Joint user scheduling, power control and beam-forming for multi-cell distributed antenna systems with imperfect CSI

Distributed antenna systems (DASs) have been shown to considerably improve the physical-layer performance of cellular wireless networks. However, several issues in the design of radio resource management algorithms (RRM) for this type of system remain open today. This paper partially fills this gap by proposing a joint user scheduling, dynamic power control and beam-forming algorithm for the down-link of multi-cell DASs under the assumption of imperfect channel state information (CSI). The algorithm aims to schedule over the same frequency band a different user attached to a different distributed node within the cell. This goal is achieved by optimizing the power levels and beam-forming vectors of each node, thereby controlling the interference created between the simultaneous transmissions of the scheduled users while maximizing the sum-rate capacity of the cell. This optimization problem is solved by means of a gradient descent iterative technique, which provides the set of optimum node-user pairs to be scheduled, as well as their optimum transmit power levels and beam-forming vectors. Outer-cell interference is computed by reusing the results of previous iterations of the central cell in the transmission parameters of the outer-cells, thus mimicking more efficiently the behavior of the algorithm at the system-level. The algorithm is also evaluated in terms of fairness by means of the first-order moment of the spatial distribution of the capacity. Capacity and fairness properties of the algorithm are shown to considerably outperform previous solutions, particularly in scenarios with good line-of-sight, optimum node location, and relatively accurate CSI.