Quote:If the input and output of LNA are both well matched, i.e., input is matched to 50Ohm, output is also matched to 50Ohm load through a source follower. Then the power gain of this LNA, or measured S21 between the input/output ports would be 20dB(Av/2/2), is that right?
Yes, if the output and input are well matched to the same impedance (50 ohms) then S21 equals the voltage gain from the input port to the output port.
Quote:And when doing sp simulation in cadence, we can get S21 as well as max. gain, available gain and transducer gain, this makes me confused, I want to know which one can reflect the actual gain most accurately?
Well, I guess it depends on what you consider to be the "actual gain?"
Power gain definitions are used because the impedances have a significant impact on actual voltages and currents when operating at high frequencies, but the definition of "real" power is a more consistent measure of the input and output signals relative to the real part of the source and load interface to the circuit under test. There are different power gain definitions according to how the circuit is intended to be used or what performance parameters are most important in a given system design.
I would think that "available power gain" is most relevant to actual voltage gain when the real part of the output and input port impedances are the same. Available power gain is defined as the available output power delivered to the load divided by the available power of the source, this implies that the output and input are conjugate matched. Available power gain is still relevant when the real part of the input and output load impedances are different, you just need to account for the impedance transformation (just like a transformer) as part of the voltage gain (output impedance> input impedance) or gain loss (output impedance <input impedance).
Hope this helps.