Research On The Application Of Compressive Sensing In MIMO Radar Parameter Estimation

As a kind of new radar system, Multiple Input Multiple Output(MIMO) radar has many advantagescompared with conventional array radar systems. However, as same as other traditional high resolution radar systems, the MIMO radar is faced with high sampling rate, large amount of data and rapid processing problems.In recent years, the sparse signal processing theory, namely the Compressive Sensing(CS), breakthrough the limitation of the traditional Nyquist sampling theorem. It turns the sampling of the signal into the sampling of the information, thus can reduce the sampling requirement of the radar systems, and also can improve parameter estimation performance. The main research content of this article is as follows:Firstly, introduced the research background and the classification of MIMO radar, and the echo signal model was established of colocated MIMO radar. Based on this model, the author studied the application of LS method, Capon, APES and CAML in the colocated MIMO radar parameter estimation. Then compared and analyzed the performance of the four methods through simulation experiments, and pointed out the advantages and disadvantages of each method.Secondly, the sparse representation of signal, sampling design and optimization of the matrix and the signal reconstruction were deeply studied, and focuses on a kind of perception matrix optimization method based on matrix decomposition. The method can effectively reduce correlation of the sample matrix.Related sparse signal reconstruction algorithms were introduced in detail, and then compared the performance of the reconstruction method.Finally, Established the sparse representation model of colocated MIMO radar echo signal, and compressive sampling theory was applied to parameter estimation in MIMO radar. Due to the faults that the robustness of reconstruction algorithm of CS used in MIMIO radar is so low, introduced the bayesian estimation model in parameter estimationof MIMO radar, and by solving an alternating optimization problem, estimated the signal of interest and the noise at the same time. Thus the scheme could greatly improve the MIMO radar parameter estimation robustness to noise, and several simulation results also proved it. And it could also achive good result for estimation under shorter coherent accumulation time than traditional methods. In order to further dig the advantages of the application of CS in MIMO radar parameter estimate, proposed a parameter estimate model of MIMO radar receiver system based on Kronecker compressive sensing. This model brought compressive sampling forward, and it is expected to reduce the complexity of the MIMO radar parameter estimation.Precise dalay estimation could be achived based on this model.