Combined ML and MMSE Multiuser Detection for STBC-OFDM Systems

Anh Tuan LE  Xuan Nam TRAN  Tadashi FUJINO  

IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E88-A    No.10    pp.2915-2925
Publication Date: 2005/10/01
Online ISSN: 
DOI: 10.1093/ietfec/e88-a.10.2915
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Information Theory and Its Applications)
Category: Communication Theory
STBC,  OFDM,  multiuser detector,  MMSE,  ML,  frequency selective fading,  

Full Text: PDF(935.3KB)>>
Buy this Article

Performance of the minimum mean square error (MMSE) detection is far below that of the maximum likelihood (ML) detection in a multiuser environment and decreases significantly as the number of co-channel users increases. In this paper, we propose a combined MMSE and ML multiuser detection scheme for space-time block coded (STBC) orthogonal frequency division multiplexing (STBC-OFDM) which has improved performance but with low complexity. In particular, we propose a reduced complexity ML post-detection (ML-PDP) scheme which can correct erroneously estimated bits from the outputs of MMSE multiuser detection. The proposed ML-PDP scheme performs sequential search to detect a predefined number of bits with higher probability of error and then uses ML detection to correct them. Upon controlling the number of corrected bits it is possible to balance the system performance with complexity associated with the ML-PDP. We show that significant improvement can be achieved at the cost of only small additional complexity compared with the MMSE multiuser detection.