Service discovery in Mobile Ad hoc networks using Association Rules Mining

Service Discovery in Mobile Ad hoc Networks (MANET) is a difficult task because of the dynamic nature of such networks. The availability of the services in MANET varies with respect to time and location. However, the service discovery requests issued by a consumer in one session are correlated, in most of the cases. If we can find out the correlation among the services and piggyback one service response with others in advance, we can gain improvement in performance. By utilizing this correlation property among service discovery requests, we have proposed a novel approach to Service Discovery in Mobile Ad hoc Networks. We have simulated the proposed approach in JIST/SWANS simulator and the results have shown significant performance improvement.