HI Eugene,

like Rob said, you would need to run the converters at 500 Msps, and then
decimate down from there, which requires a super-sample capable DDC block.
Such a block does not exist for RFNoC, at least not right now. The bigger
issue is that the 500 Msps images are built for only two channels total
(the 8-channel images top out at 250 Msps). Bottom line is, I think this is
stretching the capabilities of X440 a bit too far.

You could try building an X4_200 image with 16 DDCs (make sure you reduce
the number of halfbands to 1, and the max CIC depth to 5 to match your 500
-> 50 resampling requirement). Does that even fit onto the FPGA? If there
should be any chance of making it work, this would have to fit with
resources to spare.

--M


On Wed, Feb 4, 2026 at 8:57 PM Eugene Grayver <[email protected]>
wrote:

> Hi,
>
> I may be getting greedy but ...
>
> I want to record GPS signal with separate streams for L1 (at 1.6 GHz)  and
> L2 (at 1.2 GHz)  Each signal is about 50 MHz wide by itself, but getting
> both would require ~400 Msps.
>
> I need all 8 channels of the X440.
>
> Ideally, I would use the 200 MHz variant with **two** DDCs per channel.
> One DDC for L1 and one for L2.  That way the total sample rate would be
> only 100 Msps (vs. 400).
>
>
>    - Would UHD support that?
>    - I assume it's a custom RFNoC build
>
>
> Any suggestions on how to set that up?  Would 16 streams even work?
>
> Thanks
>
> Eugene Grayver, Ph.D.
> Principal Engineer
> 310-336-1274
> _______________________________________________
> USRP-users mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>
_______________________________________________
USRP-users mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to