Acoustic target radiation noise classification method and system based on EMD and compressed sensing

A radiation noise and compressed sensing technology, which is applied in character and pattern recognition, pattern recognition in signals, instruments, etc., can solve problems such as radiation noise recognition difficulties, and achieve good separability and good application prospects

CN112183280AActive Publication Date: 2021-01-05XI AN JIAOTONG UNIV
11 Cites 6 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
Publication Date
2021-01-05

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses an acoustic target radiation noise classification method and system based on EMD (Empirical Mode Decomposition) and compressed sensing. The method comprises the following stepsof firstly, carrying out EMD decomposition on an obtained ship radiation noise signal and carrying out line spectrum component extraction on a basic mode component by utilizing compressed sensing; and utilizing the maximum mutual information coefficient to select the line spectrum component obtained by the maximum correlation coefficient of the basic mode component and the original signal to resynthesize the signal line spectrum. According to the method, the line spectrum with relatively low amplitude in the signal can be extracted; sparse dictionary training being carried out on the extracted line spectrums, sparse coding matrixes obtained after sparse representation is carried out on different signals being distributed at different positions, subsequent classification being more facilitated, time domain, frequency domain and sparse domain feature extraction being carried out on the signal line spectrums to form an accurate feature set capable of serving as a classification basis, and classification accuracy being improved; the line spectrum of the signal can be extracted more accurately; the extracted sparse features have better separability, and have a good application prospectin identification of radiation noise classification of underwater acoustic targets such as ships and warships.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the field of underwater acoustic signal processing, and in particular relates to an underwater acoustic target radiation noise classification method and system based on EMD and compressed sensing. Background technique

[0002] The main components of the radiated noise of underwater acoustic targets such as ships include mechanical noise, propeller noise, and hydrodynamic noise, which contain more information about the movement of underwater acoustic targets, especially the radiated noise line spectrum formed by propeller noise and propeller singing. Therefore, if the radiation noise line spectrum can be accurately extracted from the radiation noise signal of the underwater acoustic target, it is of great significance for the acquisition of the motion information of the underwater acoustic target and the classification and identification of the underwater acoustic target.

[0003] Due to the complex generation mechanism of under...

Examples

Embodiment

[0113] Choose the signal of 3 kinds of different ships, specifically passenger ship letter, ocean liner signal and motorboat signal, utilize the method proposed in the present invention to classify:

[0114] The first step: use the hydrophone to collect the ship radiation noise signal, Figure 5 is the time-domain waveform diagram and spectrum diagram of the original radiation noise signal, by Figure 5 It can be seen that due to the strong background noise, the signal line spectrum is not obvious;

[0115] Step 2: After using EMD to decompose the signals of the three types of ships, use the compressed sensing method to process each intrinsic mode component, and calculate the maximum mutual information value between each intrinsic mode component and the original signal. The four largest natural mode components in the maximum mutual information value are extracted from the line spectrum, so as to select the four line spectrum components with the greatest correlation with the o...