The cross-spectrum experimental method

TL;DR

The paper introduces a cross-spectrum experimental method for accurately measuring device noise spectra by using two instruments to suppress background noise, applicable across various scientific fields.

Contribution

It presents a practical and theoretical framework for the cross-spectrum method, including estimator choices and experimental techniques, broadening its application scope.

Findings

Effective noise spectrum extraction below background levels

Versatile application across RF, photonics, and fundamental sciences

Tutorial approach accessible to practitioners and researchers

Abstract

The noise of a device under test (DUT) is measured simultaneously with two instruments, each of which contributes its own background. The average cross power spectral density converges to the DUT power spectral density. This method enables the extraction of the DUT noise spectrum, even if it is significantly lower than the background. After a snapshot on practical experiments, we go through the statistical theory and the choice of the estimator. A few experimental techniques are described, with reference to phase noise and amplitude noise in RF/microwave systems and in photonic systems. The set of applications of this method is wide. The final section gives a short panorama on radioastronomy, radiometry, quantum optics, thermometry (fundamental and applied), semiconductor technology, metallurgy, etc. This report is intended as a tutorial, as opposed to a report on advanced research, yet addressed to a broad readership: technicians, practitioners, Ph.D. students, academics, and full-time scientists.