Simple Multipath is a time domain phenomenon. It won't create a frequency shift like that. If the frequency shift changes, it could be dopler, the reflection of a moving, a truck? If one of the atmospheric paths is changing effective length, that would create a dopler shift. A constant frequency image could be caused by mixing. It looked like the image was shifted by 60 hz. LED Christmas tree lights? Or many other way 60Hz could get nonlinearly mixed.
Excellent video Rob and a very personal interest to me. I’ve noticed this phenomenon on 6m too and from what I can see on my end is Sporadic E backstatter. Basically two (or more) stations might be pointing at a common point (a cloud) but not actually pointing at each other (but still receiving each other direct, either via another cloud or via groundwave etc). What you then get are these ghost or mirror images with multipathing as you described, sometimes really strong and distorted. It’s common for me to receive a station in VK5 (approx 315 degrees heading) when I’m beaming toward New Zealand (20 degrees) and I’ll get two signals - you can actually see it more clearly on WSPR and it decodes as two signals on different frequencies. Very cool! Thanks again for the video
I think you are right - the reflections are most likely coming from sporadic E "clouds". I think what we are seeing is "delay spread" which causes the signal to spread out and look wide.
FT8 signals are a single tone that shifts in frequency over time to represent the data symbols. The most current symbol is received over the more direct path. A previous symbol is received at the same time, but over a longer path. These two may be creating mixing products, the sum and the difference. Distance does not change the frequency. Changing distance (Doppler) does, but is unlikely given your description and steady observations. To shift up and down, at the same time would require two reflecting surfaces, one approaching you and one leaving you. The waterfall display, of the decoded audio, obscures what is really going on. (WSJT-X display) It would be interesting to observe this phenomenon using a different receiver design, such as one of the SDRs with a high resolution spectrum display. You have a 991A? I think that is a triple conversion superhetrodyne, so mixing products, artifacts, may be occurring in your receiver. I am unlikely to recreate your situation here in Florida, where I can compare an Icom IC-756proIII with a FLEX-6400 using a 5-element beam, but without any mountains. -Tim, KI4TG
Hi Rob, Nice explanation of multi-path. Since an FT8 signal is narrow, when reflected it will show a change in frequency much more than a SSB signal. Stay safe. 73 WJ3U
On the AB1OC signal, you can see it decode at 2 different frequencies, but they are really close, the K1SIX signal seems to decode on two wider spread frequencies. Pretty neat!
A time delay such as multipath would cause a phase shift, Fourier Transform pairs, so it must be something else. Hard to tell since not under controlled conditions. Good that you are looking into this!
Suspect it is not for different pathways outside antenna, but for different pathways inside the antenna, meaning in the space inside the geometry of the antenna. Suspect it only happens, or mainly happens to multielement directional antennas. It is about the directivity mechanism, where axial it is phased, and from sides it is unphased. The differences in frequency that result, correspond to the size of the elements of the antenna, space delays that can result within those sizes. It has nothing to do with different pathways on the long distance propagation.
This looks like classic doppler shifting the audio frequency just like a freight train coming and going while the horn is triggered. If thats the case, there's likely a multipath-causing object that is moving at a constant rate. Or two objects causing multipath, each moving at a certain rate but in opposite directions. Pretty amazing. My guess is that the exact frequency shift of the audio and the amplitude could point you toward the object(s). One other thought - is there a heliport nearby ? If so, it could be worth checking the typical rotation speed of the props when a helicopter is on the ground, and compare that with the location (dist) and the audio shift.
The closest airport is 25 miles away. And this phenomenon happens all the time, not just for a few minutes. I think we're looking at delay spread which is similar to the Doppler spread you are talking about.
@@SevenFortyOne You might be seeing meteor scatter. Its very doable on 6m. There are specific frequencies people use for ft8, msk144, and ssb meteor work on 6.
Very well explained. I had heard of this, but you made things clear (in my mind) as to how this happens. I'm not an FT8 kind of guy ( I kinda think it's like cruise control for contacts). Take care and good DX! KC2RDU
When a signal reflects off a stationary object, it will be shifted in time only, not frequency. If it was reflected off of a moving object, you would have a frequency shift due to the doppler effect.
That’s odd and I’m confused. Unless the thing doing the reflecting is moving very fast (like a plane, satellite, or ionospheric disturbances) multipath should not alter the frequency *that much*. It will alter the time, ie latency, because it travels a longer path. It’s the same as an echo off a canyon wall - your voice’s echo isn’t deeper or higher pitched. I will have to read up on it! Update: after reading up on it, i'm convinced this is airplane scatter. Air traffic flying to or away the transmitter on the order of 600 mph can cause a doppler shift of up to 50Hz on either side of the fundamental. That lines up with the greatest extent AB1OC's signal was off-frequency at 5:29. However, the change in frequency over time is kind of slow, so you don't see it well in the ft8 waterfall, but I do notice wsjt-x fails to decode the ghost, and that makes sense - because wsjt loses frequency sync since it might be several Hz off from when it started. Also, since AB1OC's using big QRO, and the RCS of a typical jet at 50MHz is pretty massive (I have industry experience as a defense aerospace RF engineer), it's a no-brainer. K1SIX alluded to it in his monologue, but called it off for "insufficient information." The smoking gun would be to get AB1OC to send out a CW signal for a minute, and watch his ghost drift by 0.5hz/sec or so. That kind of thing is pretty hard to track along with FT8. Multipath against static reflectors like terrain should certainly not modify the frequency (only the time delay i.e. phase, which will cause amplitude fading) unless those mountains are being moved. Shout out to whoever figured out how to move mountains that fast!
I'm certainly no expert but I think we are seeing delay spread or Doppler spread but I'm not sure what's causing it. I need to do more research too....
@@N0SSC What you are saying makes perfect sense. However, this wasn't a one time occurrence. I didn't mention it in the video but I saw this on AB1OC and K1SIX every day over the span of a few hours for a week or so. Judging from the complaints K1SIX has received others are seeing it pretty consistently too. Also, If I kept the antenna pointed away from Fred the ghost images stayed fairly constant during the times I was paying attention. Signal strength faded in and out a bit but he always looked wide when the antenna was pointed away. However, we are under the main flight path between the US and Europe so there is pretty consistent air traffic overhead (30,000 - 40,000 ft) most days. Maybe there is enough traffic up there to sustain the signal spreading over several hours and occur regularly every day. I'll have set this all up again and have another look with flight radar open on another screen and see if I can draw any correlations between the signal spread and air traffic.
@@SevenFortyOne k1vlf on RU-vid has some airplane scatter spectrum captures from near your area, using VORs as sources. There’s tons of air traffic density up there, so it’s not surprising that it’s consistent but looks a bit different every time. Aircraft scatter is a highly desired phenomenon for microwave flatlanders, insofar that W3SZ wrote a program to plan AS contacts with real-time flight data.
Thanks for the heads up - I'll check out Paul's videos on the subject. I really like his videos and watch most of them when he publishes them but must have missed that one.
I don't quite grasp this. A signal arriving at different times would be a shift in phase between the two (or more) signals. But how does that cause the freq shift ? It isnt doppler because everything is stationary. What does make sense is what was mentioned above, signals somehow mixing, priducing the highs and lows. Good catch to even see this ! Good discussion also, thanks..
Great video Rob. However UFO's don't reflect signals. It is against the "The Galactic Charter of Planets" for ships being near primitive Ape planets such as ours. 🛸. I see you have a new toy in the background. FTdx10? Nice choice. 73 Joe
I am sure my point has already been raised in the comments, this is an interesting observation but as presented is not entirely convincing me. I think the multipath aspect is absolutely correct but that the reflection path is changing in distance. Perhaps a moving layer of the ionosphere. A reflection over a fixed path length should appear a the receiver at the same frequency it was transmitted on. A varying length path will create doppler shift. A very interesting video, and I do have some sympathy for a station that does get complaints from others. I operate FT8 a lot on HF, but so does another station a few hundred yards away. The phase noise from my rig and the ridiculous signal strength he got from me made his life a bit miserable when he was working some of the exotic DX, he now has my phone number so he can text me and I will change band. 73
@@SergiuCosminViorel I don't know but even a fast moving vehicle can produce noticeable doppler shift on 6m. I don't believe it is an effect within the antenna. more likely a rapidly changing path length.
@@g0fvt a car that travels with 72 km/h, it travels with 20 m/second, that in 20m band gives 1 hertz difference, in 6m band gives 3 hertz and something. Whatis observed there, is no Doppler effect. Not of any car. Airplanes? multiply! Not more than 30 times, it means bellow 90 hertz. And must be two planes, one incoming, one departing. lso must not be observed at all moments, only while you have two airplanes above. The two sides of the spread must alternate in strength over long time. Anyway, i do not know how a directive antenna works. Am just wondering about it
@@SergiuCosminViorel a moving reflector ie a car in the example can be responsible for double that doppler shift. At 50MHz you will get a shift of 50Hz if the path changes by 300m/S, ie the car doing 150m/S. Un-realistic speed for a car but not for a plane or ionised gas I would have thought.
@@g0fvt i did not exclude airplanes. Am doing more than following this video, and i cannot watch it all again, i have other things to do. I wrote here about how to consider. IF the frequency shift is of certain value, what to search for. The problem is with the shifs on both sides. Airplanes cannot do that, unless you have two, one coming, one departing. IF you have 3khz shift, then it is modulation, non-linear component... a bad contact and some stray signals can do that. Unlikely though. Too extreme. Back to the idea of highway full of cars: In this case you do not need high speed, normal speed would do fine, and you would not look for any Doppler, as the length of that highway replaces the need for high speed Doppler, the speed of the cars does not matter, the size of the cars matter, optimizes effect on 6m band. In the case of the highway, the two sides shift is given by the fact that your antenna is in the middle of some paralel to the highway. One problem with that though, a long highway would give large shifts on both sides. Or is it? As secondary emitters, they are not reflectors, they are receivers and secondary emitters, meaning your "reflection is not attenuated for near, and attenuated a lot for far". Anyway, to decide exactly what it is, more precise observation is needed
I wanted first to say it is rather Doppler, but thought more, played silly time wasting game, ate nutts with fungus, drunk a lot of coffee, looked at some porn photos, then it came to me. See? Porn is best inspiring. I do not agree is for different pathways. In any case, not that! Suspect it is not for different pathways outside antenna, but for different pathways inside the antenna, meaning in the space inside the geometry of the antenna. Suspect it only happens, or mainly happens to multielement directional antennas. It is about the directivity mechanism, where axial it is phased, and from sides it is unphased. The differences in frequency that result, correspond to the size of the elements of the antenna, space delays that can result within those sizes. It has nothing to do with different pathways on the long distance propagation.
@@SevenFortyOne Yeah, but it is not so. Just a trick. Could be, maybe, but it depends on frequency shift. Anyway, the idea, the valid idea, is like this: Take the frequency shift, multiply it by the wavelength, what you get is the size of the pathway difference. If you have 3khz difference on 6m, it gives 18 km, so you must check what is at that distance. If the shift is symetrical, must be something like come and go, that could be an highway. The effect is optimal for 6m band, which coincides with size of the cars. Around that... So i say it is a highway effect. When you put antenna on one direction, directivity disfavours the highway effect. You change it at some 30 degrees say, it favours the highway effect, which still comes weaker, because it suffered dispersion, as the cars radiate roughly non-directively the incoming snergy. If your bad is 3khz, then you have a highway at some10 to 25 km. If your band is 300hz, the distance is 1 to 3 km. there is also the possibility that you are closer to the highway, and you get signal from all along it. If it is not this, then you have something very naughty there. Maybe you are naughty.
@@SevenFortyOne something more to verify, the quality of connections to antenna, and metal-metal contacts within antenna. Partial, bad contact, oxides, maybe they can play the role of mixer, double sidebands to appear as your signal gets modulated by audio frequency, but the result would rather be fady from the center off. When i say audio frequency, i do not mean audio, but electromagnetic in audio band, which could be, should be like progressively fading specter. Unlike your example.
@@SevenFortyOne sorry, i was sleepy, and now i am even more sleepy. It is sleepy catastrophy, so i can not concentrate. back to the problem! cars, with their velocity, can produce frequency shift of only few hz, so any Doppler is out of question. Back to the highway, that very well can be a powerline of kilometers straitline or so, and also some railway. Any many kilometers long metal. With this model, your sides must be of unequal power, which is your case, if i am not mistaken. With an omnidirectional antenna, you would have symetrical sides. Now thinking more, with powerline or railway, you would have this on many bands, but if you observed it only, or mainly on 6m band, it means car sized metalic elements, those are cars on a highway. A long distribution, like a long metal element, but optimized on of around 6m
That's not what we are seeing here at all. If it was splatter the signals would splatter more when I had my antenna pointed directly at them and they were strongest. In these examples, the strong signals were narrow/normal and the weak ones were wide. Take a look at the research k1six had done for a detailed explanation
