Low-Complexity Algorithm for Near-Optimum Detection of V-BLAST Systems

For the Vertical Bell Laboratories Layered Space-Time systems, the optimum a posteriori probability detector encounters the problem of calculating Gaussian mixtures with large numbers of components. In this letter, a new detector is proposed, which consists of an minimum-mean-square-error generalized decision feedback equalization-based dominant symbol combinations selection algorithm for mixture components reduction, and a Cholesky factorization-based efficient algorithm for likelihood function calculation. The new detector can obtain near optimum performance with a complexity much lower than that of a previous sequential algorithm with Gaussian approximation.