基于节点权重和DS证据理论的WSN数据融合

Data Fusion in Wireless Sensor Network Based on Node Weight and DS Evidence Theory

为了减少无线传感器网络中传感器节点目标检测和识别的时间开销和数据冲突,研究了一种基于节点权重和DS证据融合理论的数据融合方法;首先通过计算各目标的总体信任度选择焦元从而减少焦元数目,在此基础上,通过计算各传感器节点采集数据即证据的总体信任度过滤总体信任度较低的节点以减少节点数目,然后通过计算各证据的信息熵来确定各证据权重,通过计算证据被其它证据支持的支持度来获得证据关系权重;最后将计算得到的证据权重和证据关系权重进行加权获得最终证据权重;仿真实验表明:文中方法能较为精确地进行目标识别,识别率高达100%,与其它方法相比,具有计算时间少和识别精度高的优点,具有很强的可行性。

In order to reduce the time consumption and data collision of goal recognition in wireless sensor network, a data fusion method based on node weight and DS evidence theory was researched. Firstly, the main trust of every goal was computed to reduce the numbers of goals, then the main trust of every sensor node was computed to filter the nodes which has the lower trust value, and after that, the informa tion entropy for evidence was computed to get the evidence weight, and the evidence relation weight was computed, Finally, the final evi dence weight was obtained by weighting evidence weight and evidence relation weight. The simulation results show that： the proposed meth od is accurate in goal recognition, the recognition rate is as high as 100%, and compared with other methods, it has the advantages of less time consumption and high reeolznition accuracy. So it is proved as strong feasibility.