舰船尾流的激光探测是一种新的鱼雷制导手段。水体的后向散射光信号是舰船尾流后向散射光信号检测的常见干扰,由于其在频域十分接近且信号强度大于尾流信号,因此难以用传统方法提取有用的尾流信号。针对这一问题,提出了一种基于盲源分离的处理方法,将核独立成分分析技术应用于舰船尾流后向散射光信号的提取。介绍了核独立成分分析的基本原理和具体算法,进行了仿真计算,并与传统独立成分分析算法进行比较。结果表明在盲源信号分离中,基于核空间的独立成分分析与其他独立成分分析算法相比更具有准确性。最后应用该方法对海上实验数据进行处理,提取出了舰船尾流信号,取得了良好的效果,验证了该算法的有效性。

The laser detection of ship wake provides a novel method to home target for torpedo. But the backscattering optical signal of water constantly influences the detecting of backscattering optical signal of warship wake flow, because it is very close to the needed signal in frequency and stronger than the needed signal. A method based on blind source separation(BBS) is presented, and the kernel independent component analysis(KICA) is used in the detection of wake flow backscattering optical signal. The principle and algorithm are introduced and the simulation results show that KICA is more accurate than other traditional ICA algorithm in BBS. Finally this method is used to calculate the experiment data and the ship wake signal is gotten, which demonstrates the efficiency of this method.