Mixer noise figure is also supported with relatively minor changes to procedure. From a physical setup point of view, it is important that if the DUT does not intrinsically filter LO leakage to the IF/RF (output) port sufficiently (meaning that portion that can get through the receiver assembly is below about –120 dBm), additional LO filtering should be added. It does take a large discrete LO signal leakage to dramatically affect a noise measurement. From a measurement execution point-of-view, the only real change is in how DUT gain is loaded since a .s2p file is not generally available. Instead, .txt files can be loaded to bring in conversion gain information. The only requirement is that the gain data be in trace 1.

Alternatively, any .s2p file can be modified by loading conversion gain data into the S21 fields and the system can be told an amplifier is being measured (as long as any required LO programming is handled via multiple source control). The phase of S21 is not used in this case, so it can be left in any form, but the magnitude must be correct.