基于盲源分离和自适应滤波的水下声信号降噪算法

在水下小孔径基阵测向应用中，阵元接收到的连续波（CW）信号质量直接关系到测向误差的大小，由于受到多径效应、信号起伏和水下背景噪声的影响，往往实际检测到的信号信噪比较低，相位估计结果离散性大。本文针对水下CW信号和水下背景噪声特点，提出了一种基于盲源分离和自适应滤波联合降噪的算法，该算法对接收的CW信号波形进行降噪以达到提高信噪比的目的。通过算法仿真和湖试试验证明，经本文算法输出的信号，估计器的输出结果比直接利用信号进行方位估计的结果精度高。

A Denoising Algorithm for Underwater Acrostic Signal Based on Blind Source Separation and Adaptive Filter

In the application of underwater small aperture array direction finding system, the quality of received con- tinuous wave（CW） signal directly relates to the direction finding error. Due to the influences of multipath effect, fluc- tuation of signal, and underwater background noise, the actual signal detected is usually in low signal to noise ratio （SNR）, and the phase estimation results are significantly discrete. According to the characteristics of the underwater CW signal and background noise, a denoising algorithm based on the combination of blind source separation and adaptive filter is proposed in this paper to improve SNR by denoising of received CW waveform. Simulation and lake trial show that the proposed algorithm gains a higher precision than that of the azimuth estimation method with direct application of signal.