End-to-End QoS through Integrated Management of Content, Networks, and Terminal

[Home] [Participants] [UNIKLU Team] [WP3 Member Area][Workshop]

Final feedback from reviewers on WP-3: "In WP-3 input to standardization was produced. Good initiative using LinkSys WiFi router. Good demonstrator input."

The information revolution of the last decade has resulted in a phenomenal increase in the quantity of content (including multimedia content) available to an increasing number of different users with different preferences who access it through a plethora of devices and over heterogeneous networks. End devices range from mobile phones to high definition TVs, access networks can be as diverse as GSM and broadband networks, and the various backbone networks are different in bandwidth and quality of service (QoS) support. In addition, users have different content/presentation preferences and intend to consume the content at different locations, times, and under altering circumstances.

Substantial research and standardization efforts have aimed at supporting Universal Multimedia Access (UMA) which attempts to comply with the scenarios indicated above. The primary goal of UMA is to provide the best Quality of Service (QoS) or Quality of Experience (QoE) with regard to the actual circumstances. It is highly desirable that the access is transparent and convenient in the sense that the user does not have to trigger or even does not notice that content negotiation, adaptation, and/or personalization has to take place to enable high quality media consumption under the given circumstances.

Given that (1) the multimedia content base is steadily growing and becoming richer, e.g., with respect to coding format or interactivity, (2) user and usage needs and preferences may vary widely, (3) there is a growing diversity in the devices used to access multimedia content, and (4) heterogeneous networks with dynamically changing characteristics may have to be traversed during content delivery; the realization of UMA represents a major challenge in multimedia research and standardization.

The following specific challenges are to tackle, which result in corresponding “building blocks” of a UMA-enabled system:

An overall important aspect is that solutions addressing these challenges must be interoperable such that, e.g., a network gateway can properly handle both the content description and device capabilities and deliver – potentially after adaptation – a conformant, suitable content variant to the device for play-out to the user. Standards and standardization bodies therefore play a central role for UMA to become reality.

First generation UMA systems have a strong focus on enabling the access to multimedia content anytime, anywhere and without concerning for particular formats, devices, networks, and so forth. The resulting UMA frameworks and systems comprise mainly transcoding server architectures allowing for the selection of already existing content variations based on a limited number of usage environment constraints.

The second generation of UMA systems aims at customisation of multimedia content at a high-semantic level through summarization and personalization according to the usage context. This kind of UMA architectures already include appropriate indexing techniques for retrieving summaries or views on a particular multimedia content based on the user profile, i.e., preferences, usage history, etc.

Standards supporting the concept and vision of UMA are the existing MPEG-7 and MPEG-21 standards. For example, tools like the Variation Description Scheme, the Digital Item Declaration Language, or the Usage Environment Description provide fundamental support for UMA.

Each of the aforementioned approaches and UMA generations concentrated either on low-level features, i.e., formats, devices, networks, etc., or on high-level features of the multimedia content, i.e., semantics, within a non-distributed, static, and deterministic environment. Thus, existing approaches lack of supporting end-to-end Quality of Service in a fully heterogeneous environment with dynamically changing usage environments.

One requirement for the implementation of such end-to-end quality-sensitive applications is the bounded-time access and storage of multimedia information. Current image and video transmission techniques using precise or progressive algorithms work well if sufficient processing power, network bandwidth and transmission time are guaranteed, but do not adapt properly to a reduction of one or more of these resources. The big challenge is how to adapt these applications in order not to distort but to offer the best level of quality under the given circumstances. The other major requirement for the realisation of such applications is the efficient transmission of multimedia information (video, image, audio, text) over a number of heterogeneous networks. In this respect, Service Providers are being presented with an unprecedented opportunity for delivering a wide range of services over heterogeneous networks.
ENTHRONE's approach for the next step towards achieving the ultimate goal of UMA and UME is centred on the concept of end-to-end Quality of Service by MPEG-21-enabled cross-layer Digital Item Adaptation as depicted in Figure 1. This approach should enable a worthwhile, informative experience anytime, anywhere for the user by translating QoS requirements from the service layer down to the system/network layer via the user/application layer.

On the one hand, the MPEG-21 Multimedia Framework will be adopted – in addition to the MPEG-7 Multimedia Description Schemes – with a special focus on MPEG-21 Digital Item Adaptation which aims transparent access to distributed advanced multimedia content by shielding users from network and terminal installation and management issues. The mapping of QoS requirements between service layer and user/application layer can be accomplished using these standards.
On the other hand, appropriate and emerging IETF/W3C transport and negotiation protocols (e.g., HTTP, RTP, RTCP, RTSP, SIP, SDP(ng), WSDL, SOAP, etc.) will be adopted enabling the QoS mapping from the user/application layer to the system/network layer.

Figure 1. 3rd generation UMA: end-to-end QoS by MPEG-21-enabled cross-layer Digital Item Adaptation

The Content Provider prepares the actual multimedia content as Digital Items facilitating scalable coding formats and metadata formats providing means for:

The Service Provider enriches the multimedia content with additional metadata w.r.t. to the cSLAs and pSLAs taking into account constraints imposed by access networks for service provisioning towards the customer:

The Network Provider offers QoS-based connectivity services providing reachability between network hosts:

The Customer is having a worthwhile, informative experience anytime and anywhere while interacting with the services provided by the Service Provider through his MPEG-21 compliant end-device:

Last updated: 2009-01-13 4:02 PM by christian(dot)timmerer(at)itec(dot)uni-klu(dot)ac(dot)at