Max-min Fairness-based IRS-aided SWIPT

The performance of an intelligent reflecting surface (IRS)-assisted time-switching simultaneous wireless information and power transfer (SWIPT) system is investigated from a maxmin user-fairness perspective. A series of optimization problems is formulated to maximize the minimum harvested energy and the achievable user rates via jointly optimizing the transmit powers at the base-station (BS) and phase-shifts at the IRS. The underlying optimization problems are non-convex, and thus, the efficient alternating optimization algorithms are developed to obtain sub-optimal solutions. A combination of geometric programming and convex optimization techniques has been employed in an iterative manner to solve the transmit power allocation and IRS phase-shift optimization problems, respectively. Max-min based common/system-wide harvested energy and achievable user rate are characterized when the BS adopts linear/conjugate precoding. Thereby, a max-min fairness-based energy-rate trade-off is quantified. Our numerical results validate the proposed optimization solutions and reveal the underlying performance gains of the optimized system.