Robustness of fractionally-spaced equalization by CMA to lack of channel disparity and noise

We study the fractionally-spaced equalization by the constant modulus algorithm (FSE-CMA) robustness to channel noise and lack of disparity. When there is lack of disparity, we show that, whereas other techniques such as linear prediction or subspace like methods fail, the FSE-CMA can still equalize. In particular for a long enough equalizer the FSE-CMA exhibits a "smoothing effect" which leads to an interesting trade-off between achieving zero-forcing equalization and noise enhancement.