Downlink multiuser beamforming and power control for base stations empowered by renewable energy

This work examines offline and online power control policies for efficient usage of renewable energy in the downlink of a multi-antenna wireless system. Two multiuser beamforming schemes are considered, namely, channel inversion (CI) and maximal ratio transmit (MRT) beamforming schemes. Power control policies are derived for both schemes, respectively, with the goal of maximizing the sum throughput by a deadline subject to energy causality and battery storage constraints. With CI beamforming, the power control problem can be formulated as a convex optimization problem, whose solution can be obtained using the directional water-filling algorithm. With MRT beam-forming, the power control problem becomes non-convex, due to interference between the signals intended for different users, and, thus, is difficult to solve exactly. However, an efficient solution can be obtained by performing successive approximation into a sequence of geometric programming problems, i.e., by employing the condensation method. Offline power control policies are first derived assuming non-causal knowledge of the energy arrival and channel coefficients over time. Online power control policies are then proposed based on observations gained from the offline policy. The performance of the proposed schemes are demonstrated through numerical simulations.