Research On Target Tracking Algorithm Based On Interaction Multiple Model In Wireless Sensor Networks

Advances in wireless communication technology,embedded technology and sensor technology,wireless sensor networks is widely used in military and civil fields.As one of the important application technologies of wireless sensor networks,target tracking has been widely concerned by experts and scholars at home and abroad,and has achieved a lot of achievements.In recent years,the background of target tracking has changed a lot.The emergence of some civil background greatly increases the types of maneuvering targets,which brings great challenges to the tracking system.Therefore,it is of great practical significance to study a tracking algorithm with high tracking accuracy and good real-time performance.In this thesis,the interactive multi-model target tracking algorithm is improved,and the main work is summarized as follows:In order to improve the fault tolerance and scalability of the network,a distributed wireless sensor network structure is proposed in this thesis to track the target.Aiming at the problems that the fixed noise covariance matrix and Markov transition probability matrix are used in the interactive multi-model algorithm to cause slow model switching and tracking accuracy degradation.An interactive multi-model target tracking algorithm based on real-time model probability correction is proposed in this thesis.This algorithm introduces a fuzzy neural network to dynamically adjust the measurement noise covariance matrix in the context of a distributed wireless sensor network.In addition,considering the interaction between the models,the Markov transition probability matrix is modified in real time.The simulation results show that the proposed algorithm can effectively improve the role of the matching model,suppress the impact of the non-matching model,and can effectively reduce the problem of large peak errors caused by model switching,and has higher tracking accuracy.Aiming at the problem that the adaptive grid interactive multi-model algorithm only uses angular velocity as the characteristic parameter of the model set and cannot fully characterize all the motion modes of the target.An improved variable structure interactive multi-model target tracking algorithm based on maneuver detection is proposed in this thesis.The algorithm combines angular velocity and acceleration as the characteristic parameters of the model set.At the same time,it performs maneuverdetection on the target,and adjusts the parameters of the model set according to the detection results.Finally,the simulation verifies that the proposed algorithm has a good tracking effect on the moving target.