Research On Accurate Target Tracking Algorithm Based On Underwater Wireless Sensor Networks

With its vast area and rich resources,the ocean is an important strategic space for future human survival and development.As one of the key technologies for marine development,target tracking based on underwater wireless sensor networks is of great research significance.Target tracking nodes provide measurements on targets.However,the limited energy and bandwidth of the underwater nodes causes that the amount of target information measured and transmitted by the nodes is limited,resulting in reduced target tracking accuracy.In addition,underwater sound speed variations,noise interference and other factors cause target information measurement errors,which can also affect target tracking accuracy.Therefore,this paper focuses on the above-mentioned problems and studies target accurate tracking algorithms based on underwater wireless sensing networks.Firstly,to address the problem of limited energy of target tracking nodes,a target tracking node selection algorithm based on isogradient sound speed profile is proposed.First,an isogradient sound speed profile measurement model is used to correct the measurement error.The evaluation index of the information contained in the nodes based on the isogradient sound speed profile is then constructed,and the nodes with the greatest amount of information are selected to track the target,through which the target tracking accuracy is improved while reducing energy consumption.Simulation results show that the algorithm improves the target tracking accuracy with the same number of target tracking nodes.Secondly,to address the problem of limited bandwidth of target tracking nodes,a matrix genetic based bit allocation algorithm for target tracking information transmission is proposed.First,the node bit allocation scheme is constructed as a population matrix.Then a quantile measurement model is used to calculate the amount of information contained in the quantile measurements.Finally,the matrix genetic algorithm is used to search for the bit allocation scheme with the largest amount of information to reduce the quantization error.Simulation results show that the algorithm improves target tracking accuracy with the same number of transmitted bits.Finally,to address the problem of differences amount of information of the target contained in the measurements of the target tracking nodes,a double layer weighted unscented Kalman target tracking algorithm based on the difference in information is proposed.Weights are first assigned to the node measurements based on their amount of information.The weighted measurements are then used to correct the weights of the particles characterising the prior state of the target to fully exploit the information contained in the measurements to correct the target state.Simulation results show that the algorithm improves target tracking robustness and accuracy.