# Actual measurements of the influence of fading and multipath on HF radio communication

I'm looking for papers or books that contain actual measurements of the influence one has to expect in HF radio communication. I'm interested in this, because i've simulated the influence of a physical channel using the Watterson Model on the detection performance of a feature-based approach. The approach is not robust for the parameters I've picked. Now I'm interested in how realistic the Watterson parametrzation I've picked is. For you information and the sake of completeness, the parameters I picked:

The information I'm gathering serves the purpose of anwering the question:

Is it just the average case of distortion that is to be expected in HF communication or is it a more challening scenario.

EDIT:

PSK,QAM and FSK signals were tested. The bandwidth was between 100 and 1000 Hz.The parametrization was chosen for the frequency range of 3kHZ to 30 MHz.

• The impact of physical environment on communication signal does not depend only on the environment itself, but the relation between the environment and they way the signal is modulated. You need to provide the modulation scheme, especially the bandwidth of modulating symbol and the carrier frequency for an accurate answer. Based on your table, the delay 6ms is quite large and looks like echoes in rural area. I cannot say a thing by Doppler shift without information of frequency. – AlexTP Sep 20 '17 at 14:44
• There is a lot of literature on this in the ham radio community. You can find a lot of scientific papers if you google for "HF channel model" or "ionospheric propagation/reflection" and permutations. – Attila Kinali Sep 20 '17 at 19:21
• As KotH pointed out, there's a wealth of literature – there's a lot of cold-war era ITU recommendations on long-range HF propagation; they are actually listed on the ITU website if you are persistent enough when using their search engine, but are largely unretrievable, for some reason. – Marcus Müller Sep 20 '17 at 21:34
• Hello Marcus, I picked the model according to an ITU recommendation. Is there some research papers I'm missing related to that recommendations. I've read somewhere, that even the most basic spatial-wave transmission spreads to 4 paths. Some qualitative notions like this would be helpful. But, of course I'm aware that the measurement of channels is very specific depeding on too many factors to easily generalize the results. Thanks, a lot anyway! – Marcel Sep 21 '17 at 7:44

Matlab have in his comm toolbox some scripts dedicated to HF ionospheric channel simulation, check it out

ITU-R only recommend two independently fading paths with equal mean attenuation, equal frequency spreads and nonfrequency shifts.

Suggested parameter values for general testing:

Good conditions Differential time delay: 0.5 ms Frequency spread: 0.1 Hz

Moderate conditions Differential time delay: 1 ms Frequency spread: 0.5 Hz

Poor conditions Differential time delay: 2 ms Frequency spread: 1 Hz

https://www.itu.int/dms_pubrec/itu-r/rec/f/R-REC-F.520-2-199203-W!!PDF-E.pdf