Efficient Prediction of End-to-End Delay Ranges in 5G Networks

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Predicting End-to-End (E2E) delay is essential for optimizing performance and maintaining Quality of Service (QoS) in 5G networks. This paper proposes a methodology to predict E2E delay ranges in 5G, based on the Hidden Markov Model (HMM). Due to the dynamic and complex nature of 5G environments, the HMM is tailored to capture the sequential patterns in delay data, with the Viterbi algorithm employed to predict short-term and medium-term end-to-end delay ranges. We conduct a thorough performance evaluation, assessing the prediction rates using different amounts of prior information and prediction horizons. The results demonstrate that the proposed method might achieve prediction accuracy rates up to 80% and 92%, making it effective for latency-sensitive applications in 5G networks.

Original languageEnglish
Title of host publication2024 IEEE Conference on Standards for Communications and Networking, CSCN 2024
PublisherInstitute of Electrical and Electronics Engineers (IEEE)
Pages120-123
Number of pages4
ISBN (Electronic)9798331507428
DOIs
Publication statusPublished - 2024
Event2024 IEEE Conference on Standards for Communications and Networking, CSCN 2024 - Belgrade, Serbia
Duration: 25 Nov 202427 Nov 2024

Publication series

Name2024 IEEE Conference on Standards for Communications and Networking, CSCN 2024

Conference

Conference2024 IEEE Conference on Standards for Communications and Networking, CSCN 2024
Country/TerritorySerbia
CityBelgrade
Period25/11/2427/11/24

Keywords

  • End-to-end delay
  • Estimation
  • Machine Learning
  • Neural Networks

Cite this