Wireless Information and Power Transfer: Rate-Energy Tradeoff for Equi-Probable Arbitrary-Shaped Discrete Inputs

In this paper, the fundamental rate-energy (R-E) tradeoffs are studied for simultaneous wireless information and power transfer (SWIPT). Unlike the existing results obtained using Gaussian input given an average power constraint, we focus on the equi-probable arbitrary shaped discrete inputs considering a peak power constraint as well as an average power constraint. For these power constraints, the transmit power strategies are presented and the R-E regions are derived. The optimality of different SWIPT schemes is first studied in terms of the R-E tradeoff for the case of negligible circuit power consumption in the information decoder. The results are extended to the case of non-negligible circuit power consumption in the information decoder. The asymptotic R-E regions are also derived, which give us useful insights into the fundamental R-E tradeoffs. Finally, the R-E regions of Gaussian input and discrete inputs are compared under an average power constraint.