DocumentCode
1685079
Title
Enhanced Soft Interference Cancellation Algorithm for V-BLAST Systems
Author
Luo, Zhendong ; Yang, Fan
Author_Institution
Inst. of Commun. Stand. Res., China Acad. of Telecommun. Res. of MII, Beijing
fYear
2008
Firstpage
1
Lastpage
5
Abstract
In this paper, we investigate soft interference cancellation algorithms for vertical Bell Laboratories layered space- time (V-BLAST) systems. By using Gaussian-forcing technique and fast matrix computation, we first derive a simplified version of the conventional generalized probabilistic data association (GPDA) algorithm, called the fast soft interference cancellation (FSIC) algorithm, whose complexity is only cubic in the number of transmit/receive antennas. Then, based on the FSIC algorithm, we develop an enhanced soft interference cancellation (ESIC) algorithm by making use of the properties of MIMO channels. It is shown that with efficient iterative processing, the ESIC algorithm can achieve near-optimal performance in both uncoded systems and coded systems.
Keywords
Gaussian processes; MIMO communication; interference suppression; iterative methods; matrix algebra; receiving antennas; transmitting antennas; Gaussian-forcing technique; MIMO channels; V-BLAST systems; enhanced soft interference cancellation algorithm; fast matrix computation; fast soft interference cancellation algorithm; generalized probabilistic data association algorithm; iterative processing; soft interference cancellation algorithms; transmit/receive antennas; uncoded systems; vertical Bell Laboratories layered space- time systems; Detection algorithms; Detectors; Gaussian processes; Interference cancellation; Iterative algorithms; Iterative decoding; MIMO; Receiving antennas; Silicon carbide; Transmitting antennas;
fLanguage
English
Publisher
ieee
Conference_Titel
Global Telecommunications Conference, 2008. IEEE GLOBECOM 2008. IEEE
Conference_Location
New Orleans, LO
ISSN
1930-529X
Print_ISBN
978-1-4244-2324-8
Type
conf
DOI
10.1109/GLOCOM.2008.ECP.786
Filename
4698561
Link To Document