Matrix-Lifting Semi-Definite Programming for Decoding in Multiple Antenna Systems

Recently, several quasi-maximum likelihood decoding methods have been introduced to solve the decoding problem in multiple antenna systems. Mobasher et al. [1] proposed a general method with a near optimal performance for M-ary QAM or PSK constellation. However, it is more complex compared to some o...

Full description

Saved in:
Bibliographic Details
Main Authors: Mobasher, A., Khandani, A.K.
Format: Conference Proceeding
Language:English
Subjects:
Computational complexity
Decoding
Degradation
Labeling
MIMO
Phase shift keying
Polynomials
Quadrature amplitude modulation
Symmetric matrices
Tin
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recently, several quasi-maximum likelihood decoding methods have been introduced to solve the decoding problem in multiple antenna systems. Mobasher et al. [1] proposed a general method with a near optimal performance for M-ary QAM or PSK constellation. However, it is more complex compared to some other methods specialized for a limited scenario. In this paper, we introduce a new general algorithm based on matrix-lifting Semi-Definite Programming (SDP). The new relaxation exploits the matrix structure of the system and introduces a degradation in the performance; however, the reduction in the complexity is significant. The number of variables is decreased from O(N 2 K 2 ) to O((N + K) 2 ). Moreover, this method can be implemented for any constellation and labeling method.
DOI:10.1109/CWIT.2007.375719