Research On Node Scheduling Mechanism For Target Tracking

Wireless sensor network is a self-organizing network.It consists of a number of low-cost sensors and widely used in military and civilian.Target tracking is an important application field of wireless sensor networks.Since the sensor nodes are limited to energy,communication and process capabilities,WSNs must be relying on dynamic scheduling node resources to collaboratively process information effectively and this will improve the tracking performance and prolong the network lifetime.This dissertation analyze the research status of target tracking algorithm and node scheduling mechanism in WSNs systematically,and introduce the target tracking algorithm,collaboration tracking strategy,adaptive sampling scheduling algorithm and index of performance evaluation.On this basis,this dissertation take the wild harsh battlefield environment as the backgrounds,and study on dynamic scheduling mechanism about node resource.For the existing literatures,adaptive scheduling algorithm does not consider speed and trajectory curvature,both of which have combined effect on sampling interval.This dissertation propose an adaptive sampling-based scheduling strategy for energy efficiency(ASSEE).This strategy determine the next sampling interval according to the trajectory curvature of the target and the moving speed,predict the location of the target for the next time,select the appropriate tasking sensor nodes for dynamic clustering.In order to balance the energy consumption of nodes in the clustering,the cluster head is selected based on the node centrality and the residual energy of the node.Simulation results show that AS SEE effectively reduce the total number of sampling times throughout the tracking process,reduce the overall energy consumption and extend the working life of the network.Since the error between predicted position and actual position of target exists,some sensors waked up based on the predicted position are not detecting target,resulting in receiving an invalid data.If it is processed and the estimated results are transmitted to the cluster head for data fusion,it not only will reduce the tracking accuracy,but alse will add unnecessary energy consumption.For above reasons,on the basis of the proposed ASSEE algorithm of Chapter 3,an erroneous data detection-based scheduling strategy is proposed(ASSEE+EDD),when a node detecting an erroneous data directly discards the data,preventing the transmission of incorrect calculation results to the cluster head.Simulation results show that ASSEE+EDD not only further reduce the overall energy consumption,but also appropriately improve the tracking accuracy.