It’s likely better to use continuous steaming, and use the tags inserted in the 
steam to determine applicable frequency. 

Tune time on the UBX is probably a few milliseconds.  But I must emphasize that 
none of this hardware was optimized for fast frequency hopping. 



Sent from my iPhone

> On Feb 23, 2021, at 11:19 AM, Cox, Jonathan Albert via USRP-users 
> <[email protected]> wrote:
> 
> 
> Hello USRP Users,
>  
> I’d like to understand the practicality of using a USRP with UBX board as a 
> swept spectrum analyzer to sweep a broad bandwidth, like sampling 10 MHz to 6 
> GHz fairly quickly. I don’t require extremely precise amplitude calibration 
> across that bandwidth, but the sweep speed (LO tuning and data acquisition 
> initialization) should be relatively quick.
>  
> How long does it take to command the USRP to tune to a particular LO center 
> frequency, initiate a data acquisition, and then return the result (excluding 
> the time required to perform the actual sampling)? For example, the Tektronix 
> RSA306B claims a sweep speed of 500 ms for 9 kHz to 6 GHz.
>  
> Roughly speaking, to cover 6 GHz with ~160 MHz bandwidth, you would need to 
> tune the LO probably 45 to 100 times (depending on overlap, filtering, etc.) 
> . Therefore, if you want to keep the overhead under 200 ms, each 
> tune/initiate acquisition/download step should take no more than 2 or 4 ms.
>  
> Is it reasonable to tune and acquire with the UBX board in an X300 in 2 to 4 
> ms?
>  
> Regards,
> Jonathan
> _______________________________________________
> USRP-users mailing list
> [email protected]
> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
_______________________________________________
USRP-users mailing list
[email protected]
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Reply via email to