Hi,

NI told me magnitude of IQ samples received in usrp is the voltage in volts.

Info: Office wifi 10 m distant USRP 70 dB (power) gain gives 10^-3 (volts? ie magnitude of IQ samples)

Assume wifi transmits +20 dBm, isotropic radiator, isotropic receiving antenna

Voltage expected at receiving antenna terminals in 50 ohm:

Effective area receiving antenna: lambda^2 / (4 pi) = (0.125 m)^2 / (4 pi) = 1.2e-3 m^2
Power per area at 10 m distance = 0.01 W / (4 pi * (10 m)^2) = 8e-6 W/m^2
Received power = power per area * effective area = 8e-6 W/m^2 * 1.2e-3 m^2 = 9.5e-9 W
P = V^2/R 9.5e-9 W = V^2 / 50 ohm V = 7e-4 V is voltage across antenna terminals

Voltage at ADC:

Power gain = 70 dB -> voltage gain = 45 dB.
7e-4 V at antenna terminal is amplified by factor 10^4.5, gives 22 V

Actually see 10^-3 (magnitude of IQ samples)

That's a big difference between expected and observed. If the unit of IQ samples is really volts and 10^-3 is in volts then 70 dB gain of usrp is actually much much lesser value than real 70 dB.

Further I see in wifi analyser app the power ratings are -30 dbm to -60 dbm. Those values agree with my calculated expected receive power above. I expected at 10 m range to receive 9.5e-9 W, which is -50 dBm. That's nicely in the -30 dBm to -60 dBm receive power range that the analyzer app reports.

So what’s wrong with usrp?

Regards,
Koyel
Get Outlook for iOShttps://aka.ms/o0ukef

One info this is usrp 2955
From: USRP-users usrp-users-bounces@lists.ettus.com on behalf of Koyel Das (Vehere) via USRP-users usrp-users@lists.ettus.com
Sent: Saturday, September 5, 2020 8:28 PM
To: 'USRP-users@lists.ettus.com'
Cc: Naveen Jaiswal (Vehere)
Subject: [USRP-users] USRP sensitivity

Hi,

NI told me magnitude of IQ samples received in usrp is the voltage in volts.

Info: Office wifi 10 m distant USRP 70 dB (power) gain gives 10^-3 (volts? ie magnitude of IQ samples)

Assume wifi transmits +20 dBm, isotropic radiator, isotropic receiving antenna

Voltage expected at receiving antenna terminals in 50 ohm:

Effective area receiving antenna: lambda^2 / (4 pi) = (0.125 m)^2 / (4 pi) = 1.2e-3 m^2
Power per area at 10 m distance = 0.01 W / (4 pi * (10 m)^2) = 8e-6 W/m^2
Received power = power per area * effective area = 8e-6 W/m^2 * 1.2e-3 m^2 = 9.5e-9 W
P = V^2/R 9.5e-9 W = V^2 / 50 ohm V = 7e-4 V is voltage across antenna terminals

Voltage at ADC:

Power gain = 70 dB -> voltage gain = 45 dB.
7e-4 V at antenna terminal is amplified by factor 10^4.5, gives 22 V

Actually see 10^-3 (magnitude of IQ samples)

That's a big difference between expected and observed. If the unit of IQ samples is really volts and 10^-3 is in volts then 70 dB gain of usrp is actually much much lesser value than real 70 dB.

Further I see in wifi analyser app the power ratings are -30 dbm to -60 dbm. Those values agree with my calculated expected receive power above. I expected at 10 m range to receive 9.5e-9 W, which is -50 dBm. That's nicely in the -30 dBm to -60 dBm receive power range that the analyzer app reports.

So what’s wrong with usrp?

Regards,
Koyel
Get Outlook for iOShttps://aka.ms/o0ukef

Hi Marcus,

Thanks so much for the explanation. If IQ samples are not voltages then ofcourse I can’t do this analysis. However, FYI I am using usrp 2955 and frequency is 2.4 GHz.

Regards,
Koyel

Get Outlook for iOShttps://aka.ms/o0ukef

From: USRP-users usrp-users-bounces@lists.ettus.com on behalf of Marcus D. Leech via USRP-users usrp-users@lists.ettus.com
Sent: Saturday, September 5, 2020 11:33:30 PM
To: usrp-users@lists.ettus.com usrp-users@lists.ettus.com
Subject: Re: [USRP-users] USRP sensitivity

On 09/05/2020 10:57 AM, Koyel Das (Vehere) via USRP-users wrote:
Hi,

NI told me magnitude of IQ samples received in usrp is the voltage in volts.
The I/Q samples are linearly proportional to the antenna voltage.  But you can't use an analysis like below to come to any conclusions
about absolute antenna voltage from the samples.

Info: Office wifi 10 m distant USRP 70 dB (power) gain gives 10^-3 (volts? ie magnitude of IQ samples)

Assume wifi transmits +20 dBm, isotropic radiator, isotropic receiving antenna

Voltage expected at receiving antenna terminals in 50 ohm:

Effective area receiving antenna: lambda^2 / (4 pi) = (0.125 m)^2 / (4 pi) = 1.2e-3 m^2
Power per area at 10 m distance = 0.01 W / (4 pi * (10 m)^2) = 8e-6 W/m^2
Received power = power per area * effective area = 8e-6 W/m^2 * 1.2e-3 m^2 = 9.5e-9 W
P = V^2/R 9.5e-9 W = V^2 / 50 ohm V = 7e-4 V is voltage across antenna terminals

Voltage at ADC:

Power gain = 70 dB -> voltage gain = 45 dB.
7e-4 V at antenna terminal is amplified by factor 10^4.5, gives 22 V
The ADCs are set to a reference voltage of 1.2V.  That's as much as they can reasonably see.

Also, you haven't said what USRP you're using, nor what frequency you're operating at.

Further 70dB power gain == 35dB voltage gain.

I prefer to do this type of analysis with any of the online path-loss calculator tools, and set both the TX and TX antenna gain to 0.

Further, you can't infer the total system gain from the gain setting in the API--that just sets a gain-setting hardware "variable", which in
most cases, is an attenuator.  Without knowing WHICH exact hardware you're using, you simply cannot know what the exact system gain
is ahead of the ADC.

But wait there's more.  Unless your sample-rate == ADC clock rate, you'll be getting decimated samples, which will tend to reduce the
total power in the output channel, leading to lower-magnitude samples.