Robust Stochastic Optimization For MISO Broadcasting Channel With Delayed CSIT and Limited Transmitting Antenna Citation for published version: Ratnarajah, T 2015, 'Robust Stochastic Optimization For MISO Broadcasting Channel With Delayed CSIT

This work considers the K-user multiple-inputsingle-output (MISO) broadcast channel (BC) with delayed channel state information at transmitter (CSIT) where the base station (BS) has only two antennas. Based on the MAT scheme, we propose a probabilistic-constraint optimization approach at the transmitter side to design the beamforming vectors by compromising between aligning interference and enhancing signal detectability. At the receiver side, minimal interference leakage method is proposed and the corresponding closed-form expressions for the interference suppression matrices are derived. For K = 2 case, the effective MIMO channel is derived and the closed-form expression of achievable data rate is obtained based on the proposed optimization method. Moreover, we show that the optimal scheme when K = 3 has the same number of effective channels as in the case of K = 2, which can be optimized by the probabilistic-constraint optimization method. For the case K ≥ 3, we propose a new suboptimal scheme which achieves 4K 3K−1 degrees-of-freedom and show that the new scheme can also be decomposed into the same number of effective channels as in the case of K = 2. Simulation result demonstrates interference leakage power versus the achievable data rate, where it can be seen that the achievable data rate increases slowly with the interference leakage until the optimal point and then decreases drastically. Simulation results also show that the proposed scheme outperforms the MAT scheme and dual signal-to-interferencenoise ratio (SINR) scheme for various channel variances.