ta mining are provided, and the general mul-tistep KDD process is outlined. This multistepprocess has the application of data-mining al-gorithms as one particular step in the process.The data-mining step is discussed in more de-tail in the context of specific data-mining al-gorithms and their application. Real-worldpractical application issues are also outlined.Finally, the article enumerates challenges forfuture research and development and in par-ticular discusses potential opportunities for AItechnology in KDD systems.Why Do We Need KDD?The traditional method of turning data intoknowledge relies on manual analysis and in-terpretation. For example, in the health-careindustry, it is common for specialists to peri-odically analyze current trends and changesin health-care data, say, on a quarterly basis.The specialists then provide a report detailingthe analysis to the sponsoring health-care or-ganization; this report becomes the basis forfuture decision making and planning forhealth-care management. In a totally differ-ent type of application, planetary geologistssift through remotely sensed images of plan-ets and asteroids, carefully locating and cata-loging such geologic objects of interest as im-pact craters. Be it science, marketing, finance,health care, retail, or any other field, the clas-sical approach to data analysis relies funda-mentally on one or more analysts becomingArticlesFALL 1996 37From Data Mining to Knowledge Discovery inDatabasesUsama Fayyad, Gregory Piatetsky-Shapiro, and Padhraic Smyth Copyright © 1996, American Association for Artificial Intelligence. All rights reserved. 0738-4602-1996 / $2.00areas is astronomy. Here, a notable success was achieved by SKICAT ,a system used by as-tronomers to perform image analysis,classification, and cataloging of sky objects from sky-survey images (Fayyad, Djorgovski,and Weir 1996). In its first application, the system was used to process the 3 terabytes (1012bytes) of image data resulting from the Second Palomar Observatory Sky Survey,where it is estimated that on the order of 109sky objects are detectable. SKICAT can outper-form humans and traditional computational techniques in classifying faint sky objects. See Fayyad, Haussler, and Stolorz (1996) for a sur-vey of scientific applications.In business, main KDD application areas includes marketing, finance (especially in-vestment), fraud detection, manufacturing,telecommunications, and Internet agents.Marketing:In marketing, the primary ap-plication is database marketing systems,which analyze customer databases to identify different customer groups and forecast their behavior. Business Week (Berry 1994) estimat-ed that over half of all retailers are using or planning to use database marketing, and those who do use it have good results; for ex-ample, American Express reports a 10- to 15-percent increase in credit-card use. Another notable marketing application is market-bas-ket analysis (Agrawal et al. 1996) systems,which find patterns such as, “If customer bought X, he/she is also likely to buy Y and Z.” Such patterns are valuable to retailers.Investment: Numerous companies use da-ta mining for investment, but most do not describe their systems. One exception is LBS Capital Management. Its system uses expert systems, neural nets, and genetic algorithms to manage portfolios totaling $600 million;since its start in 1993, the system has outper-formed the broad stock market (Hall, Mani,and Barr 1996).Fraud detection: HNC Falcon and Nestor PRISM systems are used for monitoring credit-card fraud, watching over millions of ac-counts. The FAIS system (Senator et al. 1995),from the U.S. Treasury Financial Crimes En-forcement Network, is used to identify finan-cial transactions that might indicate money-laundering activity.Manufacturing: The CASSIOPEE trou-bleshooting system, developed as part of a joint venture between General Electric and SNECMA, was applied by three major Euro-pean airlines to diagnose and predict prob-lems for the Boeing 737. To derive families of faults, clustering methods are used. CASSIOPEE received the European first prize for innova-intimately familiar with the data and serving as an interface between the data and the users and products.For these (and many other) applications,this form of manual probing of a data set is slow, expensive, and highly subjective. In fact, as data volumes grow dramatically, this type of manual data analysis is becoming completely impractical in many domains.Databases are increasing in size in two ways:(1) the number N of records or objects in the database and (2) the number d of fields or at-tributes to an object. Databases containing on the order of N = 109objects are becoming in-creasingly common, for example, in the as-tronomical sciences. Similarly, the number of fields d can easily be on the order of 102or even 103, for example, in medical diagnostic applications. Who could be expected to di-gest millions of records, each having tens or hundreds of fields? We believe that this job is certainly not one for humans; hence, analysis work needs to be automated, at least partially.The need to scale up human analysis capa-bilities to handling the large number of bytes that we can collect is both economic and sci-entific. Businesses use data to gain competi-tive advantage, increase efficiency, and pro-vide more valuable services to customers.Data we capture about our environment are the basic evidence we use to build theories and models of the universe we live in. Be-cause computers have enabled humans to gather more data than we can digest, it is on-ly natural to turn to computational tech-niques to help us unearth meaningful pat-terns and structures from the massive volumes of data. Hence, KDD is an attempt to address a problem that the digital informa-tion era made a fact of life for all of us: data overload.Data Mining and Knowledge Discovery in the Real WorldA large degree of the current interest in KDD is the result of the media interest surrounding successful KDD applications, for example, the focus articles within the last two years in Business Week , Newsweek , Byte , PC Week , and other large-circulation periodicals. Unfortu-nately, it is not always easy to separate fact from media hype. Nonetheless, several well-documented examples of successful systems can rightly be referred to as KDD applications and have been deployed in operational use on large-scale real-world problems in science and in business.In science, one of the primary applicationThere is an urgent need for a new generation of computation-al theories and tools toassist humans in extractinguseful information (knowledge)from the rapidly growing volumes ofdigital data.Articles38AI MAGAZINEtive applications (Manago and Auriol 1996).Telecommunications: The telecommuni-cations alarm-sequence analyzer (TASA) wasbuilt in cooperation with a manufacturer oftelecommunications equipment and threetelephone networks (Mannila, Toivonen, andVerkamo 1995). The system uses a novelframework for locating frequently occurringalarm episodes from the alarm stream andpresenting them as rules. Large sets of discov-ered rules can be explored with flexible infor-mation-retrieval tools supporting interactivityand iteration. In this way, TASA offers pruning,grouping, and ordering tools to refine the re-sults of a basic brute-force search for rules.Data cleaning: The MERGE-PURGE systemwas applied to the identification of duplicatewelfare claims (Hernandez and Stolfo 1995).It was used successfully on data from the Wel-fare Department of the State of Washington.In other areas, a well-publicized system isIBM’s ADVANCED SCOUT,a specialized data-min-ing system that helps National Basketball As-sociation (NBA) coaches organize and inter-pret data from NBA games (U.S. News 1995). ADVANCED SCOUT was used by several of the NBA teams in 1996, including the Seattle Su-personics, which reached the NBA finals.Finally, a novel and increasingly importanttype of discovery is one based on the use of in-telligent agents to navigate through an infor-mation-rich environment. Although the ideaof active triggers has long been analyzed in thedatabase field, really successful applications ofthis idea appeared only with the advent of theInternet. These systems ask the user to specifya profile of interest and search for related in-formation among a wide variety of public-do-main and proprietary sources. For example, FIREFLY is a personal music-recommendation agent: It asks a user his/her opinion of several music pieces and then suggests other music that the user might like (<http:// www.ffl/>). CRAYON(/>) allows users to create their own free newspaper (supported by ads); NEWSHOUND(<http://www. /hound/>) from the San Jose Mercury News and FARCAST(</> automatically search information from a wide variety of sources, including newspapers and wire services, and e-mail rele-vant documents directly to the user.These are just a few of the numerous suchsystems that use KDD techniques to automat-ically produce useful information from largemasses of raw data. See Piatetsky-Shapiro etal. (1996) for an overview of issues in devel-oping industrial KDD applications.Data Mining and KDDHistorically, the notion of finding useful pat-terns in data has been given a variety ofnames, including data mining, knowledge ex-traction, information discovery, informationharvesting, data archaeology, and data patternprocessing. The term data mining has mostlybeen used by statisticians, data analysts, andthe management information systems (MIS)communities. It has also gained popularity inthe database field. The phrase knowledge dis-covery in databases was coined at the first KDDworkshop in 1989 (Piatetsky-Shapiro 1991) toemphasize that knowledge is the end productof a data-driven discovery. It has been popular-ized in the AI and machine-learning fields.In our view, KDD refers to the overall pro-cess of discovering useful knowledge from da-ta, and data mining refers to a particular stepin this process. Data mining is the applicationof specific algorithms for extracting patternsfrom data. The distinction between the KDDprocess and the data-mining step (within theprocess) is a central point of this article. Theadditional steps in the KDD process, such asdata preparation, data selection, data cleaning,incorporation of appropriate prior knowledge,and proper interpretation of the results ofmining, are essential to ensure that usefulknowledge is derived from the data. Blind ap-plication of data-mining methods (rightly crit-icized as data dredging in the statistical litera-ture) can be a dangerous activity, easilyleading to the discovery of meaningless andinvalid patterns.The Interdisciplinary Nature of KDDKDD has evolved, and continues to evolve,from the intersection of research fields such asmachine learning, pattern recognition,databases, statistics, AI, knowledge acquisitionfor expert systems, data visualization, andhigh-performance computing. The unifyinggoal is extracting high-level knowledge fromlow-level data in the context of large data sets.The data-mining component of KDD cur-rently relies heavily on known techniquesfrom machine learning, pattern recognition,and statistics to find patterns from data in thedata-mining step of the KDD process. A natu-ral question is, How is KDD different from pat-tern recognition or machine learning (and re-lated fields)? The answer is that these fieldsprovide some of the data-mining methodsthat are used in the data-mining step of theKDD process. KDD focuses on the overall pro-cess of knowledge discovery from data, includ-ing how the data are stored and accessed, howalgorithms can be scaled to massive data setsThe basicproblemaddressed bythe KDDprocess isone ofmappinglow-leveldata intoother formsthat might bemorecompact,moreabstract,or moreuseful.ArticlesFALL 1996 39A driving force behind KDD is the database field (the second D in KDD). Indeed, the problem of effective data manipulation when data cannot fit in the main memory is of fun-damental importance to KDD. Database tech-niques for gaining efficient data access,grouping and ordering operations when ac-cessing data, and optimizing queries consti-tute the basics for scaling algorithms to larger data sets. Most data-mining algorithms from statistics, pattern recognition, and machine learning assume data are in the main memo-ry and pay no attention to how the algorithm breaks down if only limited views of the data are possible.A related field evolving from databases is data warehousing,which refers to the popular business trend of collecting and cleaning transactional data to make them available for online analysis and decision support. Data warehousing helps set the stage for KDD in two important ways: (1) data cleaning and (2)data access.Data cleaning: As organizations are forced to think about a unified logical view of the wide variety of data and databases they pos-sess, they have to address the issues of map-ping data to a single naming convention,uniformly representing and handling missing data, and handling noise and errors when possible.Data access: Uniform and well-defined methods must be created for accessing the da-ta and providing access paths to data that were historically difficult to get to (for exam-ple, stored offline).Once organizations and individuals have solved the problem of how to store and ac-cess their data, the natural next step is the question, What else do we do with all the da-ta? This is where opportunities for KDD natu-rally arise.A popular approach for analysis of data warehouses is called online analytical processing (OLAP), named for a set of principles pro-posed by Codd (1993). OLAP tools focus on providing multidimensional data analysis,which is superior to SQL in computing sum-maries and breakdowns along many dimen-sions. OLAP tools are targeted toward simpli-fying and supporting interactive data analysis,but the goal of KDD tools is to automate as much of the process as possible. Thus, KDD is a step beyond what is currently supported by most standard database systems.Basic DefinitionsKDD is the nontrivial process of identifying valid, novel, potentially useful, and ultimate-and still run efficiently, how results can be in-terpreted and visualized, and how the overall man-machine interaction can usefully be modeled and supported. The KDD process can be viewed as a multidisciplinary activity that encompasses techniques beyond the scope of any one particular discipline such as machine learning. In this context, there are clear opportunities for other fields of AI (be-sides machine learning) to contribute to KDD. KDD places a special emphasis on find-ing understandable patterns that can be inter-preted as useful or interesting knowledge.Thus, for example, neural networks, although a powerful modeling tool, are relatively difficult to understand compared to decision trees. KDD also emphasizes scaling and ro-bustness properties of modeling algorithms for large noisy data sets.Related AI research fields include machine discovery, which targets the discovery of em-pirical laws from observation and experimen-tation (Shrager and Langley 1990) (see Kloes-gen and Zytkow [1996] for a glossary of terms common to KDD and machine discovery),and causal modeling for the inference of causal models from data (Spirtes, Glymour,and Scheines 1993). Statistics in particular has much in common with KDD (see Elder and Pregibon [1996] and Glymour et al.[1996] for a more detailed discussion of this synergy). Knowledge discovery from data is fundamentally a statistical endeavor. Statistics provides a language and framework for quan-tifying the uncertainty that results when one tries to infer general patterns from a particu-lar sample of an overall population. As men-tioned earlier, the term data mining has had negative connotations in statistics since the 1960s when computer-based data analysis techniques were first introduced. The concern arose because if one searches long enough in any data set (even randomly generated data),one can find patterns that appear to be statis-tically significant but, in fact, are not. Clearly,this issue is of fundamental importance to KDD. Substantial progress has been made in recent years in understanding such issues in statistics. Much of this work is of direct rele-vance to KDD. Thus, data mining is a legiti-mate activity as long as one understands how to do it correctly; data mining carried out poorly (without regard to the statistical as-pects of the problem) is to be avoided. KDD can also be viewed as encompassing a broader view of modeling than statistics. KDD aims to provide tools to automate (to the degree pos-sible) the entire process of data analysis and the statistician’s “art” of hypothesis selection.Data mining is a step in the KDD process that consists of ap-plying data analysis and discovery al-gorithms that produce a par-ticular enu-meration ofpatterns (or models)over the data.Articles40AI MAGAZINEly understandable patterns in data (Fayyad, Piatetsky-Shapiro, and Smyth 1996).Here, data are a set of facts (for example, cases in a database), and pattern is an expres-sion in some language describing a subset of the data or a model applicable to the subset. Hence, in our usage here, extracting a pattern also designates fitting a model to data; find-ing structure from data; or, in general, mak-ing any high-level description of a set of data. The term process implies that KDD comprises many steps, which involve data preparation, search for patterns, knowledge evaluation, and refinement, all repeated in multiple itera-tions. By nontrivial, we mean that some search or inference is involved; that is, it is not a straightforward computation of predefined quantities like computing the av-erage value of a set of numbers.The discovered patterns should be valid on new data with some degree of certainty. We also want patterns to be novel (at least to the system and preferably to the user) and poten-tially useful, that is, lead to some benefit to the user or task. Finally, the patterns should be understandable, if not immediately then after some postprocessing.The previous discussion implies that we can define quantitative measures for evaluating extracted patterns. In many cases, it is possi-ble to define measures of certainty (for exam-ple, estimated prediction accuracy on new data) or utility (for example, gain, perhaps indollars saved because of better predictions orspeedup in response time of a system). No-tions such as novelty and understandabilityare much more subjective. In certain contexts,understandability can be estimated by sim-plicity (for example, the number of bits to de-scribe a pattern). An important notion, calledinterestingness(for example, see Silberschatzand Tuzhilin [1995] and Piatetsky-Shapiro andMatheus [1994]), is usually taken as an overallmeasure of pattern value, combining validity,novelty, usefulness, and simplicity. Interest-ingness functions can be defined explicitly orcan be manifested implicitly through an or-dering placed by the KDD system on the dis-covered patterns or models.Given these notions, we can consider apattern to be knowledge if it exceeds some in-terestingness threshold, which is by nomeans an attempt to define knowledge in thephilosophical or even the popular view. As amatter of fact, knowledge in this definition ispurely user oriented and domain specific andis determined by whatever functions andthresholds the user chooses.Data mining is a step in the KDD processthat consists of applying data analysis anddiscovery algorithms that, under acceptablecomputational efficiency limitations, pro-duce a particular enumeration of patterns (ormodels) over the data. Note that the space ofArticlesFALL 1996 41Figure 1. An Overview of the Steps That Compose the KDD Process.methods, the effective number of variables under consideration can be reduced, or in-variant representations for the data can be found.Fifth is matching the goals of the KDD pro-cess (step 1) to a particular data-mining method. For example, summarization, clas-sification, regression, clustering, and so on,are described later as well as in Fayyad, Piatet-sky-Shapiro, and Smyth (1996).Sixth is exploratory analysis and model and hypothesis selection: choosing the data-mining algorithm(s) and selecting method(s)to be used for searching for data patterns.This process includes deciding which models and parameters might be appropriate (for ex-ample, models of categorical data are differ-ent than models of vectors over the reals) and matching a particular data-mining method with the overall criteria of the KDD process (for example, the end user might be more in-terested in understanding the model than its predictive capabilities).Seventh is data mining: searching for pat-terns of interest in a particular representa-tional form or a set of such representations,including classification rules or trees, regres-sion, and clustering. The user can significant-ly aid the data-mining method by correctly performing the preceding steps.Eighth is interpreting mined patterns, pos-sibly returning to any of steps 1 through 7 for further iteration. This step can also involve visualization of the extracted patterns and models or visualization of the data given the extracted models.Ninth is acting on the discovered knowl-edge: using the knowledge directly, incorpo-rating the knowledge into another system for further action, or simply documenting it and reporting it to interested parties. This process also includes checking for and resolving po-tential conflicts with previously believed (or extracted) knowledge.The KDD process can involve significant iteration and can contain loops between any two steps. The basic flow of steps (al-though not the potential multitude of itera-tions and loops) is illustrated in figure 1.Most previous work on KDD has focused on step 7, the data mining. However, the other steps are as important (and probably more so) for the successful application of KDD in practice. Having defined the basic notions and introduced the KDD process, we now focus on the data-mining component,which has, by far, received the most atten-tion in the literature.patterns is often infinite, and the enumera-tion of patterns involves some form of search in this space. Practical computational constraints place severe limits on the sub-space that can be explored by a data-mining algorithm.The KDD process involves using the database along with any required selection,preprocessing, subsampling, and transforma-tions of it; applying data-mining methods (algorithms) to enumerate patterns from it;and evaluating the products of data mining to identify the subset of the enumerated pat-terns deemed knowledge. The data-mining component of the KDD process is concerned with the algorithmic means by which pat-terns are extracted and enumerated from da-ta. The overall KDD process (figure 1) in-cludes the evaluation and possible interpretation of the mined patterns to de-termine which patterns can be considered new knowledge. The KDD process also in-cludes all the additional steps described in the next section.The notion of an overall user-driven pro-cess is not unique to KDD: analogous propos-als have been put forward both in statistics (Hand 1994) and in machine learning (Brod-ley and Smyth 1996).The KDD ProcessThe KDD process is interactive and iterative,involving numerous steps with many deci-sions made by the user. Brachman and Anand (1996) give a practical view of the KDD pro-cess, emphasizing the interactive nature of the process. Here, we broadly outline some of its basic steps:First is developing an understanding of the application domain and the relevant prior knowledge and identifying the goal of the KDD process from the customer’s viewpoint.Second is creating a target data set: select-ing a data set, or focusing on a subset of vari-ables or data samples, on which discovery is to be performed.Third is data cleaning and preprocessing.Basic operations include removing noise if appropriate, collecting the necessary informa-tion to model or account for noise, deciding on strategies for handling missing data fields,and accounting for time-sequence informa-tion and known changes.Fourth is data reduction and projection:finding useful features to represent the data depending on the goal of the task. With di-mensionality reduction or transformationArticles42AI MAGAZINEThe Data-Mining Stepof the KDD ProcessThe data-mining component of the KDD pro-cess often involves repeated iterative applica-tion of particular data-mining methods. This section presents an overview of the primary goals of data mining, a description of the methods used to address these goals, and a brief description of the data-mining algo-rithms that incorporate these methods.The knowledge discovery goals are defined by the intended use of the system. We can distinguish two types of goals: (1) verification and (2) discovery. With verification,the sys-tem is limited to verifying the user’s hypothe-sis. With discovery,the system autonomously finds new patterns. We further subdivide the discovery goal into prediction,where the sys-tem finds patterns for predicting the future behavior of some entities, and description, where the system finds patterns for presenta-tion to a user in a human-understandableform. In this article, we are primarily con-cerned with discovery-oriented data mining.Data mining involves fitting models to, or determining patterns from, observed data. The fitted models play the role of inferred knowledge: Whether the models reflect useful or interesting knowledge is part of the over-all, interactive KDD process where subjective human judgment is typically required. Two primary mathematical formalisms are used in model fitting: (1) statistical and (2) logical. The statistical approach allows for nondeter-ministic effects in the model, whereas a logi-cal model is purely deterministic. We focus primarily on the statistical approach to data mining, which tends to be the most widely used basis for practical data-mining applica-tions given the typical presence of uncertain-ty in real-world data-generating processes.Most data-mining methods are based on tried and tested techniques from machine learning, pattern recognition, and statistics: classification, clustering, regression, and so on. The array of different algorithms under each of these headings can often be bewilder-ing to both the novice and the experienced data analyst. It should be emphasized that of the many data-mining methods advertised in the literature, there are really only a few fun-damental techniques. The actual underlying model representation being used by a particu-lar method typically comes from a composi-tion of a small number of well-known op-tions: polynomials, splines, kernel and basis functions, threshold-Boolean functions, and so on. Thus, algorithms tend to differ primar-ily in the goodness-of-fit criterion used toevaluate model fit or in the search methodused to find a good fit.In our brief overview of data-mining meth-ods, we try in particular to convey the notionthat most (if not all) methods can be viewedas extensions or hybrids of a few basic tech-niques and principles. We first discuss the pri-mary methods of data mining and then showthat the data- mining methods can be viewedas consisting of three primary algorithmiccomponents: (1) model representation, (2)model evaluation, and (3) search. In the dis-cussion of KDD and data-mining methods,we use a simple example to make some of thenotions more concrete. Figure 2 shows a sim-ple two-dimensional artificial data set consist-ing of 23 cases. Each point on the graph rep-resents a person who has been given a loanby a particular bank at some time in the past.The horizontal axis represents the income ofthe person; the vertical axis represents the to-tal personal debt of the person (mortgage, carpayments, and so on). The data have beenclassified into two classes: (1) the x’s repre-sent persons who have defaulted on theirloans and (2) the o’s represent persons whoseloans are in good status with the bank. Thus,this simple artificial data set could represent ahistorical data set that can contain usefulknowledge from the point of view of thebank making the loans. Note that in actualKDD applications, there are typically manymore dimensions (as many as several hun-dreds) and many more data points (manythousands or even millions).ArticlesFALL 1996 43Figure 2. A Simple Data Set with Two Classes Used for Illustrative Purposes.。
2009年诺贝尔奖生理学或医学奖端粒与端粒酶
组成
RNA（作为模板） 蛋白质（反转录酶）
作用
在端粒DNA的复制时，端粒酶既有模板， 又有逆转录酶这两方面的作用。其与
端粒3´端结合后，以其RNA为模板，经反 转录延长端粒，从而保护DNA双链末段
免遭降解及相互融合。
5.获奖成果的理论意义和应用前景
端粒的保护,维持和端粒的缩短形成了端粒调节 的对立统一体“ 一方面，对端粒的保护维持了遗传 物质的稳定性，而另一方面，DNA复制的内在特 性却注定了体细胞的寿命” 生殖细胞中端粒酶的活 性保证了端粒初始长度的恒定，而体细胞缺乏端粒 酶活性而注定了走向衰老和死亡“ 作为染色体末端 的”守护者，端粒的变化决定着细胞的命运，对端 粒,端粒酶的深入研究，有益于揭示遗传物质的稳 定性以及细胞的衰老,死亡和癌变的奥妙，同时也 为肿瘤,遗传病的诊断和治疗带来 了新的曙光”。
• 当端粒不能再缩短时，细胞就无法继续分裂 而死亡。伊丽莎白，布莱克本他们发现的端
粒酶，在一些失控的恶性细胞的生长中扮演 重要角色。大约90%的癌细胞都有着不断增 长的端粒及相对来说数量较多的端粒酶。
伊丽莎白·布莱克本
澳大利亚 (Elizabeth Blackburn
2009年 诺贝尔生理 学或医学获奖 美国
3.获奖成果— 端粒
3.获奖成果— 端粒
端粒
概念
组成
作用
缺陷
真核细胞线性染色体
末端的一组重复DNA 序列，通常由富含 由端粒蛋 鸟嘌呤核苷酸(G) 白和端粒 的短的串联重复 DNA组成 序列组成。
DNA复制时，负责复制的酶 不能复制线性DNA分子尾部， 对染色体有 这样就在端粒区域产生一段 保护作用 单链区域，导致部分端粒
Ӂ 2007年伊丽莎白·布莱克本因 学术成就卓著曾被美国《时代 》周刊评为年度全球最具影响 力的100个人物之一。
开启片剂完整性的窗户(中英文对照)
开启片剂完整性的窗户日本东芝公司，剑桥大学摘要：由日本东芝公司和剑桥大学合作成立的公司向《医药技术》解释了FDA支持的技术如何在不损坏片剂的情况下测定其完整性。
太赫脉冲成像的一个应用是检查肠溶制剂的完整性，以确保它们在到达肠溶之前不会溶解。
关键词：片剂完整性，太赫脉冲成像。
能够检测片剂的结构完整性和化学成分而无需将它们打碎的一种技术，已经通过了概念验证阶段，正在进行法规申请。
由英国私募Teraview公司研发并且以太赫光（介于无线电波和光波之间）为基础。
该成像技术为配方研发和质量控制中的湿溶出试验提供了一个更好的选择。
该技术还可以缩短新产品的研发时间，并且根据厂商的情况，随时间推移甚至可能发展成为一个用于制药生产线的实时片剂检测系统。
TPI技术通过发射太赫射线绘制出片剂和涂层厚度的三维差异图谱，在有结构或化学变化时太赫射线被反射回。
反射脉冲的时间延迟累加成该片剂的三维图像。
该系统使用太赫发射极，采用一个机器臂捡起片剂并且使其通过太赫光束，用一个扫描仪收集反射光并且建成三维图像(见图)。
技术研发太赫技术发源于二十世纪九十年代中期13本东芝公司位于英国的东芝欧洲研究中心，该中心与剑桥大学的物理学系有着密切的联系。
日本东芝公司当时正在研究新一代的半导体，研究的副产品是发现了这些半导体实际上是太赫光非常好的发射源和检测器。
二十世纪九十年代后期，日本东芝公司授权研究小组寻求该技术可能的应用，包括成像和化学传感光谱学，并与葛兰素史克和辉瑞以及其它公司建立了关系，以探讨其在制药业的应用。
虽然早期的结果表明该技术有前景，但日本东芝公司却不愿深入研究下去，原因是此应用与日本东芝公司在消费电子行业的任何业务兴趣都没有交叉。
这一决定的结果是研究中心的首席执行官DonArnone和剑桥桥大学物理学系的教授Michael Pepper先生于2001年成立了Teraview公司一作为研究中心的子公司。
TPI imaga 2000是第一个商品化太赫成像系统，该系统经优化用于成品片剂及其核心完整性和性能的无破坏检测。
NATURE RNA-Seq a revolutionary tool for transcriptomics
The transcriptome is the complete set of transcripts in a cell, and their quantity, for a specific developmental stage or physi-ological condition. Understanding the transcriptome is essential for interpreting the functional elements of the genome and revealing the molecular constituents of cells and tissues, and also for understand-ing development and disease. The key aims of transcriptomics are: to catalogue all species of transcript, including mRNAs, non-coding RNAs and small RNAs; to determine the transcriptional structureof genes, in terms of their start sites, 5′and 3′ ends, splicing patterns and other post-transcriptional modifications; and to quantify the changing expression levels of each transcript during development and under different conditions.Various technologies have been developed to deduce and quantify the transcriptome, including hybridization- or sequence-based approaches. Hybridization-based approaches typically involve incubating fluorescently labelled cDNA with custom-made microarrays or commercial high-density oligo microar-rays. Specialized microarrays have also been designed; for example, arrays with probes spanning exon junctions canbe used to detect and quantify distinct spliced isoforms1. Genomic tiling microar-rays that represent the genome at high density have been constructed and allow the mapping of transcribed regions to avery high resolution, from several basepairs to ~100 bp2–5. Hybridization-basedapproaches are high throughput andrelatively inexpensive, except for high-resolution tiling arrays that interrogatelarge genomes. However, these methodshave several limitations, which include:reliance upon existing knowledge aboutgenome sequence; high background levelsowing to cross-hybridization6,7; and alimited dynamic range of detection owingto both background and saturation ofsignals. Moreover, comparing expressionlevels across different experiments is oftendifficult and can require complicatednormalization methods.In contrast to microarray methods,sequence-based approaches directly deter-mine the cDNA sequence. Initially, Sangersequencing of cDNA or EST libraries8,9was used, but this approach is relativelylow throughput, expensive and generallynot quantitative. Tag-based methods weredeveloped to overcome these limitations,including serial analysis of gene expression(SAGE)10,11, cap analysis of gene expression(CAGE)12–14 and massively parallel signaturesequencing (MPSS)15–17. These tag-basedsequencing approaches are high through-put and can provide precise, ‘digital’ geneexpression levels. However, most arebased on expensive Sanger sequencingtechnology, and a significant portion ofthe short tags cannot be uniquely mappedto the reference genome. Moreover, onlya portion of the transcript is analysed andisoforms are generally indistinguishablefrom each other. These disadvantageslimit the use of traditional sequencingtechnology in annotating the structure oftranscriptomes.Recently, the development of novelhigh-throughput DNA sequencing meth-ods has provided a new method for bothmapping and quantifying transcriptomes.This method, termed RNA-Seq (RNAsequencing), has clear advantages overexisting approaches and is expected to rev-olutionize the manner in which eukaryotictranscriptomes are analysed. It has alreadybeen applied to Saccharomyces cerevisiae,Schizosaccharomyces pombe, Arabidopsisthaliana, mouse and human cells18–24. Here,we explain how RNA-Seq works, discussits challenges and provide an overview ofstudies that have used this approach, whichhave already begun to change our view ofeukaryotic transcriptomes.RNA-Seq technology and benefitsRNA-Seq uses recently developed deep-sequencing technologies. In general, apopulation of RNA (total or fractionated,such as poly(A)+) is converted to a libraryof cDNA fragments with adaptors attachedto one or both ends (FIG. 1). Each molecule,with or without amplification, is thensequenced in a high-throughput mannerto obtain short sequences from one end(single-end sequencing) or both ends(pair-end sequencing).The reads are typi-cally 30–400 bp, depending on the DNA-sequencing technology used. In principle,any high-throughput sequencing technol-ogy25 can be used for RNA-Seq, and theIllumina IG18–21,23,24, Applied BiosystemsSOLiD22 and Roche 454 Life Science26–28I N N OVAT I O NRNA-Seq: a revolutionary tool fortranscriptomicsZhong Wang, Mark Gerstein and Michael SnyderAbstract | RNA-Seq is a recently developed approach to transcriptome profilingthat uses deep-sequencing technologies. Studies using this method havealready altered our view of the extent and complexity of eukaryotictranscriptomes. RNA-Seq also provides a far more precise measurement oflevels of transcripts and their isoforms than other methods. This article describesthe RNA-Seq approach, the challenges associated with its application, and theadvances made so far in characterizing several eukaryote transcriptomes.RNA-Seq […] is expectedto revolutionize themanner in which eukaryotictranscriptomes are analysed.NATURE REVIEwS |genetics VOLUME 10 | jANUARy 2009 |57PeRSPecTiveS© 2009 Macmillan Publishers Limited. All rights reservedsystems have already been applied for this purpose. The Helicos Biosciences tSMS system has not yet been used for published RNA-Seq studies, but is also appropriate and has the added advantage of avoiding amplification of target cDNA. Following sequencing, the resulting reads are either aligned to a reference genome or reference transcripts, or assembledde novo without the genomic sequenceto produce a genome-scale transcription map that consists of both the transcrip-tional structure and/or level of expression for each gene.Although RNA-Seq is still a technology under active development, it offers several key advantages over existing technologies (Table 1).First, unlike hybridization-based approaches, RNA-Seq is not limited to detecting transcripts that correspondto existing genomic sequence. For example, 454-based RNA-Seq has beenused to sequence the transcriptome ofthe Glanville fritillary butterfly27. Thismakes RNA-Seq particularly attractivefor non-model organisms with genomicsequences that are yet to be determined.RNA-Seq can reveal the precise locationof transcription boundaries, to a single-base resolution. Furthermore, 30-bp shortreads from RNA-Seq give informationabout how two exons are connected,whereas longer reads or pair-end shortreads should reveal connectivity betweenmultiple exons. These factors make RNA-Seq useful for studying complex tran-scriptomes. In addition, RNA-Seq can alsoreveal sequence variations (for example,SNPs) in the transcribed regions22,24.A second advantage of RNA-Seqrelative to DNA microarrays is thatRNA-Seq has very low, if any, backgroundsignal because DNA sequences canbeen unambiguously mapped to uniqueregions of the genome. RNA-Seq doesnot have an upper limit for quantifica-tion, which correlates with the numberof sequences obtained. Consequently,it has a large dynamic range of expres-sion levels over which transcripts can bedetected: a greater than 9,000-fold rangewas estimated in a study that analysed 16million mapped reads in Saccharomycescerevisiae18, and a range spanning fiveorders of magnitude was estimated for40 million mouse sequence reads20. Bycontrast, DNA microarrays lack sensitivityfor genes expressed either at low or veryhigh levels and therefore have a muchsmaller dynamic range (one-hundredfoldto a few-hundredfold) (FIG. 2). RNA-Seqhas also been shown to be highly accuratefor quantifying expression levels, as deter-mined using quantitative PCR (qPCR)18 andspike-in RNa controls of known concentra-tion20. The results of RNA-Seq also showhigh levels of reproducibility, for bothtechnical and biological replicates18,22.Finally, because there are no cloning steps,and with the Helicos technology there isno amplification step, RNA-Seq requiresless RNA sample.Taking all of these advantages intoaccount, RNA-Seq is the first sequencing-based method that allows the entiretranscriptome to be surveyed in a veryhigh-throughput and quantitative man-ner. This method offers both single-baseresolution for annotation and ‘digital’gene expression levels at the genome scale,often at a much lower cost than eithertiling arrays or large-scale Sanger ESTsequencing.Challenges for RNA-SeqLibrary construction.The ideal methodfor transcriptomics should be able todirectly identify and quantify all RNAs,small or large. Although there are onlya few steps in RNA-Seq (FIG. 1), it doesinvolve several manipulation stages dur-ing the production of cDNA libraries,which can complicate its use in profilingall types of transcript.Unlike small RNAs (microRNas(miRNAs), Piwi-interacting RNas (piRNAs),short interfering RNas (siRNAs)and manyothers), which can be directly sequencedafter adaptor ligation, larger RNA mol-ecules must be fragmented into smallerpieces (200–500 bp) to be compatiblewith most deep-sequencing technologies.Common fragmentation methods includeCoding sequenceORFP e r s P e c t i v e s58 | jANUARy 2009 | VOLUME 10 /reviews/genetics© 2009 Macmillan Publishers Limited. All rights reservedRNA fragmentation (RNA hydrolysis or nebulization) and cDNA fragmentation (DNase I treatment or sonication). Each of these methods creates a different bias in the outcome. For example, RNA fragmen-tation has little bias over the transcript body20, but is depleted for transcript ends compared with other methods (FIG. 3). Conversely, cDNA fragmentation is usually strongly biased towards the iden-tification of sequences from the 3′ ends of transcripts, and thereby provides valuable information about the precise identity of these ends18 (FIG. 4).Some manipulations during library construction also complicate the analysis of RNA-Seq results. For example, many shorts reads that are identical to each other can be obtained from cDNA librariesthat have been amplified. These could bea genuine reflection of abundant RNAspecies, or they could be PCR artefacts.One way to discriminate between thesepossibilities is to determine whether thesame sequences are observed in differentbiological replicates.Another key consideration concerninglibrary construction is whether or not toprepare strand-specific libraries, as hasbeen done in two studies21,22. These librarieshave the advantage of yielding informationabout the orientation of transcripts, whichis valuable for transcriptome annotation,especially for regions with overlappingtranscription from opposite directions2,19,29;however, strand-specific libraries arecurrently laborious to produce because theyrequire many steps22 or direct RNA–RNAligation21, which is inefficient. Moreover,it is essential to ensure that the antisensetranscripts are not artefacts of reverse tran-scription30. Because of these complications,most studies thus far have analysed cDNAswithout strand information.Bioinformatic challenges.Like otherhigh-throughput sequencing technolo-gies, RNA-Seq faces several informaticschallenges, including the development ofefficient methods to store, retrieve andprocess large amounts of data, which mustbe overcome to reduce errors in imageanalysis and base-calling and removelow-quality reads.Table 1 |Advantages of RNA-Seq compared with other transcriptomics methodsP e r s P e c t i v e sOnce high-quality reads have been obtained, the first task of data analysis is to map the short reads from RNA-Seq to the reference genome, or to assemble them into contigs before aligning them to the genomic sequence to reveal transcription structure. There are several programs for mapping reads to the genome, including ELAND, SOAP31, MAQ32 and RMAP33 (information about these can be found atthe Illumina forum and at SEQanswers).However, short transcriptomic reads also only needs to be given to poly(A) tailsand to a small number of exon–exonjunctions. Poly(A) tails can be identifiedsimply by the presence of multiple As orTs at the end of some reads. Exon–exonjunctions can be identified by the pres-ence of a specific sequence context (theGT–AG dinucleotides that flank splicesites) and confirmed by the low expressionof intronic sequences, which are removedduring splicing. Transcriptome mapsa junction library that contains all theknown and predicted junction sequencesand map reads to this library19,20. A chal-lenge for the future is to develop computa-tionally simple methods to identify novelsplicing events that take place between twodistant sequences or between exons fromtwo different genes.For large transcriptomes, alignmentis also complicated by the fact that a sig-nificant portion of sequence reads matchP e r s P e c t i v e sgenes are presumably either not expressed under this condition (for example, sporu-lation genes18) or do not have poly(A) tails. Analyzing many different conditions can further increase the coverage; inS. pombe 122 million reads from six differ-ent growth conditions detected transcrip-tion from >99% of annotated genes19.In general, the larger the genome, the more complex the transcriptome, the more sequencing depth is required for adequate coverage. Unlike genome-sequencing cov-erage, it is less straightforward to calculate the coverage of the transcriptome; thisis because the true number and level of different transcript isoforms is not usually known and because transcription activity varies greatly across the genome. One study used the number of unique transcription start sites as a measure of coverage in mouse embryonic cells, and demonstrated that at 80 million reads, the number of start sites reached a plateau22 (FIG. 5b). However, this approach does not address transcrip-tome complexity in alternative splicing and transcription termination sites; presumably further sequencing can reveal additionalvariants.New transcriptomic insights Despite the challenges described above, the advantages of RNA-Seq have enabled us to generate an unprecedented global view of the transcriptome and its organi-zation for a number of species and cell types. Before the advent of RNA-Seq,it was known that a much greater than expected fraction of the yeast, Drosophila melanogaster and human genomes are transcribed2,4,36, and for yeast and humans a number of distinct isoforms have been found for many genes2,4. However, the starts and ends of most transcripts and exons had not been precisely resolved and the extent of spliced heterogeneity remained poorly understood. RNA-Seq, with its high resolution and sensitivity has revealed many novel transcribed regions and splicing isoforms of known genes, and has mapped 5′ and 3′ boundaries for many genes.Mapping gene and exon boundaries. The single-base resolution of RNA-Seq has the potential to revise many aspectsof the existing gene annotation, including gene boundaries and introns for known genes as well as the identification of novel transcribed regions. 5′ and 3′ boundaries can be mapped to within 10–50 bases by a precipitous drop in signal. 3′ boundaries can be precisely mapped by searching forpoly(A) tags, and introns can be mappedby searching for tags that span GT–AGsplicing consensus sites. Using these meth-ods the 5′ and 3′ boundaries of 80% and85% of all annotated genes, respectively,were mapped in S. cerevisiae18. Similarly,in S. pombe many boundaries were definedby RNA-Seq data in combination withtiling array data19.These two studies led to the discoveryof many 5′ and 3′ UTRs that had notbeen analysed previously. In S. cerevisiae,extensive 3′-end heterogeneity wasdiscovered at two levels: first, localheterogeneity exists in which a cluster ofsites are involved, typically within a 10 bpwindow; second, there are distinct regionsof poly(A) addition for 540 genes (FIG. 4).It is plausible that these different 3′ endsconfer distinct properties to the differentmRNA isoforms, such as mRNA localiza-tion or degradation signals, which in turnmight be responsible for unique biologicalfunctions18,19. In addition to 3′ heterogene-ity, the list of upstream ORFs within the 5′UTRs of mRNAs (uORFs) was also greatlyexpanded from 17 to 340 (6% of yeastgenes)18; uORFs regulate mRNA transla-tion37 or stability38, so these sequencesmight make a previously underappreciatedcontribution to the regulatory sophistica-tion of eukaryotic genomes. Interestingly,many mRNAs with uORFs are transcrip-tion factors, suggesting that these regulatorsare themselves heavily regulated.The mapping of transcript boundariesrevealed several novel features of eukaryo-tic gene organization. Many yeast geneswere found to overlap at their 3′ ends18.Using relaxed criteria similar to thoseemployed in a recent study18 we found that808 pairs, approximately 25% of all yeastORFs, overlap at their 3′ ends18. Likewise,antisense expression is enriched in the 3′exons of mouse transcripts22. These featuresmight confer interesting regulatory proper-ties on the affected genes. For multicellularorganisms, antisense transcription couldmodulate gene expression through theproduction of siRNAs or through dsRNaediting39,40. For yeast, which seems to lacksiRNA and dsRNA-editing functions,transcription from one gene might interferewith that from an overlapping gene, orcoordinate gene expression through othermechanisms.Extensive transcript complexity.RNA-Seqcan be used to quantitatively examinesplicing diversity by searching for readsthat span known splice junctions as wellas potential new ones. In humans, 31,618known splicing events were confirmed(11% of all known splicing events) and 379novel splicing events were discovered24.ture Reviews |Genetics′Local heterogeneityDistinct poly(A) sitesFigure 4 | Poly(A) tags from RnA-seq. A region containing two overlapping transcripts (ACT1, from the actin gene, and YFL040W, an uncharacterized ORF) from the Saccharomyces cerevisiae genome is shown. Arrows point to transcription direction. The poly(A) tags from RNA-Seq experiments are shown below these transcripts, with arrows indicating transcription direction. The precise location of each locus identified by poly(A) tags reveals the heterogeneity in poly(A) sites, for example, ACT1 has two big clusters, both with a few bases of local heterogeneity. The transcription direction revealed by poly(A) tags also helps to resolve 3′-end overlapping transcribed regions18.P e r s P e c t i v e sNATURE REVIEwS |genetics VOLUME 10 | jANUARy 2009 |61© 2009 Macmillan Publishers Limited. All rights reservedglossaryCap analysis of gene expression(CaGe). Similar to SaGe, except that 5′-end information of the transcript is analysed instead of 3′-end information. ContigsA group of sequences representing overlapping regions from a genome or transcriptome.dsRNA editingSite-specific modification of a pre-mRNa by dsRNa-specific enzymes that leads to the production of variant mRNa from the same gene.Genomic tiling microarraya DNa microarray that uses a set of overlapping oligonucleotide probes that represent a subset of or the whole genome at very high resolution.Massively parallel signature sequencing (MPSS). a gene expression quantification method that determines 17–20-bp ‘signatures’ from the ends of a cDNa molecule using multiple cycles of enzymatic cleavage and ligation.MicroRNA(miRNa). Small RNa molecules that areprocessed from small hairpin RNa (shRNa)precursors that are produced from miRNagenes. miRNas are 21–23 nucleotides in lengthand through the RNa-induced silencing complexthey target and silence mRNas containing imperfectlycomplementary sequence.Piwi-interacting RNAs(piRNa). Small RNa species that are processedfrom single-stranded precursor RNas. Theyare 25–35 nucleotides in length and formcomplexes with the piwi protein. piRNas areprobably involved in transposon silencing andstem-cell function.Quantitative PCR(qPCR). an application of PCR to determinethe quantity of DNa or RNa in a sample. Themeasurements are often made in real time andthe method is also called real-time PCR.Sequencing depthThe total number of all the sequences reads or basepairs represented in a single sequencing experiment orseries of experiments.Serial analysis of gene expression(SaGe). a method that uses short ~14–20-bp sequencetags from the 3′ ends of transcripts to measure geneexpression levels.Short interfering RNA(siRNa). RNa molecules that are 21–23 nucleotides longand that are processed from long double-stranded RNas;they are functional components of the RNai-inducedsilencing complex. siRNas typically target and silencemRNas by binding perfectly complementary sequencesin the mRNa and causing their degradation and/ortranslation inhibition.Spike-in RNAa few species of RNa with known sequence and quantitythat are added as internal controls in RNa-Seq experiments.62 | jANUARy 2009 | VOLUME 10 /reviews/genetics© 2009 Macmillan Publishers Limited. All rights reservedDefining transcription levelAs RNA-Seq is quantitative, it can be used to determine RNA expression levels more accurately than microarrays. In principle, it is possible to determine the absolute quantity of every molecule in a cell population, and directly compare results between experiments. Several methods have been used for quantification. For RNA fragmentation followed by cDNA synthesis, which gives more uniform cov-erage of each exon, gene expression levels can be deduced from the total numberof reads that fall into the exons of a gene, normalized by the length of exons that can be uniquely mapped24; for 3′-biased methods, read counts from a window near the 3′ end are used18. Gene expression levels determined by these methods closely correlate with qPCR and RNA spike-in controls.One particularly powerful advantage of RNA-Seq is that it can capture transcrip-tome dynamics across different tissues or conditions without sophisticated normali-zation of data sets19,20,22. RNA-Seq has been used to accurately monitor gene expres-sion during yeast vegetative growth18, yeast meiosis19 and mouse embryonic stem-cell differentiation22, to track gene expression changes during development, and to provide a ‘digital measurement’ of gene expression difference between differ-ent tissues20. Because of these advantages, RNA-Seq will undoubtedly be valuable for understanding transcriptomic dynamics during development and normal physi-ological changes, and in the analysis of biomedical samples, where it will allow robust comparison between diseased and normal tissues, as well as the subclassification of disease states.Future directionsAlthough RNA-Seq is still in the early stages of use, it has clear advantages over previously developed transcriptomic methods. The next big challenge for RNA-Seq is to target more complex transcriptomes to identify and track the expression changes of rare RNA isoforms from all genes. Technologies that will advance achievement of this goal are pair-end sequencing, strand-specific sequencing and the use of longer reads to increase coverage and depth. As the cost of sequencing continues to fall, RNA-Seq is expected to replace microarrays for many applications that involve determin-ing the structure and dynamics of the transcriptome.Zhong Wang and Michael Snyder are at the Departmentof Molecular, Cellular and Developmental Biology, andMark Gerstein is at the Department of Molecular,Biophysics and Biochemistry, Yale University, 219Prospect Street, New Haven, Connecticut 06520, USA.Correspondence to M.S.e‑mail: michael.snyder@doi:10.1038/nrg2484Published online 18 November 20081. Clark, T. A., Sugnet, C. W. & Ares, M. Jr.Genomewide analysis of mRNA processing in yeastusing splicing-specific microarrays. Science296,907–910 (2002).2. David, L. et al. A high-resolution map of transcriptionin the yeast genome. Proc. Natl Acad. Sci. USA103,5320–5325 (2006).3. Yamada, K. et al. Empirical analysis of transcriptionalactivity in the Arabidopsis genome. Science 302,842–846 (2003).4. Bertone, P. et al. Global identification of humantranscribed sequences with genome tiling arrays.Science 306, 2242–2246 (2004).5. Cheng, J. et al. T ranscriptional maps of 10 humanchromosomes at 5-nucleotide resolution. Science308, 1149–1154 (2005).6. Okoniewski, M. J. & Miller, C. J. Hybridizationinteractions between probesets in short oligomicroarrays lead to spurious correlations.BMC Bioinformatics7, 276 (2006).7. Royce, T. E., Rozowsky, J. S. & Gerstein, M. B.T oward a universal microarray: prediction of geneexpression through nearest-neighbor probe sequenceidentification. Nucleic Acids Res.35, e99 (2007).8. Boguski, M. S., T olstoshev, C. M. & Bassett, D. E. Jr.Gene discovery in dbEST. Science 265, 1993–1994(1994).9. Gerhard, D. S. et al. The status, quality, and expansionof the NIH full-length cDNA project: the MammalianGene Collection (MGC). Genome Res.14, 2121–2127(2004).10. Velculescu, V. E., Zhang, L., Vogelstein, B. &Kinzler, K. W. Serial analysis of gene expression.Science270, 484–487 (1995).11. Harbers, M. & Carninci, P. T ag-based approaches fortranscriptome research and genome annotation.Nature Methods2, 495–502 (2005).12. Kodzius, R. et al. CAGE: cap analysis of geneexpression. Nature Methods3, 211–222 (2006).13. Nakamura, M. & Carninci, P. [Cap analysis geneexpression: CAGE]. T anpakushitsu Kakusan Koso49,2688–2693 (2004) (in Japanese).14. Shiraki, T. et al. Cap analysis gene expressionfor high-throughput analysis of transcriptional startingpoint and identification of promoter usage. Proc. NatlAcad. Sci. USA100, 15776–15781 (2003).15. Brenner, S. et al. Gene expression analysis bymassively parallel signature sequencing (MPSS) onmicrobead arrays. Nature Biotechnol.18, 630–634(2000).16. Peiffer, J. A. et al. A spatial dissection of theArabidopsis floral transcriptome by MPSS.BMC Plant Biol.8, 43 (2008).17. Reinartz, J. et al. Massively parallel signaturesequencing (MPSS) as a tool for in-depth quantitativegene expression profiling in all organisms. Brief. Funct.Genomic Proteomic1, 95–104 (2002).18. Nagalakshmi, U. et al. The transcriptional landscapeof the yeast genome defined by RNA sequencing.Science 320, 1344–1349 (2008).19. Wilhelm, B. T. et al. Dynamic repertoire of a eukaryotictranscriptome surveyed at single-nucleotideresolution. Nature453, 1239–1243 (2008).20. Mortazavi, A., Williams, B. A., McCue, K.,Schaeffer, L. & Wold, B. Mapping and quantifyingmammalian transcriptomes by RNA-Seq. NatureMethods5, 621–628 (2008).21. Lister, R. et al. Highly integrated single-base resolutionmaps of the epigenome in Arabidopsis.Cell133, 523–536 (2008).22. Cloonan, N. et al. Stem cell transcriptome profilingvia massive-scale mRNA sequencing. Nature Methods5, 613–619 (2008).23. Marioni, J., Mason, C., Mane, S., Stephens, M. &Gilad, Y. RNA-seq: an assessment of technicalreproducibility and comparison with gene expressionarrays. Genome Res. 11 Jun 2008 (doi:10.1101/gr.079558.108).24. Morin, R. et al. Profiling the HeLa S3 transcriptomeusing randomly primed cDNA and massively parallelshort-read sequencing. Biotechniques45, 81–94(2008).25. Holt, R. A. & Jones, S. J. The new paradigm of flow cellsequencing. Genome Res.18, 839–846 (2008).26. Barbazuk, W. B., Emrich, S. J., Chen, H. D., Li, L. &Schnable, P. S. SNP discovery via 454 transcriptomesequencing. Plant J.51, 910–918 (2007).27. Vera, J. C. et al. Rapid transcriptome characterizationfor a nonmodel organism using 454 pyrosequencing.Mol. Ecol.17, 1636–1647 (2008).28. Emrich, S. J., Barbazuk, W. B., Li, L. & Schnable, P. S.Gene discovery and annotation using LCM-454transcriptome sequencing. Genome Res.17, 69–73(2007).29. Dutrow, N. et al. Dynamic transcriptome ofSchizosaccharomyces pombe shown by RNA–DNAhybrid mapping. Nature Genet.40, 977–986(2008).30. Wu, J. Q., et al. Systematic analysis of transcribed lociin ENCODE regions using RACE sequencing revealsextensive transcription in the human genome. GenomeBiol.9, R3 (2008).31. Li, R., Li, Y., Kristiansen, K. & Wang, J. SOAP: shortoligonucleotide alignment program. Bioinformatics24, 713–714 (2008).32. Li, H., Ruan, J. & Durbin, R. Mapping short DNAsequencing reads and calling variants using mappingquality scores. Genome Res. 19 Aug 2008(doi:10.1101/gr.078212.108).33. Smith, A. D., Xuan, Z. & Zhang, M. Q. Usingquality scores and longer reads improves accuracyof Solexa read mapping. BMC Bioinformatics9,128 (2008).34. Hillier, L. W. et al. Whole-genome sequencing andvariant discovery in C. elegans. Nature Methods5,183–188 (2008).35. Campbell, P. J. et al. Identification of somaticallyacquired rearrangements in cancer using genome-widemassively parallel paired-end sequencing. NatureGenet.40, 722–729 (2008).36. Manak, J. R. et al. Biological function ofunannotated transcription during the earlydevelopment of Drosophila melanogaster. NatureGenet.38, 1151–1158 (2006).37. Hinnebusch, A. G. T ranslational regulation of GCN4and the general amino acid control of yeast. Annu.Rev. Microbiol.59, 407–450 (2005).38. Ruiz-Echevarria, M. J. & Peltz, S. W. The RNA bindingprotein Pub1 modulates the stability of transcriptscontaining upstream open reading frames. Cell101,741–751 (2000).39. T omari, Y. & Zamore, P. D. MicroRNA biogenesis:drosha can’t cut it without a partner. Curr. Biol.15,R61–64 (2005).40. Bass, B. L. How does RNA editing affect dsRNA-mediated gene silencing? Cold Spring Harb. Symp.Quant. Biol.71, 285–292 (2006).41. Sultan, M. et al. A global view of gene activityand alternative splicing by deep sequencing of thehuman transcriptome. Science 321, 956–960(2008).42. Ross-Macdonald, P. et al. Large-scale analysis of theyeast genome by transposon tagging and genedisruption. Nature402, 413–418 (1999).43. Kumar, A., des Etages, S. A., Coelho, P. S.,Roeder, G. S. & Snyder, M. High-throughputmethods for the large-scale analysis of gene functionby transposon tagging. Methods Enzymol.328,550–574 (2000).AcknowledgementsWe thank D. Raha for many valuable comments.P e r s P e c t i v e sNATURE REVIEwS |genetics VOLUME 10 | jANUARy 2009 |63© 2009 Macmillan Publishers Limited. All rights reserved。
LTE wifi 共存
I NTRODUCTIONWireless communication infrastructure is facing a great challenge with the expanding demand for wireless broadband access to Internet. A recent forecast study [1] indicates that a traffic growth beyond 500-fold is expected between 2010 and 2020, assuming the same increase in data usage is maintained. In order to improve the capacity, theThird Generation Partnership Project (3GPP)standards group has been investigating the per-formance gains obtained by small cell deploy-ment in Long Term Evolution (LTE) Release 12and beyond. On the other hand, the IEEE 802.11Working Group (WG) just ratified a new IEEE 802.11ax Task Group (TGax) primarily focused on enhancing the system performance of Wi-Fi in dense deployment scenarios [2].However, some practical issues impose limita-tions on large-scale small cell deployments. First,there are increased costs for deploying and maintaining the required infrastructure. Cus-tomers are increasingly seeing wireless Internet access as a utility, and premium taxation on faster connections becomes less of an optionsince the introduction of flat rate tariffs. So, as revenue is not increasing at the same pace as expenditures [1], capacity expansion requiring larger capital expenditures (CAPEX), such as acquisition and installation of cells, and operat-ing expenditures (OPEX) (e.g., backbone main-tenance) becomes an economic challenge. The second issue relates to the diminishing availabili-ty of radio spectrum, a fundamental resource that is both finite and expensive. Modern wire-less technologies like orthogonal frequency-divi-sion multiplexing (OFDM), relaying, and spatial multiplexing allow high spectrum usage efficien-cy to be achieved, and some researchers argue that spectrum scarcity is a non-issue due to avail-able technology [3]. Nonetheless, a bandwidth shortage of 275 MHz in the United States alone is foreseen by the end of 2014 [4].To face these challenges, cellular operators are deploying complementary network infra-structure for data delivery, a technique known as mobile traffic offloading [5]. The two main tech-nological advances to enable mobile traffic offloading are the introduction of small cell net-works and the development of dynamic spectrum access techniques for operation in license-exempt radio bands.The concept of small cells, as proposed for heterogeneous networks (HetNets), is two-fold.In the data plane, the goal is enabling the dense deployment of cells with smaller coverage areas,but capable of serving high traffic loads. On the other hand, in the control plane, the main goal is diminishing the dependence on an operator’s backbone by implementing concepts like self-organization and self-adaptation. These require-ments led 3GPP to standardize LTE small cells for operation on licensed spectrum in Release 12. 3GPP also foresees the adoption of enhanced IEEE 802.11 WLANs in unlicensed spectrum as a complementary solution. In this sense, IEEE 802.11ac networks with Wi-Fi Passpoint are a good starting point, while the IEEE 802.11ax standard (currently under development) is being considered for dense deployment scenarios. It is foreseen that by 2016, up to 30 percent of broad-band access in cellular networks will be attained over traffic offloading networks [1].A BSTRACTThe expansion of wireless broadband access network deployments is resulting in increased scarcity of available radio spectrum. It is very likely that in the near future, cellular technolo-gies and wireless local area networks will need to coexist in the same unlicensed bands. However,the two most prominent technologies, LTE and Wi-Fi, were designed to work in different bands and not to coexist in a shared band. In this arti-cle, we discuss the issues that arise from the con-current operation of LTE and Wi-Fi in the same unlicensed bands from the point of view of radio resource management. We show that Wi-Fi is severely impacted by LTE transmissions; hence,the coexistence of LTE and Wi-Fi needs to be carefully investigated. We discuss some possible coexistence mechanisms and future research directions that may lead to successful joint deployment of LTE and Wi-Fi in the same unli-censed band.Fuad M. Abinader, Jr., Erika P. L. Almeida, Fabiano S. Chaves, André M. Cavalcante, Robson D. Vieira,Rafael C. D. Paiva, Angilberto M. Sobrinho, Sayantan Choudhury, Esa Tuomaala, Klaus Doppler, and Vicente A. Sousa, Jr.Enabling the Coexistence ofLTE and Wi-Fi in Unlicensed BandsWith the increasing relevance of Wi-Fi for traffic offloading in cellular networks, improving Wi-Fi efficiency in terms of end-user perfor-mance in the presence of dense deployment of APs and STAs has become more important. Recognizing this, IEEE 802 WG created the IEEE 802.11 High Efficiency WLAN (HEW) Study Group (SG) [2] in May 2013, aiming to enhance the quality of experience (QoE) of wireless users in everyday high-density scenarios. As a result of the discussions in HEW SG, the IEEE 802.11ax Task Group (TGax) was recently established to substantially increase user throughput in dense networks with a large num-ber of users and devices, dense heterogeneous networks, and outdoor deployments. TGax includes improvements to cellular offloading as one of its major requirements, and is also inves-tigating mechanisms to increase spatial capacity with PHY-MAC enhancements to the existing IEEE 802.11 standard in the 2.4 GHz and 5 GHz radio frequency bands. A first draft of TGax amendments to IEEE 802.11 is expected to be concluded by 2016.C ONCLUSIONSThe wireless communications community has been searching for solutions to handle the increasing demand for wireless broadband access. In this context of spectrum scarcity, there has been recent discussion about allowing wire-less network technologies like LTE and Wi-Fi to coexist in the same unlicensed bands. In this article, we show that Wi-Fi is severely affected by concurrent operation of LTE in the same band. This indicates a serious need for coexis-tence mechanisms to improve the performance of both systems. The applicability of some coex-istence enabling features for both LTE and Wi-Fi are discussed, and research directions for further development of inter-technology coexis-tence are presented. We also propose coexis-tence mechanisms by reusing the blank subframe approach and the UL transmit power used in LTE, and show that it can significantly improveWi-Fi performance when coexisting with LTE in the same unlicensed bands.R EFERENCES[1] Cisco White Paper, “Cisco Visual Networking Index:Global Mobile Data Traffic Forecast Update, 2011–2016,” 2012.[2] O. Aboul-Magd, IEEE 802.11 HEW SG Proposed ProjectAuthorization Request (PAR), IEEE 802 WG Std. IEEE 802.11-14/0165r1; https:///802.11/ dcn/14/11-14-0165-01-0hew-802-11-hew-sg-proposed-par.docx[3] G. Staple and K. Werbach, “The End of Spectrum Scarcity,”IEEE Spectrum, vol. 41, no. 3, Mar. 2004, pp. 48–52. [4] Deloitte, “Airwave Overload? Addressing SpectrumStrategy Issues that Jeopardize U.S. Mobile Broadband Leadership,” Deloitte Development LLC, White Paper, Sept. 2012.[5] C. Sankaran, “Data Offloading Techniques in 3GPP Rel-10 Networks: A Tutorial,” IEEE Commun. Mag., vol. 50,no. 6, 2012, pp. 46–53.[6] I. F. Akyildiz et al., “NeXt Generation/Dynamic SpectrumAccess/Cognitive Radio Wireless Networks: A Survey,”Computer Networks, vol. 50, no. 13, Sep. 2006, pp.2127–59.[7] “FCC 10-198 Notice of Inquiry,” Nov. 2010, ET Docketno. 10-237.[8] ECC, “Technical and Operational Requirements for theOperation of White Space Devices under Geo-Location Approach,” Report 186, Jan. 2013.[9] M. Matinmikko et al., “Cognitive Radio Trial Environ-ment: First Live Authorized Shared Access-Based Spec-trum-Sharing Demonstration,” IEEE Vehic. Tech. Mag., vol. 8, no. 3, Sept. 2013, pp. 30–37.[10] M. I. Rahman et al., “License-Exempt LTE Systems forSecondary Spectrum Usage: Scenarios and First Assess-ment,” IEEE Symp. New Frontiers in Dynamic Spectrum Access Networks, DySPAN, 2011, pp. 349–58.[11] Q. Ericsson, Study on LTE Evolution for UnlicensedSpectrum Deployments, 3GPP TSG RAN Meeting 62, 3GPP TSG RAN Std. RP-131 788, Dec. 2013; http://www./ftp/tsg ran/TSG RAN/TSGR 62/Docs/RP-131788.zip[12] A. M. Cavalcante et al., “Performance Evaluation of LTEand Wi-Fi Coexistence in Unlicensed Bands,” Proc. IEEE 77th VTC 2013-Spring, Dresden, Germany, June 2013. [13] T. Nihtil et al., “System Performance of LTE and IEEE802.11 Coexisting on a Shared Frequency Band,” IEEE Wireless Commun. and Networking Conf. 2013, Apr. 2013.[14] E. P. L. Almeida et al., “Enabling LTE/Wi-Fi Coexistence byLTE Blank Subframe Allocation,” Proc. IEEE ICC ’13, 2013.[15] F. S. Chaves et al., “LTE UL Power Control for theImprovement of LTE/Wi-Fi Coexistence,” Proc. IEEE VTC 2013-Fall, Las Vegas, NV, Sept. 2013.[16] T. Baykas, M. Kasslin, and S. Shellhammer, “IEEE802.19.1 System Design Document,” IEEE 802 WG, Mar. 2010.B IOGRAPHIESF UAD M. A BINADER, J R.(fuad.abinader@.br) received his B.Sc. in computer science from Federal University of Amazonas (UFAM) in 2003, and his M.Sc. in informatics from UFAM in 2006. He is currently a doctoral student in electrical engineering at Federal University of Rio Grande do Norte (UFRN) and a researcher at Nokia Institute of Technology (INdT). His current interests include mobile Internet protocols, Wi-Fi, LTE, and WiMAX, and standard-ization in IEEE, 3GPP, and IETF.E RIKA P. L. A LMEIDA(erika.almeida@.br) received her B.Sc. in telecommunications engineering and M.Sc. degrees from the University of Brasilia (UnB), Brazil, in 2007 and 2010, respectively. She has been a researcher at INdT since 2011, where she has worked on LTE, white space concepts and coexistence issues in TV white spaces. Her current research topics include Wi-Fi evolution and cognitive radio networks.F ABIANO S. C HAVES(fabiano.chaves@.br) received his B.Sc. and M.Sc. degrees in electrical engineering from the Federal University of Ceará (UFC), Brazil, and his Ph.D. degree in electrical engineering from the University of Campinas (UNICAMP), Brazil, and the École Normale Supérieure de Cachan, France, in 2010. Since 2010, he has been a research engineer at INdT, Brazil. His research inter-ests include radio resource management, signal processing and game theory for communications, and cognitive radio systems.A NDRÉM. C AVALCANTE(andre.cavalcante@.br) received his B.Sc., M.Sc., and Ph.D. degrees in electrical engineering from Federal University of Pará (UFPA) in 2001, 2003, and 2007, respectively. Since 2007 he has worked as a research engineer at INdT on several research projects and standardization activities (IEEE). His areas of interest are the evolution of Wi-Fi networks, beyond 4G networks, and systems with multiple antennas.R OBSON D. V IEIRA(robson.d.vieira@.br) received M.Sc. and Ph.D. degrees in electrical engineering from the Catholic University of Rio de Janeiro, Brazil, in 2001 and 2005, respectively. From 2005 to 2010, he worked with white space concepts, and supporting some GERAN and 802.16m standardization activities focused on sys-tem performance evaluation at INdT. Since 2010, he is an R&D technical manager at INdT. His research interests include Wi-Fi Evolution, B4G, and cognitive radio net-works.R AFAEL C. D. P AIVA(rcdpaiva@.br) has been a researcher at INdT since 2008. He obtained his Bachelor’s degree in electrical engineering from Federal University of Santa Maria (UFSM) in 2005, his Master’s degree in signalprocessing from the Federal University of Rio de Janeiro (UFRJ) in 2008, and his Doctor’s degree in acoustics and audio signal processing from Aalto University in 2013. Among his research interests are digital signal processing and new technologies of wireless networks.A NGILBERTO M. S OBRINHO(angilberto.m.sobrinho@indt. org.br) received his M.Sc. degree in computer architec-tures from the Industrial Engineering Faculty (FEI) in 1984, and his M.Sc. in industrial automation from Federal University of Campina Grande (UFCG) in 2006. He joined as a professor of the State University of Amazonas (UEA) in 2005, and since 2012 has been working as a researcher at INdT. His main areas of interest are time synchroniza-tion in packet networks, and adaptive and phased array antennas.S AYANTAN C HOUDHURY(sayantan.choudhury@) leads the wireless research and standardization activities in NRC-Berkeley. His interests include optimization of PHY and MAC layers focusing on LTE-Advanced and next generation Wi-Fi networks. Currently, he is investigating concepts to enable dense deployments of Wi-Fi and also coexistence of LTE and Wi-Fi systems in unlicensed bands. He is the co-recipient of the 2009-2010 Sharp Labs Inventor of the Yearaward, and the 2010 IEEE Transactions on Multimediaand2012 PIMRC Best Paper awards.E SA T UOMAALA(esa.tuomaala@) received his M.S.(Tech.) degree in engineering physics and mathematics from Helsinki University of Technology, Espoo, Finland, in 2002. He joined Nokia Research Center in 2000. He is cur-rently working as a principal researcher, focusing on sys-tem-level performance evaluation of next generation wireless systems and contributing to the development of relevant IEEE standards.K LAUS D OPPLER(klaus.doppler@) received his Ph.D. from Helsinki University of Technology, Finland, in 2010 and his M.Sc. in electrical engineering from Graz University of Technology, Austria, in 2003. He joined Nokia Research Center in 2002 and currently leads the Wireless Systems team in Berkeley, California. He has been recognized several times as top inventor in Nokia. He has about 75 pending and granted patent applications, and published in 30 jour-nals, conference publications, and book chapters.V ICENTE A. DE S OUSA, J R.(vicente.sousa@ct.ufrn.br) received his B.Sc., M.Sc., and Ph.D. degrees in electrical engineering from UFC in 2001, 2002, and 2009, respectively. Between 2001 and 2006, he developed solutions to UMTS/WLAN interworking for UFC and Ericsson of Brazil. Between 2006 and 2010, he contributed to WIMAX standardization and Nokia’s product as a researcher at INdT. He is now a lectur-er at UFRN, Brazil.。
彼得·阿格雷
彼得·阿格雷
彼得·阿格雷
彼得·阿格雷1949年生于美国明尼苏达州小城诺斯菲尔德，1974年在巴尔的摩约翰斯·霍普金斯大学医学院获医学博士，现为该学院生物化学教授和医学教授。
2019年来到杜克大学，担任医学院副院长。
麦金农1956年出生，在美国波士顿附近的小镇伯灵顿长大，1982年在塔夫茨医学院获医学博士，现为洛克菲勒大学分子神经生物学和生物物理学教授。
两人将分享总额为1000万克朗（约合130万美元）的奖金。
早在100多年前，人们就猜测细胞中存在特殊的输送水分子的通道。
但直到1988年，阿格雷才成功分离出了一种膜蛋白质，之后他意识到它就是科学家孜孜以求的水通道。
阿格雷于2019年被授予诺贝尔化学奖。
诺贝尔奖评选委员会说，这是个重大发现，开启了细菌、植物和哺乳动物水通道的生物化学、生理学和遗传学研究之门。
2019年诺贝尔化学奖授予美国科学家彼得·阿格雷和罗德里克·麦金农，分别表彰他们发现细胞膜水通道，以及对离子通道结构和机理研究作出的开创性贡献。
奖项: 2019年诺贝尔化学奖
获得者: 彼得·阿格雷和罗德里克·麦金农
成就: 表彰他们在细胞膜通道方面做出的开创性贡献
个人简介:
的通道。
欧洲药典7.5版
INDEX
To aid users the index includes a reference to the supplement in which the latest version of a text can be found. For example : Amikacin sulfate...............................................7.5-4579 means the monograph Amikacin sulfate can be found on page 4579 of Supplement 7.5. Note that where no reference to a supplement is made, the text can be found in the principal volume.
English index ........................................................................ 4707
Latin index ................................................................................. 4739
EUROPEAN PHARMACOPபைடு நூலகம்EIA 7.5
Index
Numerics 1. General notices ................................................................... 7.5-4453 2.1.1. Droppers...................
- 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)。
Progress In Electromagnetics Research,PIER93,307–322,2009A HETERODYNE SIX-PORT FMCW RADAR SENSOR ARCHITECTURE BASED ON BEAT SIGNAL PHASE SLOPE TECHNIQUESB.Boukari,E.Moldovan,and S.AffesCentre´EnergieMat´e riaux et T´e l´e communicationsInstitut National de la Recherche Scientifique800de la Gaucheti`e re,Montr´e al,QC H4N1K6,CanadaK.Wu and R.G.BosisioPoly-Grames Research CenterD´e partement de G´e nie´Electrique´Ecole Polytechnique de Montr´e alMontr´e al,QC H3T1J4,CanadaS.O.TatuCentre´EnergieMat´e riaux et T´e l´e communicationsInstitut National de la Recherche Scientifique800de la Gaucheti`e re,Montr´e al,QC H4N1K6,CanadaAbstract—A Heterodyne six-port FMCW collision avoidance radar sensor configuration based on beat signal phase slope techniques is presented in this paper.Digital IF circuitry has been used in order to avoid problems related to DC offset and amplitude and phase imbalance.Simulations show that the velocity and range to the target is obtained simultaneously,with very good accuracy.Results are compared to other techniques and system architectures.Corresponding author:B.Boukari(boukari@emt.inrs.ca).308Boukari et al.1.INTRODUCTIONThe paper presents new six-port architecture using frequency modulated continuous wave(FMCW)technique,to obtain range and speed information simultaneously.This information is embedded in the spectrum of the beat signal,which is the mixing product of received signal with transmitted one.Beat frequency measurement is performed conventionally,by means of frequency counting or Fast Fourier transform(FFT)techniques.For accurate measurement,these conventional techniques require a long data acquisition time which must cover several cycles of the beat signal.For data acquisition time covering less than one cycle of the beat signal for beat frequency measuring,the phase slope method can be useful.This method was used for range and speed measurement in FMCW radar in[1,2]and for Doppler frequency measurement in[3–6].The implementation of the phase slope method requires an interferometer which can be seen in millimeter-wave domain as an I/Q-mixer.For cost efficiency and compactness six-port interferometers or six-port I/Q-mixer have been used[1–6]in the homodyne architecture in radar sensor receivers.Although homodyne FM-CW radar is known for its simplicity,it incorporates several problems:DC offset,amplitude and phase imbalance,FM-AM conversion noise.DC offset and amplitude and phase imbalance are caused by the six-port imbalance and the non-linear analog devices(diodes,transistors,etc)used for the I/Q-mixer.The FM-AM conversion noise is produced by unwanted envelope components resulting from transmission line effects,mixing, and frequency response limitations of the FM modulator[7].In fact DC offset,amplitude and phase imbalance and FM-AM conversion noise cause unacceptable measuring errors.To overcome the problem related to the DC offset,amplitude and phase imbalance,six-port calibration is performed conventionally,in the frequency domain with network analyzer and important data processing means are necessary infinal determination of speed and range[4]and[5].In order to avoid this complicated calibration with network analyzer,baseband data processing techniques,known as baseband analytical calibration were proposed in[1]and[2].Both methods do not reduce the FM-AM conversion noise.In present paper,a heterodyne architecture using a digital IF I/Q mixer is proposed for the phase slope method implementation.The heterodyne architecture reduces the FM-AM conversion noise and low frequency noise as demonstrated in[7]and[8],improving thereby the receiver sensitivity.The digital IF I/Q mixer avoid DC offset and amplitude and phase imbalance.Progress In Electromagnetics Research,PIER93,2009309 2.THE HETERODYNE RADAR SENSOR CONFIGURATIONFM-AM conversion noise is the largest noise component in a homodyne radar receiver[7].The FM-AM conversion noise is produced by unwanted envelope components resulting from transmission line effects, mixing,and frequency response limitations of the FM modulator. Given that the FM-AM conversion noise and its harmonics lie in the same frequency band as the detected beat signal,the sensitivity of the radar receiver is degraded.In order to improve the sensitivity of the receiver,the frequency range of the beat signal must be separated from that of FM-AM conversion.This is achieved by a heterodyne down-conversion.In a heterodyne configuration the beat signal is in the IF band and the FM-AM conversion noise,remaining in baseband,can be filtered by the IF band-passfilter.The heterodyne configuration alone cannot solve the remaining problems,namely DC offset,amplitude and phase imbalance.They are produced on one hand by the imbalance structure of the analog passive multi-ports(such as baluns,couplers and six-ports),and,on the other hand,by the fact that the analog non-linear components used for mixers(diodes,transistors),which generate the DC component by rectifying the millimeter-wave signals, are not ideally identical.In order to overcome the problems related to DC offset and amplitude and phase imbalance,the heterodyne down-conversion must be performed in conjunction with digital IF circuitry. The DC offset problem is solved in such a way that,in place of analog mixers,digital multipliers are used.They generate only the lower and the upper sidebands,DC and other mixing components are not generated.The amplitude imbalance results from unequal power splitting in analog configurations.In digital domain,in place of power splitting,data are duplicated.The same data go to both I and Q channels.In this manner the problem of amplitude imbalance is solved.Figure1shows the simplified operating principle of the heterodyne FMCW radar with a vector receiver using a six-port double-balanced mixer[9]and a single sideband mixer in the receiver.The millimeter-wave voltage controlled oscillator(VCO)is modulated by a linearly swept frequency signal which gives a linear frequency modulated signal at its output.A portion of the generated FMCW signal is coupled to the six-port mixer at port5and serves as reference signal.The main part of generated FMCW signal is amplified and transmitted towards the target.The signal reflected from target is received by radar antenna and amplified using a low noise amplifier(LNA).The amplified receiving signal isfirst downconverted by thefirst mixer.Because of the fact that the used six-ports are not very wide band in frequency,310Boukari et al.Figure1.Block diagram of the heterodyne FMCW radar sensor with a six-port double-balanced mixer afive-port single sideband mixer. the resulting IF frequency cannot be high.Therefore,both sideband signals move closer together in frequency,forcing the selectivity of the filter to be high enough,in order to separate the two sideband signals. Due to the fact that it is impossible to design a bandpassfilter with very sharp cut-offs at millimeter-wave frequencies,a single sideband mixer has been used.Unlike conventional mixers,single sideband mixers achieve sideband rejection through phase cancellation,notfiltering,so the frequency spacing between,the desired and undesired sidebands can be small.This means that down-conversion can be accomplished withoutfiltering,and in fewer stages,saving the cost of extra mixers, amplifiers,local oscillators,andfilters.The output signal of the single sideband mixer is mixed by the six-port double-balanced mixer with the reference signal to give the IF signal.The IF signal is amplified, A/D converted,and then downconverted to baseband with a digital I/Q mixer whose sine and cosine function are numerically generated and have their frequencies equal to the IF frequency.The outputs of the I/Q mixer are processed to obtain relative speed and range of target.Progress In Electromagnetics Research,PIER93,2009311Figure2.Block diagram of the six-port junction.Figure2shows the block-diagram of the six-port used for the double-balanced mixer,a passive circuit composed of two90◦and two180◦hybrid couplers.The relationship between output and input normalized waves is highlighted.It is to be noted that various millimeter-wave six-port front-end architectures,fabrication technologies,and modulation schemes were proposed in recent years[12,13].3.RANGE AND SPEED MEASUREMENTIn order to implement the phase slope method for beat frequency measuring in FMCW radar,DC offset,amplitude and phase imbalance and FM-AM noise must be negligible.Two solutions are offered to us for solving the problem.Thefirst one is to use simple homodyne architecture(hardware)and spend much time in signal processing (software).The second one is to use a complex heterodyne or double heterodyne architecture and spend little time in signal processing. In[1]and[2]we proposed thefirst kind of solution,namely a simple homodyne architecture with a complex signal processing to overcome the problems related to DC offset and amplitude imbalance.In the present paper,we propose the second kind of solution,namely,312Boukari et al. heterodyne architecture combined with digital IF circuitry.The heterodyne architecture reduces FM-AM noise and the digital IF circuitry keeps DC offset and amplitude and phase imbalance at minimum.In fact,after analog to digital conversion,the same data go to both I and Q channels,so that the problem of amplitude imbalance is avoided.The DC offset is not generated due to the fact that,in place of analog mixers,digital multipliers are used.They produce only, lower and upper sidebands.The lower sideband is,in case,of direct conversion a baseband signal without DC offset.The upper sideband can befiltered easily by the Finite Impulse Response Filters(FIR). The local oscillator for digital I/Q mixer is a numerically controlled oscillator whose sine and cosine functions are generated accurately, i.e.,same amplitude and90◦phase shift between them,so that the phase imbalance is avoided.The complex beat signal at the digital I/Q mixer outputs (Figures1)is:¯S=SI+jS Q.(1) Due to the fact that DC offset and amplitude and phase imbalance are negligible,the I/Q signals can be written as follows:S I=A cos(2πf b t+ϕ0)=A cosϕ,(2)S Q=A sin(2πf b t+ϕ0)=A sinϕ.(3) From(2)and(3)the beat signal instantaneous phase is found tobe:ϕ=2πf b t+ϕ0=Arc tanS QS I.(4)From(4)the beat frequency can be determined as follows:f b=12πdϕdt.(5)Figure3shows the waveform of the FMCW signal.It is the same as that used in[2].The top part of thisfigure shows plots of frequency versus time for signal transmitted by radar(solid line),and return signal from target received by radar(dotted line).The bottom part of figure shows frequency of output signal generated by radar mixer with transmitted and received input signal frequencies.The frequency shift due to the range of target moving away from the radar is[8]:f R=f b(up)−f d.(6) f b(up)and f d are,respectively,the beat frequency at rising chirp and the Doppler frequency.It is to be noted that range can be calculatedProgress In Electromagnetics Research,PIER93,2009313ff∆Figure3.FMCW waveforms.using f R as follows[8]:R=cf R2f chirp∆f.(7)From f d one canfind the relative speed[8]:v r=cf d2f0(8)where:c is the speed of light in vacuum.f0,the center frequency,∆f,the bandwidth of the FMCW signal andF chirp=1/T chirp is the chirp cycle.4.SIMULATION RESULTSDesign and simulation consist of three sections as follows:(i)design and electromagnetic simulation of multi-ports:six-port junction,five-port junction and a Wilkinson power divider,(ii)large-signal harmonic balance simulation of the six-port double-balanced mixer,and the single sideband mixer,and(iii)envelope simulation of the heterodyne radar system according to Figure1.314Boukari etal.216yout of the six-portjunction.six-port double-balanced mixer.4.1.Six-port Double-balanced MixerThe six-port junction was simulated with electromagnetic simulation tool software of Agilent Technologies:Momentum of Advanced Design System(ADS).The circuit was designed in MHMIC technology on a 127µm alumina substrate with a relative permittivity of9.9.Figure4 shows the layout of the six-port junction.The outer dimensions are 6.5mm by6.5mm.In order to perform advanced system simulations,S parameters of six-port model have been saved in a datafile.This datafile is then used as data input to large-signal harmonic balance simulation of mixer along with Spice model of Microwave Device Technology(MDT)flip-chip Schottky diodes.Figure5shows the block schematic of the six-port double-balanced mixer.In order to provide a ground return at the IF frequencies,four λ/4short-circuit stubs were used.As known,these stubs are open circuits for RF and LO signal.Oneλ/4open-circuit stub is connected to the mixer output preventing LO and RF leakage in IF path.In addition,matching networks(MN)are inserted at the inputs of all Schottky diodes.Figure6shows the spectrum at the IF output port.As seen, all the harmonics are efficiently suppressed.Their levels are at least 130dB under the IF signal level.This fact is to be explained that, beside suppression of harmonics due to the balanced structure of the mixer,the six-port itself acts as afilter.Figure5(b)represents the zooming of the Figure5(a)in the neighborhood of the IF frequency. The levels of all spurious products of the form mf RF–mf LO are at least70dB under the IF signal level and theirfiltering with an extra filter is no more necessary.Progress In Electromagnetics Research,PIER 93,2009315Frequency (GHz) −−I F S p e c t r u m (d B ) (b)I F S p e c t r u m (d B ) Figure 6.IF output spectrum of the six-port double-balanced mixer.LO Power (dBm)C o n v e r s i o n l o s s (d B ) 3.56.05.55.04.54.01056789Figure 7.Conversion loss versus LO power of the six-port double-balanced mixer.Figure 7shows the conversion loss value versus the LO power.In order to improve the conversion loss,the LO power must be increased.A conversion loss of 3.6dB is obtained with 10dBm LO power.316Boukari et al.4.2.The Single Sideband MixerA diode single sideband and image-reject mixer has been analyzed in[10].This configuration is only suitable for low microwave frequency applications because at millimeter-wave frequencies the diodes must be matched and the ring configuration of the diodes does not allow matching the diodes easily.Furthermore air-bridges which cannot be designed with good performances at millimeter-wave frequencies are needed in place of path crossings.In order to overcome theses problems,we propose to use a multiport in the designing of a single sideband mixer.For this aim,the configuration of the diode single sideband mixer analyzed in[10]has been modified as follows:(i)the monopolar excitation of LO signal has been replaced by a differential excitation,then,(ii)LO and RF ports have been interchanged.Figure8shows the modified diode ring single sideband mixer configuration.The analysis of LO and RF phase relationships(at the diodes)yields to a multi-port configuration.Figure9shows the block-diagram of the multi-port single sideband and image-reject mixer.In order to provide a ground return,matching networks(MN)with short-circuit stubs are inserted at the inputs of all Schottky diodes.Figure10shows the block-diagram of resultingfive-port. Commercial full-wave software(High Frequency Structure Simulator (HFSS)version10)of Ansoft Corporation was used for the design and simulation of thefive-port.The circuit was designed for MHMIC technology on a127µm substrate with a relative permittivity of9.9.Figure11shows the lay-out of thefive-port.The outer dimensions are,approximately4.5mm by4.5mm.In order to perform advanced system simulations,S parameters offive-port model were saved in a datafile.This datafile was then used as data input to large-signal harmonic balance simulation of mixer along with Spice model of Microwave Device Technology(MDT)flip-chip Schottky diodes.Figure8.Modified diode ring single sideband mixer configuration.Figure 12shows the conversion gain value versus the LO voltage.The sideband suppression is maximal at 1.3Volts of the LO voltage.Figure 13shows the spectrum at the output port.The level of the upper sideband is 40dB under the level of the lower sideband signal.Furthermore,the spurious products of the form f RF ±mf LO are at least 40dB under the low sideband signal level and their filtering with an extra filter is no morenecessary.Figure 9.Block-diagram of the multi-port single sideband and image-rejectmixer.Figure 10.Block-diagram ofresulting five-port.21345Figure y-out of the five-port.Upconv. Downconv. RFfreq. = 77 GHz LOfreq. = 1 GHz 2.50−20−40−60−800.5 1.0 1.5 2.0LO Voltage (V)C o n v e r s . G a i n (d B ) Figure 12.Conversion gain value versus the LO voltage.Frequency (GHz) Lower S. Upper S.RFfreq. = 77 GHzLOfreq. = 1 GHz−50−10074757677787980R F s p e c t r u m (d B ) Figure 13.Spectrum at the output port of the single sideband mixer.4.3.The Heterodyne Radar SystemThe heterodyne radar system according to Figure 1has been simulated on envelope simulator of ADS.ADS system models of VCO,amplifiers,mixers,multipliers and filters have been used.In addition,models have been used for various path lengths taking into account:signal attenuation (radar equation [8]),propagation delay (between radar and target),and Doppler effect.The radar cross-section of target has been set to 1m 2.The simulated radar parameters have been for the transmitters as follows:∆f =200.0MHz,f 0=77.0GHz,f chirp =1/T chirp =83.333kHz,VCO power has been 10dBm,PA gain has been set at 10dB,the coupling factor of the directional couplers has been −10dB.f LO 1=f IF 1=1GHz,f LO 2=950MHz and f IF 2=f LO 1−f LO 2=40MHz.In addition,transmitter and receiver antenna gains are 20dBi each.S I (V)S Q (V) Figure plex beat signalfor the speed of 20m/s.S Q (V) S I (V)Figure plex beat signal for the range of 25m.N u m b e r o f s i m u l a t i o n s Range, (m) SD =0.17%282015105222324252627Figure 16.Histograms of the range for a target situated at 25m from the radar.Figures 14and 15show the complex beat signals in polar form for a relative target speed of 20m/s and target range of 20m,respectively.All these curves describe circles centered at origin of polar coordinates systems,showing that errors related to FM/AM noise,DC offset,amplitude and phase imbalance are negligible.In order to do a statistical evaluation of range and speed obtained by simulations,a range resolution equal to about half of the 77GHz wavelength and a speed resolution equal to half of the 77GHz wavelength per second were chosen.Up to 50simulations for each range and speed value were performed.Range measurement histograms are also obtained,as show in Figures 16and 17(for distances of 25m and 45m,respectively).The standard deviations are 0.044m (0.17%)and 0.097m (0.21%)for the 25m and 45m,respectively.N u m b e r o f s i m u l a t i o n s Range, (m) SD =0.21%4842434445464720151050Figure 17.Histograms of therange for a target situated at 45mfrom the radar.Speed, (m/s)N u m b e r o f s i m u l a t i o n s SD =0.045%201510502218192021Figure 18.Histograms of the speed for a target speed of 20m/s.Table 1.Statistical simulation results comparison.SD for thespeed of20m/sSD for the range of 25m SD for the range of 45m Heterodyne FMCW radar(phase slope based rangeand speed measurement)0.009m/s (0.045%)0.044m (0.17%)0.097%(0.21%)Homodyne FMCWrader in [2](phaseslope based range andspeed measurement)0.003m/s (0.015%)0.07m (0.28%)0.283m (0.6%)Rader sensor in [5](phase defferencebetween two properlyspaced frequenciesbased range measurement)– 2.8% 1.8%Figure 18shows histograms for speed of 20m/s.These histograms show negligible dispersions of range and speed values.In addition,Table 1shows standard deviations (SD)in comparison with those obtained in [2]and [6].Present results show that method of determining range by evaluating the slope of the instantaneous phase of beat signal is more accurate than past methods of determining range from phase difference between two CW signals properly spaced in frequency. Moreover,these results show that the heterodyne radar system with a single sideband mixer is more accurate is range measuring than the homodyne configuration from[2],due to the fact that,in heterodyne configurations FM-AM conversion noise is completely suppressed.5.CONCLUSIONA Heterodyne FMCW collision-avoidance radar sensor,based on phase slope techniques for range and speed measuring,has been presented.Range and speed are determined by evaluating the slope of instantaneous phase of beat signal.Simulation results show that the heterodyne radar sensor with single sideband mixer offer good performances.Range measurements derived from slope of instantaneous phase of baseband beat signal is found to be more accurate than method of determining range from phase difference between two CW signals properly spaced in frequency.Moreover, the results show that the heterodyne radar system with a single sideband mixer is more accurate is range measuring than the homodyne configurations using the same phase slope method. ACKNOWLEDGMENTThis work wasfinancially supported by the National Science Engineering Research Council(NSERC)of Canada. REFERENCES1.Boukari,B.,E.Moldovan,S.Affes,K.Wu,R.G.Bosisio,andS.O.Tatu,“Six-port FMCW radar sensor configurations,”21st Canadian Conference on Electrical and Computer Engineering CCECE2008,305–308,Niagara Falls,Canada,2008.2.Boukari, B., E.Moldovan,S.Affes,K.Wu,R.G.Bosisio,and S.O.Tatu,“A77GHz six-port FMCW collision-avoidance radar sensor with baseband analytical calibration,microwave and optical technology letters,”(submitted for publication).3.Li,J.,K.Wu,and R.G.Bosisio,“A collision avoidance radarusing six-port phase/frequency discriminator,”IEEE MTT-S International Microwave Symposium,1553–1556,1994.4.Stelzer, A., C.G.Diskus,K.Luebke,and H.W.Thim,“Amicrowave position sensor with submillimeter accuracy,”IEEE Transactions on Microwave Theory and Techniques,Vol.47,2621–2624,1999.5.Migu´e lez,G.C.,B.Huyart,E.Bergeault,and P.L.Jallet,“Anew automobile radar based on the six-port phase/frequency discriminator,”IEEE Transactions on Vehicular Technology, Vol.49,1416–1423,2000.6.Moldovan,E.,S.O.Tatu,T.Gaman,K.Wu,and R.G.Bosisio,“A new94GHz six port collision avoidance radar sensor,”IEEE Transactions on Microwave Theory and Techniques,Vol.52,751–759,2004.7.Saito,T.,N.Okubo,Y.Kawasaki,O.Isaji,and H.Susuki,“An FM-CW radar module with front-end switching heterodyne receiver,”IEEE MTT-S International Microwave Symposium, 713–716,1992.8.Chang,K.,RF and Microwave Wireless Systems,John Wiley&Sons,Inc.,2000.9.Boukari, B., E.Moldovan,S.Affes,K.Wu,R.G.Bosisio,and S.O.Tatu,“A low cost millimeter-wave six-port double-balanced mixer,”International Symposium on Signal System and Electronics,ISSSE2007,513–516,2007.10.Henderson, B. C.and J. A.Cook,“Image-reject and single-sideband mixers,”WJ Tech.Note,1985.11.Zavrell,Jr.,R.J.,“New-low power single sideband circuits,”Phillips Semiconductor Application Note AN1981,1997.12.Khaddaj Mallat,N., E.Moldovan,and S.O.Tatu,“V-bandquadrature phase shift keying de-modulator using WR-12six-port,”Progress In Electromagnetics Research Letters,Vol.6,193–199,2009.13.Khaddaj Mallat,N.,E.Moldovan,and S.O.Tatu,“Comparativedemodulation results for sixport and conventional60GHz direct conversion receivers,”Progress In Electromagnetics Research, PIER84,437–449,2008.。