On the authentication of devices in the Internet of things

Abstract

Reaping the benefits of the Internet of things (IoT) system is contingent upon developing IoT-specific security and privacy solutions. Conventional security solutions fail to meet the IoT security requirements due to the computationally limited and portable nature of IoT objects. In this paper, an object authentication framework is proposed to exploit device-specific information, called fingerprints, to authenticate objects in the IoT. The proposed framework is shown to effectively track the effects of physical environment on objects' fingerprints via a transfer learning tool to differentiate between security attacks and normal change in fingerprints. Simulation results show that the proposed framework improves the authentication accuracy.