Navigating a mobile social network

Abstract

With the emergence of mobile social networks, there has been an increasing interest in studying the methods of using social network connections, such as social network-aware routing, to improve the performance of communication networks. This raises the need for new models of combined social and communication networks with realistic link reliability and mobility properties. Going beyond graph analysis of social networks, such models are useful to have a better understanding of the algorithmic aspects in mobile social networking, for example, how to navigate a mobile social network and search for users with local information. The first step toward this goal is to develop a combined social and communication network model, where wireless communication becomes the underlay to route information with the aid of social connections. Random patterns of node mobility are included in this combined model, which is then used to analyze the average delay of search paths between source-destination pairs. By applying greedy routing over the combined network, analytical expressions are obtained to evaluate the average delay as a function of separation between source-destination pairs. The results quantify the dependence of the average delay on network mobility and on the combined use of social and communication links. Search paths are typically prone to link failures due to message drops in social and communication links. Therefore, the network model is extended with probabilistic link failures, and analytical expressions are obtained for success probability on search paths. The analysis of average delay and success probability shows how social connections can be used to reduce end-to-end delay and increase end-to-end success probability in a mobile social network.