A high-resolution fast deconvolution sound source imaging algorithm

Abstract

The invention discloses a high-resolution fast deconvolution sound source imaging algorithm, which is characterized in that the approximate spatial translation invariance of the point spread function is used in the construction process of the point spread function matrix of the deconvolution sound source imaging algorithm , calculate the point spread function of the sound source at the center position of the sound source calculation plane, and then construct the point spread function matrix by cyclically shifting the point spread function at the center position up and down, avoiding the need for all point spread functions Calculation, reducing the amount of calculation, improving the calculation speed and efficiency; secondly, in the deconvolution reconstruction process of the sound source intensity energy distribution, using the spatial sparse prior of the sound source, combined with the compressive sensing theory, through the orthogonal matching pursuit algorithm Realize the fast sparse deconvolution reconstruction of the energy distribution of the sound source intensity, reduce the number of iterations, and improve the calculation efficiency and resolution. The invention has high calculation efficiency and spatial resolution, and can identify and locate the position of the sound source in space faster and better.

Description

technical field [0001] The invention is mainly used in the field of noise source location and analysis, and is a high-resolution fast deconvolution sound source imaging algorithm. Background technique [0002] The traditional beamforming technology is based on delay summation, cross-spectrum imaging function and other methods to post-process the received signals of each sensor, so that the output of the focal point corresponding to the real sound source on the sound source calculation plane is strengthened, while the output of other focal points The output volume is attenuated, thereby identifying the sound source. However, it will not only output a "main lobe" with a certain width at the real sound source position, but also output a "side lobe" at a non-sound source position. In order to effectively eliminate the influence of the main lobe width and side lobe interference, the anti-roll Acoustic Acoustic Source Imaging Method, or DAMAS, has significantly better spatial res...

AI Technical Summary

Problems solved by technology

First of all, both the DAMAS2 method and the FFT-NNLS method are iterative algorithms. In order to obtain more accurate results, thousands of loop iterations are usually required, and because the Fourier transform introduces a winding error, in order to reduce the winding error, the calculation Influenced by the results, it is usually necessary to pad the number of measurement data points to four times the original, resulting in a sharp increase in the calculation scale, so although the DAMAS2 method and the FFT-NNLS method are superior to the original deconvolution sound source imaging method in terms of computational efficiency, However, there are still deficiencies in computational efficiency and resolution.

Images