Channel estimation for OFDM systems using adaptive radial basis function networks

We propose a new scheme for pilot-symbol-aided channel estimation in orthogonal frequency-division multiplexing (OFDM) systems in multipath fading channels, that does not require knowledge of the channel statistics (e.g., Doppler or power spectrum). It is based on using the radial basis function (RBF) network to model the dynamics of the fading process. Both one-dimensional and two-dimensional RBF networks are proposed to exploit the channel correlation in the time domain and in the time-frequency domain. The proposed RBF networks are essentially nonlinear interpolators of the pilot channels. Compared with the existing OFDM channel estimation methods based on linear filtering, the proposed new techniques offer both robustness to fading rate, and a better performance especially in relatively fast fading channels.