Fine-grained Augmentation for RF Fingerprinting under Impaired Channels

• Published in: 2022 IEEE 27th International Workshop on Computer Aided Modeling and Design of Communication Links and Networks (CAMAD) pp. 115 - 120
• Main Authors: Gul, Omer Melih, Kulhandjian, Michel, Kantarci, Burak, Touazi, Azzedine, Ellement, Cliff, D'Amours, Claude
• Format: Conference Proceeding
• Language: English
• Published: IEEE 02.11.2022
• Subjects: data augmentation; Data models; Deep learning; Fingerprint recognition; Performance evaluation; Radio frequency; radio frequency fingerprinting; Radio transmitters; secure design; Training data; unmanned aerial vehicles
• ISSN: 2378-4873
• DOI: 10.1109/CAMAD55695.2022.9966888

Critical infrastructures such as connected and au-tonomous vehicles, are susceptible to cyber attacks due to their mission-critical deployment. To ensure security by design, radio frequency (RF)-based security is considered as an effective technique for a wirelessly monitored or actuated critical infrastructure. For this purpose, this paper proposes a novel augmentation-driven deep learning approach to analyze unique transmitter fingerprints to determine the legitimacy of a user device or transmitter. An RF fingerprinting model is susceptible to various channel and environmental conditions that impact the learning performance of a machine/deep learning model. As data gathering cannot be considered as a feasible alternative, efficient solutions that can tackle the impact of varying channels on learning performance are emergent. This work aims to shed light on the RF fingerprinting problem from a different angle when 4G, 5G and WiFi data samples are collected from different transmitters by proposing a fine-grained augmentation approach to improve the learning performance of a deep learning model. Numerical results point out the promising RF fingerprinting performance when training data are augmented in a waveform-specific fine-grained manner as fingerprinting accuracy (87.94%) under the previously presented TDL/CDL augmentation can be boosted to 95.61% under previously unseen RF data instances.