Channel-PUFs: AI-assisted Channel Estimation for Enhanced Wireless Network Security

Christoph Lipps, Sachinkumar Bavikatti Mallikarjun, Mathias Strufe, Hans Dieter Schotten

In: Journal of Information Warfare (JIW) 20 1 Pages 1-2 ArmisteadTEC 2021.


Next Generation Mobile Networks (NGMNs) are entering existing and future (industrial) wireless networks, associated with advantages such as higher data throughput, low latency, operation in almost real-time, and the microcell approach. However, this development also comes with drawbacks in the form of new attack vectors and security threats. Within this work, Physical Layer Security (PhySec) methods—the Channel-based Physically Unclonable Functions (PUFs)—are applied to derive symmetric cryptographic credentials and to establish a trustworthy sound, and secure communication between interacting entities. This is done by using a real-world implementation of an NGMN testbed to evaluate the adaptions to mobile radio. Artificial Intelligence (AI) in the form of the Linear Regression Algorithm (LRA) is applied to enhance the accuracy of the estimated channel profiles.

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz