DocumentCode :
2020448
Title :
A sub-optimal soft decision PDA method for binary quadratic programming
Author :
Luo, J. ; Pattipati, K.R. ; Willett, P.K.
Author_Institution :
ECE Dept., Connecticut Univ., Storrs, CT, USA
Volume :
5
fYear :
2001
fDate :
2001
Firstpage :
3140
Abstract :
Binary quadratic programming (BQP) problems arise frequently in digital communication systems where online solutions are required. The multiuser detection (MUD) problem in code division multiple access (CDMA) communications, studied in the paper, is one such example. Due to the NP-hard nature of the BQP problem arising in MUD, only sub-optimal methods with polynomial complexities can be realistically considered. In the paper, a suboptimal algorithm based on the idea of probabilistic data association (PDA) is proposed. By treating the detection parameters as binary random variables, and by approximating the multi-modal Gaussian mixture by a single Gaussian noise, the PDA method provides near-optimal solution with a computational complexity of O(N3), where N is the problem size. Several other algorithms for the MUD problem are also considered and compared in terms of computational efficiency and the degree of suboptimality
Keywords :
code division multiple access; computational complexity; digital simulation; multiuser channels; probability; quadratic programming; NP-hard problem; binary quadratic programming; binary random variables; code division multiple access; computational complexity; digital communication systems; multi-modal Gaussian mixture; multiuser detection problem; polynomial complexities; single Gaussian noise; sub-optimal soft decision probabilistic data association method; Computational complexity; Computational efficiency; Digital communication; Gaussian noise; Multiaccess communication; Multiuser detection; Personal digital assistants; Polynomials; Quadratic programming; Random variables;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Systems, Man, and Cybernetics, 2001 IEEE International Conference on
Conference_Location :
Tucson, AZ
ISSN :
1062-922X
Print_ISBN :
0-7803-7087-2
Type :
conf
DOI :
10.1109/ICSMC.2001.972001
Filename :
972001
Link To Document :
بازگشت