A novel blind adaptive equalization technique
Xutao Yu, Z.-Y. Wu · International Conference on Acoustics, Speech, and Signal Processing · 2002
A preliminary study of a new adaptive equalization technique is addressed. The new technique consists of two cascade stages: an adaptive blind amplitude equalizer (ABAE) and a commonly used adaptive linear equalizer (ALE). The ABAE can compensate the amplitude distortion of the channel without the need for the values of the desired signals or the decision-directed symbols, so as to mitigate the burden of the subsequent decision-directed ALE. An extension of a modified Gauss-Newton algorithm, called the Levenberg-Marquardt algorithm, is established to solve the ABAE's coefficients online. Simulation results show that the new adaptive equalization technique applies to serious distortion situations where the common decision-directed adaptive equalizers do not work at all.>