Research On Energy Efficiency Optimization Of Underwater Sensor Networks For Tracking

Due to the advantages of flexible distribution,multi dimensions,inexpensive equipment and strong concealment of underwater sensor networks(USNs),the underwater target tracking based on USNs has higher tracking accuracy and wider tracking range.Affected by underwater communication and data processing,once the energy of nodes in USNs is exhausted,the lifespan of USNs will be affected.Therefore,how to effectively balance the energy consumption and tracking accuracy of USNs become the focus of underwater target tracking.To solve this problem,this thesis optimizes the tracking energy efficiency from three perspectives: node selection,bit allocation and depth adjustment.The main works are as follows:1.To solve the problems of uncertain positions of nodes under the influence of ocean currents and inconsistent selection of nodes under the single optimization criterion,a node selection algorithm based on multi-objective optimization under position floating is proposed.The characteristic of position floating is analyzed,so the position floating is transformed into floating noise by using the first-order Taylor expansion.The Fisher Information Matrix and Mutual Information under the position floating are derived,and the node selection problem under the position floating is designed by combining the number of nodes.The node selection is realized by Non-dominated sorting genetic algorithm II(NSGA-II)and Technique for Order of Preference by Similarity to Ideal Solution(TOPSIS)algorithm.The simulation results show that the node selection algorithm based on multi-objective optimization considering the position floating can overcome the impact caused by position floating,select fewer nodes,and achieve better and more stable tracking performance.2.In the process of underwater acoustic communication,the energy consumption of nodes is greatly affected by the number of bits.A non-myopic bit allocation algorithm based on dynamic programming is proposed to solve the problems of data conflict and channel congestion in myopic scene.The conditional posterior Crammer lower bound is derived and is set as the optimization criteria.The problem of bit allocation in non-myopic scene is constructed,and a double-side approximate dynamic programming algorithm is proposed to allocate bits.The first-side can allocate bits at each moment,and the second-side can realize the bit allocation of nodes on each branch.The simulation results show that the bit allocation algorithm proposed in this paper has more stable tracking performance under the premise of ensuring the computational efficiency.It also reduces the frequency of data transmission,and realizes high energy efficiency of tracking from the time perspective.3.Aiming at the problems of large transmission delay of acoustic signal,depth-affected sound velocity profile,and tracking effect affected by topology structure during depth adjustment,a node depth adjustment algorithm based on convex optimization under topology control is proposed.The influence of sound velocity profile on acoustic signal transmission is analyzed,the asynchronous particle filter based on delay estimation is improved,and the influence of node topology on tracking accuracy is discussed.The depth-related Fisher Information Matrix is set as the optimization criterion,and the node depth adjustment optimization problem is constructed.For the two scenarios of target depth known and unknown,both the analytical method and interior point method are used to solve the problem respectively,and the optimal depth adjustment strategies in corresponding scenario are obtained.The simulation results show that the proposed algorithm can fully adjust the node depth,obtain a more effective node topology and improve the tracking accuracy when only four nodes are used to participate in the tracking.