THE ANALYSIS OF COMBUSTION AUDIO SIGNALS BASED ON CONTINUOUS WAVELETS TRANSFORM
The combustion audio signal, which is a typical week non-stationary component hiding in the sound signals, is hard to separate. And this hinders the development of combustion audio using in fire forecast field. In this paper, physical characteristics of combustion audio were analyzed, and a new method based on Continuous Wavelets Transform (CWT) for combustion audio separation has been studied. Moreover, a software system based on Labview has been developed to realize the separation. Simulation contrast tests were made; signals were treated by Short Time Fourier Transform (STFT) and Wigner-Ville Distribution (WVD) and CWT. The results showed that the weak combustion audio signal was distinguished clearly by the CWT. And because of the homogenization, the STFT and WVD have a fuzzy result that testifies the practicability of the CWT method in combustion audio separation.
Combustion audio Fire forecast CWT Combustion audio feature recognition
Shu FANG Jing-xue JIANG Zhen-Wei ZHANG
Scientific research center, Shanghai Fire Institute of Ministry of Public Security, Shanghai, 200438 School of information science and technology, Donghua University, Shanghai, 200000
国际会议
深圳
英文
282-285
2011-12-09（万方平台首次上网日期，不代表论文的发表时间）