1
\$\begingroup\$

I am trying to wrap my mind around very small signals; what they mean, and how to amplify them.

For example, here is a screenshot from my cell phone. Notice the signal level received is -97dBm. Apparently, this is pretty standard for a cell phone signal.

cell signal

So, by my calcs, -97dBm is approximately 0.1 picoWatt of power. That seems like a really small amount of power. And I mean really small!

I tend to think in terms of my college physics lab whereby we had 5v supplying 1A of current into a light bulb. Or something like this, from what I remember. When that power comes down from the receive antenna and into the amplifier, what does the amplifier "see"? A very small voltage potential? It just seems so small that I'm having a hard time grasping the concept.

Now, how does it get amplified? I was looking at some simple low noise amplifiers (LNA), NXP BGU7003, to make a simple circuit that amplifies the signal. I was planning on using the app note for this chip as an example.

Is there a discreet component way this is done? Where would I look to get more insight into the circuit design of something like this? Wouldn't a tiny power RF signal like this get "lost" in the noise of an LNA?

I'm just so perplexed by this. So any help/insight is most appreciated.

-brad w.

\$\endgroup\$
  • \$\begingroup\$ Heavy filtering before amplification. Tuned resonators and heterodynes, both can be emulated with software and a good adc for reasonable frequencies but typically you tune analog (discrete or otherwise) components as a filter. \$\endgroup\$ – crasic Jun 26 '15 at 5:54
  • \$\begingroup\$ what is the input signal? A GPS signal will be around -130 dB but because of its nature (C/A code and modulation) it can be brought above noise floor once properly beated with the input noise with PRNs... \$\endgroup\$ – Umar Jun 26 '15 at 7:17
3
\$\begingroup\$

The thermal noise floor is given by kTB (watts) where k is Boltzmanns constant, T the temperature in Kelvin and B the bandwidth of the signal.

This noise at room temperature for a 25kHz bandwidth signal is approximately -130dBm.

The noise at the output of any receiver amplifier is the thermal noise, plus the noise figure of the amplifier in dB, plus the gain of the system in dB.

There are other sources of noise, but for the purpose of this answer they are not included.

What we need to achieve is an amplifier where the total system noise is below the inbound signal level.

In RF communications, the signal level is indeed very low (as you have already discovered), so we use LNAs (which really are widely available) and combine a very low noise figure with a reasonable gain.

These amplifiers are used to get the signal to a level where other processing (such as heterodyning, direct conversion and AGC to mention just 3 of the various possibilities, may be done.

HTH

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.