A radial basis function (RBF?) equalizer design process has been developed
in which the number of basis function centers used is substantially fewer t
han conventional required. The reduction of centers is accomplished in two-
steps. First an algorithm is used to select a reduced set of centers that l
ie close to the decision boundary. Then the centers in this reduced set are
grouped, and an average position is chosen to represent each group. Channe
l order and delay, which are determining factors in setting the initial num
ber of centers, are estimated from regression analysis. In simulation studi
es, an RBF equalizer with more than 2000-to-1 reduction in renters performe
d as well as the RBF equalizer without reduction in centers, and better tha
n a conventional linear equalizer.