虚拟通道扩展的引信欺骗干扰盲分离算法

为滤除引信接收信号中的欺骗干扰,研究了基于虚拟通道扩展的过渡干扰滤除FastICA(TJE-FastICA)和双系统FastICA(DS-FastICA)算法.算法利用不同周期接收信号特性以及信号非圆性来扩展单通道,并对传统FastICA进行噪声抑制改进,在此基础上提出了利用接收信号的循环相关输出幅值统计特性来剔除过渡干扰的TJE-FastICA算法,以及利用两个盲分离系统对分组信号同时分离的DS-FastICA算法.仿真结果表明,提出的算法明显优于正交投影算法和直接盲分离算法,DS-FastICA比TJE-FastICA具有2.5 dB的优势,且当信噪比R_SN>7.5 dB时,两算法分离的回波信号相似度均在80%以上,DS-FastICA的分离时间比TJE-FastICA算法快91%. 

BSS Algorithm of Fuze Deception Jamming Based on Virtual Channel Extension

Algorithms called TJE-FastICA and DS-FastICA were proposed to filter out deception jamming in fuze. First, the channels were extended by making use of the noncircularity and the characteristic of the signals in different circles, and the original FastICA was improved to suppress noise. Then, TJE-FastICA was proposed by eliminating transitional interference using the statistical characteristics of amplitude of the cycling-related outputs, DS-FastICA was proposed by using two BSS systems to separate two sets of signals. Simulation results show that the proposed algorithms are superior to algorithms of orthogonal projection and direct BSS, DS-FastICA is superior to TJE-FastICA with 2.5 dB, and when R_SN>7.5 dB, the similarity of the separated echo in the two algorithms are greater than 80%, and the separation time of DS-FastICA is nearly 91% faster than TJE-FastICA.