基于噪声信号EMD的装甲车辆识别研究

Research on Armored Vehicle Classification Based on EMD of Noise Signal

针对地面战场装甲车辆目标的被动声识别问题,选取具有代表性的两类坦克和两类履带式装甲车为对象,采集多种工况下的噪声信号,通过EMD自适应分解得到其IMF分量,利用前8个IMF分量与原信号的能量比值构建特征向量,以BP神经网络作为分类器,建立了一种装甲车辆识别方法。该方法对目标工况适应性强,识别率可达90%以上。

In order to identify the ground battlefield armored vehicle target through passive acoustic recognition,this paper selects representative objectives include two kinds of tanks and two kinds of crawler armored vehicle as the noise acquisition,and collects noise signal of target in different working conditions,and decomposes the noise signal with the EMD method,which can provide the IMF components. Using the energy ratio of first eight IMF components and the original signal as characteristic values to construct eigenvalue vector,with BP neural network as a classifier,this paper establishes an armored vehicle classification method. The method is adaptable to the target condition and classification rate can reach to more than 90%.