DocumentCode
1263556
Title
Look-ahead sphere decoding: algorithm and VLSI architecture
Author
Gamba, Micaela Troglia ; Masera, Guido
Author_Institution
Dipt. di Elettron., Politec. di Torino, Torino, Italy
Volume
5
Issue
9
fYear
2011
fDate
6/1/2011 12:00:00 AM
Firstpage
1275
Lastpage
1285
Abstract
Multiple-input-multiple-output (MIMO) systems are recognised as a key enabling technology in high-performance wireless communications; however, the implementation of high-throughput MIMO detectors still is a critical task. Among known MIMO detectors, sphere decoder algorithm (SDA) is capable of optimal performance with acceptable processing complexity. The study presents an improved SDA, which enables significant throughput increase at a very limited additional complexity and with no degradation in terms of bit error rate performance. The modified detection method, called LASDA (look-ahead SDA), is based on formal algorithm transformations, namely look-ahead, retiming and pipelining, and requires a modified tree search strategy. The VLSI design of LASDA detector supporting a 4 × 4 MIMO channel with 16 QAM modulation is detailed in the study for a 130 nm CMOS standard cell technology: synthesis results show that the proposed solution achieves an average throughput of 380 Mbps at a signal-to-noise ratio of 22 dB (Rayleigh fading channel), with an occupied silicon area of 0.18 mm2. Comparisons with a number of previous implementations are also provided.
Keywords
CMOS integrated circuits; MIMO communication; VLSI; decoding; error statistics; quadrature amplitude modulation; search problems; trees (mathematics); wireless channels; CMOS standard cell technology; LASDA; MIMO channel; QAM modulation; VLSI design; bit error rate performance; high-performance wireless communications; high-throughput MIMO detectors; look-ahead SDA; look-ahead sphere decoding; modified detection method; modified tree search strategy; multiple-input-multiple-output systems; signal-to-noise ratio; size 130 nm;
fLanguage
English
Journal_Title
Communications, IET
Publisher
iet
ISSN
1751-8628
Type
jour
DOI
10.1049/iet-com.2010.0523
Filename
5936942
Link To Document