简介：Andynamicsystemforreal-timeobstacleavoidancepathplanningofredundantrobotsisconstructedinthispaper.Firstly,theinter-framedifferencemethodisusedtoidentifythemovingtargetandtocalculatethetargetarea,thenonthebasisofcolorfeaturesandgradientfeaturesextractedfromthetargetarea,thefeaturefusionCam-Shiftmeanshiftalgorithmisusedtotracktarget,improvingtherobustnessofthetrackingalgorithm.Secondly,aparalleltwo-channeltargetidentificationandlocationmethodbasedonbinocularvisionisproposed,updatingthetarget’sthree-dimensionalinformationinrealtime.Then,adynamiccollision-freepathplanningmethodisimplemented:thesafetyrodsareremovedthroughtheintersectiontest,andtheminimumdistanceisderiveddirectlybyusingthecoordinatevaluesofthetargetinthelocalcoordinatesystemoftherod.Onthisbasis,theobstacleavoidancegainandescapevelocityrelatedtotheminimumdistanceisestablished,andobstacleavoidancepathplanningisimplementedbyusingthezerospacemappingmatrixofredundantrobot.Experimentsareperformedtostudytheefficiencyoftheproposedsystem.

简介：Thispaperaddressestheissueofreal-timedatatransmissioninvehiclestoroadside(V2R)environmentbythehybridcoordinationfunction(HCF)ofcontrolledchannelaccess(HCCA)specifiedbytheIEEE802.11pstandard.HCCAisoneofthemediumaccessmechanismsintheIEEE802.11standard,andusesthepollingschemesimilartothepointcoordinationfunction(PCF)toprovidereliablequalityofservice(QoS),whichmaycauseresourceovershootingandhightimeconsuming.ThispapertriestoimprovetheperformanceoftheHCCApollingschemeanddesignsnewdatatransmissionschedulingscheme.Thesimulationresultsoftheproposalarecomparedwiththoseofthestandardstrategy.SimulationresultsdemonstratethattheimprovedHCCAhaslowerdelay,lossrate,andhigherthroughputthanthoseofthestandardone.

简介：Storingthewholedatabaseinthemain-memoryisacommonmethodtoprocessreal-timetransactioninreal-timedatabasesystems.TherecoverymechanismofMain-memoryReal-timeDatabaseSystems(MMRTDBS)shouldreflectthecharacteristicsofthemain-memorydatabaseandreal-timedatabasebecausetheirstructuresarequitedifferentfromotherconventionaldatabasesystems.Inthispaper,therefore,weproposeamulti-levelrecoverymechanismformain-memoryreal-timedatabasesystemswithExtendibleChainedBucketHashing(ECBH).Owingtotheoccurrenceofreal-timedatainreal-timesystems,weshouldalsoconsideritinourrecoverymechanism.Accordingtoourperformancetest,thismechanismcanimprovethetransactionconcurrency,reducingtransactions'deadlinemissingrate.

简介：在这份报纸，一个频率领域决定反馈均衡器为单身的搬运人传播被建议，时间颠倒时空块(TR-STBC)编码。斜决定反馈均衡器矩阵能从频率领域隧道反应是计算的，这被显示出。在完美的反馈假设下面，建议均衡器能接近匹配的过滤器界限(MFB)。与存在时间领域决定反馈均衡器相比，建议均衡器与一样的均等复杂性展出更好的性能。

简介：Thispapergivesanewmethodintotheevaluatingsystemoftelecominvestmentprojects,I.e.RealOption.ThismayovercomethedefectsresultedfromemployingNetPresentValue(NPV),whichisnowusedintheevaluationoftelecomprojects.AtheoreticalanalysisofRealOptionisprovided,followedbyanexampleoftelecominvestmentprojecttoillustratethedifferencesbetweenthetwomethods.

简介：一种技术把重叠频率时间分割称为multiplexing(OVFTDM))在这篇文章被建议。这种技术从Nyquist系统和频率时间部门multiplexing系统被导出。没有orthogonality，信号是什么时候，简洁地及时重叠了领域，，当没有orthogonality，信号简洁地在频率领域被重叠时，这种技术被称为重叠时间部门multiplexing(OVTDM)，这种技术被称为重叠频率部门multiplexing(OVFDM)。进一步改善光谱效率，在哪个发信号的OVFTDM在频率领域两个都被重叠，及时，域被探索。OVFTDM不取决于无论什么的orthogonality及时域或在象Nyquist系统或OFDM系统一样的频率领域，但是在在信号之中的convolutional限制关系上。因此，不是仅仅光谱效率而且可靠性被改进。模拟验证这个理论的有效性。

简介：Thestatisticcharacteristicsofchaoticsequencesgeneratedbyimprovedlogistic-mapareanalyzedanditisfoundthatimprovedlogistic-mapchaoticsequenceshavegoodcorrelationandtheycanbeusedasaddresssequencesinspread-spectrumcommunication.Thediscrete-timesynchronizationofHenonmapisimmediate.Weapplythediscrete-timesynchronizationtochaoticspread-spectrumcomuunicationandproposeanoriginalcommunicationscheme.Thesimulationshowsthattheapplicationissuccessful.

简介：Accuratetimedelayestimationisveryimportantinmultichannelnoisesuppressionsystems.Thispaperhasproposedanewmethodtoreducetheeffectofnoiseontheestimationoftimedelaybetweendifferentchannels.Ournewtimedelayalgorithmwillmoreconsiderablyreducetheestimationerror,comparedwiththeconventionalautocorrelationfunctionmethod.

简介：Consideringthejointchannelestimationanddatadetectionintime-varyingorthogonalfrequencydivisionmultiplexing(OFDM)andaddressingtransmissionperformancedegradationinducedbythesevereinter-carrierinterference(ICI)atveryhighspeed,anewprogressiveiterativechannelestimationschemeisproposed.ToalleviatetheerrorpropagationoftheinaccuratedataduetoICI,themeasurementsubcarriersintheKalmanfilterisdesignedtobeextendedfrompilotssubcarrierstoallthesubcarriersprogressivelythroughtheiterations.Furthermore,initerationprocess,theinterferenceofthenon-pilotdatatothemeasurementsubcarriersisconsideredtobepartofnoiseinthemodifiedKalmanfilter,whichimprovestheestimationaccuracy.Simulationindicatesthattheproposedschemeimprovestheperformanceinfasttime-varyingsituation.

简介：Opticaltimedivisonmultiplexing(OTDM)isoneofthepromisinigwaysforthefuturehigh-speedopticalfibercommunicationnetworks.All-opticalswitchis,beingoneofthecoretechnologiesofOTDMsystemsandnetorks,crucialtorealizethevaroussignalprocessesincludingtime-divisiondemultiplexing,packerswitching,all-opticalregeneratingandsoon.Thisthesismainlystudiesvariousall-opticalswitchtechnologiesandtheirutilizationinthefieldsofall-opticalsignalprocessingsintheOTDMsystemsandnetworks.Themainjobsarelistedasfollows.(1)Anovelall-opticalultrafastdemultiplexingschemeusingthesolitonself-trappingeffectinbirefringentfiberisproposed.(2)ThedemultiplexingperformanceoftheNonlinearOPticalLoopMirror(NOLM)isthoroughlyanalyzedanditsoptimizationisfurtherdiscussed.(3)Theperformanceanalysisandtheconfigurationoptimizationoftheall-opticalswitchesbasedontheSemiconductorOpticalAmplifier(SOA)aresystematicallypresented.Thespeedlimitationoftheall-opticalSOAswitchesinducedbythefastgaindepletionofSOAisdiscussed,Besides,anovelSOAswitchisproposed,whichadoptstheasymmetricMach-ZehnderInterferomerterconfiguration.(4)The8×2.5Gb/sOTDMexperimentaltransmissionsystemalong105kmstandardfiberisrealizedusingtheNOLMdemultiplexer.(5)TheNOLMswitchisusedtorealizetheall-optical3Rregenerationof2.5G/sReturn-to-Zerosignal.(6)ThefeasibilityandlimitationoftheallopticalSOApacketswitchisdiscussed.AndadevelopedMZIconfigurationofSOApacketswitchisfurthershowntoimprovethepackerswitchingperformance.Finally,anall-opticalpacketdroppingnodesuitableinthenetworkswithringorbusconfingurationandanall-opticalpacketswithcingnodeintheShuffleNetnetworksareproposedtoshowthefeasibilityofall-opticalpacketswitchingthroughcombiningtheall-opticalswitchesandthereasonablelogicdecisions.

简介：这篇文章考虑由与利用一个单个一半双的继电器的相应目的地交流的多重来源组成的一个无线网络，而，来源机会主义地使用继电器。由与有效电容的概念集成信息理论，这篇文章在瞄准最大化继电器网络产量的无线继电器网络上建议动态时间分配策略，题目到服务(QoS)的给定的延期质量限制，在哪儿预定分割多重存取(TDMA)在继电器网络被使用。模拟结果证明建议分配策略罐头显著地作为与传统的相等的时间分配策略相比改进有效能力。

简介：与高精确性提取multipath的参数能被为时间域使用高分辨率的算法完成极端宽带(UWB)隧道建模。干净算法为UWB时间域描述被用作如此的一个高分辨率的算法。这份报纸论述压缩察觉到(CS)为时间域UWB隧道建模基于高分辨率的deconvolution算法。UWB无线隧道是长、稀少的隧道推动反应(CIR)的一个主要例子。而且，仔细匹配multipath的波形的parameterized波形的字典导致UWB隧道测量信号更简洁地被代表。由调整字典的参数，不同分辨率的CIR能被获得。匹配的追求(MP)算法为CS被用作信号重建方法并且直接输出CIR。我们也证明如果CS的字典明确地被设计，MP是单个模板的等价物干净。最后，CS-MP的计算复杂性被分析并且MP的比较并且干净被执行。模拟结果显示出那与相比干净，建议CS-MPdeconvolution算法能与许多少数完成可比较的表演采样。

简介：AnovelandefficienttechniquetodesignmodulateddiscreteFouriertransform(DFT)filterbanksisintroducedinthispaper.Theproposedmethodfirstrelaxesthelimitsthatthesynthesisfiltersarethetime-reversedversionoftheanalysisfiltersandthenadoptsthetimedomainformulaoftheperfectreconstructionpropertyasthesolutiontodesignthesynthesisfilters.TheprototypefilterinanalysisfilterbanksisdesignedbasedonFourier-Kaiserwindowapproach.Simulationresultsshowthatthedesignedfilterbanksapproximatelysatisfytheperfectreconstructionwithcontrollablereconstructionerrors.

简介：在我们调查了的这份报纸，downlink的表演概括了分布式的天线系统(GDAS)。在空间相关褪色的条件的假设下面，我们根据Bessel功能的系列导出相关系数的数字表达式，并且提起了事件的吝啬的角度的范围限制。而且，当我们从空空间使用了基础向量而不是直角的集合获得一样的系统性能时，分布式的概括分层的空间时间代码(GLST)的体系结构被认为以便完成multiplexing获得和差异获得但是与更低的复杂性。而且，以便最大化能力，Gerschgorin圆基于快天线选择算法包括那些模拟结果的讨论被评估了。

简介：是的巨大的部门操作广泛地为象图形的处理那样的设备作为最昂贵的操作之一接受了联合起来的完美地匹配的层(CPML)算法受不了的卷绕旋转的平行加速(GPU)，地可编程的门数组(FPGA)等等。在更高的效率和更低的电源消费的追求，这篇文章重游CPML理论并且建议了新快没有部门的平行CPML结构。由最佳地重排CPML内部重复进程，所有部门操作符能被recalculating消除并且代替脱机更新系数的相关的域。实验证明没有任何精确性损失，建议没有部门的结构能节省超过50%算术指令和传统的平行CPML结构的25%实行时间。

简介：Inlongtermevolution(LTE)uplinksinglecarrierfrequencydivisionmultipleaccess(SC-FDMA)system,therestrictionthatmultipleresourceblocks(RBs)allocatedtoausershouldbeadjacent,makestheresourceallocationproblemhardtosolve.Moreover,withthepracticalconstraintthatperfectchannelstateinformation(CSI)cannotbeobtainedintime-varyingchannel,theresourceallocationproblemwillbecomemoredifficult.Inthispaper,anefficientresourceallocationalgorithmisproposedinLTEuplinkSC-FDMAsystemwithimperfectCSIassumption.Firstly,theresourceallocationproblemisformulatedasamixedintegerprogrammingproblem.Thenanefficientalgorithmbasedondiscretestochasticoptimizationisproposedtosolvetheproblem.Finally,simulationresultsshowthattheproposedalgorithmhasdesirablesystemperformance.