Energy Efficient Coordinated Beamforming for Multi-cell MISO Systems

TL;DR

This paper develops a method to optimize energy efficiency in multi-cell MISO systems by characterizing the Pareto boundary of EE and proposing a decentralized beamforming algorithm, improving energy use in wireless networks.

Contribution

It introduces a novel approach to characterize the energy efficiency boundary using fractional concave programming and relates it to cognitive radio channels for decentralized implementation.

Findings

Achieves the EE Pareto boundary in simulations

Transforms the problem into fractional concave programming

Provides a decentralized beamforming algorithm

Abstract

In this paper, we investigate the optimal energy efficient coordinated beamforming in multi-cell multiple-input single-output (MISO) systems with $K$ multiple-antenna base stations (BS) and $K$ single-antenna mobile stations (MS), where each BS sends information to its own intended MS with cooperatively designed transmit beamforming. We assume single user detection at the MS by treating the interference as noise. By taking into account a realistic power model at the BS, we characterize the Pareto boundary of the achievable energy efficiency (EE) region of the $K$ links, where the EE of each link is defined as the achievable data rate at the MS divided by the total power consumption at the BS. Since the EE of each link is non-cancave (which is a non-concave function over an affine function), characterizing this boundary is difficult. To meet this challenge, we relate this multi-cell MISO system to cognitive radio (CR) MISO channels by applying the concept of interference temperature (IT), and accordingly transform the EE boundary characterization problem into a set of fractional concave programming problems. Then, we apply the fractional concave programming technique to solve these fractional concave problems, and correspondingly give a parametrization for the EE boundary in terms of IT levels. Based on this characterization, we further present a decentralized algorithm to implement the multi-cell coordinated beamforming, which is shown by simulations to achieve the EE Pareto boundary.