Fast Propagation Predictions for HF Network Simulations

The complexity of current HF radio networks (such as the SCOPE Command HF Global Communications System) has renewed interest in techniques for efficiently simulating long-haul HF radio networks. Because the dynamic propagation effects of skywave paths are the distinguishing characteristic of HF network simulations, we need to accurately simulate the connectivity among HF stations. Predictions of ionospheric propagation are typically very costly in either time or memory or both. There is a clear need for accurate, efficient techniques to predict the path losses between pairs of HF stations. This paper describes a suite of techniques that has been developed and applied to simulations of large networks of fixed and mobile radios. These techniques approximate the accuracy of the best propagation prediction programs while minimizing both the memory and CPU time used during a simulation. The paper discusses the accuracy of the predictions along with the memory requirements and running time of the new propagation prediction techniques.