Intelligent Reflecting Surface-Aided Covert Wireless Communications With Finite-Alphabet Inputs

This treatise evaluates the performance improvement attained upon integrating an intelligent reflecting surface (IRS) into covert wireless communications under finite-alphabet inputs. Specifically, we first derive an analytical expression for a lower bound of the achievable mutual information (MI). Then, we derive an analytical expression for an upper bound of the Kullback-Leibler (KL) divergence for the covertness, based on which we conceive an optimization problem to maximize the covert MI under the constraints on covertness, IRS coefficients and average transmit power. We utilize the semidefinite relaxation (SDR) technique along with the Gaussian randomization method to achieve high-quality solutions for both the transmit power and IRS reflection beamforming. We further propose a low-complexity algorithm, in which we derive analytical expressions for the transmit power and phase shift of each IRS element. Our numerical results demonstrate that, with finite-alphabet inputs, the devised IRS-aided scheme can achieve considerable covert performance gain on the basis of the scheme having no IRS.