Opportunistic Over-Provision for Minimizing Transmit Power with Long-Term Average Rate Guarantees in STAR-RIS Communication Systems

We investigate the problem of minimizing the transmit power while providing long-term average data rate guarantees for two mobile users in a simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) assisted wireless communication system. We formulate the studied problem as a sequential optimization problem in which one has to make a decision in each time slot based on current channel states and past data rates. To achieve long-term rate guarantees, we propose the opportunistic over-provision algorithm for choosing an achievable rate vector for users, finding a beamforming vector for the base station and selecting phase shifts as well as energy splitting ratios for the STAR-RIS in each time slot. Specifically, when channel gains are large enough, the opportunistic over-provision algorithm could select an achievable rate vector that is larger than the vector that consists of specified long-term average rates. Simulation results indicate that the proposed algorithm could significantly outperform a number of alternative schemes.