Multiantenna Downlink Interference Management for Next Generation Mobile Networks

Abstract

In downlink multi-input single-output (MISO) networks, achieving optimal sum-rate with limited channel state information (CSI) is still a challenge even with a single user per cell. In this dissertation, three cooperative downlink multicell MISO beamforming schemes are proposed with highly limited information exchange among the base stations (BSs) to maximize the sum-rate. In the proposed schemes, each BS can design its beamforming vector with only local CSI based on limited information exchange on CSI. Unlike previous studies, the proposed beamforming designs are non-iterative and do not require any vector or matrix feedback but require only quantized scalar information.

In the first work, the beamforming vector at each BS is designed to minimize the sum of its weighted generating-interference (WGI) with local CSI and the aid of information exchange between the BSs. The generating-interference weight coefficients are designed in pursuit of increasing the sum-rate. Simulation results show that the proposed scheme outperforms the existing scheme in the mid to high signal-to-noise ratio (SNR) regime even with much reduced amount of information exchange via backhaul.

In the second work, the proposed beamforming design is based on the combination of the maximization of weighted signal-to-leakage-plus-noise ratio (WSLNR) and WGI. The weights in WSLNR and WGI are designed via choosing a proper set of users who shall be interference-free, which has never been endeavored in the literature. Though there have been extensive studies on downlink multicell beamforming, the proposed scheme closely achieves the optimal sum-rate bound in almost all SNR regime based on non-iterative optimization with lower amount of information exchange than existing schemes, which is justified by numerical simulations. In addition, the proposed scheme achieves a better trade-off between the amount of the information exchange and the sum-rate than existing schemes.

In the third work, a beamforming vector design based on a deep neural network (DNN) is proposed for multicell multi-input single-output channels with scalar information exchange and local CSI. The beamforming vectors are designed making zero generating-interference to the selected interference-free users (IFUs). The set of IFUs is chosen from the DNN based on supervised learning where the inputs can be obtained with only local CSI and limited scalar information exchange. Simulation results show that the DNN is well-trained in estimating the unknown CSI from the inputs with only local CSI in multicell networks.