DocumentCode :
3047694
Title :
Mutual information distribution of spatially correlated MIMO-OFDM systems in the low signal-to-noise ratio regime
Author :
Letzepis, Nick ; Pollok, André ; Cowley, William
Author_Institution :
Inst. for Telecommun. Res., Univ. of South Australia, Mawson Lakes, SA
fYear :
2009
fDate :
4-7 Feb. 2009
Firstpage :
83
Lastpage :
89
Abstract :
We characterise the distribution of mutual information for spatially correlated multiple-input multiple-output channels using orthogonal frequency division multiplexing (MIMO-OFDM) in the asymptotically low signal-to-noise (SNR) regime. In particular, we derive the moment generating function (MGF) and cumulative distribution function (CDF) of the mutual information random variable as a function of the power delay profile and eigenvalues of the spatial correlation matrices. For the special case of zero spatial correlation and uniform power delay profile, we show that the mutual information is gamma distributed. In the general case, it can be expressed as a sum of weighted exponential distributions, or an infinite sum of weighted gamma distributions. The weights and the parameters of the component distributions are expressed in terms of the power delay profile and eigenvalues of the correlation matrices.
Keywords :
MIMO communication; OFDM modulation; eigenvalues and eigenfunctions; cumulative distribution function; eigenvalues; moment generating function; multiple-input multiple-output channels; mutual information distribution; orthogonal frequency division multiplexing systems; power delay profile; signal-to-noise ratio regime; spatial correlation matrices; spatially correlated MIMO-OFDM systems; weighted exponential distributions; weighted gamma distributions; Delay; Distribution functions; Eigenvalues and eigenfunctions; Exponential distribution; MIMO; Mutual information; OFDM; Power generation; Random variables; Signal to noise ratio;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Communications Theory Workshop, 2009. AusCTW 2009. Australian
Conference_Location :
Sydney, NSW
Print_ISBN :
978-1-4244-3356-8
Electronic_ISBN :
978-1-4244-3357-5
Type :
conf
DOI :
10.1109/AUSCTW.2009.4805605
Filename :
4805605
Link To Document :
بازگشت