The multimodulus blind equalization algorithm

Blind Equalizers for WDM/SCM-PON

“…The MMA error for square QAM constellations is defined as [6]: n,R= Yn,R (Yn,R M) nE,I Ynj (Ynj1RM) E:mma = Emma + j Emma where the subscripts R and I correspond to the real and imaginary components, and RM is a constant of the constellation. The modified-CMA (MCMA) error can be modified to weight the CMA and CME terms based on a data-depending weighting factor that trades off errors depending on the adaptation phase.…”

A configurable fractionally-spaced blind adaptive equalizer for QAM demodulators

“…The MMA [9,10] achieves channel equalization by penalizing the dispersion of the y 2 R (n) and y 2 R (n) components from the constant γ 2 M . The MMA error signal is defined as…”

New hybrid adaptive blind equalization algorithms for QAM signals

“…The minimization of the CMA criterion does not solve satisfactorily the equalization problem for QAM signals, since it leads to large residual errors and constellation is recovered up to a phase rotation. To overcome these drawbacks, other algorithms like the multimodulus algorithm (MMA) [11], the square contour algorithm (SCA) [12,13] and the extended CMA (ECMA) [14] have been proposed. The most interesting extensions of the CMA are those based on hybrid approaches [6][7][8], where it is proposed to augment the standard CMA cost function by a penalty term that takes into account the location of constellation symbols.…”