无线传感器网络自适应动态簇目标跟踪策略

考虑到无线传感器网络的特点及目标跟踪对实时性要求较高,提出一种基于动态簇的无线传感器网络目标跟踪策略.当目标出现时,基于节点管理机制采用层次分析法选举簇首,定义节点跟踪权值作为参与跟踪的依据,并建立动态簇.目标监视过程中,根据实际跟踪质量自适应调整动态簇规模;为了避免目标丢失,在自适应机制的基础上提出目标恢复机制.目标离开后,解散动态簇,释放节点资源.采用NS2对所提算法进行仿真测试,结果表明,基于动态簇的跟踪策略能够以较少的通信开销为代价获得很好的目标捕获率.

Adaptive Dynamic-Cluster Based Target Tracking Strategy in Wireless Sensor Networks

In view of the characteristics of wireless sensor networks (WSNs) and their high timeliness in target tracking, a tracking strategy based on dynamic clusters is proposed. When the target appears, analytic hierarchy process (AHP) is used to select a cluster head, and the tracking weight of a node is defined and used to judge whether the node should join in the cluster for tracking, based on the node management mechanism. During the tracking, the cluster activates suitable sensors to participate in the tracking task, and the range of the activation area is adaptively adjusted according to real-time tracking quality. Besides, an adaptive target recover mechanism is proposed to avoid target losing. After the target leaves, the cluster is deconstructed to release sensor resources. The NS2 simulation results indicate that the dynamic cluster strategy for target tracking can get high tracking rate at low cost of communication overhead.