Prediction of Rayleigh fading channels based on hidden Markov modeling of sequential channel decoding complexity

In fading channels that exhibit memory, errors tend to occur in blocks. Knowledge of the channel condition of the previous block can be used to predict the future channel quality and improve the performance of the channel decoding system. Sequential decoding algorithms are known to have the advantage of allowing for variable decoding complexity with changing channel conditions. On the other hand, the changing complexity is also an indicator of channel conditions. We employ the complexity of Fano (1963) sequential decoders to model the Rayleigh fading channels. Based on hidden Markov models, We propose a fast sliding window prediction approach. We empirically determine the relations between the prediction performance and the number of distinctive symbols in the model.