Hydroacoustic signal processing method based on sparse representation

Abstract

The invention provides a hydroacoustic signal processing method based on sparse representation. The method comprising: (1) constructing a Fourier dictionary D1; (2) constructing a dictionary D2 basedon a Mel filter bank; (3) constructing a composite Dictionary D=[D1 D2]; (4) constructing a sparse representation model shown in the description; and (5) using the sparse representation model to perform sparse representation on hydroacoustic signals. When the method is applied, not only are main line spectrum characteristics of the signals represented, but also the energy trend of the signals arereserved relatively obviously, and the sparse decomposition result has the smallest error; the signals are very close to target signals to be reconstructed, and the recovered signal error is minimal.

Description

Technical field [0001] The invention relates to the field of underwater acoustic signal processing, in particular to an underwater acoustic signal processing method based on sparse representation. Background technique [0002] At present, the main sparse representation method of underwater acoustic signal is the sparse representation of a single dictionary, and the sparse representation is performed by constructing a Fourier dictionary or a dictionary based on a Mel filter bank. [0003] Based on the traditional single dictionary, the complex mixed signal composed of line spectrum signal and continuous envelope cannot be sparsely represented effectively at the same time. [0004] The Fourier dictionary effectively retains some single-line spectrum points with relatively large amplitudes, but the spectrum envelope trend of the original signal is not expressed, so the Fourier dictionary has certain limitations in retaining the target continuous spectrum components. Although the signal...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to: aim at the problems existing in the prior art, provide a kind of underwater acoustic signal processing method based on sparse representation, solve the problem that the existing sparse representation method of underwater acoustic signal cannot represent the line spectrum information well, and the error of the restored signal is often bigger

Images