Research On Robust Direction-finding Algorithm Based On The Regularized Sparse Recovery

The research of robust direction finding algorithm is an important branch of signal processing,especially in unmanned driving,radar,medical imaging and so on.Most of the traditional direction finding algorithms are based on subspace algorithms,but these algorithms are difficult to estimate the angle of the target in the case of low signal-to-noise ratio?SNR?or limited number of snapshots,and when the signal is coherent,the performance of most subspace algorithms will decline or even fail to work.With the rise of compressed sensing theory,sparse recovery theory is well known and used by people.Compared with subspace algorithm,it can adapt to the harsh direction finding environment in reality.At the same time,due to the interference of electromagnetic field between array antennas,there will be unknown mutual coupling effect between array elements,which destroys the traditional structure of array receiving data model.There-fore,this paper will study the robust direction finding algorithm based on uniform linear array?ULA?and Monostatic MIMO radar background under the condition of unknown mutual coupling.Firstly,the basic concepts of array signal model and matrix analysis are introduced,and the steps and principles of subspace-based direction finding algorithm are intro-duced.Secondly,the basic theory of sparse recovery and related representative algo-rithms are introduced.Finally,the estimation performance of these algorithms is ana-lyzed and compared through the simulation experiments.Secondly,a robust array direction finding algorithm based on data domain is pro-posed to solve the problem of unknown mutual coupling between antenna elements.In this paper,a block structure model based on parameterized mutual coupling vector ma-trix is proposed to represent the received data under the condition of unknown mutual coupling.The robust direction finding algorithm of weighted block sparse recovery ar-ray is studied in detail.Finally,the robust sparse recovery performance of the algorithm under the condition of unknown mutual coupling effect is verified by simulation exper-iments.Then,aiming at the shortcomings of robust array direction finding algorithm based on data domain,several robust array direction finding algorithms based on covariance domain are proposed.Firstly,the sparse representation of covariance matrix and ₁l-SRACV algorithm are introduced.Secondly,the algorithm of weighted norm minimi-zation and weighted kernel norm minimization are proposed for block sparse matrix,which not only has block sparsity but also has rank sparsity.In order to further improve the utilization of received data information,the block sparse recovery algorithm based on the weighted subspace fitting is proposed.Finally,through simulation analysis and performance comparison,the superiority of the proposed algorithm is verified.Finally,based on the background of the Monostatic MIMO radar,a robust algo-rithm of Monostatic MIMO radar via weighted block sparse reconstruction is proposed.In this algorithm,the data model of MIMO radar is constructed by parameterizing the transmit-receive matrix,and then the DOA estimation problem of MIMO radar is trans-formed into the weighted block sparse recovery problem by constructing the weighting matrix.The algorithm can also avoid the influence of unknown mutual coupling and the loss of array aperture,thus improving the DOA estimation performance of Mono-static MIMO Radar under unknown mutual coupling.