Research On Game Model For Energy Consumption Optimization In Wireless Sensor Networks

In recent years,wireless sensor networks have been widely used in various industries and have received extensive attention from researchers.Wireless sensor network nodes are mostly powered by batteries with limited energy.The network nodes are mostly distributed in areas where humans are difficult to reach,such as harsh environments and terrain obstacles.Therefore,it is almost impossible to replace the battery,and the energy problem becomes a key factor limiting the further development of the network.Therefore,it is of great theoretical and practical significance to study energy consumption optimization of wireless sensor network nodes to improve the lifetime of wireless sensor networks.Aiming at the energy-constrained problem of wireless sensor network nodes,this thesis proposes three game model algorithms based on cooperative game theory.The first is the multi-state optimal energy-saving game algorithm of single-node communication module.The second is the multi-node cooperative game model based on a given cluster head.The third is the cooperative game model based on cluster head dynamic election.The rationality and feasibility of constructing energy-saving model by applying cooperative game theory are explained from different angles.The specific research contents are as follows:(1)A game description is performed on several working states of the communication module of a single node.The transmission,reception,idle and sleep are regarded as the players of the game respectively.They optimize the sleep time through the cooperative game strategy,and preliminarily explain the research ideas of cooperative game theory in node energy saving.(2)A multi-node cooperative game model based on a given cluster head is proposed in a cluster.Based on game theory,the residual energy payoff function is established according to the communication energy consumption of nodes,and the nodes' coalitions are constructed through the combination of exhaustion and the sub-regions division.The Nash Equilibrium of cooperative game is solved by using the nodes income distribution method based on Shapley value.The coalition's combination in this state is the optimal coalition combination,which can maximize the residual energy of the nodes in the cluster.The simulation results show that compared with the non-cooperative game model,the energy consumption optimization game model of node cooperation has better stability,and its fluctuation range of energy-saving efficiency is small under different network operation conditions.In addition,the energy-saving efficiency of the game model of energy consumption optimization with node cooperation has significant advantages.With the increase of the number of nodes,the energy-saving efficiency increases from 12.870% to 38.796%.(3)A cooperative game model based on dynamic cluster head election is constructed.Firstly,according to the geographic location of the nodes,the network is divided into several regions,each region is a cluster,and each cluster is composed of cluster head and other nodes.Secondly,the node with the largest energy in the cluster is elected as the cluster head,and the remaining nodes construct a multi-node cooperative game model based on a given cluster head,and finally form an optimal coalition combination.Finally,after each round of data transmission tasks,the remaining energy of all nodes(including the cluster head)is compared,in the next round,the node with maximum residual energy acts as the new cluster head,and the coalition combination of the last round is restructured according to the cluster head situation.The simulation results show that the model has high stability,and the energy-saving efficiency ? has little correlation with the number of running rounds r.Compared with the non-cooperative game model of dynamic cluster head election,the energy-saving efficiency is improved by 16.954%.The number of deaths with the first node between two models is different by 201 rounds.After running 1200 rounds,the number of dead nodes in the non-cooperative game model of dynamic cluster head election is 1.83 times that of the cooperative game.Compared with the cooperative game model based on a given cluster head,the number of deaths of the first node is prolonged by 604 rounds and the network energy are further balanced.Compared with LEACH protocol,the lifetime of the first node is prolonged by 467 rounds,life cycle extended by 190 rounds,and the lifetime of the network is increased by 16.58%.