基于压缩感知的弱观测雷达信号重构算法改进研究

Research on the Improvement of Weak Observation Radar Signal Reconstruction Algorithm Based on Compressed Sensing

针对雷达信号在弱观测条件下采样数据随机丢失的问题,引入一种基于压缩感知理论的雷达残缺信号修复方法,并根据传统压缩感知重构算法的不足,提出一种改进算法——JAWBMP算法。该方法利用雷达信号的稀疏特性,采用改进重构算法实现残缺信号的修复,改进算法在压缩采样匹配追踪(CoSaMP)算法框架下,通过稀疏度预估计和递归的方式逐步逼近信号的稀疏度,采用基于广义Jaccard系数匹配准则的原子匹配方式,通过弱选择和回溯的方法优化初始原子候选集。理论分析和仿真实验结果表明,改进算法可以实现盲稀疏度信号的重构,重构精度优于同类经典算法,在信号随机丢失60%的条件下仍可以较好地实现雷达残缺信号的修复。

Aiming at the random loss problem of sampling data of radar signals under weak observation conditions,a radar incomplete signal repair method based on compressed sensing theory is introduced. According to the shortcomings of traditional compressed sensing reconstruction algorithm,an improved algorithm JAWBMP algorithm is proposed. In this method,the sparse characteristics of radar signal are used,and the improved reconstruction algorithm is used to repair the incomplete signal. The improved algorithm gradually approximates the sparsity of the signal through sparsity pre-estimation and recursion mode under the framework of CoSaMP algorithm. The atomic matching method based on generalized Jaccard coefficient matching criterion is used to optimize the initial atomic candidate set by weak selection and backtracking methed. Theoretical analysis and simulation results show that the improved algorithm can reconstruct the blind sparsity signal,the reconstruction accuracy is better than that of the similar classical algorithms,and can still repair the radar incomplete signal under the condition of 60%random signal loss.