Wireless Physical-Layer Identification: Modeling and Validation

The wireless physical-layer identification (WPLI) techniques utilize the unique features of the physical waveforms of wireless signals to identify and classify authorized devices. As the inherent physical-layer features are difficult to forge, WPLI is deemed as a promising technique for wireless security solutions. However, as of today, it still remains unclear whether the existing WPLI techniques can be applied under real-world requirements and constraints. In this paper, through both theoretical modeling and experiment validation, the reliability and the differentiability of the WPLI techniques are rigorously evaluated, especially under the constraints of the state-of-the-art wireless devices, real operation environments, as well as wireless protocols and regulations. In particular, a theoretical model is first established to systematically describe the complete procedure of the WPLI. More importantly, the proposed model is then implemented to thoroughly characterize the various WPLI techniques that utilize the spectrum features coming from the nonlinear RF front-end, under the influences from different transmitters, receivers, and wireless channels. Subsequently, the limitations of the existing WPLI techniques are revealed and evaluated in details using both the developed theoretical model and the in-lab experiments. The real-world requirements and constraints are characterized along each step in WPLI, including: 1) the signal processing at the transmitter (device to be identified); 2) the various physical-layer features that originate from circuits, antenna, and environments; 3) the signal propagation in various wireless channels; 4) the signal reception and processing at the receiver (the identifier); and 5) the fingerprint extraction and classification at the receiver.