Abstract
In this paper, considering the simultaneous wireless information and power transfer scheme, we investigate the robust transceiver design in the multiuser multiple-input-multiple-output system with the energy harvesting constraints where the channel uncertainties are modeled by the worst-case model. We formulate the robust transceiver design problem as a worst-case mean-square-error (MSE) minimization problem where the transmit covariance matrix at the transmitter and the preprocessing matrices at the information-decoding (ID) receivers are alternatively optimized. With the known preprocessing matrices at the ID receivers, the transmit covariance matrix optimization problem is transformed into a semidefinite programming (SDP) by employing the -procedure. With the known transmit covariance matrix at the transmitter, the optimization problem is transformed into another SDP by converting the MSE minimization constraint into a linear matrix inequality. Simulation results have shown that our proposed robust transceiver design outperforms the non-robust one.