Heterogeneity is seen as the prevailing feature in Next Generation Networks, requiring the development of novel solutions to overcome the challenges posed by future communication environments. Thus, the major goal is to transform the diversity into an opportunity to increase user experience, which can be achieved through context-awareness. Considering this, and given the increasing demand for group-based multimedia services, this chapter focuses on providing an effective multimedia group communication in heterogeneous environments, developing an architecture driven by the available context information. It introduces a hierarchical architecture that employs the concept of abstract trees to control multiparty content distribution in a scalable way, being the trees built based on intelligent network selection schemes fed by available context information. The chapter’s ultimate goal is to find the best compromise between service personalization and resources optimization, developing an architecture flexible enough to properly react to context changes.
TopIntroduction
Future networking environments will be characterized by their diversity in features, which may be found at different levels (Figure 1), ranging from the terminals to the services: user devices have distinct processing capabilities and mobility features; the panoply of access network technologies differ on bandwidth and coverage; the co-existence of different transport technologies in the core and the variety of application requirements. Nevertheless, more than a challenge, this diversity represents an opportunity to differentiate the services provided and, as a consequence, enhance user experience. In this sense, the adoption of the increasingly popular concept of context-awareness will endow future architectures with the ability to sense surrounding conditions and take advantage of that information towards better network management decisions and user service perception.
Figure 1. Next generation network heterogeneity
Novel architectures driven by context have been proposed, focusing on adaptable content delivery according to user characteristics and network resources. Furthermore, the recent growth of group-based multimedia applications (e.g. IPTV) poses new challenges regarding the use of context towards service personalization, which can also be an added value for service providers that might differentiate and enrich their offer. This way, multiparty content transport technologies as multicast can be enriched by employing context, enabling the development of novel applications that increase the quality perceived by users. Moreover, the same information can also be used towards better network efficiency, making a better use of available resources and adapting content distribution to the network context.
The focus of this chapter relies on the challenges to provide an effective multimedia group communication on Next Generation Networks (NGN), developing an architecture that may overcome the limitations imposed by the heterogeneity inherent to these networking environments. Our approach supports a context-driven architecture, sensitive to network resources availability and user characteristics, in order to find the best compromise between two conflicting aspects of a multiparty content delivery architecture: service personalization based on user context and network resources optimization considering group communications.
Nevertheless, context-aware networks face several challenges that must be addressed: from one side, any type of context (e.g. preferences, location, mobility, resources) can be used to select the best networks for the groups of each service; from the other side, any context change may require a complete restructuring of the network and its sessions. Thus, it is critical to wisely select the context information that triggers network adaptations, so as to avoid constant re-arrangements that can impair the quality of the service provided.
To pursue these architectural goals, we present a hierarchical architecture that employs the concept of abstract trees to control multiparty content distribution in a scalable and flexible way. The abstract trees are built based on overlay nodes distributed throughout the network, being in charge of controlling the content distribution to the multiple receivers. Using this approach, it is possible to increase the flexibility of the distribution trees as well as hide network dynamics and heterogeneity of NGN. These abstract trees are built according to two main intelligent selection schemes: context-aware access and core network selection. However, group communication over heterogeneous environments increases the difficulty in deploying network selection. Considering different user interests, devices and networks characteristics, it is complex to fulfill all requirements simultaneously (different users, access networks, location).
In order to overcome this issue we introduce the grouping of interested users according to their current context, such as terminal capabilities, network conditions, location or others. This way, the same session content can be sent simultaneously to different sub-groups in distinct formats. Moreover, resilience can be improved through grouping approach, where only some sub-groups can be adapted so as to avoid affecting all interested users.