Target Node Tracking Method Based On Distributed Particle Filtering

Wireless sensor network has attracted a lot of attention and discussion in recent years because of its flexible and fast networking and easy deployment.Among them,node positioning and target tracking are widely used in intelligent breeding,environmental monitoring,and disaster relief.In order to reduce energy consumption in practical applications and ensure the effective operation of the network,research on methods that can save network resources and ensure the positioning and tracking accuracy of nodes has become one of the urgent problems to be solved.In order to better track the target node,first optimize the deployment of anchor nodes to increase network coverage,then locate unknown nodes,and finally use anchor nodes and unknown nodes to coordinate the tracking of target nodes.The main research contents of this paper are:(1)Aiming at the problem of particle filter algorithm in target tracking application with high algorithm complexity and long tracking time,which leads to the inaccuracy of the final coordinates,the simulated annealing positioning algorithm based on modified ranging is used to realize the positioning of the target node.When calculating the average hop distance,in view of the large error caused by the uneven distribution of anchor nodes,the initial position estimate obtained by the DV-Hop algorithm is used as the initial value of the simulated annealing algorithm,and the similarity is introduced to correct the distance measurement value.,Search for the global optimal solution,and modify the positioning result.Experiments show that the improved algorithm has the same number of nodes and the same communication radius.Compared with the DV-Hop algorithm and the genetic particle swarm optimization DV-Hop algorithm,the positioning accuracy is improved by 24% and 9%,respectively.Higher positioning accuracy.(2)Aiming at the problem that the DV-Hop algorithm(distance vector hopping algorithm)has a large error in estimating the average hop distance,which leads to the inaccuracy of the final coordinates,the simulated annealing location algorithm with modified ranging is used to achieve the location of the target node.When calculating the average hop distance,the initial position estimate obtained by the DV-Hop algorithm is used as the initial value of the simulated annealing algorithm,and the similarity is introduced to correct the range value,search for the global optimal solution,and modify the positioning result.Experiments show that when the number of nodes is the same and the communication radius is the same,the improved algorithm reduces the cumulative tracking error by 24% and 9% compared with DV-Hop and genetic particle swarm optimization DV-Hop?(3)Aiming at the problem of particle filter algorithm in the target tracking application of high algorithm complexity and long tracking time,the distributed particle filter method is adopted to improve the tracking efficiency.Based on the framework of non-proportional particle filter processing,under the RSSI(received signal strength)observation model,the system is modeled and the state transition process of the target is estimated.Experiments have proved that the improved target tracking algorithm reduces the cumulative tracking error by 20% compared with centralized particle filtering when the environmental parameters are the same.