基于DCS的统计MIMO雷达信号模型及参数估计

Signal Model and Parameters Estimation of Statistical MIMO Radar Based on Distributed Compressed Sensing

分布式压缩感知(Distributed Compressed Sensing,DCS)将单信号的压缩采样扩展到信号群的压缩采样,利用信号内相关性和互相关性对多个信号进行联合重构。统计多输入多输出(Multiple-Input Multiple-Output,MIMO)雷达系统通过多发多收配置,在发射机、目标以及接收机之间构成对目标的分布式探测系统。该文将DCS应用到统计MIMO雷达中,通过对该场景中目标回波的延时在距离空间稀疏性的分析,提出联合所有接收信号重构目标场景的设想,建立了接收信号的联合稀疏模型,并实现了目标参数估计的联合重构算法。仿真结果表明与基于压缩感知(Compressed Sensing,CS)的算法相比,基于DCS的算法在进一步降低采样数目的同时提高了参数估计精度,同时也验证了DCS-MIMO雷达可以有效克服目标的雷达散射截面积(Radar Cross Section,RCS)起伏。

Distributed Compressed Sensing （DCS） extends compressive sampling from single signal to multl-signal ensembles. It also enables joint recovery that exploits both intra- and inter-signal correlation structures. Statistical MIMO radar systems that are made up of widely separated transmit/receive antennas form distributed detection systems for targets among transmitters, targets and receivers. In this paper, DCS is applied to statistical MIMO radars, and through the analysis of sparisty of the delays of target echo signals in the range space, the idea is proposed to construct target scene by joining all received signals. It also establishes the joint sparsity model of received signals, and gives joint reconstruction algorithms that can estimate target parameters. Simulation results show that, compared with the algorithm based on CS, the one based on DCS increases the parameter estimation accuracy while offering a reduction in the number of measurements. It is also validated that DCS -MIMO radars can effectively overcome target RCS fluctuations.