Bayesian Monte Carlo multiuser receiver for space-time codedmulticarrier CDMA systems

摘要

We consider the design of optimal multiuser receivers fornspace-time block coded (STBC) multicarrier code-division multiple-accessn(MC-CDMA) systems in unknown frequency-selective fading channels. Underna Bayesian framework, the proposed multiuser receiver is based on thenGibbs sampler, a Markov chain Monte Carlo (MCMC) method for numericallyncomputing the marginal a posteriori probabilities of different users'ndata symbols. By exploiting the orthogonality property of the STBC andnthe multicarrier modulation, the computational complexity of thenreceiver is significantly reduced. Furthermore, being a soft-inputnsoft-output algorithm, the Bayesian Monte Carlo multiuser detector isncapable of exchanging the so-called extrinsic information with thenmaximum a posteriori (MAP) outer channel code decoders of all users, andnsuccessively improving the overall receiver performance. Severalnpractical issues, such as testing the convergence of the Gibbs samplernin fading channel applications, resolving the phase ambiguity as well asnthe antenna ambiguity, and adapting the proposed receiver to multiratenMC-CDMA systems, are also discussed. Finally, the performance of thenBayesian Monte Carlo multiuser receiver is demonstrated through computernsimulations