Maximizing Harvested Energy in Green Energy Powered Multi-user MISO RF-based WPT

• Published in: IEEE transactions on vehicular technology pp. 1 - 15
• Main Authors: Zhang, Xiang, Xiong, Ke, Wang, Qiong, Chen, Wei, Fan, Pingyi, Ai, Bo, Letaief, Khaled Ben
• Format: Journal Article
• Language: English
• Published: IEEE 2025
• Subjects: Array signal processing; Batteries; Green energy; Green energy harvesting (GEH); Internet of Things; multi-user MISO; Optimization; radio frequency (RF)-based wireless power transfer (WPT); Rail transportation; RF signals; Sensors; Simultaneous wireless information and power transfer; the total harvested energy maximization; Wireless sensor networks
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2025.3593931

This paper investigates the green energy powered radio frequency (RF)-based wireless power transfer (WPT) system, where a wireless power station (WPS) powered by natural energy wirelessly charges multiple low-power sensors via radio signals. To explore the achievable upper bound of the total energy that the sensors can harvest, an optimization problem is established by jointly optimizing the WPS's active-idle adjusting policy and transmit beamforming subject to the accumulatively arriving green energy, the finite battery capacity, and the maximum transmit power budget. To be general, the non-ideal circuit power consumption is also taken into account. To solve the problem, three propositions are proved theoretically, based on which, we equivalently transform the problem to be convex. By doing so, finding the optimization variables is equivalently transformed to seeking the optimal energy scheduling. Then, by using Karush-Kuhn-Tucker (KKT) conditions, a semi-closed-form solution is derived to characterize the structure of optimal energy scheduling over multiple time blocks, in terms of which, we design a reverse order directional water-filling (RODWF) algorithm to find the optimal solution. Simulations show that the presented RODWF achieves the same global optimal results with the well-known SDPT3 solver but with much lower time complexity. Besides, by changing the weights of the terms in the objective function, the maximum achievable performance region of the energy harvested by multiple sensors is also characterized.