[time-nuts] GNURadio USRP for Allan/Phase-noise measurement

Poul-Henning Kamp phk at phk.freebsd.dk
Wed Jun 22 19:25:34 UTC 2005


In message <42B9B959.4070203 at febo.com>, John Ackermann N8UR writes:
>Poul-Henning Kamp wrote:
>> Has anybody considered using the GNURadio USRP to measure Allan
>> variation and phase-noise of signals ?
>> 
>> It can sample two input signals at high resolution and speed
>> so it should be almost ideal for our purposes...
>> 
>> http://comsec.com/wiki?UniversalSoftwareRadioPeripheral
>> 
>
>I've *considered* it, but haven't done anything about it yet :-).
>
>I do have one of the USRP units, and it is a highly cool piece of 
>hardware.  The whole system (including the four ADCs, each 
>64msample/sec, 14 bit) is clocked from a single 64MHz oscillator, and 
>provision is made to replace the on-board oscillator with an SMA jack. 

That wouldn't really be necessary I think, because since the USRP
has two inputs you can feed your reference to one and the signal
you want to measure to the other and the 64MHz signal factors out
in the math.  (For other purposes it would be cool to have it
PLL-locked to atomics)

The idea I have is something like this:

Calculate the phase difference between the two signals as a function
of time.

Any linear component to that is a frequency difference and it
should be eliminated.

A FFT of the residual gives the phase noise or the allan variance
could be calculated from it.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.




More information about the Time-nuts_lists.febo.com mailing list