面向衛(wèi)星通信的非線性選代學習混沌通信

打開文本圖片集

關鍵詞：衛(wèi)星通信；混沌通信；選代學習；長短期記憶神經網絡；混沌同步質量；參數(shù)優(yōu)化中圖分類號：TN911.7-34；TP391.41 文獻標識碼：A 文章編號：1004-373X（2025）10-0015-05

Abstract：Inalusion tothedemand forsecure transmission performanceinsatelitecommunication，especially in applicationscenarioswithoutsecurepayloads，achaoticcommunicationmodeloptimized bytheimprovednonlinear iterative learning isproposed.Themodeltrainingof theencryptedsignalmixedwithchaoticcariersignalsandrawinformationina specificratioisconductedbymeansof thelngshort-term memoryneuralnetwork（LSTM）toobtainneuralnetworkmodelhighly consistentwiththeparametersofthelasertransmiter，soastosolvetheproblemofincompletematching betweenthereceiver andtransmitersystemparametersinchaoticcommunicationsystems.Inordertofurtherimprovethesynchronizationqualityof chaoticsignals，iterativelearningisintroducedtoconducttheparameterotimizationoftheLSTM.Thedecryptionrecognition rateof theproposednonlinearchaoticcommunicationsynchronizationmodelbasedonimprovedLSTMisfinallstableat 94.07% ，which is 4.03% and 1.82% higher than those of the radial basis function （RBF） neural network chaotic secure communicationmodelandthechaoticsecurecommunicationmodelbasedonconvolutionalneuralntwork，respectivelyerifying that the proposed communication model has good comprehensive performance.

Keywords：satelitecommunication;chaoticcommunication；iterativelearning;longshort-termmemoryneuralnetwork; chaotic synchronization quality;parameter optimization

0 引言

近年來，全球各國航天技術發(fā)展迅猛，以星鏈和千帆為代表的低軌道通信衛(wèi)星成為研究熱點。（剩余5731字）