Anwire

New advanced services impose an underlying network able to support the increased requirements in terms of connectivity and Quality of Service (QoS) support. It is now widely recognised that the next generation wireless systems will be multi-access networks with the concept of “Always Best Connected (ABC)”. This can be achieved by the convergence, integration, and/or inter-working of all existing and emerging fixed and mobile (wired and wireless) networks including broadcast. The IP technology is a solution for this challenge. However, the structure and functionality of the IP layer model may not be sufficient and/or efficient in achieving the requirements of such integrated networks.

Additionally, the introduction of a generic architecture enabling efficient and always-on connectivity, distributed processing of reconfigurability management related features in various layers and an environment empowering adaptability in service creation, introduction, deployment and execution will be investigated based on integrated thematic efforts. This requires the combination of the currently defined methods known from classical protocol design, mobile radio design and computer science. From this synergy will emerge many novel concepts but also new questions currently not formulated. Ideally, one unique access technology would be sufficient for next generation wireless systems. But the wide range of services, environments and user requirements to be covered, ask for a family of access systems that seamlessly inter-operate with each other, forming the overall system. At this system, end-to-end QoS should be guaranteed at any time to any user. To do this, currently proposed QoS techniques for fixed IP networks, such as IntServ, DiffServ and MPLS, should be extended, combined and integrated into a unified QoS provision system. In short, a complete traffic management system is required, able to satisfy the required QoS of the user.

Concerning IP transmission over wireless access systems, there are a number of proposals to maintain QoS. Basically, wireless IP systems are considered access systems to the core network, and for this reason IntServ techniques can be used, without ignoring the peculiarities of the wireless medium. To maintain the agreements of active connections, the time-varying status of the wireless channel should be taken into account, in order to control the multiple access in such a way that the requirements of each mobile user are fulfilled. Wherever the capabilities of the access system are insufficient in guarantying the agreed QoS, the use of Performance Enhancing Proxies (PEPs) can be considered. A PEP improves the performance of the Internet protocols on network paths where native performance suffers due to characteristics of a link or sub-network on the path. In principle, a PEP implementation may function at any protocol layer but typically it functions at one or two layers only. It can contain scheduling functions, advanced error control mechanisms, connection establishment signalling, compression and fragmentation techniques, etc.

Seamless mobility support is a key issue in Wireless Internet, especially concerning QoS provision during handover. For IP micro-mobility (mobility within the same access network) a considerable number of proposals exist in the literature. In order for these techniques to support QoS, they should be extended or combined with mechanisms that consider QoS. In general, coupling QoS techniques with micro-mobility mechanisms can minimise set-up delays and packet losses. It can also improve scalability and overhead, since less update messages are sent or they are better localised to only the affected areas. Coupling techniques are distinguished to closely and loosely coupled. According to the closely coupled approach, the same signalling mechanism is used to carry both the mobility and QoS information, leading to reduced signalling, compared to loosely coupled techniques. The disadvantage is that this limits the QoS implementation to be dependent on a specific micro-mobility mechanism, which might need certain extensions to support the required functionality. On the other hand, the loosely coupled approach uses mobility events to trigger generation of QoS messages. QoS messages can be triggered as soon as the new routing information has been installed in the network. This has the advantage of keeping mobility and QoS signalling independent to each other, but cannot attain the same delays as the closely coupled techniques.

The research area can be further divided into research tasks, such as:

• Optimisation on all levels
• End-to-end Quality-of-Service
• Integration of heterogeneous access technologies
• Mobility management (terminal, personal, session mobility)

The aim of this workpackage is to co-ordinate work that will focus on studying and integrating existing and future proposals in the area, in order to result in a full QoS and mobility support system that guarantees efficient and always on connectivity. Intermediate and final results will be fed into promotion and dissemination activities.