Title
ET-FSD: A feasible scheme of MIMO detection to exploit scalar effects
Abstract
In this paper, the early-termination fixed-complexity sphere detector (ET-FSD) that is a channel-adaptive version of FSD, is developed to exploit scale effects for the multiple-antenna system which has to detect signals of multiple users under the constraint of sum complexity (e.g., the base-station systems always encounter the run-time limit of signal detection of all the users). In order to apply the concept of scale effects in microeconomics to the study of ET-FSD, the corresponding mathematic model primarily based on the large deviation principle is established. The key evidence that tells the existence of scale effects in ET-FSD is presented instead of the complete proof due to the space limitation. We also argue that within multiple-user systems the scale effects can be utilized to realize ET-FSD that has both the optimal performance in the point view of diversity order and the polynomial time unit complexity in the worst-case sense (one can think of unit complexity as the complexity per user in multiple-user scenario). Finally, we provide the numerical results to support the theoretical analysis.
Year
DOI
Venue
2012
10.1109/PIMRC.2012.6362725
PIMRC
Keywords
Field
DocType
multiple- antenna system,early-termination fixed-complexity sphere detector,multiple-input multiple-output (mimo),numerical analysis,large deviation principle,mathematical analysis,fixed-complexity sphere decoder (fsd),microeconomics,sum complexity constraint,antenna arrays,diversity order,wireless channels,channel-adaptive version,polynomial time unit complexity,mimo communication,complexity,multiple-user system,base-station system,scalar effect exploitation,et-fsd,mathematic model,signal detection,polynomials,mimo detection,mimo,detectors,signal to noise ratio,decoding,tin
Mathematical optimization,Detection theory,Polynomial,Computer science,Scalar (physics),Signal-to-noise ratio,MIMO,Decoding methods,Time complexity,Rate function
Conference
Volume
Issue
ISSN
null
null
2166-9570 E-ISBN : 978-1-4673-2568-4
ISBN
Citations 
PageRank 
978-1-4673-2568-4
1
0.37
References 
Authors
12
4
Name
Order
Citations
PageRank
Rongrong Qian14913.30
Yuan Qi22415.41
Tao Peng339146.51
Wenbo Wang410.37