Research On Incentive Scheme For Opportunistic Networks Based On Game Theory

Opportunistic Networks is a new kind of network, which is derived from the traditional Self-organizing Network. Different from traditional wireless network, it does not rely on a fixed infrastructure and transmits data between nodes by the mode of "storage-carry-forward". Because of the highly flexible networking features, Opportunistic Networks has a wide range of application prospect. Therefore, it has attracted more and more scholars’ attention and research.At present, most of the research on opportunity networks has a premise hypothesis, that all nodes in the network will take the initiative to forward data for other nodes, and will not reject the forwarding and drop the packets. However, in the actual network environment, nodes are often intelligent devices, which are limited by the resources of energy, memory space, CPU processing capacity and so on. When the node’s energy is insufficient, to maintain the survival, it will give up cooperation, and then shows the selfish behavior. The existing research results show that the selfish behavior of a small number of nodes in the network is very harmful to the performance of the network. Therefore, this paper focuses on the problem of selfish behavior of opportunistic networks, analyzes and combines the existing solutions, and puts forward two incentive mechanisms. The main research contents are as follows:(1) The existing mechanism blindly encourage nodes to cooperate, without taking into account the actual situation of the node itself, which results in excessive consumption of nodes and death and greatly reduces the network service life. To solve this problem, this paper proposes a gaming based energy-constrained incentive protocol. In this mechanism, firstly, the concept of energy level is introduced, and the nodes are divided into three classes, which are rich, medium and poor according to the remaining energy. Secondly, using the knowledge of game theory analyses the different behavior of nodes, and make the rational nodes under the drive of interests to collaborate. Finally, the mechanism takes different incentives for the nodes in different energy levels. When the node is in a rich state, it will receive the most severe punishment for the selfish behavior. When the node is in the middle and poor state, its selfish behavior can be tolerated in a certain degree. Thus, on the one hand, it restrains the selfish behavior of nodes, and promotes the cooperation of nodes. On the other hand, it extends the service life of the network.(2) In opportunistic networks, the source node in the selection of the transmission lines, often has great randomness, which may cause the increase of network latency. In addition, when the nodes in the optimal transmission line appear selfish behavior, it will cause the low success rate of the node delivery, and seriously affect the performance of the network. To solve above two problems, this paper proposes an incentive protocol based on non-cooperative game. In this mechanism, the concept of node priority is introduced, and the node priority is defined by the three attribute values of the node’s running speed, energy and message forwarding success rate. Secondly, Dijkstra algorithm is used to select the optimal link of message transmission based on the node priority. Again, the game theory knowledge is introduced to analyze the optimal link in the node behavior. The exogenous subsidies is used to change the income and promote the cooperation of nodes. Finally, through the detection of node behavior, once the node performance is found out of privacy, it will be excluded from the network. In this way, the nodes in the optimal link can be made to cooperate to ensure that the message is transmitted in accordance with the optimal link, which improves the success rate of node message delivery, and greatly reduces the network delay.