LoS/NLoS Detection based Authentication for IoT Systems

Abstract

The number of IoT devices is increasing significantly these days, especially in private spaces. The data collected in personal space may include private information. Because a lot of IoT devices use wireless channels to communicate with other devices, illegal devices (attackers) can easily access the channel. The attacker can access IoT devices and pretend to be legitimate devices. This paper proposes a novel LoS/NLoS detection based authentication scheme that aims to guarantee security while providing proper user convenience. We propose a new metric called Gap Difference of Phase (GDP) that helps to differentiate between LoS and NLoS channels. We evaluate the performance of the GDP-based LoS/NLoS detection method that also uses Support Vector Machine (SVM). As a result we confirm that our proposed method successfully differentiates between LoS/NLoS channels and achieves a balance between a high level of security and user convenience.