Exploiting Schelling behavior for improving data accessibility in mobile peer-to-peer networks

In 1969, Thomas Schelling proposed one of the most cited models in economics to explain how similar people (e.g. people with the same race, education, community) group together in American neighborhoods. Interestingly, we observe that the analogy of this model indeed exists in numerous real world scenarios where co-located people communicate via their personal wireless devices (e.g. cell phones, PDAs, Zune) in Peer-to-Peer (P2P) fashion. Schelling’s model therefore can potentially serve as a mobility model and offer a unique opportunity to efficiently disseminate messages in mobile P2P networks. In this paper, we exploit the natural grouping behaviors of humans presented by Schelling to expedite data dissemination in such networks. Particularly, we design a push-based scheme for dense network areas to maximize query hit and a pull-based scheme for sparse network areas to utilize network bandwidth. We ensure that our scheme is lightweight since query and response are automatically limited within groups of mobile nodes carried by similar people. Moreover, we avoid broadcast storms by assigning each message a broadcast timer and applying overhearing mechanism to reduce redundant transmissions. Our scheme also allows leaving nodes and arriving nodes to collaboratively answer queries and thus further improve data accessibility. Finally, our experiment results show that the proposed data dissemination scheme improves the query hit ratio significantly while utilizing network bandwidth efficiently and avoiding broadcast storms.