简介：Theobjectiveofsteganographyistohidemessagesecurelyincoverobjectsforsecretcommunication.Howtodesignasecuresteganographicalgorithmisstillmajorchallengeinthisre-searchfield.Inthisletter,developingsecuresteganographyisformulatedassolvingaconstrainedIP(IntegerProgramming)problem,whichtakestherelativeentropyofcoverandstegodistributionsastheobjectivefunction.Furthermore,anovelmethodisintroducedbasedonBPSO(BinaryParticleSwarmOptimization)forachievingtheoptimalsolutionofthisprogrammingproblem.Experimentalresultsshowthattheproposedmethodcanachieveexcellentperformanceonpreservingneighboringco-occurrencefeaturesforJPEGsteganography.

简介：Inradartargettrackingapplication,theobservationnoiseisusuallynon-Gaussian,whichisalsoreferredasglintnoise.Theperformancesofconventionaltrackersdegradeseverelyinthepresenceofglintnoise.Animprovedparticlefilter,MarkovchainMonteCarloparticlefilter(MCMC-PF),isappliedtocopewithradartargettrackingwhenthemeasurementsareperturbedbyglintnoise.Trackingperformanceofthefilterisdemonstratedinthepresentofglintnoisebycomputersimulation.

简介：Targettrackingisoneofthemainapplicationsofwirelesssensornetworks.Optimizedcomputationandenergydissipationarecriticalrequirementstosavethelimitedresourceofthesensornodes.Aframeworkandanalysisforcollaborativetrackingviaparticlefilterarepresentedinthispaper.Collaborativetrackingisimplementedthroughsensorselection,andresultsoftrackingarepropagatedamongsensornodes.Inordertosavecommunicationresources,anewGaussiansumparticlefilter,calledGaussiansumquasiparticlefilter,toperformthetargettrackingispresented,inwhichonlymeanandcovarianceofmixandsneedtobecommunicated.BasedontheGaussiansumquasiparticlefilter,asensorselectioncriterionisproposed,whichiscomputationallymuchsimplerthanothersensorselectioncriterions.Simulationresultsshowthattheproposedmethodworkswellfortargettracking.

简介：Acceleratingtheconvergencespeedandavoidingthelocaloptimalsolutionaretwomaingoalsofparticleswarmoptimization(PSO).TheverybasicPSOmodelandsomevariantsofPSOdonotconsidertheenhancementoftheexplorativecapabilityofeachparticle.Thusthesemethodshaveaslowconvergencespeedandmaytrapintoalocaloptimalsolution.Toenhancetheexplorativecapabilityofparticles,aschemecalledexplorativecapabilityenhancementinPSO(ECE-PSO)isproposedbyintroducingsomevirtualparticlesinrandomdirectionswithrandomamplitude.Thelinearlydecreasingmethodrelatedtothemaximumiterationandthenonlinearlydecreasingmethodrelatedtothefitnessvalueofthegloballybestparticleareemployedtoproducevirtualparticles.TheabovetwomethodsarethoroughlycomparedwithfourrepresentativeadvancedPSOvariantsoneightunimodalandmultimodalbenchmarkproblems.ExperimentalresultsindicatethattheconvergencespeedandsolutionqualityofECE-PSOoutperformthestate-of-the-artPSOvariants.

简介：Thenovelfan-shapedself-scanningphotodiodearray(SSPA)hastheadvantagesofthehighsensitivitywithsmalldevicesizeandtheserialvideooutputmode.UsingnovelSSPAinsteadofthepotodiodearayofasemicircuarannulardetectorsinthelaserdiffractionparticlesizeanalyzer,thesecorrectnessoftheresultsareverifiedbyvariousparticlesamplesmeasuringandthepracticalrunningovertenthousandhoursinthedesuplhurationtower.

简介：Inordertodesignacomplexlaserresonatorwithmulti-parameters,themethodofparticleswarmoptimization(PSO)algorithmisemployed.Theparametersinfluencingtheresonatorstabilityandmodesizedistributionaretakenintoconsideration,andthestabilitycriteriaindexandthemodesizedistributionareusedastargetvalues.TheabsolutevaluesofthedifferencesbetweenpracticalandthetargetvaluesaresetasthefitnessfunctionforthePSO.Byminimizingthefitnessfunction,alaserresonatorwiththeoptimizedcavityparameterscanbefound.TheanalysesforthedesignexampledemonstratethefeasibilityandvalidityofthePSOmethodinthecomputeraideddesignofmul-ti-parameterslaserresonator.ApplyingPSOalgorithmintheintelligentdesignofsolidstatelaserresonatorscanrealizethe.transitionfrommanualtrial-and-errortocomputerintelligentdesignofthelaserresonators.

简介：Energyconsumptionofsensornodesisoneofthecrucialissuesinprolongingthelifetimeofwirelesssensornetworks.Oneofthemethodsthatcanimprovetheutilizationofsensornodesbatteriesistheclusteringmethod.Inthispaper,weproposeagreenclusteringprotocolformobilesensornetworksusingparticleswarmoptimization(PSO)algorithm.Wedefineanewfitnessfunctionthatcanoptimizetheenergyconsumptionofthewholenetworkandminimizetherelativedistancebetweenclusterheadsandtheirrespectivemembernodes.Wealsotakeintoaccountthemobilityfactorwhendefiningtheclustermembership,sothatthesensornodescanjointheclusterthathasthesimilarmobilitypattern.Theperformanceoftheproposedprotocoliscomparedwithwell-knownclusteringprotocolsdevelopedforwirelesssensornetworkssuchasLEACH(low-energyadaptiveclusteringhierarchy)andprotocolsdesignedforsensornetworkswithmobilenodescalledCM-IR(clusteringmobility-invalidround).Inaddition,wealsomodifytheimprovedversionofLEACHcalledMLEACH-C,sothatitisapplicabletothemobilesensornodesenvironment.SimulationresultsdemonstratethattheproposedprotocolusingPSOalgorithmcanimprovetheenergyconsumptionofthenetwork,achievebetternetworklifetime,andincreasethedatadeliveredatthebasestation.

简介：Ahierarchicalparticlefilter(HPF)frameworkbasedonmulti-featurefusionisproposed.TheproposedHPFeffectivelyusesdifferentfeatureinformationtoavoidthetrackingfailurebasedonthesinglefeatureinacomplicatedenvironment.Inthisapproach,theHarrisalgorithmisintroducedtodetectthecornerpointsoftheobject,andthecornermatchingalgorithmbasedonsingularvaluedecompositionisusedtocomputethefirstorderweightsandmakeparticlescentralizeinthehighlikelihoodarea.Thenthelocalbinarypattern(LBP)operatorisusedtobuildtheobservationmodelofthetargetbasedonthecolorandtexturefeatures,bywhichthesecond-orderweightsofparticlesandtheaccuratelocationofthetargetcanbeobtained.Moreover,abacksteppingcontrollerisproposedtocompletethewholetrackingsystem.Simulationsandexperimentsarecarriedout,andtheresultsshowthattheHPFalgorithmwiththebacksteppingcontrollerachievesstableandaccuratetrackingwithgoodrobustnessincomplexenvironments.

简介：Awayofresolvingspreadingcodemismatchesinblindmultiuserdetectionwithaparticleswarmoptimization(PSO)approachisproposed.IthasbeenshownthatthePSOalgorithmincorporatingthelinearsystemofthedecorrelatingdetector,whichistermedasdecorrelatingPSO(DPSO),cansignificantlyimprovethebiterrorrate(BER)andthesystemcapacity.Asthecodemismatchoccurs,theoutputBERperformanceisvulnerabletodegradationforDPSO.Withablinddecorrelatingscheme,theproposedblindDPSO(BDPSO)offersmorerobustcapabilitiesoverexistingDPSOundercodemismatchscenarios.

简介：Aimingtoreducethecomputationalcostsandconvergetoglobaloptimum,anovelmethodisproposedtosolvetheoptimizationofacostfunctionintheestimationofdirectionofarrival(DOA).Inthismethod,ageneticalgorithm(GA)andfuzzydiscreteparticleswarmoptimization(FDPSO)areappliedtooptimizethedirectionofarrivalandpowerparametersofthemodesimultaneously.Firstly,theGAalgorithmisappliedtomakethesolutionfallintotheglobalsearching.Secondly,theFDPSOmethodisutilizedtonarrowdownthesearchfield.InFDPSO,achaoticfactorandacrossovermethodareaddedtospeeduptheconvergence.Thisapproachhasbeendemonstratedthroughsomecomputationalsimulations.ItisshownthattheproposedalgorithmcanestimateboththeDOAandthepowersaccurately.Itismoreefficientthansomepresentmethods,suchastheNewton-likealgorithm,Akaikeinformationcritical(AIC),particleswarmoptimization(PSO),andgeneticalgorithmwithparticleswarmoptimization(GA-PSO).

简介：Afuzzyparticleswarmoptimization(PSO)onthebasisofelitearchivingisproposedforsolvingmulti-objectiveoptimizationproblems.First,anewperturbationoperatorisdesigned,andtheconceptsoffuzzyglobalbestandfuzzypersonalbestaregivenonbasisofthenewoperator.Afterthat,particleupdatingequationsarerevisedonthebasisofthetwonewconceptstodiscouragetheprematureconvergenceandenlargethepotentialsearchspace;second,theelitearchivingtechniqueisusedduringtheprocessofevolution,namely,theeliteparticlesareintroducedintotheswarm,whereastheinferiorparticlesaredeleted.Therefore,thequalityoftheswarmisensured.Finally,theconvergenceofthisswarmisproved.TheexperimentalresultsshowthatthenondominatedsolutionsfoundbytheproposedalgorithmareuniformlydistributedandwidelyspreadalongtheParetofront.

简介：Arationalapproximationmethodofthefractional-orderderivativeandintegraloperatorsisproposed.TheturningfrequencypointsarefixedineachfrequencyintervalinthestandardOustaloupapproximation.IntheimprovedOustaloupmethod,theturningfrequencypointsaredeterminedbytheadaptivechaoticparticleswarmoptimization(PSO).TheaveragevelocityisproposedtoreducetheiterationsofthePSO.Thechaoticsearchschemeiscombinedtoreducetheopportunityoftheprematurephenomenon.Twofitnessfunctionsaregiventominimizethezero-poleandamplitude-phasefrequencyerrorsfortheunderlyingoptimizationproblems.Somenumericalexamplesarecomparedtodemonstratetheeffectivenessandaccuracyofthisproposedrationalapproximationmethod.