[USRP-users] [Discuss-gnuradio] usrp b210 and mimo

Balint Seeber balint.seeber at ettus.com
Wed Dec 10 02:47:53 EST 2014


Hi Juha,

Ian and I have been busy looking into this.

For the moment, I suggest running a 'finite_acquisiton' after creating the
B210 device instance and setting time/clock sources, but before setting the
times on each device. E.g. just the short bit of Python code:

u.finite_acquisition(16384)

Part of the problem is that the RX streamer is created on-the-fly (if it
doesn't already exist) when you start streaming in GR, and this does some
more device configuration. If you do a finite_acquisition, this will force
the creation of the streamer before you run the flowgraph later (as
normal). After the acquisition, you can set the times and everything should
be aligned as you'd expect (we saw this working today).

We'll now look at making things 'just work'.

Hope that helps.

Kind regards,
Balint

On Fri, Dec 5, 2014 at 5:50 AM, Juha Vierinen <x at mit.edu> wrote:

> I've slowly gathered the courage to try out the B210 devices. I was
> impressed by the fact that I could get 50 MHz of bandwidth out of it.
> However, I immediately ran into problems with absolute timing.
>
> I've been using the B210 for mimo with USRP N200 devices for testing. The
> N200 devices all nicely synchronize themselves to the external PPS and
> after multiple restarts, their timing will be accurate. However, the
> starting time of the B210 seems to have a random component with a standard
> deviation of about 100 microseconds. Every time I restart sampling, the
> offset between the N200 and the B210 is slightly different. There is no
> change in the timing offset between the N200 devices.
>
> Is this a known issue, or am I doing something wrong?
>
> This is how I configure the devices:
>
> u.set_clock_source("external")
> u.set_time_source("external")
> tt = time.time()
> while tt-math.floor(tt) < 0.2 or tt-math.floor(tt) > 0.3:
>     tt = time.time()
>     time.sleep(0.01)
> print("Latching at "+str(tt))
> u.set_time_unknown_pps(uhd.time_spec(math.ceil(tt)+1.0))
> time.sleep(1)
> u.set_start_time(uhd.time_spec( starttime ) )
>
> juha
>
> _______________________________________________
> Discuss-gnuradio mailing list
> Discuss-gnuradio at gnu.org
> https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20141209/b214fc4e/attachment-0002.html>


More information about the USRP-users mailing list