chaotic communication systems an iterative learning perspective - mingxuan sun

《chaotic communication systems an iterative learning perspective - mingxuan sun》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

1、2010InternationalConferenceonElectricalandControlEngineeringChaoticCommunicationSystems:AnIterativeLearningPerspectiveMingxuanSunCollegeofInformationEngineeringZhejiangUniversityofTechnologyHangzhou,310023,ChinaEmail:mxsun@zjut.edu.cnAbstractInthispaper,the

2、problemofchaoticsecurecom-overafinitetimeinterval.Inourproposedscheme,nonlinearmunicationisaddressedfromaniterativelearningperspective,chaoticmaskingisadoptedandthedifficultyintheinversionwheretheplaintextspansoverapre-specifiedfinite-timeinterval.computingisov

3、ercome.TheplaintextisassumedtobegivenInthecryptographicscheme,theplaintextismodulatedthroughoverapre-specifiedtimeinterval.Thelearningalgorithmuseschaoticparametermodulation,andthenchaoticallymaskedbyanonlinearmechanism.Anefficientlearningalgorithmisexploited

4、onlythetransmittedsignal,andtheplaintextcanberecoveredfordecryption,andthesufficientconvergenceconditionisderivedfromtheciphertextthroughlearning.Ontheotherhand,bywhichthelearninggaincanbechosen.Furthermore,twonosynchronizationoccurs.Theiterativelearningmeth

5、odisfundamentaltechniquesaregiveninordertoenhancesecurity.comparabletotheinversionsystembasedapproach[10].ourCasestudiesarecarriedouttodemonstratetheeffectivenessofproposedlearningalgorithmissimpleandneedsnotinversiontheproposedmethod.computing.Comparedwith

6、theexistingparameterestimationI.INTRODUCTIONalgorithms,I)ourproposedschemedoesnotassumethatAchaoticsystemexhibitsunpredictableandrandom-theparameterstobeestimatedareslowlytimevarying,orseemingbehavior.Nevertheless,itsstatevariablesaredifferentiable;andII)th

7、ePEconditionisnotrequiredforbounded,andevolvewithtimeinadeterministicmanner.ourscheme,butacheckableconvergenceconditionasanBecauseofthisreason,chaosisconsideredadesirableprop-alternativeisgivenforthelearninggainselection.ertyinpotentialapplicationinsecureco

8、mmunication.InII.CHAOTICCRYPTOSYSTEMtheapplicationofchaostocryptography,thedrive-responsesynchronizationmethodwassuggestedin[1],whichhasgoneIterativelearningoffersanefficienttoolforestimatingtime-throug