Efficient Wi-Fi Device RF Fingerprint Detection and Classification Using Wigner-Ville Distribution

In the realm of Internet of Things (IoT) applications, Radio Frequency Fingerprint Identification (RFFI) is emerging as a robust technique for enhancing wireless security based on the inherent hardware attributes in wireless devices. We proposed and demonstrated a robust scheme achieved by the statistical extraction method based on Wigner- Ville Distribution analysis for Wi - Fi devices. The proposed framework attains full authentication using a low-complexity feature extractor and employs K-nearest neighbor algorithm for both detecting rogue devices and classifying legitimate devices. This work shows the suitability of statistical feature extraction for rogue device detection. An excellent recognition accuracy of 97.64 % is experimentally demonstrated, with a false alarm rate of 3.29% and a miss alarm rate of 0%, using 11 wireless network interface controllers. Notably, the computational efficiency of our RFFI scheme, achieved through lightweight and straightforward feature extraction method, positions it as a viable solution for security enhancement in resource-constrained wireless communications.