DoA Methods Based On Bayesian Compressive Sensing

Array signal processing is an important branch in the field of signal processing.Its purpose is to extract the signal received by the array and its characteristic parameters,moreover,suppress interference,noise or uninterested information.DOA estimation is an important branch of array signal processing,which is widely used in radar,sonar,seismic exploration,radio astronomy and so on.It is through a group of antenna array to receive the signal in the space to determine the incident angle of the signal,so as to estimate the azimuth of the signal.After years of research by mathematicians,there are some mature algorithms,such as space spectrum estimation class algorithm,base tracking algorithm(BP),orthogonal matching tracking algorithm(OMP),etc.,but these algorithms inevitably have some shortcomings,how to be more rapid improving the DOA estimation accuracy effectively has always been an important research content in DOA estimation methods.Compared with traditional estimation methods,Bayesian compressed sensing has some advantages in accuracy or time.How to use Bayesian compressed sensing to estimate DOA is the main content of this thesis.Firstly,aiming at the poor estimation accuracy of multiple signal classification algorithm(MUSIC)and the time-consuming problem of classical compressed sensing algorithm,a new algorithm based on Bayesian theory is proposed to estimate the signal angle of arrival(DOA),that is,Bayesian compressed sensing based on Laplace prior.This method can estimate the signal angle of arrival more quickly and effectively.In the process of solving,the compressed sensing theory is used to sparse representation,compressed sampling and signal reconstruction of signal angle.The estimation method combined with Bayesian theory is using to estimate the signal source with stronger energy angle,which avoids the time-consuming problem caused by solving each angle.Secondly,based on Bayesian compressed sensing(BCS),the single task and multi task Bayesian compressed sensing theory is studied,and the DOA of wideband signal is estimated by using MT-BCS algorithm,so that the signal direction and bandwidth can be estimated at the same time.By referring to the sparse representation of compressed sensing theory,dividing the signal bandwidth,and the signal bandwidth is determined by estimating the number of signals in the bandwidth.The multi task Bayesian compressed sensing theory is using to avoid the difficulty of manually selecting parameters when estimating each sample value.At the same time,it makes the algorithm more stable,and has strong anti noise ability.Besides,makes the reconstructed signal direction angle clearer.Results of numerical simulation and experimental test verify the effectiveness of the algorithm.Finally,the DOA estimation problem of four-dimensional time modulated array(TMA)has studied systematically.Time is introducing into the DOA estimation method of antenna array as a parameter.In the linear time modulated array,compressed sensing algorithm and classical DOA estimation algorithm are using to estimate the DOA The simulation results show that the DOA estimation is effective in different scenarios.