基于高阶累积量的二次雷达应答信号特征分析

摘要

二次监视雷达应答信号混扰将导致解码错误,在解码前需进行有效分离.信号的高阶累积量被广泛应用于盲分离算法,主要利用了信号的非高斯性.建立二次雷达应答信号模型,进而根据其零恒模特性分析高阶累积量特征,得到其三至五阶累积量近似为零,表明二次雷达应答信号具有高斯特征.对独立成分分析算法、投影算法等分选算法的性能进行了分析比较.仿真结果表明,基于高阶累积量的盲分离算法不适用于二次雷达应答信号分选.%To overcome decoding errors problem of secondary surveillance radar(SSR)due to the phenomenon of garble,effective signal separation should be done before decoding. High order cumulants have been widely used in blind source separation algorithms,which mainly use the non-Gaussian feature of signals. Firstly,a signal model for the SSR replies is established. Then,high order cumulants are analyzed according to the zero constant modulus(ZCM)property,and the cumulants of order three-five are proved to be almost equal to zero,which shows that the SSR replies have the character of Gaussian. Finally,the separation performance is compared among independent component analysis(ICA)algorithm,projection algorithm(PA)and so on.Numerical simulations show that the algorithms based on high order cumulants are not suitable for SSR replies.