基于小波能量特征的尾流光信号特征研究

Laser Wake Signal Feature Extraction Based on Wavelet Package Energy Feature

针对舰船尾流激光探测的信号检测与分析问题,提出了利用小波域能量特征进行尾流特征分析的新方法。首先设计了变遗忘因子的自适应抵消器滤除水体后向散射信号与系统噪声,有效地选取了含有目标的信号;其次采用小波包分解方法并按频段的能量形成提取目标特征;最后借助主成分分析方法对特征向量进行降维。通过对试验数据的分析表明,提取的特征能有效地揭示不同性质下的尾流后向散射光信号,提高了对尾流分析的能力。

For detection and recognition of the laser wake of a surface ship, a novel method for detecting the wake backscattering signal by making use of the wavelet package energy feature vector is presented. An adaptive noise canceller based on variable forgivable factor is designed to filter the backscattering signal and system noise, so as to effectively select the target signal. The wave package energy feature vector is extracted by using wavelet package decompo- sition method and different energy formation of frequency range. The dimension of the feature vector is reduced with the principal components analysis method. Experimental results show that the proposed method can effectively recognize the features of different wake backscattering signals of a surface ship.