Propagation Day by Day: CRI Kashi, 15.260 kHz

Signal strengths of CRI/Kashi, day by day, from 08:58 UTC to 09:58 UTC on the nine consecutive days March 15 to March 23, 2020; see text.

Propagation on HF differs from day to day. The nine diagrams at the top show the signal strengths of China Radio International’s Kashi transmitter, 500 kW, beaming to Romania; 08:58 UTC to 09:58 UTC from March 15 to March 23. The basic resolution (black grey points in the background) is 100 milliseconds, whereas the blue line marks the moving average with 601 points. The “moving average” can be best understood as a lowpass filter, revealing possible trends on a coarser scale. In this case, you cannot see such a trend.

If you compare a part of each transmission on a much finer scale, you even see sheer chaos, as the Figure below is showing:

All nine signal levels drawn together into one diagram (top), and a small part of it zoomed (bottom) reveals sheer chaos.

There seems to be no visible correlation on any scale in this case. There are other cases where, however, some correlation can be found – to which I will come back in some future entries.

The last diagram at the bottom of this pages shows a much more forgiving picture of the signal: the average level changes not more than ±4 dB between best and worst days. This so-called box diagram illustrates best the actual receiving quality of the broadcast, demodulated with an synchronous detector to largely avoid severe distortion by selective fading. The difference of deciles 90% and 10% marks the fading range, a key figure in describing the quality of reception – see “Ionospheric Radio” by Kenneth Davies [London, 1990/96, pp. 232].

The box plot shows very similar signal strengths, day by day. You should concentrate on their each center of gravity. You will also see that the distribution of the signals strengths relative to the center is not symmetrically, with a clearly visible advantage to the percentalge below the average strengths. THis will be covered in some future entry.

Analyzing signal strenghts, is an interesting tool to get to know more about propagation. I will continue this topic – stay tuned!

Receiver: Winradio’s Excalibur Sigma
Antenna: active vertical dipole (2 x 5 m) MD300DX
Software: V3 by Simon Brown, G4ELI, QtiPlot


  • Nice! And how about an SDRC S-meter option to add mean and σ (for a roughly 70% band, or nσ, configurable), sampled at a fairly coarse rate over a running 1-minute (configurable) period?

  • Good idea – but: if you have all data down to 100ms, you can do these – and many other! – statistical calculations more flexible afterwards with e.g. QtiPlot. That’s (my) best practice to navigate through many ten thousands data lines … 73 Nils, DK8OK

  • Thank you, Nils, but my immediate need is for real-time display. Also I don’t want to learn new software just at the moment.
    But I do appreciate the idea!

  • The idea is very good but quite complicated to join them in all in once . Yesterday i tried to join two parallel frequency plots with QTI in around 20 minutes…

  • Hi, Zacharias – thanks, and it took some time also with me, until I had the feeling to control QtiPlot a bit more than QtiPlot was controlling me 😉 …
    It will work with a bit of practice, and is worth it: 73 Nils DK8OK

Leave a Reply to DK8OK Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s