Title :
On the Ergodic Capacity of the Wideband MIMO Channel
Author :
Chatzinotas, Symeon ; Imran, Muhammad Ali ; Hoshyar, Reza
Author_Institution :
Centre for Commun. Syst. Res., Univ. of Surrey, Guildford
Abstract :
In the ergodic capacity literature, the majority of results preserve the assumption of a flat-fading Gaussian channel. However, the actual wireless channels in current communication systems are often wideband and therefore time dispersive and frequency selective. In this direction, we study the ergodic capacity of a wideband MIMO channel. Starting from a tapped delay line model for the time domain, the frequency domain model is derived by employing the Fourier transform. However, in both cases the resulting channel matrix is characterised by non-separable correlation amongst the Gaussian blocks. The asymptotic eigenvalue probability density function and the channel capacity are calculated by specializing a theorem originating in operator-valued free probability. Finally, numerical results are presented to verify the validity of the approach and study the effect of the channel model parameters.
Keywords :
Fourier transforms; Gaussian channels; MIMO communication; broadband networks; channel capacity; correlation methods; dispersive channels; eigenvalues and eigenfunctions; fading channels; frequency selective surfaces; frequency-domain analysis; probability; time-domain analysis; Fourier transform; Gaussian blocks; asymptotic eigenvalue probability density function; channel matrix; channel model parameters; ergodic channel capacity; flat-fading Gaussian channel; frequency domain model; frequency selective channel; nonseparable correlation; operator-valued free probability; tapped delay line model; time dispersive channel; time domain model; wideband MIMO channel; wireless channels; Communication systems; Delay lines; Dispersion; Eigenvalues and eigenfunctions; Fourier transforms; Frequency domain analysis; Gaussian channels; MIMO; Probability density function; Wideband;
Conference_Titel :
Vehicular Technology Conference, 2009. VTC Spring 2009. IEEE 69th
Conference_Location :
Barcelona
Print_ISBN :
978-1-4244-2517-4
Electronic_ISBN :
1550-2252
DOI :
10.1109/VETECS.2009.5073731