[USRP-users] USRP N200 + octoclock MIMO channel sample time alignment

Matt Ettus matt at ettus.com
Mon Sep 9 12:30:23 EDT 2013


Thanks for the update.  I'm glad it is working well for you now.  I look
forward to hearing more about your passive radar work.


On Mon, Sep 9, 2013 at 7:13 AM, Juha Vierinen <x at mit.edu> wrote:

> Hi,
> I now switched from the FM band as the test signal to white noise and now
> I'm getting nicely aligned relative sample delays and relative phases
> across all the channels. The problem was in my test setup: the ambiguity
> function of a random 1 MHz sliver FM band really isn't good enough for this
> type of a measurement. I should have known better.
> juha
> On Fri, Sep 6, 2013 at 12:09 PM, Matt Ettus <matt at ettus.com> wrote:
>> Juha,
>> You should expect the timing on all the devices to be aligned to much
>> better than 1ns.  If it is not, then there is another issue.  Can you tell
>> me how you are computing the time difference?  Is it by RF phase or by
>> correlation?  Also, what daughterboards are you using?  Are all the N200s
>> the same rev and running the same rev of the FPGA image?
>> Matt
>> On Fri, Sep 6, 2013 at 7:29 AM, Juha Vierinen <x at mit.edu> wrote:
>>> Hi,
>>> I've been testing MIMO alignment of a usrp N200 setup. I've got four
>>> usrps hooked up to a octoclock-g, which is showing the gps lock light. I
>>> then periodically start 8 channel recordings with external 10 MHz and PPS
>>> locking. All channels are fed the same signal, originating from a 8 way
>>> splitter.
>>> The relative phase differences between the channels is relatively
>>> stable, although over tens of hours, I can detect some drift in the
>>> relative phase differences between the channels. Also, from some reason, a
>>> few of the USRPs seem to be consistently either 2 microseconds behind, or 2
>>> microseconds ahead of each other, ie., the samples aren't aligned for some
>>> reason. The timing cables aren't exactly equal length, but the  differences
>>> in the cable lengths is the order of 1 m or less, not 300 m.
>>> What kind of timing alignment should I expect for a four usrp N200
>>> system? Is it expected that in a MIMO setup there will be this much
>>> difference in sample alignment? The rx_time tags suggest perfect channel
>>> alignment.
>>> juha
>>> _______________________________________________
>>> USRP-users mailing list
>>> USRP-users at lists.ettus.com
>>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20130909/31c8e3fd/attachment-0002.html>

More information about the USRP-users mailing list