An Adaptive Equalizer Using A Stabilized Extended Kalman Filter

The adaptive equalizer described in this paper attempts to remove intersymbol interference in a digital communication channel. The channel is modeled as an n-tap FIR filter and the coefficients of the channel model are updated through the stochastic Newton Algorithm (SNA). This particular Algorithm is a stabilized version of the Extended Kalman Filter (EKF), overcoming the divergence problems associated with the EKF. However, as expected, the improvement is at the expense of tuning procedures, additional memory space and computation time. © 2006 IEEE