Master Thesis MSTR-2023-96

BibliographyRichter, Marcel: Machine-learning based Mobility Prediction for Resource Allocation in Mobile Edge Computing.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Master Thesis No. 96 (2023).
65 pages, english.
Abstract

The rapid evolution of wireless communication technology within the last decade, fuelled by the popularity of the smartphone, has made applications possible that have never been before. One of those applications is mobile cloud computing, which addresses the issue that mobile devices are inherently resource-poor compared to their static counterparts. The concept of mobile cloud computing is to deploy a cloud application in a powerful remote data centre, that supports the mobile device in storage and computation. Mobile cloud computing is a bad match for low latency applications, though, because communication with a remote data centre causes a lot of propagation delay within the network. Multi-access Edge Computing was introduced to alleviate this issue by bringing the cloud to the edge of the network in proximity of the user. However, it faces on key challenge. Maintaining the cloud in proximity of the user while they are moving to provide low latency applications is far from a trivial problem and requires thoughtful placement of the cloud. The research field that studies this particular problem is called resource allocation. One increasingly popular approach of resource allocation algorithms is to predict the user's future position with a mobility prediction algorithm. In this thesis, we study the literature on resource allocation with mobility prediction, derive the basic algorithms that they are based on and evaluate them in a simulation. Thereby, we provide the reader with the necessary information to deploy or develop such algorithms in the future.

Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Distributed Systems
Superviser(s)Becker, Prof. Christian
Entry dateFebruary 22, 2024
   Publ. Computer Science