Weighted Sum Rate Maximization for MIMO-OFDM Systems with Linear and Dirty Paper Precoding

Many sophisticated resource allocation strategies are based on the maximization of the weighted sum of data rates for a given transmit power. While this problem can be easily solved for orthogonal multiple access schemes like TDMA, it is much more complicated if users are separated in space using multiple antennas at the base station due to the mutual coupling. In this paper, we propose a new projected conjugate gradient algorithm for the optimization of the transmit filters. The power constraint is taken into account in the calculation of the search direction by projecting the gradient onto a tangent hyperplane. Our method features excellent convergence properties when applied to dirty paper precoding, and it may also be used for the optimization of linear precoders.