正弦函数基原子库微弱被动鱼声信号的稀疏检测  

Sparse detection for weak passive fish acoustic signal based on sine wavelets

作　　者：陈功[1] 常睿[1] 于海平[1] 杜玉华[1] 吴雪芬[1] 王平波[2] 

机构地区：[1]常州工学院,江苏常州213022 [2]海军工程大学,湖北武汉430033

出　　处：《中国测试》2015年第3期108-112,共5页China Measurement & Test

基　　金：江苏省自然科学基金青年基金项目(BK20130245);常州工学院自然科学研究基金项目(YN1311)

摘　　要：为实现海洋环境中微弱被动鱼声信号的检测,针对单频正弦信号稀疏分解用于微弱信号检测的局限性,采用正弦函数基拟合被动鱼声信号,构建不同幅值、频率和初相位的正弦波信号作为过完备原子库,通过稀疏分解,检测出淹没在强噪声环境中的微弱正弦信号的幅度、频率和初相位参数,从而恢复出待检测的被动鱼声信号。实验表明:该项技术在-40 d B条件下可以实现任意形式的鱼声信号检测。In order to detect weak passive fish acoustic signal in the oceans, the numerical method was presented for matching the passive fish acoustic signal based on sine wavelets. Based on an over-complete dictionary from sine wavelets signal of different frequency, amplitude and phase instead of a single sine signal, sparse decomposition algorithm was realized. Parameters of weak passive fish acoustic signal can be detected under low signal-to-noise ratio(SNR)surroundings. Experimental results show that an arbitrary passive fish acoustic signal can be detected under-40 d B SNR.

关 键 词：稀疏分解 正弦函数 拟合 被动鱼声 

分 类 号：TN911.23[电子电信—通信与信息系统]