MDPAS: Markov Decision Process Based Adaptive Security for Sensors in Internet of Things

Nowadays chipped based sensors and RFID tags are widely employed in Internet of Things; however, for those devices, effective and flexible security mechanisms lack. In this paper we study the security requirement and propose an adaptive security framework for sensors in Internet of things, which provides dynamic confidentiality, authenticity and integrity in the networks with relative suitable overhead by context aware computing, decision making and dynamic enforcement of policies. We employ Markov Decision Process to make the decisions of security actions and adopt aspect-oriented programming technique to enforce the security policies dynamically in the working networks. We made simulations of our framework, and the performance is encouraging.