一种MEMS水听器信号去噪的组合算法研究

针对MEMS水听器采集的数据"淹没"在强噪声场中的问题,提出采用LMS自适应噪声对消与Fourier变换滤波相结合的组合算法实现MEMS水听器的信噪分离。在信号频率已知的情况下,设计了一种自适应噪声对消和Fourier变换滤波组合算法的滤波器,对提取后的信号与理想信号做性能对比。仿真实验表明:该组合算法在-15 dB的强噪声场中仍有较高的分辨精度和提取效果,对搜寻类似于"黑匣子"等情况比较适宜,并将设计的滤波器用于中北汾机测试实验的信噪分离中,结果验证了该算法具有良好的高效性和实用性。

A Combination Algorithm for MEMS Hydrophone Signal and Noise Separation

According to data collected by MEMS hydrophone "drowned" in strong noise field, LMS adaptive noise cancellation and Fourier transform filtering combining fusion algorithm is proposed to achieve MEMS hydrophone signal and noise separation. Designs a stickogram based on noise and signal cross validation of LMS Adaptive Noise Cancellation and Fourier transform combined algorithm when the frequency of the signal, compare the performance of extraction of signal and the ideal signal. Simulation results show that: this combination algorithm has a good extracting effect and high precision on the search such as the black boxes, and can be to adaptation for more complex noise scenarios. The fenji test experiment result also shows that this algorithm has good efficiency and practicability.