Finding temporal periodicity with chi-squared test in mobile social networks

Mobile opportunistic networks (MONs) provide data dissemination services in challenged networks where end-to-end data paths are unstable and difficult to remain in a long-term. Opportunistic routing in MONs takes advantage of contact opportunities between mobile nodes to disseminate messages in unstructured networks. Recent studies intend to utilise mobility patterns and select the best relay nodes to improve message forwarding schemes in MONs. In this paper, we exploit the dynamic contact patterns among mobile nodes from the temporal perspective and propose a periodic contact-based routing (PCBR) scheme. To detect periodic communications between two nodes, we analyse the inter-contact time, adopt the chi-squared test to identify definite periodicity, and use such periodic information to enhance the message delivery rate cost-effectively. We conduct a synthetic simulation to examine the PCBR scheme with the SLAW mobility model. Performance results show its efficiency on the successful delivery rate with lower message overhead as compared with typical epidemic and PRoPHETv2 schemes.