Automatic Extraction Method of Marine Engineering Noise Signal Based on Sound Exposure Level Vector

Abstract

In order to extract the ocean engineering noise signal quickly and accurately from the massive ocean original noise data, this paper proposed an automatic signal extraction method based on acoustic exposure level vector. The results showed that the core of the method was to calculate the standardized Euclidean distance of the acoustic exposure level vector between the suspected signal and the sample signal. By combining the techniques of variable step size interception analysis, double-threshold trigger based on half-wave envelope and center position trigger, the method could effectively improve the efficiency and accuracy of signal extraction. Compared with the traditional manual extraction method, this automatic extraction method was robust, accurate and efficient, which could replace the manual extraction method and be widely used.

Keywords

• automatic signal extraction
• sound exposure level vector
• dual-threshold trigger
• standardized euclidean distance
• ocean noise