Collaborative selfish node detection with an incentive mechanism for opportunistic networks

Selfish nodes in an opportunistic network are nodes that do not want to participate in the routing process for various reasons, such as low resources (e.g. battery, memory, CPU, network or bandwidth), fear of malicious data from unknown users, or even lack of interest in helping nodes from other communities. This may lead to messages sent in the network being delayed or even lost. Therefore, these types of nodes should be detected and avoided. Moreover, incentive mechanisms that reward nodes when they actively take part in the network (i.e. relay messages for other nodes) and punish them when they do not, should be employed where possible. In this paper, we propose a novel collaborative content and context-based selfish node detection algorithm and an incentive mechanism which aim to reduce the issues of selfish nodes in an opportunistic network. Since local information may not be sufficient to reach an informed decision, nodes running our algorithm collaborate through gossiping, for the final goal of detecting selfish nodes, later on to punish and avoid them. We compare our approach to an algorithm entitled IRONMAN and show that it behaves better in terms of network performance and detection accuracy.