Performance of RBF equalizer in data storage channels

In order to compensate for severe intersymbol interference (ISI) and combat nonlinear distortions in digital magnetic recording systems, Bayesian equalizer using the radial basis function (RBF) network, REQ, is applied as a channel equalizer to a data storage system using partial erasure (PE) model. It can be seen from various computer simulation results that the REQ has more complex structure than conventional linear equalizer (LE) while the bit-error-ratio (BER) of the REQ is lower than that of the LE. Using various simulation parameters such as the width of the basis function, the updating algorithm for the centers in REQ, the recording density of a data storage system, D_u, and so froth, the performance in terms of BER of the REQ is examined.