Enhancing Fairness-Aware Massive Wireless Powered IoT Connectivity by IRS

This paper explores an intelligent reflecting surface (IRS) enhanced wireless powered Internet of Things (WP-IoT) network, wherein massive IoT nodes are wirelessly charged by radio frequency signals and then transmit information by means of an IRS to promote the system performance. To evaluate the network performance, we aim at maximizing the total throughput while adhering to constraints pertaining to fairness-aware individual signal-to-noise ratio (SNR), the time allocations (TAs) as well as the unit-modulus IRS phase shifts. However, the intricate coupling of these variables renders the optimization problem nonconvex, thus posing a challenge for direct solution. To deal with this dilemma, we first resort to employing the Lagrange dual method and Karush-Kuhn-Tucker (KKT) conditions to transform the sum of logarithmic objective function into sum of fractional counterpart, and further derive the analytical solutions of TAs for downlink wireless energy transfer (WET) and uplink wireless information transfer (WIT). Then, the Riemannian manifold optimization (RMO) is utilized to iteratively derive the IRS phase shifts in term of semi-closed-form expression. Lastly, numerical simulations are conducted to examine the efficacy of the proposed algorithm in enhancing performance in comparison to the existing benchmark schemes.