Two-Dimensional Interpolation-Assisted Channel Prediction for OFDM Systems

Channel prediction can provide up-to-date information about the state of a channel in OFDM systems at the expense of increasing the complexity of wireless systems. Therefore, this paper proposes a novel approach for optimizing the shape and intervals between samples in two-dimensions (2-D) for an orthogonal frequency-division multiplexing (OFDM) system to reduce the complexity of predicting the channel frequency response (CFR). Furthermore, the proposed prediction scheme has the flexibility to integrate with an adaptive algorithm which does not require any statistical prior knowledge and is able to track nonstationary channel statistics. The number of multipliers that are required to predict the CFR is reduced by predicting the CFRs at particular subcarriers and incorporating interpolation methods. This paper also deals with the proposal of a 2-D curve fitting (CF) interpolator to be used for predicting the CFR that allow improving the accuracy of the CFR prediction. Simulation and analysis results demonstrate that the 2-D CF interpolator combined with the proposed scheme has lower complexity than the other approaches to predicting CFR, without compromising the mean square error (MSE) or bit error ratio, even in a noisy and fast-fading channel environment.