Three new versions for the Newton type adaptive filtering algorithm

In this work, we present three new versions for the Newton type adaptive filtering algorithm FNTF. The first version is based on a temporal division of the basic transverse structure by using local filtering errors for actualization of the different sections of the transverse filter, whereas the second version differs from the first in the use of the final filtering error to update the coefficients of the transverse filter. The third version is based on mixing the second version with the introduction of a scalar parameter into the filtering part. These three versions were proposed to improve the convergence speed of the original Newton algorithm version for the identification of the acoustic impulse response and to improve, also, the tracking capability when these systems vary in time.