摘要
针对目标所采用的辐射源识别方法对复杂体制雷达信号识别效率低或无法识别的现象 ,提出了一种新的辐射源识别方法。首先 ,利用小波包可对信号进行多维多分辨率分析的特点 ,对辐射源信号进行信号特征的提取 ,然后将各辐射源的信号特征作为ART2神经网络的训练样本 ,对其进行辐射源类型的识别。通过计算机仿真 ,对上述方法进行了验证。试验结果表明 ,这种基于小波 神经网络的辐射源识别方法在识别复杂体制雷达信号的应用中不仅克服了目前识别方法识别效率低的弊端 。
According to the current method whose recognition result is inefficient and can not recognize the unknow radar signal, this paper presents a new method of the radar recognition. At first, based on the advantage of multidimensional, multi-scaling decomposition of wavelet packets, the features of signals are extracted. Then these features are inputted into ART2 neural network to recognize the type of the radar. The paper validates the new method through the computer-aided experiment. The analysis of experiment data proves that the proposed method is efficient.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2002年第5期55-57,共3页
Systems Engineering and Electronics
