Probabilistically-Constrained Approaches to the Design of the Multiple Antenna Downlink

We consider the downlink of a cellular system in which the base station is equipped with multiple antennas and each user has a single antenna. We study the design of linear precoders with probabilistically-constrained Quality of Service (QoS) requirements for each user, in scenarios with uncertain channel state information (CSI) at the transmitter. Our goal is to design the precoder so as to minimize the total transmitted power subject to the satisfaction of the QoS constraints with a maximum allowed outage probability. We consider two stochastic models for the uncertainty in the channel coefficients of each user. The first is a Gaussian model that is appropriate for uncertainty that results from estimation errors. The second one is uniform model that is appropriate for the quantization errors in systems with quantized feedback of channel state information. We formulate the design problem as a chance constrained optimization problem, in which each chance constraint involves randomly perturbed second order cone constraints. We adopt a conservative approach that yields (deterministic) convex and efficiently-solvable design formulations that guarantee the satisfaction of the probabilistic QoS constraints. Furthermore, based on these convex formulations, we propose computationally-efficient algorithms that can reduce the level of conservatism in the initial formulations. Our simulations indicate that the proposed methods can significantly expands the range of QoS requirements that can be satisfied in the presence of uncertainty in the CSI.