Robust Transceiver Design for the Multi-User Interference Channel

Ali Tajer, Narayan Prasad, Xiaodong Wang

Abstract: We consider the problem of designing robust linear transceivers for a memoryless narrowband Gaussian interference channel (GIC) where M multi-antenna sources communicate with their respective single-antenna receivers. The design of such linear transceivers heavily depends on the accuracy of the channel state information (CSI) available at the transmitters. In practice, the transmitters can acquire only imperfect or noisy CSI. We adopt a popular noisy CSI model which assumes that the noise terms (i.e., errors in the CSI) lie within known hyper-ellipsoids and design transceivers that optimize a worst-case quality of service measure. In particular, we focus on maximizing the worst-case weighted sum-rate as well as the worst-case minimum rate. For obtaining such transceiver designs, we exploit semidefinite programming methods and offer efficient centralized and distributed algorithms that entail different levels of information exchange among the transmitters.