Phase Shift Matrix Optimization and Channel Quantization Alternating in RIS-Assisted Physical Layer Key Generation

In the quasi-static environment, the physical layer key generation (PLKG) is limited due to the low entropy of the channel. To overcome this challenge, we use reconfigurable intelligent Surfaces (RIS) to reconfigure channels to aid key generation. In this paper, we introduce a RIS-assisted key generation protocol and focus on minimizing bit disagreement ratio (BDR). We analyze the influencing factors of BDR, namely, the reconfiguration error during channel reconfiguration and the estimation error during channel probing. In our proposed scheme, we first select the corresponding quantizer according to the channel state information (CSI) in order to utilize the channel range as much as possible. Then, to minimize the BDR caused by the estimation error, we propose an optimization scheme of the phase shift matrix and give a closed-form solution. Meanwhile, we use channel quantization alternating (CQA) to reduce the impact of the reconfiguration error on BDR. Compared with existing benchmark schemes, the simulation results verify the effectiveness of our proposed scheme.