On 2022-10-19 19:01, Minutolo, Lorenzo wrote:
Hi All,
I'm trying to develop a rfnoc4 block that changes the data rate and
the packet size.
Specifically, my block is very similar to the keep-1-in-N block
operated in packet mode with the exception that the first arbitrary M
packets are discarded, the output packet-in-N is an average of the
packets that arrived in the meantime and the output packed is cropped
respect to the input packet length.
I'm trying to understand how to develop this block.
The first approach that comes to my mind is to copy the keep-1-in-N
and modify it. First question:
1. How do I port the existing keep-1-in-N block to a new OOT block?
The second question is relative to how to deal with rate change and
packet size change in a OOt block. Except for looking into files, I
cannot find a comprehensive guide.
2. Is there a hands-on guide on how to develop these blocks other
than the gain example?
I'm a physicist with some FPGA knowledge, not a FPGA developer:
3. are there companies/contractors that can design these blocks?
Thanks,
Lorenzo
Are you considering the FPGA path because you feel that the sample-rates
are too high for an ordinary computer to cope with, or
because it just "seems right".
I'll note that the CASPER folks (at U C Berkeley) routinely do
FPGA-based signal-processing chains for radio science--they
might be a source of some help for FPGA development.
I'm not an RFNOC person myself. My preference is to avoid radio-science
applications where "FPGA" is the only answer.
But, that approach isn't, I admit, universally applicable....
_______________________________________________
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com
