Physical Layer Security in Bistatic Backscatter Communications for Passive Internet of Things

In this paper, we consider a bistatic backscatter communication system for passive Internet of Things, which includes a carrier transmitter (CT), a dual-antenna backscatter device (BD), and a legitimate user (LU). Due to the risk of being eavesdropped, one of the BD’s antennas is utilized to reflect artificial noise to disturb the eavesdropper (Eve), while the other reflects secret message symbols to the LU. First, to analyze the system’s reliability and security, we derive the outage probability and intercept probability in (semi-)closed form. Then, to enhance the system’s physical layer security, we maximize the secrecy rate by jointly optimizing the CT’s beamforming vector and the BD’s power reflection coefficient, subject to the CT’s maximum transmit power and BD’s power reflection coefficient constraints. To solve this non-convex problem, we propose an alternating-optimization based algorithm that alternately optimizes the variables using the dinkelbach’s transform and successive convex approximation technique. Numerical simulations show that: 1) the analytical results coincide with the simulation results, and the performance trade-off of reliability and security can be realised by optimizing the CT’s transmit power; 2) the proposed optimization algorithm can significantly improve the system’s secrecy rate compared to the benchmarks; 3) the system’s secrecy rate increases for a longer Eve-to-BD distance and a larger number of CT antennas.