Radial basis function aided space-time equalization in dispersive fading uplink environments

A novel radial basis function network (RBFN) assisted decision-feedback aided space-time equalizer (DF-STE) designed for receivers employing multiple antennas is proposed. The bit error rate (BER) performance of the RBFN aided DF-STE is evaluated when communicating over correlated Rayleigh fading channels, whose channel impulse response (CIR) is estimated using a Kalman filtering based channel estimator. The proposed receiver structure outperforms the linear minimum mean-squared error bench marker and it is less sensitive to both error propagation and channel estimation errors.