Achieving Max-Min Fairness for MU-MISO with Partial CSIT: A Multicast Assisted Transmission

Abstract: We address the max-min fairness design problem for a MU-MISO system with partial Channel State Information (CSI) at the Base Station (BS), consisting of an imperfect channel estimate and statistical knowledge of the estimation error, and perfect CSI at the receivers. The objective is to maximize the minimum Average Rate (AR) among users subject to a transmit power constraint. An unconventional transmission scheme is adopted where the Base Station (BS) transmits a common message in addition to the conventional private messages. In situations where the CSIT is not accurate enough to perform interference nulling, individual rates are assisted by allocating parts of the common message to different users according to their needs. The AR problem is transformed into an augmented Average Weighted Mean Square Error (AWMSE) problem, solved using Alternating Optimization (AO). The benefits of incorporating the common message are demonstrated through simulations.