Research On Acoustic Vector Array Azimuth Estimation Method Based On Compressed Sensing Theory

With the rapid development of small underwater vehicle, the traditional sound pressure sensor will not fully meet the high underwater vehicle equipment requirements, so the stable performance of the vector hydrophone will be more widely used in underwater acoustic engineering fields. Acoustic vector sensors each contain one omnidirectional hydrophone measuring pressure and three orthogonal geophones measuring the component of particle velocity. Vector sensors include directional information,and they have the potential to improve the performance of passive sonar systems. Acoustic vector sensor dipoles not only frequency independent directivity, suitable for detecting and measuring the low-frequency signals, and compared with the pressure-sensor array composed at technical indicators to achieve the same premise, its smaller size and weight. Vector-sensor measurements provides more information than pressure-sensor measurements. Using this additional information to improve performance is the role of vector-sensor processing.According to sparsity of direction of arrival of underwater target in its search space,DOA estimation based on acoustic vector array is realized using sparse decomposition theory in small samples case. Estimate the direction of arrival by Smooth l0 Norm Combined with the characteristics of compressed sensing theory, and this method has fast speed and high resolution features. The simulation results show that this method improves the anti-interference ability of the azimuth estimation comparing with traditional sound pressure sensor array DOA estimation method, and implements the comprehensive estimate direction. The proposed algorithm provides a higher spatial resolution and estimation accuracy in comparison to many other current algorithms.The existing Compressed sensing is conducted based on the spatial structure characteristics of signal to estimate the direction of the arrival. The temporal correlation between the sources, however, results in poor performance and accuracy. Existing sound pressure sensor array DOA estimation method can not distinguish between right and left. To overcome this problem, we propose a DOA estimation algorithm based on Compressed Sensing Theories with acoustic vector sensor array. The high-accuracy DOA estimation method reduces the impact of noise by adopting the idea of Muti-vectors Sparse Bayesian Learning. Muti-vectors Sparse Bayesian Learning algorithm is used to reconstruct the signal space spectrum, and the CS-MMV model of the unknown sparse vector signal source is established to estimate the DOA. The simulation results show that this method improves the anti-interference ability of the azimuth estimation comparing with traditional sound pressuresensor array DOA estimation method, and implements the comprehensive estimate direction. The proposed algorithm provides a higher spatial resolution and estimation accuracy in comparison to many other current algorithms.