The convergence of Artificial Intelligence (AI) technologies and digital transformation has significantly amplified the importance of understanding and measuring user experience in real-time video communication applications. Therefore, this dissertation presents novel approaches for Quality of Experience (QoE) prediction in Web Real-Time Communication (WebRTC)-based applications through multimodal analysis of naturally available user feedback signals. The research develops comprehensive frameworks for both quality estimation and sustainable service delivery, addressing two critical challenges: the limitations of traditional quality prediction methods and the growing environmental impact of video communications. The first research direction establishes a hybrid framework combining network metrics with human-centric features, specifically facial expressions and speech characteristics, achieving QoE estimation accuracy of 93% through advanced data fusion techniques. The second direction investigates QoE-aware energy efficiency, demonstrating that content characteristics, device configurations, and environmental conditions can be optimized to significantly reduce energy consumption while maintaining acceptable quality thresholds which establishes quantitative frameworks for balancing user experience with environmental impact. The methodological rigor of this work is evidenced through extensive statistical validation, comprehensive experimental design, and thorough performance analysis. The research combines theoretical innovation with practical applicability, providing concrete solutions for implementing energy-aware video communication systems. The frameworks developed enable service providers to make informed decisions about resource allocation, quality management, and energy optimization. The research makes contributions to both QoE prediction methodologies and sustainable video delivery systems, presenting practical solutions for implementing energy-efficient video communications without compromising user satisfaction. The thesis advances the theoretical under standing of QoE evaluation while providing actionable insights for developing user centric, environmentally conscious multimedia communication systems. In addition, the research outcomes have been validated through extensive experimentation and disseminated through multiple peer-reviewed international conferences and journals. The publication of comprehensive datasets supports reproducibility and facilitates further investigation in this emerging field.

Advancing Video Communication: From WebRTC Quality Prediction to Green Applications

BINGOL, GULNAZIYE
2025

Abstract

The convergence of Artificial Intelligence (AI) technologies and digital transformation has significantly amplified the importance of understanding and measuring user experience in real-time video communication applications. Therefore, this dissertation presents novel approaches for Quality of Experience (QoE) prediction in Web Real-Time Communication (WebRTC)-based applications through multimodal analysis of naturally available user feedback signals. The research develops comprehensive frameworks for both quality estimation and sustainable service delivery, addressing two critical challenges: the limitations of traditional quality prediction methods and the growing environmental impact of video communications. The first research direction establishes a hybrid framework combining network metrics with human-centric features, specifically facial expressions and speech characteristics, achieving QoE estimation accuracy of 93% through advanced data fusion techniques. The second direction investigates QoE-aware energy efficiency, demonstrating that content characteristics, device configurations, and environmental conditions can be optimized to significantly reduce energy consumption while maintaining acceptable quality thresholds which establishes quantitative frameworks for balancing user experience with environmental impact. The methodological rigor of this work is evidenced through extensive statistical validation, comprehensive experimental design, and thorough performance analysis. The research combines theoretical innovation with practical applicability, providing concrete solutions for implementing energy-aware video communication systems. The frameworks developed enable service providers to make informed decisions about resource allocation, quality management, and energy optimization. The research makes contributions to both QoE prediction methodologies and sustainable video delivery systems, presenting practical solutions for implementing energy-efficient video communications without compromising user satisfaction. The thesis advances the theoretical under standing of QoE evaluation while providing actionable insights for developing user centric, environmentally conscious multimedia communication systems. In addition, the research outcomes have been validated through extensive experimentation and disseminated through multiple peer-reviewed international conferences and journals. The publication of comprehensive datasets supports reproducibility and facilitates further investigation in this emerging field.
21-mar-2025
Inglese
ATZORI, LUIGI
Università degli Studi di Cagliari
File in questo prodotto:
File Dimensione Formato  
Gulnaziye_Bingol_PhD_Thesis_.pdf

accesso aperto

Dimensione 8.24 MB
Formato Adobe PDF
8.24 MB Adobe PDF Visualizza/Apri

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14242/208582
Il codice NBN di questa tesi è URN:NBN:IT:UNICA-208582