It is a matter of fact that QoE is central in the design, creation, provisioning, and management of current and future multimedia services, especially when these are provided over the Internet, which is the most common case. The time it takes for a video-on-demand service to start playout, the number of stalling events experienced during a video streaming service, the resolution of the pictures shown on a social network site, how different images are combined in a news portal and the type of content they convey, as well as many other system factors, all impact perceived quality. However, not only are system factors important but also those belonging to the human, context, and business domains. Indeed, the mood of the user, the expected quality, and the place where the service is consumed are some elements that should be taken into account when estimating user perception and inferring consumer willingness to keep paying for a service or move to another provider.

Unfortunately for the service providers and fortunately for the researchers working in the field, there is a lot to be done to reach a satisfactory level of understanding on how to perform the estimation and manage it. Even the definition of this novel concept is not agreed upon by the relevant research and industrial communities, as highlighted in Chapter 2. Indeed, QoE is a complex concept that lies at the junction of several, mostly unrelated, scientific, technical, and human disciplines. It is probably for this reason that research in this domain is evolving, and mostly toward the user. This poses conceptual and practical difficulties, but it is a necessary step to take if QoE is to establish itself as a mature field of study. QoE is, after all, all about the user, and the user is the most important element of the end-to-end chains of service creation and provisioning.

In the QoE arena, one of the areas where significant results have already been obtained is the definition of perceptual quality metrics for quality assessment for various types of content, as highlighted in Chapter 3. Indeed, during the past 10 years, some perceptual quality metrics have gained popularity and been used in various signal processing applications, such as the Structural Similarity Measure (SSIM). In the past, a lot of effort has been focused on designing FR metrics for audio or video. It is not easy to obtain good evaluation performance with RR or NR quality metrics. However, effective NR metrics are greatly desired, with more and more multimedia content (such as image, video, or music files) being distributed over the Internet today. The widely used Internet transmission and new compression standards bring many new challenges for multimedia quality evaluation, such as new types of transmission loss and compression distortion. Additionally, various emerging applications of 3D systems and displays require new quality metrics. Depth perception in particular should be investigated further for 3D quality evaluation. Other substantial quality evaluation topics include the quality assessment for super-resolution images/video and High Dynamic Range (HDR) images/video. All these emerging content types and their corresponding processing methods bring about many challenges for multimedia quality evaluation.

In recent years, MPEG-DASH has proved to be one of the most important standardization activities related to multimedia content streaming, whose impact on QoE needs to be carefully investigated. Indeed, the correct configuration of the DASH systems needs to take into account the impact of each packet on end-user perceived quality; additionally, information must be exchanged between the server and the client. This introduces several issues that require further work, as highlighted in Chapter 4: (i) development of evaluation methodologies and performance metrics to accurately assess user QoE for DASH services; (ii) DASH-specific QoS delivery and service adaptation at the network level, involving the development of new policy and charging control guidelines, QoS mapping rules, and resource management techniques over radio access network and core IP network architectures; (iii) QoE/QoS-based adaptation schemes for DASH at the client, network, and server (potentially assisted by QoE feedback reporting from clients), to jointly determine the best video, transport, network, and radio configurations, toward realizing the highest possible service capacity and end-user QoE; (iv) DASH-specific transport optimizations over heterogeneous network environments, where content is delivered over multiple access networks such as WWAN (Wireless Wide Area Networks) unicast. Accordingly, in the future, we expect to see major work in this area.

No-reference or blind image and video quality assessment are reviewed in detail in Chapter 5. The existing algorithms mostly differ in the amount of information available on the types of artifacts that affect the visual data. Even in the face of almost no information regarding the distortion type or human opinion on quality, certain algorithms exist that predict visual quality with a fair degree of accuracy. However, while great progress has been observed in the field of image quality assessment, no-reference video quality assessment has received less attention. Indeed, this is due to the complexity of motion and its interaction with distortion.

Methodologies for assessing the QoE are of fundamental importance and should consider, among other aspects, the following issues. First and foremost, the picture quality perceived; thereafter, the ratings given by human observers are affected by what they have seen or experienced prior to a specific subjective test. Second, it has long been acknowledged that human perception and judgment in a psychophysical measurement task usually perform better in comparison tasks than casting an absolute rating. Third, an issue not altogether disassociated with the previous one is HVS (Human Visual System) response under two distinctive picture quality assessment conditions – that is, where artifacts and distortions are at visibility sub-threshold or around the threshold (usually found in high-quality pictures) and at supra-threshold (commonly associated with medium and low-quality pictures). Finally, it is becoming increasingly clear that assessment of QoE requires more than evaluation of picture quality alone, with the need to differentiate measurement of the perceived resemblance of a picture at hand to the original and that of usefulness of the picture to an intended task. All these issues are discussed in Chapter 6, together with the available approaches and models that consider all or part of these.

One of the major activities in QoE management is dynamic rate control, which can be performed at the application layer in order to allocate the available resources according to user requirements and transmission conditions. Whereas originally this approach was used with the intention of achieving a constant bit rate, now it adapts dynamically the transmission to the available bandwidth as well as to the variable channel and network conditions. Additionally, recent research has proposed a QoE-based control scheme which relies on a cross-layer approach to optimize the overall system configuration according to the quality as perceived by the end-user, as highlighted in Chapter 7. Clearly, this requires constant monitoring of the played content and how it is affected by the experience of the user. For this purpose, only RR and NR metrics can be used, with the former allowing us to achieve better results at the expense of an increase in the information that is sent to the receiver.

Summarizing, this book has addressed several aspects related to QoE: its definition, how to establish appropriate metrics, how the evaluation methodology works, which control and management aspects are to be considered. These represent only some of the key aspects of the subject, which is still in its infancy, according to the editors' opinion. Indeed, QoE is something that is destined to evolve over time, as ICT services are continuously changing – introducing new challenges and benefits for society. Accordingly, much research activity is expected in the years to come, for which we hope this book will form a foundation.