Joint power allocation and power splitting for MISO SWIPT RSMA systems with energy-constrained users

Abstract

Simultaneous wireless information and power transfer (SWIPT) has been widely used in multi-input single-output (MISO) systems to transmit information and energy simultaneously from the base station (BS) towards the users. The traditional approach used in the literature is based on space-division multiple access (SDMA) method by using the multi-user linear precoding technique, where the information decoding is done by considering the multi-user interference as noise. Recently, the novel rate-splitting multiple access (RSMA) method, based on partially decode the multi-user interference and partially treat that interference as noise, has been shown to outperform the SDMA method in multi-user MISO systems. Motivated by the superior performance of RSMA, we consider a multi-user MISO SWIPT system applying the RSMA method, where multiple energy-constrained users are equipped with a power-splitting structure to harvest energy and decode information simultaneously. In particular, we investigate the optimal precoders and power-splitting ratios design to minimize the total transmit power at the BS, subject to constraints of the minimum data rate for users, minimum energy harvesting by users, and maximum power at the BS. The proposed solution for the formulated non-convex problem is based on two phases. First, we convert the non-convex problem into bilevel programming where the upper optimization problem is solved by using particle swarm optimization. Second, we propose two algorithms to solve the inner optimization problem based on a semidefinite relaxation method or a successive interference cancellation method. Numerical results show that RSMA achieves significant improvement over SDMA in reducing the total transmit power.