An improved 2-dimensional pilot-symbols assisted channel estimation in OFDM systems

Abstract

In orthogonal frequency division multiplexing (OFDM) systems, the pilot-symbols (PS) are utilized to provide an accurate estimation of the channel response. Based on the channel response of PS, 2-dimensional Wiener filtering can provide the optimal channel estimation. However, for the shortage of PS the bit error rate (BER) performance of the side subcarriers obviously worse than that of the other subcarriers. In this paper the periodical character of the discrete time-frequency channel response in frequency domain is observed and the new frequency autocorrelation function (acf) with the periodical property is derived. Then it is applied to obtain the new coefficients of the Wiener filtering. Through the simulation it is verified that the 2-dimensional Wiener filtering based on the periodical frequency acf can efficiently enhance the accuracy of the channel estimation at side subcarriers and it is more accurate optimal PS assisted channel estimation scheme in discrete OFDM systems.