RIS-assisted Resource Allocation under Base Stations' Non-cooperation Scheme

In this paper, we focus on reconfigurable intelligent surface (RIS)-aided resource allocation under base stations (BSs)' non-cooperation scheme, where the RIS is solely controlled by one BS and should not affect the communication of the adjacent BS. The minimum quality-of-service (QoS) of users served by the RIS-aided BS, minimum effects on the channel quality of the adjacent BS, the sub-channel assignment rule, and the total transmit power constraint are taken into account. Based on these constraints, the sum-rate of users served by the RISaided BS is maximized by jointly optimizing the RIS passive beamforming, power allocation, and sub-channel assignment. To tackle the non-convex problem, an efficient algorithm exploiting the techniques of block coordinate descent (BCD) is developed. A two-sided matching (TSM) algorithm is firstly applied to solve the discrete sub-channel assignment optimization. Then the power allocation and RIS passive beamforming are optimized iteratively. To address the non-convexity in optimizing the power allocation and RIS beamforming, a successive convex upper bound approximation method and a multi-ratio fractional programming (FP) with Taylor series approximation-based successive cancellation algorithm (SCA) are used, respectively. The convergence of simulation results proves the validity of our proposed algorithm, and the effects of the numbers of RIS elements and the total transmit power are studied.