Nonrecursive adaptive decision-feedback equalization from channel estimates

The performance and complexity of channel estimate-based equalizer adaptation are considered. It is shown via simulation of terrestrial mobile radio channels that both initialization and tracking performance of a fractionally spaced decision feedback equalizer (DFE) are improved over conventional, recursive adaptation algorithms. A plausible explanation for the improvement is provided, and the computational requirements of this approach are compared with those of conventional adaptation methods. Those requirements can be minimized by taking advantage of the structure of the DFE coefficient equations. The complexity is compared with that of a conventional adaptive DFE