Research On Node Localization And Target Tracking Algorithm In Underwater Wireless Sensor Networks

The 21 st century is the century of the ocean,and the ocean will become a new space for human survival and development.Target tracking in Underwater Wireless Sensor Networks(UWSNs)has become a research hotspot as an important means of continuous monitoring in a wide range of water.Since nodes in UWSNs provide measurement for target tracking,the position information of nodes is crucial for target tracking.Therefore,this thesis first studies the node localization in UWSNs,then studies the target tracking in UWSNs.The research work of this thesis mainly includes the following aspects:(1)Aiming at the problem of fast localization of sensor nodes in a certain area in UWSNs,a mobile anchor node assisted received signal strength localization algorithm is proposed.Firstly,to improve the location accuracy,a support vector regression-based interpolation method is designed.This method can increase the accuracy of the nonlinear regression model of noisy measured data,and decrease the estimation error caused by the discreteness of measured data.The location accuracy is improved by the reduction of the projection position error of the sensor node on the linear trajectory.Then,to improve the location speed,a curve matching method is designed.By estimating the perpendicular distance from sensor nodes to the linear trajectory of the mobile anchor node,sensor nodes can locate themselves based on the projection and the perpendicular distance after only one trajectory is driven by the mobile anchor node,and the location speed is improved.(2)Aiming at the problem of accurately locating nodes under the condition of the asynchronous clock in UWSNs,an asynchronous localization algorithm based on signal propagation delay is proposed.By constructing the relationship between signal propagation delay and node location,the influence of the asynchronous clock is eliminated.The optimal estimation function of node location which minimizes the sum of all measurement errors is given,and the function is solved to obtain the accurate location of nodes.In addition,for the case that the time measurement will be wrong,a node localization algorithm based on the received signal strength is proposed.The location error caused by the time measurement error is reduced by fusing the location results of the two algorithms according to the weight.(3)Aiming at the balance of tracking accuracy and energy consumption in UWSNs target tracking process,a target tracking algorithm based on adaptive sampling is proposed.Firstly,to maximize the energy efficiency,an Adaptive Sampling Interval Adjustment(ASIA)method is designed.This method uses a two-input-single-output fuzzy logic controller to adaptively adjust the sampling interval,ensures the tracking uncertainty of each step equal to the uncertainty threshold,the maximum reduction in energy consumption is achieved by minimizing the sampling frequency.Then,to balance the energy consumption in different parts of the tracking area,a dynamic uncertainty threshold adjustment method is designed.The single-input-single-output fuzzy logic controller is used to dynamically adjust the uncertainty threshold in the ASIA method according to the location of the target.The change in sampling frequency achieves the balance of energy consumption in different parts.(4)In order to verify the designed algorithms through experiments,an algorithm experiment platform is developed and algorithms are verified.Firstly,the overall design of the experimental platform,the system design of the underwater acoustic communication machine,and the system design of the autonomous underwater vehicle are expounded respectively.Then,using the developed experimental platform,the node asynchronous localization algorithm based on signal propagation delay and the node asynchronous localization algorithm based on received signal strength are experimentally verified.