A game-theoretic approach to transmitter covariance matrix design for broadband MIMO Gaussian interference channels

A game-theoretic approach to transmitter covariance matrix design for broadband MIMO Gaussian interference channels Anandkumar, A.J.G. Lambotharan, S. Chambers, J.A. Dept of Electron. & Electr. Eng., Loughborough Univ., Loughborough, UK This paper appears in: Statistical Signal Processing, 2009. SSP '09. IEEE/SP 15th Workshop on Publication Date: Aug. 31 2009-Sept. 3 2009 On page(s): 301 - 304 E-ISBN: 978-1-4244-2711-6 Location: Cardiff ISBN: 978-1-4244-2709-3 INSPEC Accession Number:10961923 Digital Object Identifier: 10.1109/SSP.2009.5278580 Current Version Published: 2009-10-06 Abstract A game-theoretic approach to the maximization of the information rates of broadband multi-input-multi-output (MIMO) Gaussian interference channels is proposed. The problem is cast as a strategic noncooperative game with the MIMO links as players and the information rates as payoff functions. The Nash equilibrium solution of this game is a waterfilling operation and sufficient conditions for its existence and uniqueness are presented. A distributed algorithm which requires no cooperation among the users is presented along with conditions for guaranteed global convergence of the proposed algorithm. The efficacy of the proposed scheme is confirmed through a design example.