Estimation method of two-dimensional array music decoherence parameters of acoustic vector sensor

Abstract

The invention provides an acoustic vector sensor two-dimensional array MUSIC decoherence parameter estimation method. An array receives K far-field narrowband coherent signals, and N times of snapshot data outputted by all the array elements are acquired by using the receiving array; four subarray data of acoustic pressure and x-axis, y-axis and z-axis acoustic velocity are extracted, and a data covariance matrix after decoherence is constructed by a data covariance matrix before transformation, a covariance matrix after data transformation and a cross-covariance matrix before and after transformation; singular value decomposition is performed on the data covariance matrix after decoherence so as to obtain a noise subspace, the noise subspace is divided into x-axis and y-axis noise subspaces according to the characteristics of an L-shaped array and the corresponding MUSIC space spectrum is constructed, and direction cosine estimation matrixes are obtained by two one-dimensional searches; and matching operation is performed on the direction cosine estimation values of the x-axis and y-axis directions by using the noise subspaces corresponding to the full array so as to achieve the estimation value of the angle of arrival. According to the method, one two-dimensional search is divided into two one-dimensional searches so that the computational burden can be greatly reduced and the parameter estimation accuracy can be enhanced.

Description

technical field [0001] The invention belongs to the technical field of signal processing, and in particular relates to a method for estimating a two-dimensional angle of arrival of a coherent source of an acoustic vector sensor array. Background technique [0002] In the actual information transmission process, due to multipath reflection and human interference, correlated signal sources have become a common phenomenon. In the case of strong correlation or coherent signals in the space, the subspace method represented by MUSIC fails. Estimation of spatial angle will produce a large error, and even the angle of arrival of the signal cannot be estimated. The research on direction of arrival estimation of coherent sources is an important research topic of signal processing. [0003] The decoherence method represented by spatial smoothing reduces the array aperture, increases the beam width of the array, and reduces the resolving power of the array. Moreover, spatial smoothing ...

AI Technical Summary

Problems solved by technology

The existing methods are mainly one-dimensional parameter estimation methods for one-dimensional uniform linear arrays. There are few studies on two-dimensional angle of arrival estimation methods, and they are often very complicated. The paper published by Wu Xiaoqiang "Two-dimensional DOA estimation method based on ESPRIT algorithm Research" (Harbin Engineering University 2008 master's degree thesis) studied the improved two-dimensional ESPRIT algorithm and the two-dimensional ESPRIT algorithm based on fourth-order cumulant processing and space-time processing. This method has a certain ability to solve the coherence, but the method It is necessary to construct a very complicated matrix, and the follow-up algorithm is also very complicated; the present invention proposes a non-uniform L-shaped acoustic vector sensor array decoherence MUSIC method for the deficiencies of the existing methods, which fully utilizes the rotational invariance of the acoustic vector sensor array itself Variable property decoherence, a method called rotational decoherence

Images