Graduation Project: Machine Learning and Optical Communications

Optical fiber Communications

Optical fibers carry more than 99.9% of the global internet traffic. The maximum data rate in these fibers is currently capped by nonlinear effects. The propagation of the optical fieldthrough optical fibers is modelled via a stochastic nonlinear partial differential equation. Although this model is very precise, it is still unclear how to design good detectors that take care of the nonlinear effects.

Project Description

Your tasks:

(1) Apply detection theory to optical communications

(2) Learn about machine learning

(3) Develop improved detector algorithms Implement and test them using state-of-the-art GPUs

Environment

You will be working in the ICT Lab of the Signal Processing Systems (SPS) Group, TU/e Eindhoven. The ICT Lab collaborates with Bias Lab (Marco Cox).

Your Background

Your basic signal processing background should be enhanced with knowledge about wireless communication and information theory. Course 5LSF0, “Applications of Information Theory”, Q4, provides this extra knowledge.

Contact


Interested applicants may contact:

Dr. Alex Alvarado (a.alvarado@tue.nl).