Effectively Identifying Wi-Fi Devices through State Transitions
Abstract
Wi-Fi management frames reveal structured communication patterns that persist even under randomization of MAC addresses. Prior approaches to associating randomized MAC addresses with devices primarily focus on probe requests, overlooking the broader set of management frames and their transition dynamics. This narrow focus limits their robustness in dense, real-world environments with high device mobility, where probe activity alone fails to yield stable and distinctive signatures. In this paper, we present a novel framework for fingerprinting Wi-Fi devices based on behavioral dynamics extracted from passively observed management frames. We model each device's behavior as a finite state machine and introduce matrix-based representations that encode both structural (state transition frequencies) and temporal (inter-state delays) characteristics. These matrices are embedded into compact feature vectors, enabling efficient similarity comparison. Through extensive evaluation in diverse real-world settings, our method achieves over 86% identification accuracy for non-randomized devices using only Wi-Fi management frames, with further improvements observed through temporal burst aggregation. Our findings are sufficient to uniquely and consistently characterize devices at scale, outperforming the state-of-the-art.