基于独立元分析的水下激光雷达后向散射噪声去除方法  

Elimination of Backscatter Noise of Underwater LiDAR Using Independent Component Analysis Algorithm

作　　者：齐若伊 李坤[1,2] 杨苏辉 高彦泽[1,2] 王欣 李卓 Qi Ruoyi;Li Kun;Yang Suhui;Gao Yanze;Wang Xin;Li Zhuo(School of Optics and Photonics,Beijing Institute of Technology,Beijing 100081,China;Beijing Key Laboratory for Precision Optoelectronic Measurement Instrument and Technology,Beijing 100081,China)

机构地区：[1]北京理工大学光电学院,北京100081 [2]精密光电测试仪器及技术北京市重点实验室,北京100081

出　　处：《光学学报》2021年第4期25-32,共8页Acta Optica Sinica

基　　金：国家自然科学基金(61835001)。

摘　　要：模拟了调制脉冲激光雷达在水下的探测信号,利用快速独立元分析方法(FastICA)的迭代算法将探测信号中的目标与后向散射信号分离,恢复出浑浊水域被强后向散射淹没的弱目标反射回波信号,极大地提升了信噪比。将解调信号设置为调制频率及三倍调制频率余弦函数之和,目标回波信号只需一次相干解调即可获得,其峰值位置与目标和探测器之间的延时相对应。经过独立元分析方法(ICA)迭代分离后,水下激光探测距离可以提高4~5个衰减长度。We simulate the detection signal of a modulated pulsed underwater light detection and ranging(LiDAR)system and use a fast independent component analysis iterative algorithm to separate the target in the detected signal from the backscatter signal.Our approach succeeds in recovering the echo signals of the weak target in muddy water submerged by strong backscattering and in improving the signal-to-noise rate.To obtain the time delay of the target reflection with respect to the transmitting signal in one step,we used single and triple modulation frequencies as references to demodulate the LiDAR echo signals.The peak position corresponds to the time delay between the target and the detector.After the iterative separation by independent component analysis,the underwater laser detection range can be increased by four to five attenuation lengths.

关 键 词：海洋光学 后向散射 水下激光雷达 盲信号分离 独立元分析 

分 类 号：TN911.[电子电信—通信与信息系统]