Irregular Array Direction Finding Precision Technology

Spatial spectrum estimation is a kind of high precision direction-finding technology, and it has been widely used in civil and military equipment. But the spatial spectrum estimation algorithm is also face with many problems. Firstly, Most of the algorithms are suitable for uniform line array (ULA), but it’s no longer applicable for irregular array. Secondly, compared with the direction finding for linear array in one-dimension, the amount of calculation increased significantly in the2-D. Thirdly, many algorithms performance is lower or completely failure in dealing with incoherent signal. Finally, the accuracy of algorithm is related to many factors, such as the array element number, the structure of array, the channel error, snapshots, SNR(Signal to Noise Ratio) and so on. Three problems are mainly analyzed in this paper. First is the relationship between the structure of planar array and the accuracy of direction finding. Another is the2-D DOA estimation of MUSIC algorithm in coherent signal situation. Third is the large amount of calculation in the ML algorithm.In allusion to problems above, there are some new solutions below.(1) Based on the differential geometry of the array manifold, a circular array design method is proposed. The feasibility of the method is verified in computer simulation. And the active calibration and passive calibration have been studied for2-D DOA estimation of irregular array.(2) Based on the traditional MUSIC algorithm, the coherent signal cannot be processed. Then, the modify MUSIC (MMUSIC) algorithm is used to realize the2-D DOA estimator for the coherent signal. Applied range of the modify MUSIC algorithm has been extended from1-D uniform linear array (ULA) to2-D centre-symmetric array.(3) In order to overcome the problem existing in the majority of solution of maximum likelihood estimation, a new improved CE-ML algorithm of2-D DOA estimator is proposed. Comparing with the PSO (Particle Swarm Optimization) algorithm, simulation results show that the improved CE-ML algorithm performs better, and iteration number of improved CE-ML algorithm is one-third of PSO algorithm.(4) Taking the floating-point TS201S series DSP chip as the platform, modified MUSIC algorithm and improved CE-ML algorithm have been used to accomplish the DOA estimation. A variety of methods have been adopted to reduce the single measuring time of DOA estimation, and this has had very good results.