基于粒子群优化的大规模传感器网络节点调度策略

针对资源受限条件下大规模无线传感器网络中协作目标跟踪问题,提出一个基于粒子群优化的节点调度方案.该方案利用高斯粒子滤波算法和方差交叉融合算法获得目标状态预测信息,进而选择下一时刻簇成员节点,并构造了通信能耗的代价函数,利用粒子群优化方法选择最佳的簇头节点,减少了节点调度的计算复杂度,同时保持了较好的跟踪精度.仿真结果验证了所提出方案的有效性.

Sensor selection via particle swarm optimization in large-scale wireless sensor networks

Under resource-constrained conditions, sensor selection scheme based on particle swarm optimization is proposed
for collaborative target tracking in large-scale wireless sensor networks. By using Gaussian particle filtering and covariance
intersection, the proposed scheme can predict the target’s state next time. Based on the state prediction, this scheme can select
the cluster member nodes, design the cost function for communication energy consumption, and obtain the optimal cluster
head node by particle swarm optimization. Simulation results show that this scheme reduces the computational complexity,
and keeps the good tracking performance.