Sum Power Minimization for TDD-Enabled Full-Duplex Bi-Directional MIMO Systems Under Channel Uncertainty

In this paper we address a sum power minimization problem for a bi-directional and full-duplex (FD) communication system, where the required rate constraints are imposed on the guaranteed communication rates in each direction. In this regard, the impact of channel-state information (CSI) error, as well as the signal distortion due to hardware impairments are jointly taken into account. In order to ensure backwards compatibility to an equivalent half-duplex setup, we assume a time-division-duplex capable system where the FD communication process takes place in multiple independent time segments. Due to the intractable structure of the resulting optimization problem, a weighted minimum mean squared-error based method is applied to cast the power minimization problem into a separately convex structure, which can be iteratively solved with a guaranteed convergence. The resulting computational complexity of the algorithm is then discussed analytically. Finally, the performance of the proposed algorithm is numerically evaluated over different levels of rate demand, CSI error and transmitter/receiver dynamic range.