DocumentCode :
3045327
Title :
Outage probability approximation for the wiener filter SINR in MIMO systems
Author :
Kammoun, Abla ; Kharouf, Malika ; Hachem, Walid ; Najim, Jamal
Author_Institution :
ENST, Paris
fYear :
2008
fDate :
6-9 July 2008
Firstpage :
171
Lastpage :
175
Abstract :
This paper studies the fluctuations of the post-processing SNR at the output of the linear MMSE receiver in (receive) correlated multiple input multiple output (MIMO) systems. Although it is known that asymptotically, the SNR behaves like a Gaussian random variable, this approximation may yield to inaccurate estimates for small dimension. In order to circumvent this, we use gamma and generalized gamma distributions to approximate the probability distribution of the SINR. The first three asymptotic moments of the SNR are computed and are used to adjust gamma and generalized gamma distributions and to accurately approximate the bit error rate (BER) and its outage probability. We provide simulations which strongly support the gamma approximation, even for a small number of emitting/receive antennas.
Keywords :
Gaussian processes; MIMO communication; Wiener filters; approximation theory; error statistics; gamma distribution; least mean squares methods; random processes; BER; Gamma distribution; Gaussian random variable; MIMO system; Wiener filter; bit error rate; linear MMSE method; minimum mean square error receiver; multiple input multiple output system; probability approximation; receiving antenna; Bit error rate; Computational modeling; Distributed computing; Fluctuations; MIMO; Probability distribution; Random variables; Signal to noise ratio; Wiener filter; Yield estimation; Bit Error Rate; Large Random Matrices; MIMO systems; outage probability;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing Advances in Wireless Communications, 2008. SPAWC 2008. IEEE 9th Workshop on
Conference_Location :
Recife
Print_ISBN :
978-1-4244-2045-2
Electronic_ISBN :
978-1-4244-2046-9
Type :
conf
DOI :
10.1109/SPAWC.2008.4641592
Filename :
4641592
Link To Document :
بازگشت