Service Discovery and Trust in Mobile Social Network in Proximity

Abstract: Service-oriented Mobile Social Network in Proximity (MSNP) lets participants establish new social interactions with strangers in public proximity using heterogeneous platforms and devices. Such characteristic faces challenges in discovery latency and trustworthiness. In a public service-oriented MSNP environment, which consists of a large number of participants, a content requester who searches for a particular service provided by other MSNP participants will need to retrieve and process a large number of Service Description Metadata (SDM) files, associated semantic metadata files and identifying the trustworthiness of the content providers. Performing such tasks on a resource constraint mobile device can be time consuming, and the overall discovery performance will be affected and will result in high latency. This paper analyses the service discovery models of MSNP and presents corresponding solutions to improve the service discovery performance of MSNP. We firstly present and analyse the basic service discovery models of service-oriented MSNP. To follow up, we apply a context-aware user preference prediction scheme to enhance the speed of the semantic service discovery process. Later, we address the trustworthiness issue in MSNP and propose a scheme to reduce the latency of the trustworthy service discovery for MSNP. The proposed scheme has been tested and evaluated on MSNP application prototype operating on real mobile devices and MSNP simulation environments.