基于循环平稳度准则的多路雷达信号识别算法

Multichannel Radar Signal Recognition Algorithm Based on DCS

在日益复杂的电磁环境中分选识别出雷达信号,是电子对抗发挥功用的先决因素。关于雷达信号调制样式与信号参数的先验信息有限,难以为信号分选提供充足的情报支撑,且信号交叠严重制约着信号分选的效能。将上述需求转换为盲源分离问题,通过Givens变换构造高阶分离矩阵,将适用于两路信号的基于3阶循环量的循环平稳度(DCS)盲源分离算法拓展到适用于具有不同循环平稳频率的多路信号。通过理论推导证明了该方法的可行性,并推导出构造Givens矩阵参数确定的方法。利用循环平稳理论提取雷达信号在循环平稳域的特征,结合DCS分离准则进行仿真验证。仿真结果表明,该算法能够实现对多路雷达信号的有效分选。

Recognizing the radar signal in complex electromagnetic environment is the necessary prerequisite for electronic countermeasures to play a role. The priori information about signal modulation and signal parameter is limited,which cannot provide enough intelligence support for signal sorting. In addition,the mixture of signals restricts the effectiveness of signal sorting. The issue mentioned above is converted to a blind source separation. A high-order disjunction matrix is established with Givens transform,and the blind source separation algorithm with degree of cyclostationarity（ DCS） based on the third-order cyclic statistics which is suitable for two channel signals is expanded to the multichannel signals with different cyclostationarity frequencies. The feasibility of the proposed method is proved by theoretical derivation,and the method for establishing the parameters of Givens matrix is derived. The features of radar signal in cyclostationary domain are extracted with cyclostationarity theory. The method is simulated with DCS separation principles. The simulated results show that the algorithm can realize the effective sorting of multichannel radar signals.