Thank you Trip! As far as I can understand, the same “quick and dirty” approach was used to derive the fudge factor: it is right? Maurizio. Da: James Humphries [mailto:james.humphries@ettus.com] Inviato: mercoledì 9 marzo 2016 02:29 A: Crozzoli Maurizio Cc: USRP-users@lists.ettus.com Oggetto: Re: Re: [USRP-users] Questions on set_rx_gain Yup, just a visual approach to see distortion of frequency domain data. Like I said, it was a very quick and dirty measurement just to get a rough idea. -Trip On Mon, Mar 7, 2016 at 6:15 PM, Crozzoli Maurizio <maurizio.crozzoli@telecomitalia.it<mailto:maurizio.crozzoli@telecomitalia.it>> wrote: "The lines end where I started to see distortion": just a "visual" approach or did you measure somehow the distortion? ---- James Humphries ha scritto ---- Hi Maurizio, As a disclaimer, it is discouraged to input power levels this high to your USRP as it can easily cause damage. I recommend at least a 20dB attenuator if you wire directly. For an input, I used an RF signal generator (Agilent 8648C) with a simple sine output at 915MHz. I used this to control the input power to the USRP. I did as you suggested to calculate the magnitude of the signal, and then used a fudge factor to 'calibrate' it to the known input power (Although, you can see that the fudge factor is slightly off in this data). This was only a quick test to get a rough estimate of saturation points, noise floor, and linearity. -Trip On Mon, Mar 7, 2016 at 10:56 AM, Crozzoli Maurizio <maurizio.crozzoli@telecomitalia.it<mailto:maurizio.crozzoli@telecomitalia.it>> wrote: Trip, thank you for your answer! We are using an E310. What should I do to obtain a curve similar to the one you sent to us? Which kind of input signal did you use? Of course I know what is the power level of the input signal (dBm) but how do I get the absolute power level of collected/acquired signals? I can calculate the average of (I^2+Q^2) and express it in dB but what is the “reference” level of the acquired signal so that “dB” becomes “dBm”? TIA! BR, Maurizio. Da: James Humphries [mailto:james.humphries@ettus.com<mailto:james.humphries@ettus.com>] Inviato: lunedì 7 marzo 2016 16:01 A: Crozzoli Maurizio Cc: USRP-users@lists.ettus.com<mailto:USRP-users@lists.ettus.com> Oggetto: Re: [USRP-users] Questions on set_rx_gain Hi Maurizio, It really depends on the application as to what you set the gain to. What kind of input power do you expect on the RX? Is this for a B200/210, if so, a good starting point is around 20-30dB. Here is some data I took a while back (see attached photo). The lines end where I started to see distortion with too much input power. Maybe this will help as a rough guide. Note the high noise floor at 0dB gain, you need some gain to improve the noise figure. -Trip [cid:image001.png@01D179F4.13130610] ​ On Mon, Mar 7, 2016 at 8:32 AM, Maurizio Crozzoli via USRP-users <usrp-users@lists.ettus.com<mailto:usrp-users@lists.ettus.com>> wrote: Martin Braun via USRP-users <usrp-users@...<mailto:usrp-users@...>> writes: > > That is correct; an analog gain is applied. Note it is still possible to > "do damage" this way, e.g. by driving amps or ADC into saturation. > So what's the rule? Setting the correct gain seems to be really critical but I cannot understand what is the correct approach to decide a reasonable value: is it just a cut&try approach? AGC might help from a theoretical point of view. But just theoretical, I'm afraid, as far as I have understood, because most likely you tend to get damages when AGC is on. Nevertheless AGC would be really needed when your signal level fluctuates along a wide range of values. And if the best suggestion is to avoid AGC, we go back to the initial questions. Comments, suggestions, even tutorial from a practical point of view are more than welcome (they might even be worth of becoming a FAQ for SDR not-so-experienced users). Volunteers? TIA! Maurizio. _______________________________________________ USRP-users mailing list USRP-users@lists.ettus.com<mailto:USRP-users@lists.ettus.com> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com Questo messaggio e i suoi allegati sono indirizzati esclusivamente alle persone indicate. La diffusione, copia o qualsiasi altra azione derivante dalla conoscenza di queste informazioni sono rigorosamente vietate. Qualora abbiate ricevuto questo documento per errore siete cortesemente pregati di darne immediata comunicazione al mittente e di provvedere alla sua distruzione, Grazie. This e-mail and any attachments is confidential and may contain privileged information intended for the addressee(s) only. Dissemination, copying, printing or use by anybody else is unauthorised. If you are not the intended recipient, please delete this message and any attachments and advise the sender by return e-mail, Thanks. Rispetta l'ambiente. Non stampare questa mail se non è necessario.

Yes, it was a simple offset to calibrate the USRP to a known input power. -Trip On Wed, Mar 9, 2016 at 5:08 AM, Crozzoli Maurizio < maurizio.crozzoli@telecomitalia.it> wrote: > Thank you Trip! > > > > As far as I can understand, the same “quick and dirty” approach was used > to derive the fudge factor: it is right? > > > > Maurizio. > > > > *Da:* James Humphries [mailto:james.humphries@ettus.com] > *Inviato:* mercoledì 9 marzo 2016 02:29 > *A:* Crozzoli Maurizio > *Cc:* USRP-users@lists.ettus.com > *Oggetto:* Re: Re: [USRP-users] Questions on set_rx_gain > > > > Yup, just a visual approach to see distortion of frequency domain data. > Like I said, it was a very quick and dirty measurement just to get a rough > idea. > > > > -Trip > > > > On Mon, Mar 7, 2016 at 6:15 PM, Crozzoli Maurizio < > maurizio.crozzoli@telecomitalia.it> wrote: > > "The lines end where I started to see distortion": just a "visual" > approach or did you measure somehow the distortion? > > > > ---- James Humphries ha scritto ---- > > > > Hi Maurizio, > > > > As a disclaimer, it is discouraged to input power levels this high to your > USRP as it can easily cause damage. I recommend at least a 20dB attenuator > if you wire directly. > > > > For an input, I used an RF signal generator (Agilent 8648C) with a simple > sine output at 915MHz. I used this to control the input power to the USRP. > I did as you suggested to calculate the magnitude of the signal, and then > used a fudge factor to 'calibrate' it to the known input power (Although, > you can see that the fudge factor is slightly off in this data). This was > only a quick test to get a rough estimate of saturation points, noise > floor, and linearity. > > > > -Trip > > > > > > > > On Mon, Mar 7, 2016 at 10:56 AM, Crozzoli Maurizio < > maurizio.crozzoli@telecomitalia.it> wrote: > > Trip, > > thank you for your answer! > > > > We are using an E310. > > > > What should I do to obtain a curve similar to the one you sent to us? > > > > Which kind of input signal did you use? > > > > Of course I know what is the power level of the input signal (dBm) but how > do I get the absolute power level of collected/acquired signals? I can > calculate the average of (I^2+Q^2) and express it in dB but what is the > “reference” level of the acquired signal so that “dB” becomes “dBm”? > > > > TIA! > > > > BR, > > Maurizio. > > > > *Da:* James Humphries [mailto:james.humphries@ettus.com] > *Inviato:* lunedì 7 marzo 2016 16:01 > *A:* Crozzoli Maurizio > *Cc:* USRP-users@lists.ettus.com > *Oggetto:* Re: [USRP-users] Questions on set_rx_gain > > > > Hi Maurizio, > > > > It really depends on the application as to what you set the gain to. What > kind of input power do you expect on the RX? Is this for a B200/210, if so, > a good starting point is around 20-30dB. > > > > Here is some data I took a while back (see attached photo). The lines end > where I started to see distortion with too much input power. Maybe this > will help as a rough guide. Note the high noise floor at 0dB gain, you need > some gain to improve the noise figure. > > > > -Trip > > > > > ​ > > > > On Mon, Mar 7, 2016 at 8:32 AM, Maurizio Crozzoli via USRP-users < > usrp-users@lists.ettus.com> wrote: > > Martin Braun via USRP-users <usrp-users@...> writes: > > > > > That is correct; an analog gain is applied. Note it is still possible to > > "do damage" this way, e.g. by driving amps or ADC into saturation. > > > > So what's the rule? > Setting the correct gain seems to be really critical but I cannot > understand > what is the correct approach to decide a reasonable value: is it just a > cut&try approach? > > AGC might help from a theoretical point of view. But just theoretical, I'm > afraid, as far as I have understood, because most likely you tend to get > damages when AGC is on. > > Nevertheless AGC would be really needed when your signal level fluctuates > along a wide range of values. And if the best suggestion is to avoid AGC, > we > go back to the initial questions. > > Comments, suggestions, even tutorial from a practical point of view are > more > than welcome (they might even be worth of becoming a FAQ for SDR > not-so-experienced users). Volunteers? > > TIA! > > Maurizio. > > > _______________________________________________ > USRP-users mailing list > USRP-users@lists.ettus.com > http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com > > > > Questo messaggio e i suoi allegati sono indirizzati esclusivamente alle > persone indicate. La diffusione, copia o qualsiasi altra azione derivante > dalla conoscenza di queste informazioni sono rigorosamente vietate. Qualora > abbiate ricevuto questo documento per errore siete cortesemente pregati di > darne immediata comunicazione al mittente e di provvedere alla sua > distruzione, Grazie. > > *This e-mail and any attachments is confidential and may contain > privileged information intended for the addressee(s) only. Dissemination, > copying, printing or use by anybody else is unauthorised. If you are not > the intended recipient, please delete this message and any attachments and > advise the sender by return e-mail, Thanks.* > > *Rispetta l'ambiente. Non stampare questa mail se non è necessario.* > > > > >