基于神经网络与信息融合的雷达辐射源识别

Information Fusion of Radar Type Recognition Based on Neural Network and D-S Envidence Theory

提出了一种关于雷达辐射源识别的信息融合模型,将概率神经网络、多元属性融合算法和D-S证据理论有效地结合起来。该模型先用概率神经网络对雷达侦察设备所获得的雷达参数进行识别,当识别结果不唯一时,再用多元信息融合算法针对神经网络识别出的可能的雷达型号进行再识别,从而确定最终的识别结果。实验结果表明,该模型可以提高雷达型号识别的识别率、可靠性和抗噪性。

A model of information fusion is presented on the radar emitter recognition which make the probabilistic neural network,multiple attribute fusion algorithm and D-S envidence theory combined effectively. At first,the RBPNN is used to indentify the radar parameters obtained by the radar reconnaissance equipment. When recognitio result is not unique,using multiple information fusion algorithm to identify possible radar type recognition,so as to determine the final recognition results. The experimental results show that the model can improve the recognition rate,the reliability and the noise immunity.