Throughout Maximization for IRS-Assisted WPCN With Hybrid TDMA-NOMA Scheme

This work leverages an Intelligent Reflecting Surface (IRS) to improve throughput performance of a wireless powered communication network (WPCN). A novel hybrid multiple access strategy is conceived, where the IRS set the time-division multiple access (TDMA) to allocate its phase shifts, each of which performs the non-orthogonal multiple access for multiple Internet of Things (IoT) devices. These devices firstly collect energy in a non-linear manner disseminated by a power station (PS) at downlink wireless energy transfer (WET) stage. The gathered energy is then utilized by these devices to convey data information to an access point (AP) by employing the TDMA protocol with flexible time slots at the uplink wireless information transfer (WIT) stage. During each time slot, the NOMA protocol is utilized for all IoT devices' data transmission, thus creating a hybrid TDMA-NOMA framework. To assess the overall performance of the system model under investigation, we aim to maximize the network throughput, subject to the constraints of the downlink WET and uplink WIT time scheduling, the power allocation for each time slots, and unit-modulus IRS phase shifts of downlink WET and uplink WIT, thereby highlighting energy harvesting and information transmission capabilities. The formulated problem is inherently not jointly convex due to the coupled nature of the variables, making it difficult to resolve directly. To deal with this challenge, a power allocation factor is first introduced for each IoT device, and then an alternating optimization (AO) algorithm is proposed to jointly design the power allocation, IRS phase shifts, and time scheduling in an alternated fashion. To be specified, the optimal power allocation factor can be obtained using interior-point methods for given time scheduling and IRS phase shifts. Also, the time slots of downlink WET and uplink WIT stages are theoretically derived in closed-form expressions using the Lagrange dual method and the Karush-Kuhn-Tucker (KKT) conditions. Additionally, the semi-closed-form expressions of IRS phase shifts during WET and WIT stages are iteratively achieved through Riemannian Manifold Optimization (RMO) and quadratic transformation (QT)-based Alternating Direction Method of Multipliers (ADMM) algorithms. Finally, the numerical results are demonstrated to highlight the effectiveness of the proposed algorithm, the optimal IRS phase shift design, and the optimal WET time scheduling design in comparison to the benchmarks.