基于DS证据理论和压缩感知的WSN数据融合策略设计

Design of Data Fusion Strategy in Wireless Sensor Network Based on DS Evidence Theory and Compressing Sensing

摘要:

为了降低WSN数据量和延长网络生命周期,设计了一种基于DS证据理论和压缩感知的WSN混合数据融合策略;首先,在分簇协议的基础上引入了基于DS证据理论和压缩感知的混合模型,然后,采用改进的DS对所有簇成员节点的基本信度分配函数进行加权处理,在簇头处采用加权和归一化的信度分配函数计算证据对各命题的支持程度,将支持程度较大的若干命题作为DS融合结果,在此基础上采用压缩感知方法通过构造测量矩阵对融合结果进行稀疏化表示,并在基站处对稀疏信号进行重构;仿真实验表明,文中方法能有效地实现数据融合,且和其他方法相比,具有重构误差较小和网络生命周期较长的优点,具有较大的优越性。

Abstract:

In order to reduce the amount of data and delay the network life cycle, a compound data fusion strategy based on DS evidence and compressing sensing was introduced based on the clustered protocol, then the improved DS is used to weight over the basic credit assigning function, the weight and normalized credit assign function is used to compute the support extent of evidence to proposition, using the larger value of support extent as the fusion result of DS, then using the CS and constructing the measure matrix to realize the sparse representing, and the construction for sparse signal is used in sink node. The simulation result shows the method in this paper can realize the data fusion effectively, and compared with the other methods, it has less construction error and long life cycle, so it has big priority.