Using QoE Metric as a Decision Criterion in Multimedia Heterogeneous Network Optimization: Challenges and Research Perspectives

Table 2

Summary of challenges and research opportunities associated with the use of QoE in network management.

References	Domain	Application considered	Challenges	Research opportunity
[30]	QoE management	Multimedia services	Subjective measurement of QoE, management (QoS) and real-time QoE, diversity of users and devices, and environmental factors	SDN, NVF, ML, mega data
[20]	QoE assessment	Multimedia services	Data collection and processing and the complexity of ML techniques	Data-driven approaches
[34]	QoE estimation	Streaming video, voice-over IP, and online gaming	Commonly used parameters are MOS and OS (opinion score), but the challenge is the need for accurate reference data and model parameters	N/A
	Modeling, measurement, and prediction	Multimedia services	Various factors (QoS) (UX), environmental factors, and user characteristics	The use of ML, hybrid models, and new approaches to measuring QoE
[10]	QoE management	N/A	Bandwidth management, quality of service (QoS) management, and network resource management
[9]	QoE management	Streaming video services	Lack of large-scale datasets and benchmarks	Using ML for QoE management
[33]	QoE management	Network services, streaming	Bandwidth management, quality of service (QoS) management, and network resource management	Modeling technique and prediction
[35]	QoE measurement	Multimedia communication	Measure and optimize QoE in wireless communications	QoE management in wireless networks
[36]	QOE measurement	Multimedia	Data collection, modeling, and prediction, design of a complete, efficient, reliable, and accurate open big data analysis and processing prototype system	N/A
[37]	QoE monitoring	ISP	Optimization of user QoE	N/A
			Encryption and privacy aspects
[19]	Load balancing	N/A	The growth of big data implementation of millimeter wave (mm-wave) security and privacy	SDN/network functions virtualization (NVF) deep learning transfer learning (tl)
[11]	QoE prediction	N/A	Domination of ML over DL	N/A
			Adaptive deep learning for enhanced video delivery
			Computational cost and interoperability
			Self-healing networks and failure recovery
[21]	QoE management	Multimedia	(1) Which key quality parameters to optimize and control?; (2) Where to control (e.g., client side, server side, and network side)?; (3) When to perform QoE optimization and control (e.g., during service, i.e., online or offline control)?; (4) How often to monitor and optimize QoE?	N/A