Joint Optimal Power Control and Transmit-Receive Beamforming in a Multi-User MIMO Environment

This master thesis considers joint optimal power control and transmit-receive beamforming for a downlink system in a multi-user MIMO (Multiple Input Multiple Output) environment. The goal of the transmit strategy presented here is to minimize the total transmitted power subject to some desired Quality of Service criteria, which are based on the operator’s perspective. This results in a multi-variable joint optimization problem with three sets of variables: the transmitter powers, the transmit beamforming vectors and the receive beamforming vectors. Two iterative algorithms will be proposed. The first one is based on the virtual uplink concept, and the solution to this multi-variable optimization problem will be determined by optimizing with respect to one set of variables while having the other two fixed. The second algorithm is based on the uplink-downlink duality for MIMO systems and is solved by minimizing the maximal eigenvalue of an extended coupling matrix. Both algorithms will require global system knowledge and are therefore mainly intended as benchmarks in system simulations. Simulation results will show that both algorithms are producing solutions that might be globally optimal. This is motivated by the fact that the resulting sets of variables are not affected by the initialization process.