TDMA with GPS synched Nodes

[brad12]

I want to implement tdma using GSP on USRPs. For now only two Nodes (as I have obly two usrps now) have to transmit for exact 8ms duration after every one second. Anyone done this before? I know that there is example VI on TDM, but that uses MIMO, I don't want to use MIMO cable. How much accuracy can I acheive? how to go about it? Any help would be appreciated.

[ErikL]

Hello Brad -

We have some code that's not quite ready for prime time, but what you are describing is exactly the goal.  I'll share it with you... but I warned you!

The code has a high level VI called main.vi inside of it we run the same subVI for each USRP.  This allows us to develop once and deploy to many.

Accuracy is a very tough thing to measure, but its been accurate enough to do mimo.  Its not coherent per say but the GPS's maximum possible accuracy is about 50nS.   There are a lot of assumptions wrapped up in that number:

- Stationary GPS

- Antennas have clear view of entire constellation above

- Multiple sync'ed devices share satellites used for timing

- Antennas are of high quality with electrically matched cable lengths

As for what you're describing, if you achieve synchronized sampling clocks, then an IQ rate of 1 MS/s Tx is 1uS.

[ErikL]

Here is the code I mentioned.

https://decibel.ni.com/content/docs/DOC-29883