Shifting Signatures: The Ephemeral Nature of the Radio Fingerprint on the USRP X310

Radio Frequency Fingerprinting (RFF) offers a unique method for identifying devices at the physical (PHY) layer based on their RF emissions due to intrinsic hardware differences. Nevertheless, RFF techniques depend on the ability to extract information from the PHY layer of the radio spectrum by resorting to Software-Defined Radios (SDR). Previous works have highlighted that the radio power cycle of SDRs leads to fingerprint mutations. In this work, we first show that fingerprint mutations also appear when reloading the firmware of the transmitter, not only when power cycling the device. Moreover, we introduce a new methodology for abstracting fingerprint mutations into a graph and provide a theoretical framework for assessing fingerprint reliability. Our results show that the common assumption that the RF fingerprint is unique and persistent is incorrect. Through real-world measurements collected via the popular USRP X310 and using state-of-the-art deep learning techniques, we exper-imentally demonstrate that the considered SDR (USRP X310) features multiple fingerprints that can be clustered according to shared features. Moreover, we show that the RF fingerprint is a time-independent probabilistic phenomenon, which requires the collection of multiple samples across various reloads to achieve the necessary reliability.