2.3 Measurement of receiver noise temperature.
Specifying the behaviour of a heterodyne system in terms of a noise temperature is convenient because noise temperature values are easy to measure. The fact that the value depends upon many factors — quantum efficiency, shot noise level, thermal noise level, amplifier noise, etc — doesn't matter when we're just comparing systems looking for the best (i.e. lowest noise) one. We can just measure their noise temperatures and choose the one we prefer. To see how we can measure a system's noise temperature consider the situation shown in figure 2.2.
This shows a complete heterodyne system used for making radiated power measurements on a thermal source. The IF output is amplified and passed to a second ‘square law detector’. The output from this is passed through an output lowpass filter or time constant. This ‘smooths’ away any swift fluctuations to provide an output, 
, which proportional to the IF power averaged over some moderate time interval. Hence the system provides an output voltage proportional to the power falling on the mixer.
The noise generated in the receiver is treated as if it came from an ‘imaginary’ thermal source of temperature 
which adds its noise power to the thermal signal power coming from the actual source. Hence when the input comes from a thermal source of temperature, 
, the smoothed output voltage from the receiver will be
where 
is the IF amplifier's power gain and G is the mixer's conversion gain.
Similarly, when presented with a thermal source of temperature, 
, the system will produce a smoothed output voltage of
combining these two equations and rearranging we can get the result
This equation tells us how to measure the noise temperature of a superheterodyne receiver. We can fill its field of view with two thermal sources of known temperatures, 
& 
, note the smoothed output voltages they produce, 
& 
, and use equation 2.24 to work out the Receiver Noise Temperature. In general, we can use this method to compare receivers and choose the one with the lowest noise temperature without taking any real interest in where the noise is actually coming from.
Content and pages maintained by: Jim Lesurf (jcgl@st-and.ac.uk)
using
TechWriter Pro and HTMLEdit on a StrongARM powered RISCOS machine.
University of St. Andrews, St Andrews, Fife KY16 9SS, Scotland.