系统工程与电子技术 ›› 2018, Vol. 40 ›› Issue (12): 2629-2635.doi: 10.3969/j.issn.1001-506X.2018.12.02

• 电子技术 • 上一篇    下一篇

基于改进时域证据融合的目标识别

高晓阳, 王刚   

  1. 空军工程大学防空反导学院, 陕西 西安 710051
  • 出版日期:2018-11-30 发布日期:2018-11-30

Target recognition based on improved combination of temporal evidence sources

GAO Xiaoyang, WANG Gang   

  1. Air and Missile Defense College, Air Force Engineering University, Xi’an 710051, China
  • Online:2018-11-30 Published:2018-11-30

摘要:

为了实现防空反导目标识别,根据证据折扣理论和证据可靠性评估的思想,提出了一种改进的时域证据组合方法。首先,提出了具有时域信息累积的信息融合模型;然后,通过基于信任度和虚假度的方法对累积的时域信息进行预处理,用可信度衰减模型和基于Einstein算子改进的冲突因子分别进行实时可靠性和相对可靠性评估;最后,基于Dempster准则和证据折扣准则对证据进行合成。仿真分析表明,该方法能够充分利用历史信息,体现了时域信息融合的动态性,能够有效处理信息冲突,具有较强的抗干扰能力。

Abstract:

To improve target recognition in air and missile defense, the improved temporal evidence combination is proposed, in accordance with the idea of evidence reliability and evidence discounting. Based on temporal evidence accumulation, a multisource information fusion model is firstly presented. Then the accumulation temporal evidences are combined under weighted average combination rules based on the degree of credibility and falsity. Meanwhile, the realtime reliability factor is yielded by the credibility decay model, and the relative reliability factor is acquired by Einstein operator. Finally, the temporal evidence is combined based on the evidence discounting theory and the Dempster combination rule. The simulation results demonstrate that the proposed method makes good use of accumulation information, which can reflect the dynamic nature of temporal information fusion. Besides, it is illustrated that this method has good performance when dealing with conflict in temporal fusion, and can enhance the antiinterference performance of the target identification fusion system.