ENR Compensation of a Noise Source Using Calibration Coefficients With Variation of the Ambient Temperature

Abstract

The excess noise ratio (ENR) of a noise source is largely dependent on the ambient temperature variation. In this paper, we discuss a method to compensate for the ENR of a noise source with respect to ambient temperature variations. From the measured quantities of the receiver and noise source from the setup, calibration coefficients of the noise source are extracted according to the relationship between bias voltage at the noise diode, the ENR, and the physical temperature of noise diode in the noise source by means of a least square estimation. From the result, the ENR of the noise source is compensated by the bias voltage at the noise diode controlled by a microcontroller and a digital-to-analog converter in the noise source. The compensated ENR variation is 0.167 dB for 1.4 dB ENR at 5.4 GHz in an ambient temperature range of 290–320 K. Specifically, the calibrated noise source shows the characteristics of 0.0016 dB/° C in a short ambient temperature range of 310–320 K. The time stability is about 0.007 dB/h for 1.4 dB ENR at 5.4 GHz. The uncertainty is evaluated in the experimental results, and the critical uncertainty contribution is confirmed by reading the power from a spectrum analyzer.