论文部分内容阅读
针对由于各种信号干扰和传感器误差导致辐射源个体正确识别率较低的问题,提出一种多传感器融合识别算法进行复杂电磁环境中的通信个体识别。该算法将Dempster-Shafer证据理论和特征提取结合起来,充分利用侦测的信号特征,减少了识别过程中的不确定信息。该融合识别算法提取侦测信号中的个体特征,使用基于决策向量的自适应证据融合方法将由个体特征转化而来的多个证据相融合,最后再根据判决准则得到最终的识别结果。分别对自适应融合方法和融合识别算法进行仿真分析,结果表明自适应证据融合方法可以综合考虑融合过程的计算效率和融合结果的合理性,在二者之间达到平衡。与现有的识别方法相比,多传感器融合识别算法可以提高复杂电磁环境中个体识别的稳定性和正确识别率。
Aiming at the problem of low correct recognition rate of individual radiation sources due to various signal interferences and sensor errors, a multi-sensor fusion recognition algorithm is proposed to identify individuals in complex electromagnetic environment. The algorithm combines Dempster-Shafer theory of evidence and feature extraction to make full use of the signal characteristics of detection and reduce the uncertain information in the recognition process. The fusion recognition algorithm extracts the individual features of the detection signal, uses the adaptive evidence fusion method based on the decision vector to fuse the multiple evidences transformed by the individual features, and finally obtains the final recognition result according to the judgment criterion. The simulation results show that the adaptive evidence fusion method can consider the computational efficiency of the fusion process and the rationality of the fusion results in order to strike a balance between the two. Compared with the existing identification methods, the multi-sensor fusion recognition algorithm can improve the stability and correct recognition rate of individuals in complex electromagnetic environment.