• Article
  • Ingénierie & Outils numériques

Article : Articles dans des revues sans comité de lecture

The Internet of things (IoT) is an indispensable part of our daily lives, bringing us many conveniences, including e-commerce and m-commerce services. Unfortunately, IoT networks suffer from several security issues, such as privacy, access control, and authentication. However, due to the limited computation resources, remote authentication between IoT devices and servers is vulnerable to being attacked over an insecure communication channel. Many authentication schemes have been proposed, but generally, they are based on traditional cryptographic techniques. Unfortunately, most of them are vulnerable to physical attacks since they rely mainly on a stored secret key in the device’s local memory. However, recently Physically unclonable functions (PUFs) have been classified as solid security primitives that could guarantee the three pillars of security (confidentiality, authenticity, and privacy) of sent or received information by IoT devices. PUFs extract unique information from the physical characteristics of the IoT device. Nevertheless, a Fuzzy extractor (FE) should be considered to extract correct and reproducible cryptographic keys from a noisy source. This paper proposes a mutual authentication and a session key establishment protocol for IoT devices based on Silicon PUFs using Arbiter chips. We also validate our developed protocol regarding its resistance to attack scenarios. By relying on formal verification using VerifPal, we found that the proposed authentication mechanism is secure and suitable for resource-constrained IoT devices. Furthermore, our scheme is more efficient than the existing ones in terms of attack robustness. Finally, the experiments have been validated on an Arbiter PUF dataset.