I’ve always been fascinated by radio propagation. For me, there’s something magical about the way signals can sometimes travel halfway around the world, while at other times they barely make it out of the county. Predicting propagation has never been an exact science, but that’s part of the challenge and the fun.
The idea behind this project – the UK HF Predictor – was simple enough: to create a tool that would help me understand propagation better, while at the same time offering something that other UK based amateurs might find useful. I wanted to experiment with different prediction techniques, compare them against real-world data, and bring everything together into a format that was easy to use.
Prediction Meets Observation
To do this, I built a web-based dashboard that combines two very different kinds of information: prediction models and observed activity. On the one hand, the dashboard pulls in real-time indices such as the Solar Flux Index (SFI), the Kp index, estimates of geomagnetic storm activity, and ionosonde-based measurements of the Maximum and Lowest Usable Frequencies (MUF and LUF). These give a sense of what the physics suggests should be possible.
But models on their own are never enough. What really matters is what’s happening on the air right now. That’s where the Reverse Beacon Network (RBN) comes in. By filtering to look only at UK stations being reported around the world, the predictor can show how many spots are being received, how many countries are hearing us, and at what average distances. It means the dashboard reflects actual openings from the UK, rather than just theoretical ones.
To add another layer of context, I’ve also included near real-time data from WSPR.live. This isn’t UK-specific, so it isn’t used in the main scoring, but it provides a global “sense check” of how bands are behaving. By comparing WSPR activity levels with the blended score on each band, you get an extra perspective on whether conditions are lining up or diverging.
A Simple but Flexible Dashboard
All of this information is presented in a web dashboard designed to be clear and adaptable. You can choose between a Summary view for a quick snapshot, a Detail view with full explanatory notes, or a Matrix view that lays everything out in a compact band-by-distance table. The idea is to let users decide how much or how little information they want at any given time.
The site is responsive, so whether you’re on a desktop, tablet, or mobile phone, it should render neatly without any special effort. And if you want to know more about what the numbers mean, the About page goes into detail on the data sources, scoring rules, and thresholds that sit behind the dashboard.
Built with AI
Although the overall design and ideas are mine, one of the most interesting aspects of this project has been the way it was built. Every line of the code behind the predictor has been created with the assistance of artificial intelligence. I’ve been using AI to translate ideas into working code, to refine layouts, and to explore approaches that would have taken me much longer to write by hand. For me, this has been as much an experiment in modern development tools as it has been an experiment in radio propagation.
A Work in Progress
I want to stress again that this is not an official forecasting service. It’s a personal, experimental project, and it remains very much a work in progress. Features may change, scoring rules may be adjusted, and at times the page may be unavailable while I make changes or try out new approaches. That’s part of the process.
My aim is simply to increase my own understanding of propagation while creating something that might be useful for others. If you find it helpful, I’d love to hear from you. If you think I’ve got something wrong, I’d love to hear that too. Constructive feedback is always welcome, and you can reach me at g4eid at hotmail dot com
In Closing
The UK HF Predictor is about curiosity, learning, and experimentation. It brings together real-time data and practical observations, and it does so with a UK focus that I felt was often missing from other online tools. I don’t pretend it’s perfect – it isn’t – but I do hope it adds something useful to your own understanding of the HF bands.
You can try it for yourself at hf-predictor.g4eid-km8h.net.