Impact of social-aware forwarding on traffic distribution in social opportunistic networks

Abstract

Social opportunistic networks (SONs) are delay-tolerant MANETs that exploit human mobility to enable message delivery in the networks. Humans tend to move in a way that is influenced by their social relations. Knowledge of social relationships therefore can be exploited to build social-aware routing protocols. These algorithms typically favour higher (social) ranking nodes as better relays for message transfers. The combination of this forwarding heuristic and the social network structure, which exhibits a non-uniform connectivity distribution with the existence of a few highly-connected nodes, leads the routing algorithm to direct most of the traffic through these hub nodes. Unbalanced traffic distribution therefore results in the network. This paper presents an analysis of traffic distribution in SONs when social-aware routing protocols are applied in the networks. Initially, we survey state-of-the-art social-aware routing protocols. We next investigate the topology characteristics of real-life SONs. Furthermore, we apply three forwarding strategies on these networks, categorising these strategies into social-aware forwarding and social-oblivious forwarding. The social-aware forwarding strategies consider node ranking when choosing traffic relays and the node ranking here is measured by degree and betweeness centralities. The social-oblivious forwarding, however, disregards node ranking on the forwarding decision and selects a relay node randomly. We show that the social-aware forwarding strategies result in very poor traffic distribution fairness, where a few (hub) nodes process a large fraction of the network traffic. Finally, we discuss the strategies for improving traffic distribution fairness in SONs.