A suboptimum linear receiver derived from SVD based channel estimates

The reduced-rank least-squares (RRLS) method, which is based on fractional T-spaced sampling of the channel output and singular value decomposition (SVD) of a data matrix is shown to have certain advantages over conventional least-squares (LS) methods for data-adaptive equalization. The performance of a suboptimum, linear, minimum mean-squared error (MMSE) data communication receiver is investigated when the channel estimates from the RRLS and LS channel identification algorithms are used in the design. When the channel estimates are obtained in severe noise environments, the receiver that is conditioned on the RRLS channel estimates shows significant improvements in error performance when compared to the design based on the LS channel estimates