[USRP-users] Sequence Errors on USRP N200

Jahshan Bhatti jabhatti91 at gmail.com
Sun Jun 24 21:26:29 EDT 2012


On Sun, Jun 24, 2012 at 7:18 PM, Jahshan Bhatti <jabhatti91 at gmail.com> wrote:
> On Sun, Jun 24, 2012 at 7:12 PM, Jahshan Bhatti <jabhatti91 at gmail.com> wrote:
>> On Sun, Jun 24, 2012 at 7:06 PM, Jahshan Bhatti <jabhatti91 at gmail.com> wrote:
>>>
>>> into two timing receievers and monitoring the PPS. Is there some delay
>>> inside the USRP that isn't accounted for?
>>
>> On the order of milliseconds of course.
>>
>
> Would the difference of the delay between when the signal enters the
> RX connector to RX timestamp (thru downcoversion and digitization) and
> the delay between the TX timestamp to when the signal leaves the TX
> connector (thru signal generation and upconversion) be on that order?
>
> ~Jahshan

Ok, instead of using the timing receivers, I simply cross-correlated
the signals with a MIMO set of USRP N200s. From the cross-correlation,
I can see that the delay is stable on multiple runs but not exact.
Here are my results:

command 50 ms delay @ 5 MS/s results in error of -5.75 us
command 50 ms delay @ 4 MS/s results in error of -4.75 us
command 100 ms delay @ 5 MS/s results in error of -5.75 us
command 100 ms delay @ 4 MS/s results in error of -4.75 us
command 200 ms delay @ 5 MS/s results in error of 121.75 us
command 200 ms delay @ 4 MS/s results in error of 154.5 us

The resolution of my cross-correlation is 0.25 us.

~Jahshan




More information about the USRP-users mailing list