Research On Method Of Narrow-band Signal DOA Estimation Based On Compression Sensing

Direction of Arrival(DOA)estimation is an important research content in the field of array signal processing.The method of using DOA estimation to obtain the parameter information of the signal,such as the incoming direction of the signal,has been widely used in military and civilian fields such as radar and wireless communication.With the increase of the signal frequency and the bandwidth,the current DOA estimation methods based on spatial spectrum estimation gradually fail to meet the requirements of accuracy.As a new signal processing method,compressed sensing(CS)can break through the limitation of Nyquist sampling to sample signals.In general,the signal is sparse in the airspace.Therefore,the compressed sensing theory is applied to DOA estimation,and the sparse reconstruction algorithm is used to recover the original signal to complete the DOA estimation of the signal.The optimize algorithm has good estimation performance under low signal-to-noise ratio and low snapshot number.This paper studies the content of narrow-band signal DOA estimation based on compressed sensing.Firstly,the array signal receiving model is explained briefly.A classic DOA estimation method is introduced,and a simulation analysis is performed.At the same time,the theory of compressive sensing is studied from three aspects.The theoretical basis of compressive sensing applied to DOA estimation is expounded,and the DOA estimation model based on compressed sensing is established.Secondly,in the theory of DOA estimation based on compressed sensing,the array flow pattern matrix is used as a measurement matrix.This paper analyzes the properties of the array flow pattern matrix based on two kinds of spatial grid division methods: equal angle division and equal sine division.For the array flow pattern matrix established under the equal angle division method,the matrix decomposition is used to optimize it,and the linear independence of the matrix is improved.The optimization method is compared with the signal-to-noise ratio and the number of array elements.The simulation results show that the optimized method improves the accuracy of DOA estimation.Thirdly,several common greedy reconstruction algorithms are selected for DOA estimation simulation,and the advantages and disadvantages of each algorithm are analyzed.The simulation results show that the GOMP(Generalized Orthogonal Matching pursuit,GOMP)algorithm has the best direction finding performance when performing DOA estimation.In order to improve the performance of DOA estimation.Aiming at the limitation of the least square used in the signal reconstruction of GOMP algorithm,this paper modifies GOMP algorithm.The improved algorithm uses the steepest descent method to reconstruct the signal iteratively,which can obtain smaller recovery error in the process of signal reconstruction.At the same time,the complexity of the improved algorithm is reduced due to the simple operation of the steepest descent algorithm.Simulation results demonstrate that the improved algorithm has good DOA estimation performance.Finally,considering the influence of grid spacing on DOA estimation performance,the paper analyzes the influence of grid spacing selection on DOA estimation accuracy and system running time.Using grid multi-level search method to estimate DOA can reduce the operation time of the system on the premise of ensuring the accuracy of DOA estimation.