[USRP-users] B200 RX Additional Functionality Questions

Ian Buckley ianb at ionconcepts.com
Sat Apr 4 17:52:17 EDT 2015

B200 uses AD9364 not AD9361…essentially they are identical in  programming and performance but with only one radio channel.
The reason I raise the 30.72MHz limit for running in 2 channel mode is that these radios only have a single LO (each) for RX and TX.
So if you want to capture 2 narrow band signals separated by 45MHz you'll need to run in single channel mode regardless of which radio version is on your USRP because you'll need to center your LO roughly between your 2 signals so they both fall within the complex sampling bandwidth.

On Apr 4, 2015, at 12:32 PM, Derek Murphy <cyrus104 at gmail.com> wrote:

> Thanks, that's great information. My millage right now on uhd and verilog is very low. I will start to look into it.
> I thought that since the ad9361 was the same then maybe the firmware could be modified to use both the a:rx2 and a:tx/rx could be used at the same time. The 30MHz limit is not a big deal because the signal is super narrow.
> On Apr 4, 2015 3:13 PM, "Ian Buckley" <ianb at ionconcepts.com> wrote:
> Derek,
> (You are specifically talking about B200 right? Not B210?)
> In theory this can be done, but the stock FPGA image won't do it. You would want to use to use the radio in a 1 RX mode because this would allow the use of (upto) a 61.44MHz sample clock between radio and FPGA, rather than the 30.72MHz max when in a 2 RX mode. (B200 doesn't have 2 RX mode anyhow)
> You would then want to take the B210 FPGA design, strip out the TX DUC's and buffer RAM so that it would fit in the B200's smaller FPGA, and then make one small alteration so that both RX DDC's are fed from the same (1 RX) sample bus. (At this point it would look very like the RX design of N210 with 2 RX DDC chains.).
> From there you have a few details to deal with in UHD (2 DDC's operational, but AD9364 delivering a single RX stream). And finally a flow graph that uses offset LO tuning such that you position the LO at (for example) 22.5MHz between your frequencies of interest and tune the 2 CORDICs to +/- 22.5MHz, then decimate both RX streams until they are the right rate to contain your tuned signals of interest but pass freely over USB2.0 bandwidth.
> It would be at least a days work for me to get that working well I suspect, and I'm very familiar with the FPGA code. Extrapolate your own milage accordingly. You should need very little new code, just a lot of shuffling of existing code, but you'll want a decent understanding of the workings of UHD and some verilog skills.
> -Ian
> On Apr 4, 2015, at 7:49 AM, Derek Murphy via USRP-users <usrp-users at lists.ettus.com> wrote:
>> I've done a lot of google research and have not found a solution yet.
>> I wanted to check and see if there is any way to use the B200 in 2 RX mode. I understand that you would not be able to TX at that time and the bandwidth on each RX channel would be lower than the full 56MB.
>> The use case that I am trying to get working in GnuRadio to process two signals that are 45Mhz a part which wouldn't be terrible except the systems that I must use only have USB 2.0. If I could have the B200 use 2 RX each at low bandwidth that seems like it would work, the other option could be to have the B200 set to receive the full span of 45Mhz but filter out the unwanted regions and send back the much smaller result.
>> Thanks
>> _______________________________________________
>> 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/20150404/21fbe133/attachment-0002.html>

More information about the USRP-users mailing list