Exploiting hopping pilots for parametric channel estimation in OFDM systems

Abstract

In this letter, we investigate the effect of hopping pilots on the parametric channel estimation in orthogonal frequency-division multiplexing (OFDM) systems. Channel estimation algorithms based on parametric channel modeling require the multipath delays to be estimated at the receiver. The use of a fixed pilot pattern leads to longer training overhead in multipath delay estimation for slow fading channels. However, if hopping pilot patterns are available, then we show that the normalized mean-squared error (NMSE) convergence rate of the channel estimates can be significantly improved. We also show that hopping pilot patterns in OFDM systems effectively allow the eigenvectors of the delay subspace (of the autocorrelation matrix) to be estimated faster. Simulation results are provided to show the faster convergence rate of the NMSE for the hopping pilot pattern over the fixed pilot pattern.