Beamforming Design Based on Virtual SINR Maximization for Interference Networks

In this paper, we propose beamforming techniques based on virtual signal-to-interference-plus-noise ratio (VSINR) maximization for weighted sum-rate (WSR) maximization in multiple-input single-output (MISO) interference channels. In the earlier work by Zakhour and Gesbert, it was shown that all Pareto-optimal beamformers can be expressed as a solution to the VSINR problem. However, how to choose the weight coefficients in the VSINR expression is not addressed when solving the WSR maximization problem. Thus, we provide a method of computing the weight terms to achieve a certain desired WSR maximizing point. Since the beamforming vectors in the proposed scheme should be computed as a function of global channel state information (CSI), we also propose a decentralized approach which shows a performance close to the centralized scheme with a significant reduction in the CSI exchange overhead.