DOD-DOA Joint Estimation Of Bistatic MIMO Radar Based On Compressed Sensing

The traditional radar Angle estimation is based on the signal processing of Nyquist sampling theorem in the background of gaussian noise.In order to suppress the influence of noise,the method of cumulant is usually used to process the signal.Although the DOD-DOA joint estimation of MIMO radar based on cumulant method can achieve better parameter estimation performance,it has some problems such as complex algorithm,large computation and poor real-time performance.Based on this,some scholars applied the compressed sensing technology to the Angle estimation of MIMO radar.This algorithm broke through the limitation of Nyquist signal sampling and reduced the computation to some extent.Due to the high requirement of real-time performance of bistatic MIMO radar,the classical compressed sensing method needs to be further optimized to reduce computation and simplify hardware space resource occupation.Aiming at the above problems,this paper improves the compressed sensing technology and applies it to the target angle estimation of MIMO radar.First of all,this paper introduces the improvement on a smooth function reconstruction algorithm,and on this basis,using a smooth l₀ norm with threshold reconstruction algorithm,the method by setting a threshold in the iterative process,and to compare the relative changes of every iteration step to determine iterative efficiency,skip the meaningless iteration,to accelerate the algorithm,improve the performance of the algorithm.To solve this problem,a reconstruction algorithm based on the sparse cyclic matrix of Toeplitz block is adopted,which has high recovery accuracy,low storage capacity and is hardware friendly.Simulation results show that under the background of Gaussian noise,the smoothing l₀ norm reconstruction algorithm with threshold value adopted in this paper can speed up the signal processing speed and the algorithm converges quickly.The reconstruction algorithm based on the sparse cyclic matrix of Toeplitz block can greatly reduce the computation and the hardware burden.The two methods have high recovery precision to the signal.