[USRP-users] time sync 2 ettus x310

Michael West michael.west at ettus.com
Tue Apr 21 20:29:34 EDT 2015


Hi Sanjoy,

The distortion may be due to the fact that some frontend components are
disabled or powered down while not transmitting or receiving and take a
little time to settle when enabled or powered up.  For the X310, there are
3 ways around this:

1)  Start TX and RX earlier and throw away RX data (as you have done)
2)  Modify ATR settings dynamically so the frontend components stay powered
on and enabled.  This can be done by using the multi_usrp::set_gpio_attr()
method.  The banks are RX0, RX1, TX0, and TX1.  You can accomplish this by
adding the following lines to your code:
    usrp->set_gpio_attr("RX0", "ATR_0X", usrp->get_gpio_attr("RX0",
"ATR_XX", 0), ~0, 0);
    usrp->set_gpio_attr("RX1", "ATR_0X", usrp->get_gpio_attr("RX1",
"ATR_XX", 0), ~0, 0);
    usrp->set_gpio_attr("TX0", "ATR_0X", usrp->get_gpio_attr("TX0",
"ATR_XX", 0), ~0, 0);
    usrp->set_gpio_attr("TX1", "ATR_0X", usrp->get_gpio_attr("TX1",
"ATR_XX", 0), ~0, 0);
    usrp->set_gpio_attr("RX0", "ATR_0X", usrp->get_gpio_attr("RX0",
"ATR_XX", 1), ~0, 1);
    usrp->set_gpio_attr("RX1", "ATR_0X", usrp->get_gpio_attr("RX1",
"ATR_XX", 1), ~0, 1);
    usrp->set_gpio_attr("TX0", "ATR_0X", usrp->get_gpio_attr("TX0",
"ATR_XX", 1), ~0, 1);
    usrp->set_gpio_attr("TX1", "ATR_0X", usrp->get_gpio_attr("TX1",
"ATR_XX", 1), ~0, 1);
*Note that changing the antenna or gain settings may overwrite these values.
3)  Modify ATR settings in UHD code.  This is done by editing the UHD code
for the specific daughter board to change the ATR settings.

Option #3 is the one that has worked the best for most users.  If you let
me know what daughter card and UHD version you are using, I may be able to
provide you with a UHD patch.

Regards,
Michael



On Tue, Apr 21, 2015 at 1:34 AM, Sanjoy Basak via USRP-users <
usrp-users at lists.ettus.com> wrote:

> Dear Mr. Martin,
> Thanks for the reply. Yes, right now I am not using octoclock, so the
> problem while using 2 USRPsis understandable.
>
> However, regarding distortion, yes I am having distortion at the beginning
> when I am using one usrp  for SISO case for x310. I found this problem even
> in b210 as well. I wonder why no one discussed or how did they actually
> solved for radar uses. But I hope I would get my solution here.
>
>
> [image: Inline image 1]
> This is a figure of xcorr of rx and tx signal. The red marked distortion I
> am having almost always. I don't know the reason. If I drop few bins in my
> binary file and then start receiving my signal then I am getting an
> undistorted signal looks like the following.( I am transmitting and
> receiving using binary file).
> However, dropping bins is undesirable for radar application.
> [image: Inline image 3]
> If I use the distorted frame, as obvious I get distorted result. However,
> after dropping bins the fresh frame gives me the exact result after signal
> processing in matlab.
> Can you please tell me how to avoid this distortion at the beginning?
>
> best regards
> Sanjoy
>
>
>
>
>
>
>
>
> On Tue, Apr 21, 2015 at 12:10 AM, Martin Braun via USRP-users <
> usrp-users at lists.ettus.com> wrote:
>
>> Sanjoy,
>>
>> if you want to do radar, you need to make sure both devices are
>> connected to the same external reference. For radar, you can't run off
>> of separate clocks.
>>
>> We've confirmed timing accuracy across devices on the nanosecond scale
>> in the past (using OctoClocks for clock distribution).
>>
>> Cheers,
>> M
>>
>> On 17.04.2015 06:00, Sanjoy Basak via USRP-users wrote:
>> > Hi all,
>> > I am trying to sync 2 ettus x310 in time. However, I am always having a
>> > distortion at the beginning after receiving, and there is always a
>> > delay of few microseconds between the tx and rx. I put tx in one x310
>> > and rx in another x310 and its in a wired connection. GPS antenna is
>> > connected to one usrp(with Tx) and another one is on external ref(Rx).
>> >
>> > So now if I check the last pps on both usrp; it shows tx:0; rx:2s;
>> > then I am compensating it making it set_start_time both at for 4s. So
>> > ideally I should get time synced tx and rx signal.
>> >
>> > But I am getting distortion in Rx signal for 2seconds at the beginning;
>> > this is constant for every sampling freq; so if in that region I have my
>> > rx signal I can't retrieve that as the distortion is already introduced
>> > there.
>> > As I am trying to implement radar with x310, this is a serious problem
>> > for me. As within first 2 seconds and few microseconds I can't receive
>> > anything.
>> >
>> > I would really appreciate any kind of suggestion that come through. If
>> > anyone have question regarding setup or python code please ask.
>> > Thanks for reading.
>> >
>> > Best regards
>> > Sanjoy
>> >
>> >
>> >
>> > _______________________________________________
>> > USRP-users mailing list
>> > USRP-users at lists.ettus.com
>> > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>> >
>>
>>
>> _______________________________________________
>> USRP-users mailing list
>> USRP-users at lists.ettus.com
>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>>
>
>
> _______________________________________________
> 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/20150421/1f3077ee/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: untitled.jpg
Type: image/jpeg
Size: 122891 bytes
Desc: not available
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20150421/1f3077ee/attachment.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: untitled2.jpg
Type: image/jpeg
Size: 18723 bytes
Desc: not available
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20150421/1f3077ee/attachment-0001.jpg>


More information about the USRP-users mailing list