MU-MIMO power control under statistical CSI and probabilistic constraints

Uplink sum power minimization problem with required Quality of Service (QoS) under statistical Channel State Information (CSI) and probabilistic constraints in solved for a multicell multiuser scenario. To handle the probabilistic constraints and analyze the system performance, financial risk management measures Value at Risk (VaR) and Conditional Value at Risk (CVaR) are applied. The resulting expressions may involve functions that are non-invertible and a combination of higher order functions, so obtaining closed form solutions may not be possible. To find the optimal value of the optimization problem and also estimate the worst case performance for a given probability, Extreme Value Theory (EVT) is applied. Simulation results show that the EVT approach generates best bounds when compared with existing methods.