E2E Throughput Maximisation in SWIPT aided Cooperative Communications with Time-Varying Channels

Simultaneous wireless information and power transfer (SWIPT) has been considered as a promising technique to address energy shortage of communication devices deployed in Internet of Everything (IoE). In this paper, we consider a multi-relay aided cooperative communication network with SWIPT, where relay stations (RSs) receive RF signal from a source node (SN) for energy harvesting and information reception by power splitters. A single activated RS then decodes and forwards information to the destination using the harvested energy. Selective-decode-and-forward (S-DF) protocol is adopted, where the activated RS forwards only when information is correctly decoded. By considering time-varying channels, we maximise the end-to-end (E2E) throughput by jointly designing the transmit beamformers for both SN and RSs, optimising transmit power and power splitters for RSs, as well as the RS selection. The original formulated non-convex optimisation problem with coupled variables is solved by invoking an iterative algorithm. Numerical results demonstrate that our design with SDF protocol outperforms that with the decode-and-forward (DF) protocol. Moreover, the impact of the imperfect channel state information (CSI) on the E2E throughput is also evaluated.