Research On Spatial Spectrum Direction-of-arrival Estimation And Beamforming For Generalized Co-prime Arrays

Direction-of-Arrival(DOA)estimation and beamforming(BF)are two branches of array signal processing,which have been applied in many fields.In practical applications,the performance of classical algorithms is degraded due to the existence of imperfect factors,such as the off-grid DOA estimation,the mutual coupling effect of array elements,the coherent sources,the destruction of array elements,etc.Moreover,in the current complex signal environment,the number of signal processing often exceeds the number of physical elements.Therefore,it is worthy of people's attention to improve the freedom of the array,reduce or even eliminate the non ideal factors of the array.In recent years,a variety of non-uniform arrays have been proposed,such as nested array,coprime array,semi-coprime array,augmented nested array and so on,but they don't have a unified concept.Therefore,we first introduce the concept of generalized coprime array and describe three kinds of commonly non-uniform array structures.Then we focus on the non ideal factors of the array and the sidelobe level of the physical array.The main work and research contents are as follows:1.In this thesis,we first give the concept of generalized coprime array,and introduce the structure of three commonly generalized coprime arrays.Then,three commonly used DOA estimation methods are studied.Due to the problem of grid division in angle search,this thesis analyzes this problem and studies the algorithm based on grid estimation error.In addition,this thesis also analyzes the performance of several algorithms.2.There are various errors in the actual received data model,such as the mutual coupling effect of array elements,the coherence between sources,and so on.Therefore,this thesis makes a deep analysis of the mutual coupling estimation model of the array,and proposes a sparse Bayesian DOA estimation algorithm based on compressed sensing,which can effectively solve the problem of DOA estimation in the presence of mutual coupling of the receiving data.and the performance is better than the above algorithm.Compared with the algorithms commonly used in compressed sensing theory,the performance of the proposed algorithm is very good.3.At present,there is little research on the model of the coherence between the sources under the generalized coprime array.Therefore,in this thesis,the DOA estimation algorithms based on the iterative adaptive approach and the N-root through covariance matrix are introduced.Because of the above algorithm's shortcoming,this thesis presents a novel way to estimate the DOA of coherent and uncorrelated sources for the generalized coprime array.In this proposed algorithm,the problem of coherent sources is solevd effectively and the degree of freedom(DOF)of generalized coprime array is improved fully.Besides,in addition to these two innovations,the proposed algorithm can also detect which sources are coherent.It is not existed in the current academic field.4.Aiming at a series of problems of beamforming in the current academic field,this thesis focuses on how to carry out beamforming on the physical array,while the sidelobe level is very low.This thesis introduces the beamforming algorithm based on forward linear prediction,studies two static beamforming algorithms,and analyzes the depth of reducing the side lobe level.Through the simulation experiment,the reliability and applicability of the algorithm are proved.