基于反向训练模型的网络通信安全传输研究

Research on Secure Transmission of Network Communication Based on Reverse Training Model

针对正向训练模式下的0/1式跳空收发技术在无线网络通信传输信息泄露后存在的安全隐患,论文提出了权值反馈型反向训练技术。该技术通过在发射方和目标用户之间进行反向训练,使目标用户获得与正向训练模式下一致的权值,而侦听方即便窃取了无线网络通信传输信息,由于无线网络信道差异性,仍然无法正确解调,这一点完全不同于传统的跳频扩频方案。反向训练技术以从根源上解决无线网络通信安全问题为出发点,充分利用了无线网络信号丰富的时空谱资源,保证了物理层安全传输。仿真结果表明,在无线网络通信传输信息和反馈的权值都泄露的情况下,侦听方的误符号率始终极高,根本无法侦听信息。

ince 0/1 space hopping technique in forward training mode has security problems after the leak of network communication transmission information,the backward training technique with weight-feedback is proposed. Through backward training between the transmitter and the target user,the target user can obtain the weights equivalent with that from forward training,while the eavesdropper cannot demodulate correctly because of the diversity of the wireless channel even if she steals the network communication transmission information,which is completely different from traditional frequency hopping spread spectrum technique. The proposed technique,which aims at solving the problem of secure wireless communications fundamentally,makes full use of the abundant space-time spectrum resources of the wireless signals and ensures the secure physical layer transmission. Simulation results show that even when both the network communication transmission information and the feedback weights are leaked,the eavesdropper keeps a high symbol error rate and cannot intercept information.