Title :
Asymptotic normality of linear CDMA multiuser detect
Author :
Guo, Dongning ; Verdú, Sergio ; Rasmussen, Lars K.
Author_Institution :
Dept. of Electr. Eng., Princeton Univ., NJ, USA
Abstract :
This paper studies the asymptotic distribution of the output of a family of linear multiuser receivers. Every receiver in the family is a function of the channel correlation matrix of a simple form. It is proved that conditioned on almost all spreading codes, the output decision statistic of such a receiver converges to a Gaussian law in distribution as the system size goes to infinity. This set of receivers encompasses a matched filter, MMSE receiver, parallel interference cancelers, and many other linear receivers of interest. The normality principle solidifies the importance of the signal-to-interference ratio in the many-user analysis of linear multiuser detectors
Keywords :
Gaussian distribution; code division multiple access; interference suppression; least mean squares methods; matched filters; multiuser channels; receivers; signal detection; spread spectrum communication; Gaussian law; MMSE receiver; asymptotic distribution; asymptotic normality; channel correlation matrix; linear CDMA multiuser detect; linear multiuser detectors; linear multiuser receivers; linear receivers; many-user analysis; matched filter; normality principle; output decision statistic; parallel interference cancelers; signal-to-interference ratio; spreading codes; system size; H infinity control; Interference cancellation; Matched filters; Matrix decomposition; Multiaccess communication; Multiple access interference; Multiuser detection; Signal analysis; Statistical distributions; Vectors;
Conference_Titel :
Information Theory, 2001. Proceedings. 2001 IEEE International Symposium on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-7123-2
DOI :
10.1109/ISIT.2001.936170
