Implementation of a Technique for Computing Antenna System Noise Temperature Using Planar Near-Field Data

This paper presents the second phase of the development of a new measurement technique to determine antenna system noise temperature using data acquired from a planar near-field measurement. In the first phase, it was shown that the noise temperature can be obtained using the plane-wave spectrum of the planar near-field data and focusing on the portion of the spectrum in the evanescent region or “imaginary space”. Actual evanescent modes are highly attenuated in the latter region and therefore the spectrum in this region must be produced by “errors” in the measured data. Some error sources such as multiple reflections will produce distinct localized lobes in the evanescent region and these are recognized and correctly identified by using a data point spacing of less than $\lambda/2$ to avoid aliasing errors in the far-field pattern. It has been observed that the plane wave spectrum beyond these localized lobes becomes random with a uniform average power. This region of the spectrum must be produced by random noise in the near-field data that is produced by all sources of thermal noise in the electronics and radiated noise sources received by the antenna. By analysing and calibrating this portion of the spectrum in the evanescent region the near-field noise power can be deduced and the corresponding noise temperature determined. In the current phase of tests, planar near-field data has been acquired on a measurement system and the analysis applied to determine the system noise parameters. Measurements have been performed with terminations inserted at three different locations in the RF receiving path: the IF input to the receiver, the input to the mixer and the input to the probe that is transmitting to a centre-fed reflector antenna. The terminations consist of either a load that serves as the “cold” noise source or a noise source with a known noise output for the “hot” noise source.