Research On Underwater Acoustic Array Signal Processing Technology Based On Compressive Sensing

The underwater acoustic array is an important device for active sonar to receive underwater information.In order to obtain more information in a complex marine environment,improve the angular resolution of the target and the gain of the receiving array,the underwater acoustic array is developing in the direction of broadband and large aperture.The result is an increase in the number of array elements and an increase in the amount of data,which not only increases the cost of hardware,but also brings a great burden to the storage,transmission and processing of signals,and even makes it difficult to build.Aiming at this problem,this paper studies an array processing method based on compressed sensing theory,which uses fewer array elements and data volume to achieve high-resolution orientation estimation of the target,thereby increasing the gain of the receiving array.The main research work of this paper is as follows:(1)Research on receiving model and processing method of underwater acoustic array signal.Sparsity in space is the prerequisite for the application of compressed sensing method.Firstly,the narrow-band and wide-band mathematical models of underwater acoustic array receiving signals are established,and then the basic theory of typical target direction of arrival(DOA)and beamforming is explained.Then,multiple signal classification(MUSIC)algorithm combined with the Incoherent signal subspace method(ISSM)to form the ISSM-MUSIC method for processing broadband array signals,and the respective advantages and characteristics of the MUSIC algorithm are analyzed through simulation verification.(2)Application research of compressive sensing theory in array signal processing.Based on the underwater acoustic array signal receiving model,the compressed sensing method is integrated into the array signal processing,and a wideband DOA estimation method based on array compressed sampling is proposed.First,using the natural sparsity of the array,the array manifold matrix is established as the sparse dictionary of the array signal;secondly,the Gaussian random measurement matrix is used to compress the array,and the sensing matrix is composed of the Gaussian measurement matrix and the array manifold matrix;Finally,the adaptive reconstruction algorithm uses its sensing matrix to reconstruct the target orientation information to realize the DOA estimation of the underwater target.The experimental simulation verification shows that,compared with the ISSM-MUSIC method,this method can still achieve high-precision azimuth estimation for the target in a small amount of array observation data.(3)A method for constructing deterministic measurement matrix for compressed sampling of arrays.Aiming at the problem that the random measurement matrix has extremely high requirements on hardware and is difficult to implement,on the basis of the Hadamard matrix,through the method of compressing and zeroing,the compressed and zeroed measurement matrix based on the sequential part of Hadamard is obtained,and an improved Hadamard matrix is proposed.A method for constructing a deterministic measurement matrix and verifying its performance by simulation.Simulation results show that the proposed improved Hadamard deterministic measurement matrix has better estimation accuracy and robust performance than other deterministic matrices.(4)Performance verification of array signal processing method based on measured data.The performance of the proposed method is mainly verified by the measured data obtained in the static and dynamic test scenarios.In the static test,the performance of the improved Hadamard deterministic matrix construction proposed above is mainly verified.The results show that the improved Hadamard matrix not only has better highresolution performance under high compression ratio,but also has a lower compression ratio of 44%.It also maintains good object orientation estimation performance.In the dynamic test,the combination of the proposed improved Hadamard deterministic measurement matrix and the broadband DOA estimation method under array compressed sampling is mainly used to realize the high-resolution azimuth estimation of the target.It realizes the high-precision resolution of the bright spot of the underwater target in the environment of low signal-to-reverberation ratio,thereby increasing the gain of the receiving array.