In this paper, we investigate a dual-hop simultaneous wireless information and power transfer (SWIPT) based amplifying-and-forward (AF) multiple-input multiple-output (MIMO) relay communication system where the relay node harvests energy based on radio frequency (RF) signals transmitted from the source node through the hybridized power-time splitting-based relaying (HPTSR) protocol to forward information to the destination node. The joint optimization of the time-switching (TS) factor, source and relay precoding matrices, and the power-splitting (PS) ratio vector is proposed to maximize the mutual information (MI) between the source and destination nodes. We derive the optimal structure for the source and relay precoding matrices to simplify the transceiver optimization problem. Two algorithms based on the upper bound and lower bound of the objective function are proposed to efficiently solve the optimization problem with low computational complexity. Numerical examples demonstrate that the proposed algorithms provide a better MI performance compared with TS based and PS based energy harvesting (EH) relay systems. INDEX TERMS Amplify-and-forward (AF) relay, energy harvesting, hybridized power-time switching relaying (HPTSR), multiple-input multiple-output (MIMO) relay, simultaneous wireless information and power transfer (SWIPT)

In this article, the transceiver design optimization problem is investigated for multi-hop multicasting amplifyand-forward (AF) multiple-input multiple-output (MIMO) relay systems, where multiple source nodes broadcast their message to multiple destination nodes via multiple serial relay nodes. Multiple antennas are installed at the sources, relays, and the destination nodes. In the transceiver design, we consider the mismatch between the true and the estimated channel state information (CSI), where the CSI mismatch follows the Gaussian-Kronecker model. A robust transceiver design algorithm is developed to jointly optimize the source, relay, and destination matrices to minimize the maximal weighted mean-squared error (WMSE) of the received message at all destination nodes. In particular, the WMSE is made statistically robust against the CSI mismatch by averaging through the distributions of the true CSI. Moreover, the WMSE decomposition is exploited to reduce the computational complexity of the transceiver optimization. Numerical simulations show a better performance of the proposed robust transceiver design against the channel mismatch.