Energy-efficient transmission for MIMO interference channels

In previous work on multiple-input multiple-output (MIMO) interference channels, it has usually been assumed that the transmit power at each transmitter is fixed. Power is equally allocated among different streams, and MIMO beamforming is applied only to mitigate interference and improve the system performance. In this paper, we design the transmit powers, power allocation among streams, and beamforming matrices jointly for each transmit-receive pair (link) to maximize the energy efficiency in a MIMO interference channel. Centralized and decentralized energy-efficient beamforming algorithms are developed based on global and local channel state information (CSI) at each transmitter, respectively. A distributed beamforming algorithm that combines minimum mean squared error (MMSE) and two power allocation algorithms is also developed; this algorithm only requires the information of the desired link. The decentralized and distributed schemes can be combined with scheduling to achieve good performance when the interference among links cannot be canceled. Simulation results show that the proposed algorithms can achieve good performance close to the upper bound, and the decentralized algorithm can perform as well as the centralized scheme. The distributed algorithm is suboptimal, but requires much less signaling. In addition, we show that the decentralized and distributed schemes result in a fairer allocation than the centralized approach.