Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancing Quality of Experience in Telecommunication Networks: A Review of Frameworks and Machine Learning Algorithms

Parsa H. S. Panahi, Amir H. Jalilvand, Abolfazl Diyanat

TL;DR

This paper surveys QoE in telecommunication networks, focusing on frameworks for measuring user-perceived quality and the integration of ML methods to enhance these tools. It clarifies QoE definitions, factors, and the QoS-QoE distinction, and reviews both closed-source and open-source QoE measurement platforms aligned with ITU standards such as P.1203. It catalogs ML algorithms from DNN/CNN to RNN/LSTM, TCN, GANs, and clustering, showing how they map network/video features to MOS-based QoE scores with high accuracy. The authors highlight major challenges, including data collection, generalizability, device diversity, interpretability, benchmarking, and privacy, and outline future directions toward standardized, privacy-preserving, and interpretable QoE modeling. The work aims to guide researchers and practitioners in deploying robust QoE measurement and ML-augmented optimization in next-generation networks.

Abstract

The Internet service provider industry is currently experiencing intense competition as companies strive to provide top-notch services to their customers. Providers are introducing cutting-edge technologies to enhance service quality, understanding that their survival depends on the level of service they offer. However, evaluating service quality is a complex task. A crucial aspect of this evaluation lies in understanding user experience, which significantly impacts the success and reputation of a service or product. Ensuring a seamless and positive user experience is essential for attracting and retaining customers. To date, much effort has been devoted to developing tools for measuring Quality of Experience (QoE), which incorporate both subjective and objective criteria. These tools, available in closed and open-source formats, are accessible to organizations and contribute to improving user experience quality. This review article delves into recent research and initiatives aimed at creating frameworks for assessing user QoE. It also explores the integration of machine learning algorithms to enhance these tools for future advancements. Additionally, the article examines current challenges and envisions future directions in the development of these measurement tools.

Enhancing Quality of Experience in Telecommunication Networks: A Review of Frameworks and Machine Learning Algorithms

TL;DR

This paper surveys QoE in telecommunication networks, focusing on frameworks for measuring user-perceived quality and the integration of ML methods to enhance these tools. It clarifies QoE definitions, factors, and the QoS-QoE distinction, and reviews both closed-source and open-source QoE measurement platforms aligned with ITU standards such as P.1203. It catalogs ML algorithms from DNN/CNN to RNN/LSTM, TCN, GANs, and clustering, showing how they map network/video features to MOS-based QoE scores with high accuracy. The authors highlight major challenges, including data collection, generalizability, device diversity, interpretability, benchmarking, and privacy, and outline future directions toward standardized, privacy-preserving, and interpretable QoE modeling. The work aims to guide researchers and practitioners in deploying robust QoE measurement and ML-augmented optimization in next-generation networks.

Abstract

The Internet service provider industry is currently experiencing intense competition as companies strive to provide top-notch services to their customers. Providers are introducing cutting-edge technologies to enhance service quality, understanding that their survival depends on the level of service they offer. However, evaluating service quality is a complex task. A crucial aspect of this evaluation lies in understanding user experience, which significantly impacts the success and reputation of a service or product. Ensuring a seamless and positive user experience is essential for attracting and retaining customers. To date, much effort has been devoted to developing tools for measuring Quality of Experience (QoE), which incorporate both subjective and objective criteria. These tools, available in closed and open-source formats, are accessible to organizations and contribute to improving user experience quality. This review article delves into recent research and initiatives aimed at creating frameworks for assessing user QoE. It also explores the integration of machine learning algorithms to enhance these tools for future advancements. Additionally, the article examines current challenges and envisions future directions in the development of these measurement tools.
Paper Structure (25 sections, 16 figures, 4 tables)

This paper contains 25 sections, 16 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Key Concepts
  3. Definition of QoE
  4. Impact Factors in QoE
  5. Distinguishing QoS and QoE
  6. QoE Measurement Tools
  7. QoE Optimization platform
  8. Closed-source platforms
  9. Open-source platforms
  10. ML Algorithms
  11. DNN and CNN
  12. RFR, SVR, and DT
  13. RNN and LSTM
  14. TCN
  15. RBFN
  16. ...and 10 more sections

Figures (16)

  • Figure 1: The QoE improvement system operates through a feedback loop mechanism. Users engage with servers to access services, resulting in specific experiences. These experiences, coupled with user-defined data, form the input for the AI system. The AI analyzes this data and generates predictions regarding future user experiences. These predictions are then utilized to optimize the server services offered creating a closed-loop system. This system ensures that user experiences continuously inform ongoing service improvements.
  • Figure 2: Four impact factors on user QoE with the corresponding exampleskougioumtzidis2022survey.
  • Figure 3: Network quality, QoS, and QoE domain within a cellular network infrastructure.
  • Figure 4: Streamlining service quality assessment: Leveraging the ITU P.1203 standard for video streaming metrics analysis.
  • Figure 5: MOS analysis conducted to assess stalls occurring in video playback under varying network conditions sr-1.
  • ...and 11 more figures