USRP Software Radio

cancel
Showing results for 
Search instead for 
Did you mean: 

Problem for introducing delay

Hello,

I am using USRP-2920 based on an OFDM system, using labview program. When I introduce a delay at the transmitting side, I cant see it at the receiving side, it doesnt work? also, I tried to connect TX1 with RX2 and with different wire lengths  and still cant see the delay effect at the receiving side, I get always a constant value. can anyone help for that?

I attached a pic with different length cabels where the delay looks constant at the receiving side which is not logical , the problem even if I introduce delay at the transmitting side, I cant see it.

 

Thank you in advance

Download All
0 Kudos
Message 1 of 5
(5,144 Views)

Hello bound2u,

 

Thanks for your post!

 

I am going to do my best to try to help you, but I will need more information to get a better grasp of your issue:

1) How do you analyze the delays? Do you do this by the measured channel impairments cluster's delay indicator ?

2) What would you expect as delay output?
     Or in other words:
     Are you lookin for the delay caused by the change in cable or the delay caused by the complete transmit/receive set-up?

3) What code are you using?

    Just based on a picture it is difficult to make any conclusions (without making guesses).

    Please also make any settings that you are using the Default Values of the Controls.
    This way I will see how you have configured everything when I open the VI.

4) What are your different cable lengths?

     Do you have a picture of how everything is connected to your USRP?

5) What types of cable are you using?

 

Kind Regards,
Thierry C - CLA, CTA - Senior R&D Engineer (Former Support Engineer) - National Instruments
If someone helped you, let them know. Mark as solved and/or give a kudo. 😉
0 Kudos
Message 2 of 5
(5,126 Views)

Hello ThiCop,

 

I would appreicate your help very much, since this problem confused me. 

 

1) Firstly lets assume that I am using the simulation mode of my VI, if I introduce a delay using the delay indicator in the channel impairments cluster, I can see it very clearly at the receiving side and without any problem. While introducing a delay in the real measurment doesnt work, I get almost a constant delay 2.99 e-3.

 

Since the delay that I introduced using the delay indicator didnt work,  I wanted to experiment the delay using different cable length, this cables are the same as the ones used for HF application (expensive ones) , I attached pics of them.  and still getting the same problem of getting almost a constant delay.

 

2) I am using OFDM based-system, I expect by introducing a delay to see my signal delayed, disturb the constellation, and the platuea of my synchronization VI to be shiffted i.e. very close resuts from the simulation ones.

 

3) I am using OFDM based-system, actully I introduced some codes as , QAM modulation/demodulation , and synchronization I changed schimdle and coax algorithm. and since in the simulation every thing seem to work fine, I guess what I have introduced is correct.

 

4) I included a pic of the cables and the all connection.

 

5) I use the SMA that comes with the USRP device, and I connect them with HF cables. sometimes I just us th SMA or HF cables alone, since I am experminting different kind of channels at the moment. I use attunators sometimes as well, and trying to get my signal correctly by changing the gain and trigging level of  transmitting and receiving sides.

 

Please let me know if I missed something.


Best regards,

Download All
0 Kudos
Message 3 of 5
(5,119 Views)

Hello bound2u,

 

Following our private messages:

Is there really no code that you can share (that isn't protected)?

 

If there is a mismatch between your simulation and the actual physical measurement, then there could be several reasons.

 

One of them could be that your simulation doesn't simulate exactly the complete set-up you are using.

For example:

- What is defined as a "channel" in your simulation? (does this include delays caused by software/driver calls and measurement set-up or is it only looking at the cable length)

- What is defined as a "channel" in your hardware? (if driver calls are a part of your measured delay, then this could explain why the difference seems constant)

- How are you measuring the delay? (eg. Which functions do you use, where in your code do you use them, ...)

- Do you have any measurement with more digits shown than "2.99 E-3"?

 

For example:

- You could have 2.99444444444 E-3 for the long cable.

- You could have 2.99400000000E-3 for the short cable. (it could even be a bigger difference depending on the rounding)

 

In your original post I see delays of  in the order of 2.7 E-5.

Which one of both (2.7 E-5 or 2.9 E-3) is correct?

Or is one (or both) of them a simulated value?

If yes, then which one?

 

How much difference in delay would you expect between both cables from your simulations/calculations?

How are you doing your simulations?

Are you reusing a parts of the code that you have for your simulation in your measurement code?

 

Which cables are you using?

I am talking about brand, type, length, ....

RF cables are expensive, but this does not really give me much information to go from.

 

Since I don't know how you are exactly determining this delay in your VIs, I am currently only able to guess.

 

Take the analogy of a car "not able to drive anymore":

- The mechanic only has a picture of a part of the outside of the car.

- The mechanic cannot look in the car.

- The mechanic cannot look at the engine.

- The mechanic cannot verify anything.

 

And the question I'm asking the mechanic is: "Why don't I move when I press my gas pedal?"

You can understand that this way it is very difficult to tell what is wrong with the car.

It's kind of a riddle... 😉

 

To make it clear: It is not my goal to make your question seem unusual.

I just want to make sure that you understand that at this moment we don't have nearly enough information to fully grasp what you're doing and which things you are comparing with eachother.

 

A good starting point for troubleshooting could be to:

- Take 2 cables of the same type with significant different length

- Write down their full specifications (or provide pdf's that give this information)

- Do measurements with the both of them.

  Also show the delay with more digits than the current 3 digit approach.

  Note/Log/Write these down as accurately as possible.

- Document all the measurement results you see for both cables.

- Now do the same for your simulation and see what results you get for both lengths of cable of that type.

 

I think it might be best that you contact your local support department via e-mail or phone (if you have an active SSP contract):

ni.com/ask

 

Especially if you cannot share any code over here, then it will be very difficult to explain the code in forum posts and pictures.

 

PS: Just some LabVIEW terminology that can be quite useful: An indicator will show/indicate/display the value a control will set/control the value.

Kind Regards,
Thierry C - CLA, CTA - Senior R&D Engineer (Former Support Engineer) - National Instruments
If someone helped you, let them know. Mark as solved and/or give a kudo. 😉
0 Kudos
Message 4 of 5
(5,095 Views)

Hello bound2u,

 

Any updates at this moment?

Kind Regards,
Thierry C - CLA, CTA - Senior R&D Engineer (Former Support Engineer) - National Instruments
If someone helped you, let them know. Mark as solved and/or give a kudo. 😉
0 Kudos
Message 5 of 5
(5,013 Views)