PhD defence by Holger Heebøll

Abstract

Lasers emit light at an extremely precise frequency, which can be thought of as a very exact color. This precision is vital in fiber-optic internet, where data is encoded by modulating the laser’s frequency. Lasers are also integral to some of the world’s most accurate clocks, which rely on their stable frequencies. Measuring and understanding fluctuations in laser frequency is essential for improving energy-efficient fiber-optic communication and ultra-precise clocks.

This PhD project explores how machine learning, such as Kalman filtering and subspace tracking, can measure and analyze these small fluctuations. The project shows that these techniques are powerful tools for revealing laser behavior, though careful attention is needed to ensure that the results reflect the laser's true physical properties.

Supervisors

• Principal supervisor: Professor Darko Zibar, Department of Electrical and Photonics Engineering, DTU, Denmark
• Co-supervisor: Associate Professor Francesco Da Ros, Department of Electrical and Photonics Engineering, DTU, Denmark
• Co-supervisor: Associate Professor Michael Galili, Department of Electrical and Photonics Engineering, DTU, Denmark

Evaluation Board

• Associate Professor Jesper Lægsgaard, Department of Electrical and Photonics Engineering, DTU, Denmark
• Professor Magnus Karlsson, Chalmers Univeristy of Technology, Sweden
• Professor Bostjan Batagelj, University of Ljubljana, Slovenia

Master of the Ceremony

• Senior Researcher Yunhong Ding, Department of Electrical and Photonics Engineering, DTU, Denmark