From Friday, April 19th (11:00 PM CDT) through Saturday, April 20th (2:00 PM CDT), 2024, ni.com will undergo system upgrades that may result in temporary service interruption.

We appreciate your patience as we improve our online experience.

USRP Software Radio

cancel
Showing results for 
Search instead for 
Did you mean: 

USRP: Time and phase synchronization for source localization ??

Solved!
Go to solution

Hello,

I used 2 USRP in Reception. They are connected between them via Mimo cable.

I want to localize transmitter source. So i need to synchronize the two receivers.

 

I did the constellation with the two USRP:

1-with identical cables         2-  with different cables.

(figure :attached files)

 

 

 

Have I synchronized my receivers (in time and  phase)?

 

 

please    help me

 

Best regards

 

 

 

 

0 Kudos
Message 1 of 27
(12,647 Views)

The MIMO cable with synchronize the USRPs frequency oscillator. There will still be some fixed phase offset between the two devices.

 

Can you explain more on "identical cables" and "different cables"

 

Anthony F.
Staff Software Engineer
National Instruments
0 Kudos
Message 2 of 27
(12,632 Views)

Identical cables: means that  the cables who connect each receiver to the transmitter have the same length.

 

Can I give to the two receivers the same phase offset to have two superposed signals when I use Identical cables?

 

Best regards

 

 

0 Kudos
Message 3 of 27
(12,623 Views)

the two receivers with the transmitter:

MimoUSRP_opt.jpg

0 Kudos
Message 4 of 27
(12,604 Views)

Thank you for the excellent picture.

 

The 10 MHz reference clock will ensure that all connected USRPs will remained locked in both frequency and phase, however it is still possible that the initial phase offset between USRPs is different.  I would suspect this is what you are seeing.

 

A quick reference into your communications book should show you that multiplying by e ^ (j * theta) can correct for these offsets. In LabVIEW this is easy with the Polar to Complex function. Take a quick look at this code snippet:

 

Clipboard02.png

 

A MathScript version (which was mentioned in another thread) is contained here as well: https://decibel.ni.com/content/docs/DOC-19209

Anthony F.
Staff Software Engineer
National Instruments
Message 5 of 27
(12,592 Views)

Thank you for your reponse.

I tried to mutplied the signal by exp(j*phaseOffset), but the problem is that the phase offset isn't constant !??


I did two measures: (phae offset isn't constant!)


mimo.jpgmimo2.jpg

 


this problem does not allow me to do localization:Smiley Sad 

0 Kudos
Message 6 of 27
(12,583 Views)
The following example sends a calibration signal into the second input. Leakage between the two RX ports is used to detect and correct for the phase offset. From there MUSIC is used for position detection.

https://decibel.ni.com/content/docs/DOC-25716
Message 7 of 27
(12,576 Views)

Thank you for your reponse

 

Can you tell me if I can remplace the USRP REF TX by a generator to produce the calibration signal?

 

Best regards

0 Kudos
Message 8 of 27
(12,556 Views)
Yes. Any RF tone should work near the center frequency of the radio.
0 Kudos
Message 9 of 27
(12,540 Views)

Thank you

I used RF generator that produced the calibration signal (near center frequency). But i have the same problem (phase offset not removed).

I have to add a code to remove phase offset?

Best regards

0 Kudos
Message 10 of 27
(12,517 Views)