Localization verification and distinguishability degree in wireless networks using received signal strength variations

A wireless network is a collection of mobile hosts forming a temporary network using limited infrastructure, or even without the aid of any established infrastructure or centralized administration. This flexibility in space and time induces new challenges towards the security infrastructure needed to support secure communications within such environment. Indeed, wireless networks are subject to attacks due to their characteristics and their vulnerabilities ; one of the most compromising attacks is the sybil nodes attack. We present in this context an intrusion detection technique, based on received signal strength variations, allowing a node to verify the authenticity of nodes which communicate with it, according to their localizations. In addition, we define an evaluation metric of the distinguishability degree between two nodes, allowing to determine sybil and malicious ones within wireless networks. The applicability of our contributions is validated through analysis, simulations and real measurements.