Research On Selfish Node Detection And Punishment Mechanism Based On Game Theory In DTN

Delay tolerant network(DTN)is a kind of network structure which does not need a complete communication path between the source node and the destination node.Nodes adopt the "store carry forward" mode,and use their own mobility to complete the message transmission process by meeting other nodes.Due to the flexibility and large degree of freedom of nodes in Delay Tolerant Networks,the delivery rate of messages and the stability of links become uncontrollable.When a node is limited by some objective conditions,such as insufficient cache space,low energy,and no idle message queue,it usually loses packets or refuses to receive messages from other nodes in order to deliver the messages it has carried or "lighten the load".The behavior that a node does not actively help other nodes to transmit messages in order to get more delivery opportunities for its own messages is called selfish behavior,which is called selfish node.The existing research shows that the selfish behavior of nodes will seriously affect the overall network efficiency.If there are a large number of selfish nodes,it may lead to network paralysis.This paper analyzes the causes of selfish behavior of nodes,the impact on the overall efficiency of the network and the reasons for non cooperation between nodes.Through the help degree of nodes and the relevant principles of game theory,we can find a balance between selfish behavior of nodes and network performance optimization.The concept of help degree is introduced,and the game model is constructed by using help degree.In the game model,the weight factor is introduced to analyze the influence of message delivery rate and node help degree on the overall performance of the network.Finally,the selfish degree of nodes is calculated by weighting.The game model is used to punish the selfish nodes,encourage them to participate in the behavior of helping other nodes forward messages,and finally optimize the network efficiency.The main improvements are as follows:(1)In view of the fact that the nodes in the network may fail to complete the successful forwarding of messages due to objective reasons,such as insufficient energy,insufficient cache space,insufficient message remaining time and so on,this paper proposes to determine the selfishness of nodes based on the ratio of the number of times to help other-nodes forward messages and the number of times to actually contact with other nodes.According to the trajectory of the node,the number of times to contact the node is recorded,and each contact is regarded as a message transmission opportunity.In the message qucae of the node,every successful help for other nodes to forward messages is counted,which is called help times.The ratio of help times and contact times is obtained.When the ratio is less than the specified value,the node is regarded as a selfish node,so as to solve the problem of only sending messages in the prophet algorithm Depending on the encounter probability between the node and the destination node to determine the selection of the next hop node may lead to the transmission failure of the node due to selfish behavior in the process of transmitting messages.(2)Aiming at the problem that selfish nodes in the network will lead to the overall performance degradation and low message delivery rate,an incentive mechanism for selfish nodes based on game theory is proposed.After the node is determined to be selfish,all nodes in the network limit the node to send its own messages actively,and take measures such as sending other nodes’ messages first or sending several other nodes’messages before sending one of its own,so as to urge the selfish node to help other nodes transmit messages as much as possible,so as to return to the ranks of non selfish nodes and get the right to forward messages freely.Finally,the selfish behavior in the network is reduced,and the packet loss rate in the process of message transmission is reduced,so as to increase the reliability of message transmission.In order to verify that the game model proposed in this paper can stimulate the cooperative behavior of nodes and improve the overall network performance to a certain extent.Select one platform to complete the experiment to verify its effectiveness,and compare the experimental data through MATLAB software.Finally,through a large number of experiments on the simulation platform,the results are compared with the classical algorithms,such as prophet,epidemic,spray&wait,and in different aspects of comparative analysis,it is found that the improved algorithm has better performance in delivery rate and flat hop number than the original algorithm.The simulation results show that the incentive mechanism based on the game model can effectively reduce the impact of selfish node behavior on the network,promote node cooperation and improve the message delivery rate.