Minimum bit error rate multiuser detection for multiple antenna aided uplink OFDM

Abstract

In this treatise, we employed the so-called Space Division Multiple Access (SDMA) principle in the context of Orthogonal Frequency Division Multiplexing (OFDM) for multiuser transmission by exploiting the system's capability of differentiating multiple users by their unique, user-specific spatial signatures constituted by their different impulse responses between the different transmit and receiver antennas. The classic method of detecting the different users is by using the well-known Minimum Mean Squared Error (MMSE) linear multiuser detector (MUD). However, the MMSE MUD family does not necessarily minimise the achievable Bit Error Rate (BER) of a system, although achieving the minimum BER is our ultimate aim. Therefore, we propose a novel Minimum Bit Error Rate (MBER) type MUD that is capable of directly minimising the BER cost function of a given system. We have demonstrated that the MBER MUD performs better than the MMSE MUD in the scenarios investigated. In the case of the MMSE MUD, different users will achieve a different BER performance and as the amount of multiuser interference increases, the MMSE MUD performs more poorly in distinguishing the different users. This is due to the fact that the performance of the MMSE MUD relies on the system matrix constituted by the different users' channel transfer functions in determining the detected user's BER performance. By contrast, all users employing the MBER MUD experience a fairly similar BER. Moreover, we showed that the MBER MUD is capable of supporting more users than the number of receiver antennas, because it will aim for finding the minimum point on the BER cost function, regardless of the system matrix. We have also employed the Genetic Algorithm (GA) search method in solving for the weight values of the MBER MUD, as an alternative to the normal conjugate gradient search method. In the case where the channel is unknown, adaptive MBER method can be used to search for the weight values of the MBER MUD.

More information

Identifiers

Catalogue record

Export record

ASCII CitationAtomBibTeXData Cite XMLDublin CoreDublin CoreEP3 XMLEndNoteHTML CitationHTML CitationHTML ListJSONMETSMODSMPEG-21 DIDLOpenURL ContextObjectOpenURL ContextObject in SpanRDF+N-TriplesRDF+N3RDF+XMLRIOXX2 XMLReferReference ManagerSimple Metadata