Efficient block implementation of the LMS-based DFE

An efficient block adaptive implementation of the widely used decision feedback equalizer is developed. Both the feedforward and the feedback filters of the equalizer are updated once every K sample time intervals, with K being the block length. However this block adaptation is done in such a way that the resulting filters, as well as the decisions, are identical to those computed by the conventional sample by sample LMS based DFE (LMS-DFE). That is, the new block adaptive algorithm is mathematically equivalent to the LMS-DFE algorithm (and hence it features identical performance). At the same time the proposed algorithm offers substantial computational savings as compared to the sample by sample LMS-DFE. Thus the new block DFE turns out to be particularly suitable for applications in which long equalizers are required