Research On Direction-of-arrival Estimation Algorithm Of Space Targets Based On Sparse Sensing

As one of the most important issues in modern signal processing, array signalprocessing is widely applied and rapidly developed in military and economic field likeradar, sonar, communication, seismic prospecting，radio astronomy and medical imaging.DOA (direction of arrival) estimation of spatial targets has become a widely research topicwhich plays a major role in array signal processing. Some limitations exist in traditionalDOA estimation methods: angular resolution of beam-forming technique relies on theamount and inner spacing of array elements, namely two far-field sources can bedistinguished only when their angle interval is larger than beam width, which is calledRayleigh limit; for spatial spectrum estimation, large snapshots are needed to acquiresufficient statistic on covariance matrix of observed signals, which needs observationperiod for sources or targets, meanwhile, performance of this method will be dramaticallydecreased in low SNR or coherent signals situation; interferometer angle estimation method,which has been widely used in practice, is only competent for single targets estimation andneeds ambiguity resolution, although it has a simple principle and practical application.In the recent years, compressed sensing method based on sparse signal reconstructionhas been deeply researched, which indicates that if signal is sparse or compressed, thereconstruction can be achieved by much lower sampling than Nyquist rate. Owing to thesparsity of signals in spatial domain, DOA estimation based on compressed sensing (CS)becomes more and more popular. So CS-DOA method in different snapshot situation isstudied in this paper, and modified algorithm is proposed for DOA estimation when largenumber of snapshots occur, which can get higher angular resolution and reconstructionprobabilty in multiple coherent sources situation and needs lower calculation complexity.