Synchronizing multiple GPSDO equipped 6000s

  • 5
  • Question
  • Updated 3 years ago
I'm interested in the possibility to synchronize data from multiple Flex 6500/6700 radios using the GPS time + location + I/Q streams for diversity reception experiments - is this even possible?  Worth the effort? Fun?  I invesion this as a Flex Radio community effort with hams having some software development desire.  Anyone else have similar ideas?

Note: This topic was created from a reply on the Installation of GPSDO topic.
Photo of Larry da Ponte

Larry da Ponte

  • 159 Posts
  • 15 Reply Likes

Posted 4 years ago

  • 5
Photo of Andrew Russell

Andrew Russell

  • 253 Posts
  • 33 Reply Likes
G'Day Larry,
Your ideas are similar to mine.
When I saw the specs for GPSDO I just had to have it because I could.
I suspect 2 6500s at different sites could be better than a 6700 at one for diversity reception.
I also imagine a world wide net of GPS synchronized radios (while having spare capacity or idle) becoming a planet sized wide band HF radio telescope or even radar.
It's a dream that is at the moment is beyond my capability.
Andrew
Photo of Walt - KZ1F

Walt - KZ1F

  • 3040 Posts
  • 641 Reply Likes
As I recall, that was /is precisely what Steve Hicks described a year or more ago in one of his recorded talks. As it was part of a larger talk, it would be difficult to isolate.

Steve, should you read this, would you elaborate on your thoughts for a geographical diversity reception?
Photo of Larry da Ponte

Larry da Ponte

  • 159 Posts
  • 15 Reply Likes
Let's say we did this and had the data streams lined up, what would be the next step?
Photo of Tim - W4TME

Tim - W4TME, Customer Experience Manager

  • 9032 Posts
  • 3431 Reply Likes
DON'T CROSS THE STREAMS!  Total protonic reversal is bad, very bad.
https://www.youtube.com/watch?v=jyaLZHiJJnE
(Edited)
Photo of Walt - KZ1F

Walt - KZ1F

  • 3040 Posts
  • 641 Reply Likes
It would be prohibitively expensive to ship an entire time-stamped spectrum around. However, a set frequency might be doable. You'd have to timestamp each packet and ship, I suppose via TCP vs UDP as the packet loss would be prohibitive. The local stream would need to be cached until the remote stream arrived, combined to be a single stream and them played. The frequency would need to be precise so one 6000 would need to be slaved to the other. This would be good for the master, not so good for the slave operator.


That's all doable though from a programming perspective, however it is not clear how intelligent the result would be. It wouldn't be a beginner project. I suspect it would only really be practical for people like Howard who have multiple station sites, geographically separate.
(Edited)
Photo of Walt - KZ1F

Walt - KZ1F

  • 3040 Posts
  • 641 Reply Likes
The other problem is the GPSDO time being as accurate as it is reflects the time, actually prior to, the received GPS status message. The GPS status messages do not update the system clock so any received packet arriving will be off the time given in the GPS status message. There are environmental factors that would affect the latency of the GPS status from the point of creation to the point of decoding. In other words, when the GPS status message arrives will always be different from when it was generated. There will always be a lag and that lag time will oscillate around a central point. One would have to measure, extremely accurately, the delta between the GPS status message and when a IF packet arrives to adjust it's timestamp. I suspect ms time would be way too wide and to measure nanosecond time in Windows is expensive.  It would be a very interesting science project.

Again, though, I suspect Steve H would have some really good insight into the practicalities of this endeavor.
(Edited)
Photo of Andrew Russell

Andrew Russell

  • 253 Posts
  • 33 Reply Likes

Larry Walt et al.

I am not even a beginner for this.

However to start off it need not be in real time.

 I wonder about 2 DAX IQ streams being combined in a client app on a pc or in the radio.

Might need to start saving for another 6500 or 6700.

Do have a mate with a 6700 a few km away perhaps we could stream to each other.

Andrew VK5CV