A simple partitioned Markov chain model was proposed recently for characterization of the HF channel. The model is uniquely determined by the gap distribution. In this paper a method is derived to calculate the burst distribution from the model. The calculated result is compared with the burst distribution simulated from the actual error data collected from the HF communication link between Chicago and San Diego by IIT Research Institute for the Naval Electronics Laboratory Center. The model is also applied to a random channel and the result is compared with the simulation from the error sequence generated by a random number generator. In both cases there is a good agreement between the calculated and simulated results. It is illustrated that the model can be used for evaluating and designing the error detecting and correcting codes for the HF channel. The model is flexible as the number of states is not rigidly fixed hence it could be applied to a wide variety of HF channels. There is some indication that the model correlates to the multipath reduction factor (MRF). If verified it could add significance to the model since the maximum usable frequency (MUF) can be determined many months ahead by the available propagation prediction. Copyright © 1969 by The Institute of Electrical and Electronics Engineers, Inc.
