基于贝叶斯网络的通信对抗目标识别

Target Recognition of Communication EW Based on Bayesian Network

当前,通信对抗侦察可以依托测向定位、解调消息中携带的位置和个体信.息等,结合态势将侦察信号与目标进行关联,通过整编与目标关联數据,为通信抗目标识别奠定基础。在现有侦察数据基础上,根据目标识别原理建立贝叶斯网络模型,对目标识别预处理得到的数据进行融合计算,推理目标类型,并对不同维度下的识别效果进行分析比对。实验结果初步验证了通信对抗侦察情报在目标识别领域应用的可行性,以朴素贝叶斯网络为算法的目标识别数据处理,能够保证数据量较少时的处理效率。

The current communication countmeasure reconnaissance can rely on the location and individual information obtained by DF and demodulation,combinedwith the situation to associate the signal with target.Through the integration of the data associated with the target,the foundation for target recognition is laid.On the basis of the existing reconnaissance data,this paper establishes Bayesian network structure according to target recognition theory.The data obtained by the pre-process of object recognition are fusion-calculated to deduce the target.The efficiencies and recognition effect of different dimensions are compared.Experimental results show that the application of communication countmeasure reconnaissance intelligence in the field of target recognition is feasible.The data processing of target recognition based on Bayesian network algorithm can ensure the processing efficiency when the amount of data is small.