|
|
|
Institute of Communication Networks and Computer Engineering (IKR)
|
|
Project description |
|
|
|
|
|
|
|
|
|
Master thesis No. 1050
(In work) [pdf]
|
Network Traffic Prediction with Machine Learning using Gaussian Processes
|
|
|
Methods
|
Topics
|
Machine learning
Gaussian processes
Bayesian statistics
|
Traffic modeling
Traffic prediction
|
|
|
|
|
|
|
|
|
Background
|
In today's demanding networking landscape, network operators face the dual challenge of upholding stringent QoS (Quality of Service) standards while managing costs effectively. Balancing OpEx (Operational Expenses) reduction with efficient equipment utilization is crucial, and accurate traffic predictions play a pivotal role in achieving this optimization. The ability to anticipate future traffic patterns empowers network operators to design and implement strategies that maximize network utilization and minimize unnecessary investments in new hardware.
|
Gaussian Processes (GPs) have emerged as a prominent ML (Machine Learning) technique, employing multivariate normal distributions to effectively model stochastic processes under the Bayesian paradigm. Their remarkable ability to quantify uncertainty across a wide range of disciplines has cemented their position as a powerful tool.
|
Problem Description
|
In the context of this thesis, you are called to use GPs to model network traffic demands. More specifically, the thesis consists of the following steps:
|
•
|
getting familiar with GPs and related Julia packages
|
•
|
investigation of network traffic patterns
|
•
|
implementing GPs for modeling the backbone network traffic
|
•
|
evaluation of the models
|
Acquired Knowledge and Skills
|
In this thesis, you will enrich your knowledge of GPs and probabilistic inference, a methodology gaining growing focus. You will also experiment with the scientific programming language Julia. Finally, you will get a great insight into networking and network services.
|
|
|
|
Requirements
|
Desirable knowledge
|
Communication Networks Architecture and Design
Programming Experience
|
Programming Experience in Julia
|
|
|
Contact
|
Dipl.-Ing. Filippos Christou,
room 1.319 (ETI II),
phone 685-67968, [E-Mail]
|
|
|
|
|
|
|