Analysis of Optical Subsystems and Links using Statistical and Machine Learning Methods
Author(s)
Lippiatt, Daniel James
Advisor(s)
Editor(s)
Collections
Supplementary to:
Permanent Link
Abstract
The internet has seen exponential growth of traffic demands in recent years as internet access has become more available and device connectivity has become an integral portion of the global market (e.g., “Internet of Things”). Optical networks serve as the backbone of the internet’s infrastructure and thus have received tremendous funding related to both deployment as well as research and development. To meet these increasing traffic demands, optical networks have continued to push the boundaries in terms of reach and data rates. This growth has been supported by the ongoing development of a variety of technologies such as optical amplifiers and optical filters as well as applications of these technologies such as wavelength division multiplexing. Likewise, computing power has increased and become more cost effective which has enabled the use of powerful digital signal processing algorithms to digitally compensate for optical impairments which previously limited link performance. However, performance monitoring of modern optical links has become a concern due to the increased complexity that results from these improvements – traditional monitoring methods have become obscured or difficult to realize. The objective of this research is to develop performance monitoring techniques for next generation optical subsystems and links that estimate key device and network parameters, identify the major impairments limiting performance, and localize these impairments to specific devices or locations within the optical link.
Sponsor
Date
2024-04-27
Extent
Resource Type
Text
Resource Subtype
Dissertation