Analytic framework for the mutual information cumulants of different MIMO channels
Paper i proceeding, 2010

In this paper, we present a general analytical framework for the exact mutual information (MI) cumulants of multiple-input multiple-output (MIMO) systems with perfect receiver channel state information (CSI) and no transmitter CSI. Our derivation is based on a recent parameterization of the joint ordered eigenvalue probability density function (PDF) [1], that encompasses both uncorrelated/semi-correlated Rayleigh channels as well as uncorrelated Rician channels. In addition, we extend our framework to account for the cumulants of doubly-correlated Rayleigh channels and also to deduce tractable expressions in the high Signal-to-Noise ratio (SNR) regime. The cumulants are particularly useful to study all high-order statistics (HOS) of the MI; in fact, they can be used to express the MI mean and variance as a finite sum of determinants. Our analytical expressions are then validated via Monte-Carlo simulations with the attained accuracy being excellent in all cases.


Paul de Kerret

Rheinisch-Westfalische Technische Hochschule Aachen

Michail Matthaiou

Signaler och system, Signalbehandling och medicinsk teknik, Signalbehandling

Rudolf Mathar

Rheinisch-Westfalische Technische Hochschule Aachen

Josef A. Nossek

Technische Universitat Munchen

53rd IEEE Global Communications Conference, GLOBECOM 2010; Miami, FL; United States; 6 December 2010 through 10 December 2010

Art. no. 5683814-


Informations- och kommunikationsteknik


Elektroteknik och elektronik