Line spectrum extraction method based on target characteristics

Abstract

The invention provides a line spectrum extraction method based on target characteristics, and the method comprises the steps: enabling a broadband target signal to be converted into a space-time two-dimensional power spectrum density function, extracting frequency domain power spectrum line information through processing of a frequency domain power spectrum density function, and obtaining a targetsignal strong line spectrum; selecting an estimated scanning azimuth angle on the spatial-temporal power spectral density without the influence of the strong line spectrum, and calculating a mean value of energy distribution in a strip along a wave number direction to obtain a power spectrum estimation map of a scanning azimuth target signal; and extracting line spectrum information of each orientation on the power spectrum estimation graph, and compounding the obtained scanning orientation line spectrum and the target signal strong line spectrum in the orientation direction to obtain the line spectrum characteristics of a specific target. According to the method, target power spectrum information does not need to be predicted in advance; the application range is wide, and extraction of strong line spectrum components and weak target line spectrum components can be included; the speed is high, the real-time performance is good, and the algorithm robustness is high; simulation and seatest data prove that the method is effective.

Description

technical field [0001] The invention belongs to the field of sonar digital signal processing, in particular to a line spectrum extraction method based on target characteristics. Background technique [0002] It is an important research content in the field of underwater acoustics to use ship radiation noise to detect targets remotely. The analysis of underwater acoustic workers shows that ship noise contains a large amount of voiceprint information, and the low-frequency line spectrum voiceprint features, which are unique and stable, have a long transmission distance and high detectability, are important for long-range detection and target detection of ship targets. Identification provides a feasible way of thinking. [0003] When the ship propeller rotates, the periodic and quasi-periodic components of the cavitation collapse in the uneven wake field at the stern and the mechanical vibration produce a modulation envelope, and the wake changes along the circumferential dire...

AI Technical Summary

Problems solved by technology

In document 4 [High-Resolution Frequency-Wavenumber SpectrumAnalysis.Proceedingsof the IEEE.1969], J. Capon regards array sampling as a space-stationary random field sampling, and uses the Fourier transform of the space-time two-dimensional correlation function to obtain the space-time power spectral density. It is used to describe the wide stationary seismic wave signal received by the array; the f-k analysis method has been widely used in the measurement of seismic wave propagation velocity, literature 5 [Wang Fang, Wang Baoshan. Development of frequency-wavenumber domain method and its application in array data analysis. China Earthquake. 2017; 2:191-202], surface wave damage detection literature 6 [Tian Z, Yu L. Lamb wave frequency–wavenumber analysis and decomposition. Journal of Intelligent Material Systems and Structures, 2014; 25(9): 1- 17] and literature [7Poggi V.Time–Frequency–Wavenumber Analysis of Surface Waves Using the Continuous Wavelet Transform.Pure Applied Geophysics.2013; 170(3):319–335] research on surface wave dispersion, in the field of sonar, literature 8[Flow noise analysis oftowed sonar arrays.Udt Europe.1999], Beerens P et al. used f-k to analyze the cause of dragging gust noise, and achieved good results; while using the energy distribution of the space-time two-dimensional power spectral density in the frequency-wavenumber parameter space There are few studies on line spectrum extraction of features

Images