Tervo, OskariOskariTervoTran, Le-NamLe-NamTranJuntti, MarkkuMarkkuJuntti2021-05-202021-05-202015 IEEE2015-12-169781479975914http://hdl.handle.net/10197/12188The 3rd IEEE Conference on Signal & Information Processing, Orlando, Florida, United States of America, 14-16 December 2015We study energy-efficient decentralized coordinated beam-forming in multi-cell multiuser multiple-input single-output system. The problem of interest is to maximize the weighted sum energy efficiency subject to user-specific quality of service constraints. The original problem is iteratively approximated as a convex program according to successive convex approximation (SCA) principle. The convex problem at each iteration is then formulated as a general global consensus problem, which is solved via alternating direction method of multipliers (ADMM). This enables base stations to independently and in parallel optimize their beamformers relying only on local channel state information and limited backhaul information exchange. In addition to waiting for the ADMM to converge as conventionally when solving the approximate convex program, we propose a method where only one ADMM iteration is performed after each SCA update step. Numerical results illustrate the fast convergence of the proposed methods and show that performing only one ADMM iteration per each convex problem can significantly improve the convergence speed.en© 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.Resource allocationOFDMA systemsMIMOConference Publication10.1109/GlobalSIP.2015.74184262021-05-1713/RC/2077https://creativecommons.org/licenses/by-nc-nd/3.0/ie/