TUHH Open Research
Help
  • Log In
    New user? Click here to register.Have you forgotten your password?
  • English
  • Deutsch
  • Communities & Collections
  • Publications
  • Research Data
  • People
  • Institutions
  • Projects
  • Statistics
  1. Home
  2. TUHH
  3. Publications
  4. Bandwidth prediction for volatile networks with informer
 
Options

Bandwidth prediction for volatile networks with informer

Citation Link: https://doi.org/10.15480/882.16513
Publikationstyp
Conference Paper
Date Issued
2023-12-01
Sprache
English
Author(s)
Denizer, Birkan  
Landsiedel, Olaf  
TORE-DOI
10.15480/882.16513
TORE-URI
https://hdl.handle.net/11420/61011
Citation
2nd Workshop on Machine Learning & Netwoking, MaLeNe 2023
Contribution to Conference
2nd Workshop on Machine Learning & Netwoking, MaLeNe 2023  
5G networks provide high throughput and low latency connections, crucial for remote monitoring and control of mission-critical operations. Managing buffer levels and accurate bandwidth estimations are essential for low-latency applications. However, wireless networks are susceptible to fluctuations in quality metrics due to mobility and interference, impacting link utilization. Sudden quality deterioration can lead to lower Quality-of-Experience (QoE). To address this, we propose a neural network-based bandwidth prediction system. Our system utilizes historical data for time-series forecasting using the Informer model. It achieves 10% lower errors on a publicly available LTE dataset and 51% lower errors on a publicly available 5G dataset. Future work includes multivariate predictions and the creation of a new 5G dataset.
Subjects
Bandwidth prediction
LTE
5G
Informer
MLE@TUHH
DDC Class
621.3: Electrical Engineering, Electronic Engineering
004: Computer Sciences
Lizenz
https://creativecommons.org/licenses/by/4.0/
Publication version
publishedVersion
Loading...
Thumbnail Image
Name

2023_Denizer_BandwidthPredictionForVolatileNetworksWithInformer.pdf

Type

Main Article

Size

830.3 KB

Format

Adobe PDF

TUHH
Weiterführende Links
  • Contact
  • Send Feedback
  • Cookie settings
  • Privacy policy
  • Impress
DSpace Software

Built with DSpace-CRIS software - Extension maintained and optimized by 4Science
Design by effective webwork GmbH

  • Deutsche NationalbibliothekDeutsche Nationalbibliothek
  • ORCiD Member OrganizationORCiD Member Organization
  • DataCiteDataCite
  • Re3DataRe3Data
  • OpenDOAROpenDOAR
  • OpenAireOpenAire
  • BASE Bielefeld Academic Search EngineBASE Bielefeld Academic Search Engine
Feedback