Secret Key Generation With Intelligent Reflecting Surface Under the Pilot Contamination Attack

Secret key generation (SKG) is a promising approach to improving communication security, but it faces many attacks such as eavesdropping and pilot contamination. This letter investigates intelligent reflecting surface (IRS) assisted SKG under the pilot contamination attack. First, we acquire the lower bound of the secret key rate on the basis of our system model. Next, an optimization problem of the IRS’s reflecting coefficients is formulated to maximize the lower bound. It is challenging to solve it due to the non-convex quadratic and quartic constraints. We, therefore, propose an efficient algorithm to solve it on the basis of successive convex approximation. Numerical results demonstrate that the IRS can improve the lower bound to counter the pilot contamination attack, and the increase in the correlation coefficient between legitimate and eavesdropping channels does not always decrease the lower bound. Additionally, the secret key rate decreases as Eve’s transmitting power increases.