In frequency domain measurements, electronic noise shows up in two distinctly different ways. The first case is when the measurement is affected by the presence of unwanted noise, with noise being a nuisance. For example, we could be measuring the distortion of an amplifier with the amplifier's noise degrading the measurement. The second case occurs when the noise present in the system is the parameter to be measured. In that same amplifier, we may want to measure the noise at the output. Many of the same principles apply to both cases, but it is important to know whether the noise is the measurement or whether it degrades the measurement. The electronic noise present in our measurements may come from the device under test (DUT) that is being measured or may be generated internally by the analyzer. In the general case, the analyzer internal noise must be significantly lower than the noise of the DUT. However, techniques that compensate for the noise in the analyzer can lower the measurement floor of the analyzer.