Self-referenced nonlinear interferometry for chromatic dispersion sensing across multiple length scales

TL;DR

This paper introduces a fiber-based nonlinear Sagnac interferometer that enables fast, precise, and calibration-free measurement of chromatic dispersion across various fiber lengths by exploiting cascaded second-order nonlinear processes.

Contribution

The work presents a novel, fully fiber-integrated nonlinear interferometer that achieves odd-order dispersion cancellation without active stabilization, suitable for a wide range of fiber lengths and applications.

Findings

Successfully measured dispersion from 25 cm to 4 km fibers.

Achieved high-visibility fringes and calibration-free spectra.

Operates at telecom wavelengths with rapid data acquisition.

Abstract

Chromatic dispersion critically impacts the performance of numerous applications ranging from telecommunication links to ultrafast optics and nonlinear devices, yet fast and precise measurements are challenging, especially for short length-dispersion products. We present a fully fiber-integrated nonlinear Sagnac interferometer that exploits cascaded second-order processes to generate frequency-anticorrelated idler light and achieve odd-order dispersion cancellation without active stabilization. The measurement is intrinsically self-referenced, as the dispersion-induced phase is extracted from the interference between counter-propagating nonlinear processes within the same Sagnac loop, eliminating the need for an external reference arm or prior calibration. Operating entirely at telecom wavelengths and read out on a standard optical spectrum analyzer, the device produces instantaneous, high-visibility fringes and calibration-free spectra using dual-port normalization. We demonstrate chromatic dispersion measurements on fiber samples ranging from 25 cm to 4 km, spanning short fiber segments to long-haul links. This architecture combines self-stability, broadband compatibility, and rapid acquisition, offering a practical metrology tool for both research and industry.