PeopleNet

People often seek information by asking other people even when they have access to vast reservoirs of information such as the Internet and libraries. This is because people are great sources of unique information, especially that which is location-specific, community-specific and time-specific. Social networking is effective because this type of information is often not easily available anywhere else. In this paper, we conceive a wireless virtual social network which mimics the way people seek information via social networking. PeopleNet is a simple, scalable and low-cost architecture for efficient information search in a distributed manner. It uses the infrastructure to propagate queries of a given type to users in specific geographic locations, called bazaars. Within each bazaar, the query is further propagated between neighboring nodes via peer-to-peer connectivity until it finds a matching query. The PeopleNet architecture can overlay easily on existing cellular infrastructure and entails minimal software installation. We identify three metrics for system performance: (i) probability of a match, (ii) time to find a match and (iii) number of matches found by a query. We describe two simple models, called the swap and spread models, for query propagation within a bazaar. We qualitatively argue that the swap model is better with respect to the performance metrics identified and demonstrate this via simulations. Next, we compute analytically the probability of match for the swap model. We show that the probability of match can be significantly improved if, prior to swapping queries, the nodes exchange some limited information about their buffer contents. We propose a simple greedy algorithm which uses this limited information to decide which queries to swap. We show via simulation that this algorithm achieves significantly better performance. Overall our results demonstrate that PeopleNet, with its bazaar concept and peer-to-peer query propagation, can provide a simple and efficient mechanism for seeking information.