Institute of Electrical & Electronic Engineers
This paper addresses a robust downlink beamforming optimization problem for the multi-group multicast scenario, when only imperfect Channel State Information (CSI) is available at the transmitter. The authors consider two different optimization criteria: minimizing the total transmits power subject to Quality of Service (QoS) constraints at each receiver; Max-Min Fair (MMF) Signal-to-Interference-plus-Noise Ratio (SINR) subject to total power constraint. With the aid of S-lemma, the infinite non-convex QoS constraints of robust downlink beamforming problem are transformed into finite Linear Matrix Inequalities (LMI). By applying the Semi-Definite Relaxation (SDR) method, the robust downlink beamforming problem can be relaxed and solved efficiently.