An adaptive access control model for medical data in Wireless Sensor Networks

Wireless Sensor Networks (WSNs) have recently attracted a lot of interest in the research community. The security mechanism with large overhead of computation and communication, are infeasible to apply in WSNs due to many constraints such as limited energy, resource and memory, and low computation capability. Current access control models cannot make an effective access decision in many events because access decisions are based on predefined access policies and roles. Sometimes, users may need to access important data urgently but apart from those predefined access policies, other user request will not be granted. An adaptive access control model is proposed aiming to provide a flexible and an effective access decision on user access request at any time. The proposed model is developed in Ponder2 framework with additional extensions to adapt the unexpected events by using privilege overriding and also adjust its decision based on users' behaviour trust value. A medical scenario is used as an example application to develop and evaluate the proposed model in Body Sensor Networks (BSNs) and WSNs. In this paper, detailed design, implementation phase, evaluation result and policies testing for the proposed adaptive access control model are presented. Based on an evaluation result, all the modules in the proposed access control model are cooperated to make an effective access decision.