Research On Direction Of Arrival Estimation Method For Wideband Array Signals Based On Sparse Representation

Array signal processing technology has been widely used in various applications such as radar,sonar and wireless communications.Direction of Arrival(DOA)estimation,as an important branch of array signal processing,has become a hot research topic.Due to the increasingly complex transmission environment,wideband signals have received more and more attention.This is because they bear a large amount of information,which is beneficial to target detection,signal positioning and parameter estimation.Traditional narrowband signal estimation methods cannot be directly applied to wideband signal estimation.Therefore,it is necessary to investigate DOA estimation methods specifically for wideband signals.This thesis mainly studies the DOA estimation methods of wideband signals.We first introduce the classical incoherent signal subspace method(ISSM)and coherent signal subspace method(CSSM)in wideband array signal processing.The ISSM method requires the narrowband signal DOA estimation method to be performed for each narrowband data model,which in general poses high computational complexity,while the CSSM method only needs to apply the narrowband signal processing method once on the signal subspace of the reference frequency point after passing a focusing operation.CSSM methods can not only estimate coherent sources,but also can improve estimation performance and reduce computational complexity compared to ISSM methods.Next,the thesis introduces a weighted wideband DOA estimation method,that is based on group sparse characteristics.This method first reconstructs the array signal model using the group sparse characteristics,then uses the weighted L1 SVD method for estimation of the DOA of each sub-band signal,and finally averages the results from all sub-bands to obtain the DOA estimation result of the wideband signal.Our method introduces weighting operation to improve the sparsity of solutions.Compared with the weighted wideband DOA estimation method of joint sub-bands,the proposed method uses the group sparsity to reduce the sampling data and reduce the computational complexity.Finally,the thesis proposes a DOA estimation method for wideband signals based on atomic norm minimization(ANM).This method is composed of two parts: focus processing and DOA estimation.Firstly,the focus processing uses a focusing method without angle estimation to process the data at each frequency point;secondly,the ANM method is employed to estimate the DOA of wideband signals.This method can be applied to coherent signal source estimation,and the performance advantage gets more prominent when the signal sources are highly correlated.