Division-based Receiver-agnostic RFF Identification in WiFi Systems

TL;DR

This work tackles cross-receiver degradation in WiFi RF fingerprinting by introducing a division-based, receiver-agnostic RFF extraction framework. It presents two regimes: (i) a reference-device based RD method for flat fading and (ii) a division-based HL method for frequency-selective fading, both achieving high-dimensional, channel-invariant features with only a single training receiver and no calibration or stacking. The approach yields significant accuracy gains over state-of-the-art baselines in both simulated and real-world experiments, with noted improvements up to 15.5% in flat fading and 28.45% in frequency-selective fading, and demonstrates practical viability for closed-set device identification. The combination of high-dimensional RFF features and practical training requirements positions this method as a robust and scalable solution for WiFi device authentication under varied receiver and channel conditions.

Abstract

In physical-layer security schemes, radio frequency fingerprint (RFF) identification of WiFi devices is susceptible to receiver differences, which can significantly degrade classification performance when a model is trained on one receiver but tested on another. In this paper, we propose a division-based receiver-agnostic RFF extraction method for WiFi systems, which removes the receivers' effects by dividing different preambles in the frequency domain. The proposed method requires only a single receiver for training and does not rely on additional calibration or stacking processes. First, for flat fading channel scenarios, the legacy short training field (L-STF) and legacy long training field (L-LTF) of the unknown device are divided by those of the reference device in the frequency domain. The receiver-dependent effects can be eliminated with the requirement of only a single receiver for training, and the higher-dimensional RFF features can be extracted. Second, for frequency-selective fading channel scenarios, the high-throughput long training field (HT-LTF) is divided by the L-LTF in the frequency domain. Only a single receiver is required for training and the higher-dimensional RFF features that are both channel-invariant and receiver-agnostic are extracted. Finally, simulation and experimental results demonstrate that the proposed method effectively mitigate the impacts of channel variations and receiver differences. The classification results show that, even when training on a single receiver and testing on a different one, the proposed method achieves classification accuracy improvements of 15.5% and 28.45% over the state-of-the-art approach in flat fading and frequency-selective fading channel scenarios, respectively.

Figures (25)

• Figure 1: Non-HT PPDU structure.
• Figure 2: HT-MF PPDU structure.
• Figure 3: Diagrams illustrating the proposed receiver-agnostic RFF extraction and classification method: (a) Under flat fading channel scenarios and (b) under frequency-selective fading channel scenarios.
• Figure 4: The CSI without RFF effects and the extracted RFF features from different WiFi devices, captured by different receivers in the L-LTF part under flat fading channel scenarios in simulations: (a) The CSI without RFF effects and (b) the extracted RFF features.
• Figure 5: Normalized RFF features of different devices, obtained by different receivers in the L-LTF part under the flat fading channel scenario during experiments: (a) Dev5, Rx3; (b) Dev11, Rx3; (c) Dev5, Rx5; and (d) Dev11, Rx5.