DocumentCode :
428192
Title :
Coherent Rake reception using noisy channel estimates [DS-CDMA cellular radio]
Author :
Bottomley, Gregory E. ; Cozzo, Carmela
Author_Institution :
Ericsson Res., Research Triangle Park, NC, USA
Volume :
3
fYear :
2004
fDate :
26-29 Sept. 2004
Firstpage :
1801
Abstract :
With traditional Rake reception, channel estimates are used to combine despread values. In this paper, a maximum-likelihood approach is used to reformulate the Rake combining weights, accounting for the channel estimation error. Specifically, Ling´s formulation is extended to include fading, noise, and channel estimation error correlation between fingers. The resulting combining weights are similar to the generalized Rake receiver, except that the known channel response is replaced by a minimum mean-square error (MMSE) estimate, and the noise covariance matrix is augmented by a term related to channel estimation. The performance of this and other combining approaches are compared, showing how performance depends on the available energy for channel estimation and the Rake finger placement strategy.
Keywords :
cellular radio; channel estimation; code division multiple access; covariance matrices; fading channels; least mean squares methods; maximum likelihood estimation; radio receivers; spread spectrum communication; DS-CDMA cellular radio; Ling formulation; MMSE estimated channel response; Rake combining weights; Rake finger placement strategy; available channel estimation energy; channel estimation error; coherent Rake reception; finger output noise; maximum-likelihood method; noise covariance matrix; noisy channel estimates; Channel estimation; Colored noise; Covariance matrix; Estimation error; Fading; Fingers; Maximum likelihood estimation; Multiaccess communication; Multipath channels; RAKE receivers;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Vehicular Technology Conference, 2004. VTC2004-Fall. 2004 IEEE 60th
ISSN :
1090-3038
Print_ISBN :
0-7803-8521-7
Type :
conf
DOI :
10.1109/VETECF.2004.1400346
Filename :
1400346
Link To Document :
بازگشت