Research On Incentive Scheme For Opportunistic Networks

Opportunistic Network (ON) is a kind of Wireless Self-organizing Networks (WSNs) derived from DTN where there may be no path between the source node and the destination node. So ONs utilize the communication opportunities from node movement to forward messages. Due to sparse and asymmetry node distribution, nodes always meets opportunistically, so a new routing named "Store-Carry-Forward" is used in ON. Obviously, the transmission of messages depends on the intermediate nodes. However, most nodes of ONs are resource-constrained, such as battery, CPU, memory and bandwidth and so on. If these nodes are rational, they may be reluctant to forward messages for others nodes if they are not interested in these messages. These nodes are named "Selfish Nodes", a lot of literatures show that if the majority of users behave selfishly, the message delivery rate decreases and the quality of the network decreases accordingly.The problem introduced above is the motivation for proposing an incentive scheme. In the WSNs, many incentive schemes are proposed. However, ONs have some inherent characters, such as long delay, intermittent disconnection, those incentive schemes may be not applicable in ONs. So the paper proposes an Exchange-based Reputation Incentive Scheme (ERIS):when two nodes of high reputation value get in the vicinity of each other, they download messages from each other, we hope that it is worth for the users collecting messages even if they are not interested in them to exchange them later for messages that they are interested in. The messages are expected to disseminate faster in the network, so that the delay of messages decreases. If the users do not cooperate, they may lose the messages that they are interested in and make their reputation value declines. If their reputation value is lower than the threshold, they will be out from the network, so that other nodes can’t help them forward messages.We analyze our proposed scheme using game theory techniques, and we simulate it using ONE simulator. The results show that our scheme indeed promotes cooperation among nodes, and the nodes realize that it is better to choose cooperation so that they can get the messages they are interested in and can improve their reputation value. Moreover, the message delivery rate and the quality of the network are improved considerably when the mobile nodes follow the Nash Equilibrium strategy.