多频海底声学原位测试信号消除干扰研究  

Study on Signal Denoising of Multi-Frequency Seabed in Situ Sediment Acoustic Measurement

作　　者：李红星[1] 陶春辉[2] 刘财[1] 邓显明[2] 周建平[2] 张金辉[2] 顾春华[2] 何拥华[2] 

机构地区：[1]吉林大学地球探测科学与技术学院,长春130026 [2]国家海洋局第二海洋研究所/海底科学重点实验室,杭州310012

出　　处：《吉林大学学报（地球科学版）》2007年第5期1034-1037,共4页Journal of Jilin University：Earth Science Edition

基　　金：国家"863"计划项目(2002AA615130);国家自然科学基金项目(NSFC49906004);国家海洋局重点青年基金项目(20005002);浙江省151工程项目

摘　　要：多频海底声学原位测试系统是由国家海洋局第二海洋研究所自行研制的国内首套海底声学原位测试系统。对多频海底声学原位测试系统的测试信号处理方法进行了初探。海底声学原位测试信号难免会包含干扰信号,其将影响到海底沉积物特性提取的准确性。通过对多频海底声学原位测试系统的测试信号进行基于小波分解的自相似性分析,发现信号在不同尺度上都相似于其本身,说明干扰信号与有效信号在频谱分布上并没有太多差异,通过频率域滤波难以消除干扰信号。将小波分析方法引入海底声学原位测试信号消除干扰信号处理中,利用小波对信号进行分解并经过小波系数选择与信号重构后,消除了干扰信号,并保留了有效信号。Multi-frequency seabed in situ sediment acoustic measurement system developed by the Second Institute of Oceanography , SOA, is the first system of same type in the country. The authors focuse on the primary results on the signal processing of the system. When contaminated by noise, the accurate identification of seabed sediment characteristics will be not easy. From the results of wavelet decomposition from measured signal, it can be found that the signal is similar to itself in different scales, which means that there is little difference in the signal and noise, leading to a hard denoising in frequency domain. Therefore, wavelet transform is applied to denoise signals of the in situ acoustic system. By wavelet method, the signal is decomposed and the wavelet coefficients are calculated. The reconstructed signal reveals true signal.

关 键 词：多频海底声学原住测试 小波变换 信号处理 

分 类 号：P631.4[天文地球—地质矿产勘探]