Re: [Discuss-gnuradio] E100 Performance

[Matthias Wilhelm]

Am 30.06.2011 um 11:27 schrieb Ralf:
> Am 29.06.2011 16:34, schrieb Marcus D. Leech:
>> On 29/06/2011 9:11 AM, Ralf wrote:
>>> Hi,
>>> 
>>> the simple GRC in the attachment creates lots of underflows on our E100 
>>> ("U" on console)
>>> and dropouts when looking at the spectrum.
>>> 
>>> Is this as expected or how can this overload of the embedded Linux be 
>>> avoided?
>>> 
>>> Thanks,
>>> Ralf
>> Well, for one, 60Khz isn't a proper divisor of the 128MHz sample rate of the 
>> DAC, which means it can't be properly interpolated.
>> 
>> The minimum sample rate that you can deliver to the USRP-E100 is 128MHz/512 
>> = 250kHz, so you'll have to interpolate your
>>  data stream up to 250kHz prior to "presentation" to the UHD sink block.
>> 
> 
> Hi Marcus, thanks for your reply.
> 
> That is very important for us to understand. How is the divider 512 
> determined? It is in the FPGA I suppose.
> Where can read more about the FPGA besides this small text?
> http://www.gnu.org/software/gnuradio/doc/exploring-gnuradio.html#fpga
> 
> Can we only put proper divisors of 128MHz in terms of datarate into the UHD 
> sink even above 250kHz?
> 
> Actually we need to put ~10kbit/s of data onto a ~400kHz carrier. Is the GRC 
> repeater the right functionality
> to interpolate the small datarate as shown in the attached setup?
> 
> Regards,
> Ralf


There is a collection of lots of internal details of the USRP1 under
www.gnuradio.org/redmine/attachments/129/USRP_Documentation.pdf

They might also apply to the operation of the E100's FPGA logic.

You can also look at the source code directly
http://ettus-apps.sourcerepo.com/redmine/ettus/projects/uhd/repository/revisions/master/show/fpga

Regards,
Matthias