An analytic approach to modeling and estimation of OFDM channels

Modeling and estimation are investigated for orthogonal frequency division multiplexing (OFDM) channels in the frequency domain. The use of the discrete Fourier transform (DFT) enables the periodic extension of the channel impulse response (CIR) and channel frequency response (CFR), and allows the development of periodic random processes, and the related parametric dynamic systems, leading to interesting analytical results on modeling and estimation for OFDM systems. It is shown that low order parametric models are adequate for modeling OFDM channels in the frequency domain, and there exists an analytic relation between the power spectral density (PSD) of the CFR, and the power profile of the CIR. Based on the results of parametric modeling, an efficient design and implementation algorithm is developed for MMSE channel estimation. Simulation results show that the proposed simple channel estimation method is robust to channel statistics mismatch and is able to track time-varying dispersive fading channels.