Research On DOA And Channel Parameter Estimation For Underwater Multipath Environment

With the increasing emphasis on the development of marine resources,the underwater acoustic positioning technology has been paid much attention and has never been developed.However,the underwater environment is complex and changeable,especially in shallow water environment,which brings great challenge to the underwater acoustic localization algorithm.When there is a small delay in the underwater multipath signal,it can be regarded as a coherent signal,and the traditional super-resolution DOA estimation algorithm will encounter great difficulty in locating the target.The classical solution is a spatial smoothing class algorithm,but this results in a loss of array aperture and a high computational complexity.Therefore,it is of great practical significance to find a multi-path DOA algorithm which does not reduce the aperture of the array and low complexity.The results show that the discrete Fourier transform can be used as an efficient,robust and asymptotically optimal spatial combiner for multipath channel and uniform linear array.Therefore,the multipath DOA estimation problem can be switched from time domain to frequency domain.In this paper,we introduce a variety of classical DOA algorithms,and then propose a blind estimation algorithm for underwater multipath DOA and channel parameters based on spatial discrete Fourier transform,frequency estimation and rational function approximation.In this paper,based on the analysis and summarization of the existing algorithms,the problem model is transformed into the frequency domain,and the idea of using the frequency estimation method to estimate the rational function of the parameter information is proposed.Then the estimation parameters are obtained by rational function approximation.In addition,the computational complexity analysis shows that this is a low complexity algorithm,and gives DOA and channel parameter simulation performance analysis based on various frequency estimation algorithms.On the basis of this blind estimation algorithm,this paper assumes that the a priori information of the source signal is known,using the a priori information to improve the performance of the algorithm,and improve the blind estimation algorithm into the non-blind estimation algorithm.In addition,under the known conditions of the prior information of the source signal,it is assumed that the average Gaussian process can improve the robustness of the algorithm,So we change the source signal from the single frequency signal to the multi frequency signal,using the amplitude estimation method instead of the frequency estimation method,So that the algorithm performance has been further improved.Simulation results show that the performance of the algorithm can be improved by using the prior information of the source signal and changing the frequency number of the source signal.Finally,CRB analysis of the parameters of the multipath model is carried out.