现代电子技术2025，Vol.48Issue(10)：15-19,5.DOI:10.16652/j.issn.1004-373x.2025.10.003
面向卫星通信的非线性迭代学习混沌通信
Nonlinear iterative learning chaotic communication for satellite communication
摘要
Abstract
In allusion to the demand for secure transmission performance in satellite communication,especially in application scenarios without secure payloads,a chaotic communication model optimized by the improved nonlinear iterative learning is proposed.The model training of the encrypted signal mixed with chaotic carrier signals and raw information in a specific ratio is conducted by means of the long short-term memory neural network(LSTM)to obtain neural network model highly consistent with the parameters of the laser transmitter,so as to solve the problem of incomplete matching between the receiver and transmitter system parameters in chaotic communication systems.In order to further improve the synchronization quality of chaotic signals,iterative learning is introduced to conduct the parameter optimization of the LSTM.The decryption recognition rate of the proposed nonlinear chaotic communication synchronization model based on improved LSTM is finally stable at 94.07%,which is 4.03%and 1.82%higher than those of the radial basis function(RBF)neural network chaotic secure communication model and the chaotic secure communication model based on convolutional neural network,respectively,verifying that the proposed communication model has good comprehensive performance.关键词
卫星通信/混沌通信/迭代学习/长短期记忆神经网络/混沌同步质量/参数优化Key words
satellite communication/chaotic communication/iterative learning/long short-term memory neural network/chaotic synchronization quality/parameter optimization分类
信息技术与安全科学引用本文复制引用
李华,王广昌,陆凌蓉,陈承,叶乐清,郑小芳..面向卫星通信的非线性迭代学习混沌通信[J].现代电子技术,2025,48(10):15-19,5.基金项目
江苏省自然科学基金项目(BK20230108) （BK20230108）
