[USRP-users] Improving Sensitivity of the Radio.
Marcus D. Leech
mleech at ripnet.com
Tue Nov 14 22:37:00 EST 2017
On 11/14/2017 10:31 PM, Kyeong Su Shin wrote:
> Dear Marcus D. Leech:
> Thank you for the reply.
> Is the only advantage of the TwinRX the pre-selector filter bank? Or,
> can the superheterodyne design of the radio also affect the data
> quality in some ways?
A Low-IF superhet design has some advantages, but the UBX has the same
overall architecture. With a Zero-iF design, mixer balance is extremely
important, and not always easy to achieve "perfectly". In the end,
it's a matter of where your mixer images end up.
> We do turn down the gain to get optimal results. (The 'noise figure'
> that I mentioned was the noise figure that is observed by us, after
> such adjustments). We will definitely try adding filters.
Keep in mind that filters "out front" necessarily add to the noise
figure of your receiver, since they aren't loss-free. RF system design
is all about
> Kyeong Su Shin
> On Tue, Nov 14, 2017 at 7:12 PM, Marcus D. Leech via USRP-users
> <usrp-users at lists.ettus.com <mailto:usrp-users at lists.ettus.com>> wrote:
> On 11/14/2017 10:00 PM, Kyeong Su Shin via USRP-users wrote:
>> Hello Everyone:
>> We are using USRPs for spectrum sensing (on pretty much any
>> frequencies). Before asking my faculty to order a TwinRX
>> daughterboard, however, I would like to see if there are any ways
>> to improve the sensitivity of the hardware that we currently own
>> Our USRP + UBX configuration works well, except that we cannot
>> really increase the gain level much without putting the RF
>> frontend into the nonlinear region. Because of this, the noise
>> figure of our data is about ~30dB in the worst case. I believe
>> that the main problem is that we are using a wideband outdoor
>> discone antenna for this - dumping quite a lot of RF energy to
>> the RF frontend.
>> In this case, what could I try to improve the sensitivity of the
>> radio? I think one option could be adding additional filters to
>> the chain (when we know that we are only looking at a certain
>> frequencies), but I wonder if there are anything else that I can
>> Also, I wonder what differences we can expect if we switch our
>> daughterboard to a TwinRX. Would it be worth it? What noise
>> figures did people experience when a wideband outdoor antenna was
>> connected to the board?
>> Kyeong Su Shin
>> USRP-users mailing list
>> USRP-users at lists.ettus.com <mailto:USRP-users at lists.ettus.com>
>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com <http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com>
> In the case of overload, the noise-figure is NOT the dominating
> factor in determining system sensitivity.
> If you are "sensing" across a limited band, then a filter that
> covers that band, and that band only, will definitely help keep
> the very first gain stages,
> which are "wide open" from going into overload.
> The TwinRX has some significant advantage here, since it has
> pre-selector filters, but those filters may or may not have band
> edges that correspond
> adequately to the band that your are "sensing".
> Basically, there's a "tension" in small-signal RF amplifiers.
> They can either have very high dynamic range, or they can have
> very-low noise figure.
> That general axiom still applies, although things are getting
> somewhat better in this regard. But taking the output of an
> outdoor disc-cone antenna
> in a normal urban environment is pretty-much begging for
> non-linearity. You might try turning *down* the gain.
> USRP-users mailing list
> USRP-users at lists.ettus.com <mailto:USRP-users at lists.ettus.com>
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the USRP-users