结合多信号模型与遗传算法的板级电路测点选取方法-豆柴文库

您所在位置：网站首页 / 结合多信号模型与遗传算法的板级电路测点选取方法.docx / 文档详情

在线预览结束，喜欢就下载吧，查找使用更方便

5 金币

下载文档

/ 2

下载提示

如果您无法下载资料，请参考说明：

1、部分资料下载需要金币，请确保您的账户上有足够的金币

2、已购买过的文档，再次下载不重复扣费

3、资料包下载后请先用软件解压，在使用对应软件打开

文本预览

结合多信号模型与遗传算法的板级电路测点选取方法
Introduction
Inelectronics,themeasurementofcircuitperformanceisofutmostimportancetoensurethatthedesignspecificationsaremet.ItisparticularlyimportantinthecaseofPCB(printedcircuitboard)leveltestingwheretheentirefunctionalityofacomplexcircuitistested.Theprocessofcircuittestingisdonethroughtheuseoftestpoints.Testpointsarespecificareasinthecircuitthatcanbeprobedfortestingpurposes.Theaccuracyofelectricalmeasurementsdependsonthetypeandplacementofthesetestpoints.
Inthispaper,weproposeanovelapproachtotestpointselectionusingmultiplesignalmodels(MSMs)andgeneticalgorithms(GAs).TheaimofthisapproachistooptimizethelocationandnumberoftestpointsonaPCB.TheproposedmethodwillimprovecircuitreliabilityandwillreducethecostandtimerequiredtotestPCBs.
MultipleSignalModels
Theconventionalmethodsoftestpointselectionrelyontheanalysisofindividualsignalsandtheirrespectivenoiselevels.However,circuitscontainmanysignals,andeachsignalhasitsowncharacteristics.TheMultipleSignalModel(MSM)techniqueovercomesthislimitationbyconsideringmultiplesignalsintheirentirety,ratherthanjustindividualsignals.
TheMSMtechniqueisbasedontheuseofmathematicalmodelstodescribethecircuitbehavior.Eachsignalismodeled,andthemodelisusedtopredictthebehaviorofthesignalunderdifferentconditions.Themodelscanbeusedtosimulatethecircuitresponsetovariousinputsandprovideinsightintothecircuitbehavior.
TheMSMapproachhasbeenshowntobeeffectiveinoptimizingthelocationandnumberoftestpointsindifferenttypesofcircuits,includingDCcircuits,RFcircuits,andmixed-signalcircuits.Theapproachcanbeappliedtobothanaloganddigitalcircuits.
GeneticAlgorithms
Geneticalgorithm(GA)isanoptimizationalgorithmthatmimicsthenaturalprocessofevolution.Itisaprobabilisticsearchalgorithmthatcanefficientlysolvecomplexoptimizationproblems.GAsareparticularlyeffectiveinsolvingproblemswherethesearchspaceislargeandtheproblemisnon-linear.
TheGAalgorithmworksbygeneratingapopulationofcandidatesolutionsandthenevolvingthepopulationoversuccessivegenerations.Fitnessfu