Research On Underwater Target Tracking Algorithm Based On Consensus Bayesian Estimation

Target tracking is an important topic in the application research of underwater acoustic sensor networks.The underwater target tracking system aims to achieve accurate position and persistent tracking of underwater target through underwater communication between sensor nodes and target.This will be provided important practical significance for improving the ocean observation capability and maintaining the maritime rights in our country.However,the uncertainty of the underwater environment,propagation delay,high energy consumption and strong noise interference characteristics make target tracking more challenging.In this paper,the underwater acoustic sensor network node is used to obtain the target position.On the basis of this,underwater moving target tracking problem is further studied.The main research work of this paper is as follows:1?The Bayesian theorem and related contents are briefly discussed,and Bayesian filtering is applied to the target tracking by constructing the dynamic equation of the target and the measurement equation of the sensor nodes.Meanwhile considering the general target tracking in the case of clock synchronization problem,the single Bayesian algorithm is designed to effectively track the target.2?In the case of clock synchronization,aiming at the uncertainty of underwater environment and strong noise interference,a single-node Bayesian algorithm is designed to improve the confidence of the sensor nodes detection target.Compared with the set target probability threshold,it is determined the presence of target that detected by sensor nodes.Based on this,a consensus-based on Bayesian algorithm is designed to reduce the measurement errors and improve the tracking accuracy of underwater moving target.3?For the clock asynchronous problem between the sensor nodes and the target,we first construct the asynchronous clock model,including clock drift and clock offset.And on the base of this,in order to eliminate the asynchronous clock of positioning caused by the underwater acoustic weak communication during the phase of position acquisition,the asynchronous positioning algorithm is designed by utilizing the sensor networks to establish the relationship between clock and asynchronous measurement to obtain the underwater target position.4?Aiming at the high energy consumption problem in underwater acoustic sensor networks,a duty-cycle strategy of sensor nodes is proposed from the perspective of energy optimization.In the persistent tracking phase,based on the strong noise interference characteristics of underwater,a consensus-based Bayesian algorithm is designed by the information fusion with the neighboring sensor node groups to improve the target tracking accuracy,which will reduce the network communication links and extend the network lifetime.