面向智能电网的无线传感器网络目标探测算法

Target detection algorithm of wireless sensor network in smart grid

文中提出一种结合压缩感知(CS)路由算法进行目标源定位的方法。该路由算法通过对网络节点进行分簇,将簇内节点的信息集中在簇头上。然后对无线传感器网络进行拓扑优化,并采用多跳路由策略完成簇头节点与汇聚节点间的通信。最后利用次梯度投影算法,从来自簇头的少量数据中恢复汇聚节点信号源,用于目标源信号的定位。采用一种自适应处理机制优化压缩感知算法的信号重构性能。仿真结果表明,算法能准确探测目标,减少计算量,并具有良好的重构性能。

In order to effectively detect the signal target in the monitoring area of the wireless sensor network, a routine algorithm combining with the compressive sensing ( CS) is proposed in this paper, in which network clustering routing algorithm and CS are integrated. In this paper, the network nodes are clustered by the routine algorithm, and information within the cluster nodes is concentrated in cluster-head. After the optimization of wireless sensor network topology, communication between the cluster head node and the sink node is realized by multi-hop routing. Then subgradient projection method is used to restore source signal from few cluster head data. To improve the signal reconstruction performance of CS, a new adaptive processing mechanism is also proposed in this paper. Simulations demonstrate that the proposed algorithm which has excellent good reconstruction performance can accurately detect the target and the computational complexity is reduced effectively.