[USRP-users] Receive two channels on the B210 with different sample rates

Michael West michael.west at ettus.com
Wed Apr 22 20:11:39 EDT 2015


Hi Hunter,

My apologies to you and Marcus for the confusion.  I have looked at the UHD
source code and the B210 can support different rates on each channel as
well.  Your set up looks correct, but it is difficult to say what may be
happening without further information.  Can you elaborate on what sample
rates you are using, what master clock rate you are using, and exactly what
errors you are seeing?

Regards,
Michael

On Wed, Apr 22, 2015 at 11:46 AM, Marcus D. Leech via USRP-users <
usrp-users at lists.ettus.com> wrote:

>  On 04/22/2015 02:05 PM, Marcus D. Leech via USRP-users wrote:
>
> On 03/17/2015 01:55 PM, Hunter DeJarnette wrote:
>
>  Marcus,
>
>  Thanks for your quick reply.  Is this supported on other USRP models?
> Are there any plans for this to be supported on the B210?
>
>  Hunter
>
> No, there's no support for this in other models, either.  While it's
> conceptually possible, it isn't actually implemented, partially because
> there's
>   no way to setup the master clock (on, for example, the X310) in cases
> where there's no common master-clock rate that can produce both
>   sample rates.
>
> The best approach is to bring both streams in at the same rate, and
> resample the lower-rate stream.  If you use Gnu Radio, there are various
>   resamplers, including a rational and fractional resampler.
>
>
>  I was mis-lead by R&D, and have now been re-lead :)
>
> On the *X310 in particular*, you can have two different sample rates from
> the two "sides" but those streams *must* be packaged inside
>   separate multi_usrp objects,  and they both must, of course, satisfy the
> integer fraction rules with respect to the master clock rate.
>
> The streams inside a multi-usrp object are required to be at the same
> sample-rate, because a multi-usrp "object" does sample time
>   synchronization which is maximally awkward if the streams are running at
> two different rates....
>
>
>
>
> On Tue, Mar 17, 2015 at 12:51 PM, <mleech at ripnet.com> wrote:
>
>>  This is not a supported configuration on the B210.
>>
>> If you need different sample-rates on the two streams, you'll have to
>> re-sample in software.
>>
>>
>>
>>
>>
>>
>> On 2015-03-17 12:12, Hunter DeJarnette via USRP-users wrote:
>>
>>  I'm working on an application with the B210 where I would like to use
>> both rx dsp's, but at different sample rates. I understand that each of the
>> channels attached to a single rx_streamer must have the same sample rate.
>> So to get around this issue I tried instantiating two usrp sources each
>> with a different channel like so:
>>
>>         self.uhd_usrp_source_0 = uhd.usrp_source(
>>             ",".join(("type=b200", "")),
>>             uhd.stream_args(
>>                 cpu_format="fc32",
>>                 channels=[0],
>>             ),
>>         )
>>         self.uhd_usrp_source_1 = uhd.usrp_source(
>>             ",".join(("type=b200", "")),
>>             uhd.stream_args(
>>                 cpu_format="fc32",
>>                 channels=[1],
>>             ),
>>         )
>>         self.uhd_usrp_source_0.set_subdev_spec("A:A A:B", 0)
>>         self.uhd_usrp_source_0.set_samp_rate(samp_rate_0)
>>         self.uhd_usrp_source_1.set_samp_rate(samp_rate_1)
>>
>> I found that when the sample rates for the two sources are the same I get
>> similar performance to the using two channels on a single usrp_source.
>> However, when I try to use different sample rates I start dropping samples.
>> Am I setting this up right?
>>
>>
>>
>> Thanks,
>> Hunter
>>
>>  _______________________________________________
>> USRP-users mailing listUSRP-users at lists.ettus.comhttp://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>>
>>
>
>
>
> _______________________________________________
> USRP-users mailing listUSRP-users at lists.ettus.comhttp://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/20150422/4345c381/attachment-0002.html>


More information about the USRP-users mailing list