Frequency error in USRP

GNU Radio
1

Hi Matt, Eric, and my dear friends,

I received third set of USRP for my research work, but the third USRP is
not
able to lock/demodulate the desired frequency. When i transmit a signal
on
2.4GHz from one of the old USRPs I need to lock the new USRP at 2.39989
GHz
else it does not demodulate the signal. Although the verbose mode
displays
that there is no frequency error in the system, but if I am not able to
demodulate on 2.4GHz then where’s the problem. I do understand that the
local oscillators may not be perfect, because two devices can not have
same
precision all the time. But, the question is how do I recover the signal
if
my receiver clock does not synchronize to the desired frequency?

If you can illuminate something on it, it would be great help.

Thanks in advance.

Tarun
University of Texas-Dallas

2

Tarun T. wrote:

the local oscillators may not be perfect

They definitely will note be perfect.

But, the question is
how do I recover the signal if my receiver clock does not synchronize to
the desired frequency?

In my application (GPS), I don’t even know the desired frequency due to
the doppler effect on the fast moving satellites. I use an acquisition
routine to locate the frequency which uses FFTs in my custom block.

Chris

3

Hi Chris,
Thank you for your reply.

On 4/10/07, Chris S. [email protected] wrote:

In my application (GPS), I don’t even know the desired frequency due to
the doppler effect on the fast moving satellites. I use an acquisition
routine to locate the frequency which uses FFTs in my custom block.

Your acquisition routine need any pilot for clock recovery or what? Are
you
using FFT to look for highest signal power? And, if your code is under
GPL,
can you please help me out with providing that to me and this group?

Chris

Tarun

4

On 4/10/07, Tarun T. [email protected] wrote:

Hi Matt, Eric, and my dear friends,

I received third set of USRP for my research work, but the third USRP is not
able to lock/demodulate the desired frequency. When i transmit a signal on
2.4GHz from one of the old USRPs I need to lock the new USRP at 2.39989 GHz
else it does not demodulate the signal.

The 64MHz oscillator on the board has a maximum 50PPM error. This
gives an error of up to 3200Hz.

To get to 2.4GHz, that’s 37.5x the 64MHz local oscillator.

3200Hz * 37.5 = 120kHz.

2.4e9 - 120e3 = 2.399880e9

So it is pretty possible that your oscillators are just that far off?

Brian

5

On 4/10/07, Tarun T. [email protected] wrote:

Can you please enlighten me, how to deal with this problem, as this is
pretty obvious in a practical system.

What type of modulation are you using and what is your application?

Brian

6

Hi Brian,
Thank you for reply.

On 4/10/07, Brian P. [email protected] wrote:

The 64MHz oscillator on the board has a maximum 50PPM error. This
gives an error of up to 3200Hz.

To get to 2.4GHz, that’s 37.5x the 64MHz local oscillator.

3200Hz * 37.5 = 120kHz.

2.4e9 - 120e3 = 2.399880e9

So, in my case its OK and in the range right… i.e.

2.4 GHz - 2.39989 GHz = 110 kHz

So it is pretty possible that your oscillators are just that far off?

Can you please enlighten me, how to deal with this problem, as this is
pretty obvious in a practical system.

Brian

Tarun

7

On 4/10/07, Brian P. [email protected] wrote:

What type of modulation are you using and what is your application?

GMSK modulation for Relay channels. Currently I am trying to relay the
voice
on two hop system.

Brian

Tarun