A stochastic resonance of line-spectrum extraction method for underwater target recognition

被引：0

作者：

Li, Nan ^{[1
,2
]}

Li, Xuikun ^{[1
,3
]}

机构：

[1] College of Underwater Acoustic Engineering, Harbin Engineering University, Harbin

[2] College of Information Engineering, Northeast Dianli University, Jilin

[3] National Key Laboratory for Underwater Acoustic Technology, Harbin Engineering University, Harbin

来源：

Journal of Information and Computational Science | 2014年 / 11卷 / 18期

关键词：

Empirical Mode Decomposition; Line Spectrum; Ship Radiated Noise; Stochastic Resonance;

D O I：

10.12733/jics20105262

中图分类号：

学科分类号：

摘要：

Because of development of noise reduction technology, the complexity of the underwater acoustic channel and increasingly human activities, passive acoustic detection of underwater target signal is facing serious challenges, especially in the low signal to noise ratio. In this paper, the method of combining the empirical mode decomposition with stochastic resonance is applied to detect the weak periodic signal. Firstly, empirical mode decomposition algorithm is used to stationary handling of ship radiated noises. Secondly, the basic mathematical model of stochastic resonance is improved and then sub-band signal after decomposition and smoothing is sent to the improved model of stochastic resonance. By changing internal noise intensity and parameters of the stochastic resonance system, stochastic resonance conditions are met. Finally, the weak periodic signal is enhanced under the synergies of the system, the signal and noise. The simulation results show that line spectrum value of period signal is improved nearly 25 dB relative to the background noise for actual underwater signal. Copyright © 2014 Binary Information Press.

引用

页码：6437 / 6446

页数：9