Research On Energy Efficiency Optimization Algorithm Based On Multi-target Tracking In Wireless Sensor Network

With the continuous innovation and development of wireless sensing technology and MEMS,wireless sensor networks have been widely used in environmental monitoring,transportation,military intrusion detection and tracking,smart home and other fields.Therefore,research on multi-target tracking technology in large-scale wireless sensor networks has important theoretical significance and application value.Due to the limitations of the wireless sensor network energy resources,the main problem faced by multi-target tracking technology is how to balance the effect of sensor node life and target positioning accuracy on tracking performance.This thesis mainly aims to improve the energy efficiency of wireless sensor networks by equalizing the residual energy distribution of sensor nodes in the network when the target positioning accuracy is guaranteed.The target node is constantly moving in the monitoring area,so the network needs to activate different sensor nodes to complete the continuous tracking of the target.Due to noise interference,the target node's motion and the sensor's observation data will be affected.Therefore,in order to solve the above problems,this thesis first constructs an accuracy model for node selection that is using the Mahalanobis distance to describe the effect of sensor observation data on the posterior distribution of the target position,while using the sum of the triangle area to approximate the optimal deployment position of the sensor node subset.Then fuzzy logic is used to calculate the probability of activating nodes and dormant nodes,and standard deviation is used to measure the role probability of nodes to build an energy balance model.The simulation results show that the residual energy difference of sensor nodes in the network can be reduced by the method.At the same time,the energy efficiency of wireless sensor networks can be improved on the basis of ensuring high accuracy of multiple target positioning.In the process of target tracking,the communication energy consumption of sensor nodes occupies most of the network energy consumption.In order to further optimize the energy efficiency of the network,this thesis presents a multi-target tracking energy efficiency optimization scheme based on measurement compensation.The algorithm reduces the communication energy consumption from two aspects.First,the data forwarding frequency of the active node is controlled according to the measurement residual size.At the same time,to ensure that the positioning accuracy of multiple targets does not decrease,the algorithm automatically generates an approximate amount of missing observation data,and the average measurement residuals of the nodes are used to dynamically adjust the number of activations.Finally,the situation where a group of sensor nodes track multiple targets at the same time is analyzed,and the Fisher information matrix combination of multiple targets is used to reconstruct the node selection accuracy model.Under the circumstance that the targets are close to each other,multiple dynamic cooperative clusters are fused and turned into single cluster tracking.The simulation results show that the scheme can maintain good tracking accuracy,reduce the number of active nodes,and optimize the network energy efficiency.