一种基于盲分离的欺骗干扰抑制算法

An Algorithm of Deception Jamming Suppression Based on Blind Signal Separation

该文针对单天线雷达的欺骗干扰,研究了一种基于盲分离的欺骗干扰抑制方法。该方法在对接收信号进行分段提取的基础上,借助特征矩阵联合对角化(JADE)的盲分离技术获得多段分离信号,然后利用分选技术获取完整源信号,最后根据数字射频存储技术(DRFM)的相位量化特性鉴别目标和干扰。仿真结果表明,在高信噪比(SNR)下算法对欺骗干扰具有较好的抑制性能,且当SNR>15 dB时,目标回波相似系数大于85%;相位量化位数低于4位时,鉴别概率接近100%,目标延迟时间测量误差小于0.3μs。

This paper presents a suppression method based on Blind Signal Separation （BSS） to counter deception jamming in the single channel. First, the received signal is cut into 3 sections. Next, the mixed one is separated into two parts by the Joint Approximate Diagonalization of Eigenmatrices （JADE） BSS technology. Then, those 4 parts are combined with each other to form two complete signals, namely, echo and deception jamming. Finally, the target and deception jamming are identified by the character of phase quantization. The experiments validate the effectiveness of the proposed algorithm. This approach has good performance under high SNR. When SNR〉 15 dB, the similarity coefficient of target echo is larger than 85%. Under the condition of 4 quantization bits, the probability of identification is almost close to 1 and the measurement error of target delay is below 0.3 μs.