Semantic-based publish/subscribe for M2M

Abstract

The number of connected devices is expected to soar in the coming years, each one of them collects and distributes real-world information though various systems. As the number of such connected devices grows, it becomes increasingly difficult to store and share all these new sources of information. Several context representation schemes try to standardize this information, however none of them have been widely adopted. Publish/ subscribe paradigm has proven to be an adequate abstraction for large scale information dissemination, but none of current variations is well suited for context information. In a previous publication we addressed these challenges, however our solution has some drawbacks: poor scalability and semantic extraction. The aim of this paper is twofold. First, we discuss an efficient way to deal with representation schemes diversity and propose a d-dimensional context organization model. Second, we propose a semantic-based publish/subscribe system that is well suited for M2M scenarios. Our evaluation shows that d-dimensional organization model outperforms our previous solution in both speed and space requirements.