Blind Multiuser Detection in MC-CDMA: Schmidt-Orthogonalization and Subspace Tracking Kalman Filtering

Abstract

We consider multiuser detection (MUD) to combat multiple access interference (MAI), the detrimental impairment for a multi-carrier code division multiple access (MC-CDMA) system. The celebrated method of projection approximation subspace tracking deflation (PASTd) has been shown to achieve a good trade-off between performance and complexity. To further reduce the complexity and improve the performance especially in the presence of multi-path fading, we propose a new, adaptive, blind MUD algorithm by jointly exploiting the useful signal processing tools of Schmidt-orthogonalization and subspace Kalman-filtering. The improved algorithm, termed Schmidt-orthogonalized PASTd with subspace Kalman-filtering (SPASTd&K), not only performs well in single- and multi-path fading alike, but is also robust against parameter-mismatch, and particularly advantageous in the present of strong multiple access interference.