Scientific programme (dated back to 2002)

Despite the intense research and standardisation tasks performed during the last few years, there are still many issues to be solved before advanced MWNs can be deployed on a large scale. These networks are expected to provide a wide range of applications (many of them still unknown) with different traffic characteristics (bit-rate, burstiness) and QoS requirements (e.g. end-to-end delay, jitter, losses). In addition, QoS parameters such as network reliability and service availability might become even more important in the future.

The solution to these problems is not simple, especially when trying to meet other desirable goals such as the efficient resource usage, seamless transition between access technologies and standards, the interworking among fixed or wireless networks both terrestrial and satellite based, the migration to ‘all IP’ network architectures, and the support of, and with time comprehensive transition to, IPv6 technologies. Further down the R&D road the transition of the ‘always on’ concept to the ‘always best connected’ concept with its massive implications for the development of terminal and network reconfigurability capacities, application and service adaptability, and associated standards. All these aspects lead to the necessity of performing further deep R&D works for the definition of future MWNs.

The scarcity and high cost of radio spectrum necessitate that network providers maximise the profit/Hz ratio. Becoming better known are the specific constraints in relation to coverage and capacity that were not apparent in earlier networks. With the expected growth of the predominance of data traffic in future mobile networks, with the resultant high and less predictable fluctuation of bit-rates in both radio access and core networks new research into adaptive approaches will be needed in both cases.

The aim of this COST Action is to face these important R&D challenges and evolve ways towards the definition of novel MWNs.

The scientific objectives of this COST Action can be summarised into four broad areas that encompass different interrelated aspects for future MWNs. In particular, important research tasks to be addressed are: the analysis of the impact of multi-class traffics on network design and dimensioning; the definition of techniques to provide adequate QoS differentiation among traffic classes; the identification of new network planning criteria to cope with the integration of heterogeneous traffic flows; the definition of new services able to provide the requested multimedia contents to users on the move. A detailed description of these research areas is provided below.

Traffic Engineering

Traffic engineering encompasses the application of scientific principles and technology to the measurement, modelling, characterisation, and control of multi-media multi-class traffic and the application of such knowledge and techniques in order to achieve specific performance objectives, including the planning of network capacity under QoS guarantee, and the efficient, reliable transfer of information.

The need to allocate and balance resources among different traffic classes to accomplish the best use of network resources is a crucial traffic engineering problem. As a matter of fact, traffic engineering and QoS issues need to be jointly considered. The major objective of traffic engineering is to improve network performance while maintaining the QoS requirements through the optimisation of network resources, with the main focus of the optimisation being the minimisation of the over-utilisation of capacity in certain parts of the network while other capacity is available/under-utilised in the same network. MPLS type concepts if expanded and translated into mobile scenarios may provide one route towards TE and QoS solutions.

The optimisation objective depends on the specific goal of network operators, which may include minimising congestion, minimising packet loss/delay, or minimising the blocking probability. Network management and control can be considered very complex, and, thus, will require robust, possibly intelligent, control methodologies to obtain satisfactory (if feasible, optimal) performance. The development of efficient and effective management and control techniques may include issues regarding resource management, congestion control, connection admission control, and active queue management.

For traffic engineering of multimedia-enabled wireless networks, advanced measurement and monitoring technologies should be applied with special focus on IPv6 mobile services, QoS enabled and context-dependent character of mobile applications, users' movements, handover methods and QoS provision techniques due to integration of heterogeneous wireless networks (into all-IPv6 environment), innovative resource management algorithms and optimisation studies.

A further aim is the evolution of techniques for anomality end event analysis based on QoS and traffic monitoring in multimedia mobile networks such as detection of specific sources for events, malformed packets, security violations and attacks with special consideration of mobile networking, as for instance vulnerabilities occurring at the translation point between the wireless protocols and the wireline (fixed) protocols.

It should be emphasised that traffic patterns generated by IP multimedia services are quite different from traditional Poisson models used for circuit-switched voice traffic. As a result, the network parameters can be underestimated if inadequate traffic models and analytical approaches are adopted. Therefore, within MWNs traffic engineering problems, a particular problem is that of the performance analysis of network elements taking into account the self-similar nature of multi-service traffic. Hence, it is necessary to derive, for instance, upper and lower bounds of a service provision rate of 3G and in systems beyond 3G. Among the latter one could already count wireless LANs and mobile ad-hoc network clouds as access networks, so relevant considerations for traffic engineering should already be extending to include Wireless Local Area Networks (WLANs) and Mobile Ad-hoc NETworks (MANETs).

Analytical and simulation models developed within this Action will be instrumental for the definition of correct techniques for the design and planning of multi-service wireless IP networks with QoS guarantees.

QoS provisioning for multimedia traffic in wireless environment

The provision of QoS guarantees is becoming a pressing need in wired and wireless networks and in distributed computing systems, particularly to support multimedia-enabled applications. Throughput, timeliness, reliability, and perceived quality are the foundations of what is known as QoS. The combination of QoS and wireless environment is one of the hot topics in telecommunications nowadays.

The research community is now directing its interest towards unified ways of looking at system design, optimisation, and QoS issues to satisfy the requirements of next generation mobile and wireless IP-based networks (e.g. UMTS). The implementation of all IP-based future mobile and wireless networks implies that IP QoS architectures and mechanisms will need to be developed, as the existing best-effort based mechanisms are unable to cope with the requirements implied by the all IP-based network architectures.

By 2005 it is hoped that 3G systems will have an experienced significant take-up in Europe on the basis of new network infrastructures, new user terminals and novel applications. To meet this deadline and to provide a research basis for the definition of 4G systems, much work has to be done. In particular, the QoS provision for each service and the identification of suitable schemes to guarantee high capacity are of particular concern in this COST Action. Hence, suitable techniques must be identified to guarantee high capacity of simultaneous users and the fulfilment of QoS levels for the different traffic classes.

Note that QoS provisioning in wireless environment involves mechanisms, algorithms and schemes at various layers of the OSI Reference Model; in particular, physical layer, Medium Access Control (MAC) layer, IP layer and transport layer. The basic idea to be pursued in this Action is that QoS support requires joint collaboration among all these layers; how it may be done, new definitions of cross-layer protocols and so forth are important areas for collaborative studying and research. In particular, the following aspects will be addressed:

• Physical layer characteristics and mode adaptivity;
• Definition of novel Radio Resource Management (RRM) protocols that include Medium Access Control (MAC) and Usage Parameter Control (UPC) mechanisms for the QoS provision under fairness constraints;
• Definition of smooth handoff layer 3 schemes with low latency;
• Interaction between layer 2 and 3 to make handoffs fast and smooth;
• Support for efficient roaming, i.e. hand-over among different provider networks;
• Charging and accounting mechanisms for roaming users based on Authentication, Authorisation, Accounting (AAA) architectures;
• Distributed denial of service detection and prevention;
• Investigation for the use of existing/new L2 and L3 QoS architectures (e.g., MPLS and DiffServ) in a mobile environment;
• TCP modifications to make it more suitable for a wireless scenario;
• Evaluation of the impact of MAC choices/rules on the TCP layer throughput.

One outstanding topic related to QoS guarantees in mobile ad-hoc networks has to be investigated in this Action. Ad-hoc functionality such as self-configurability and independence of existing infrastructure are the key issues in this context.

The investigated techniques will be able to manage multimedia traffic with different characteristics in terms of burstiness, QoS, load, etc. Also regarding this topic a huge number of issues for investigation exists, among which statistical traffic models for MWNs, mobility and location awareness, dynamic resource allocation mechanisms and adaptive MAC protocols depending on traffic load and channel propagation conditions and based on QoS requirements, mobile Virtual Private Networks (VPN), security, handoff techniques, etc.

Network planning and dimensioning

The traditional task of network planning and dimensioning with QoS support is a multi-step process that involves the identification of the following aspects: (i) identification of network node location; (ii) definition of the link topology; (iii) definition of a routing strategy accounting for external input traffics; (iv) capacity allocation to the links so that suitable QoS metrics (end-to-end delay, jitter and loss ratio) are fulfilled. Many of these steps are interrelated; for instance, capacity allocation to links depends on traffic loads on the links and then on traffic routing. However, also traffic routing can be adapted to account for traffic bottlenecks, which result from capacity shortage on some links. As it is evident from these examples, network planning is a quite complex optimisation process.

Traditional multi-service networks have been developed within the controlled environment of large telecom operators. A significant knowledge is available for the design of Asynchronous Transfer Mode (ATM) multi-service backbone networks with QoS guarantees. Network planning and dimensioning problems for such kind of networks (where bandwidth availability may not be always a problem, as it is in the case of MWNs) are almost solved and represent a valuable knowledge basis for future work in this COST Action.

MWNs most probably will adopt “all IP” approaches to integrate different traffics and then different related applications and services. “All IP” will bring different challenges and require different solutions to the network planning and dimensioning problems. For IP-based packet-switched networks a number of QoS-assurance techniques have been proposed. But, providing QoS guarantees in IP networks is a difficult task, because the communication paradigm adopted by Internet was not originally conceived for a multi-service QoS context, and in particular for services which would find a more suitable “home” in circuit switched or at least network-layer connection oriented network infrastructure. Multi-service networks greatly require strategic planning for network growth for today and tomorrow. The techniques for design and planning of multi-service IP networks are far from being settled, for a number of reasons, such as, for instance:

• The diversity of applications and traffic flows;
• The variety of proposals that have been submitted for the integrated management of traffic flows with different QoS requirements (architectures, protocols);
• The difficulty in the characterisation of traffic sources in probabilistic terms;
• The uncertainty about the QoS metrics to be adopted for planning purposes;
• Multimedia wireless network integration into all-IPv6 networking environment, considering intra- and inter-domain issues of capacity planning and optimisation for mobile multimedia services.

MWNs planning and dimensioning tasks must take account of all these aspects.

In particular, the estimation of the necessary transmission resources must be based on the characteristics of the traffic generated by future mobile services (bit-rates, burstiness, packet sizes, protocol overheads) as well as on the possibility of obtaining bandwidth savings by statistical multiplexing.

Since multimedia services will be provided to mobile users, the evolved core network will have to support aggregated traffics, exhibiting Long-Range Dependent (LRD) and self-similar characteristics, as it occurs in the Internet. A self-similar traffic trace shows structural likeness for a wide range of time aggregations. In turn, LRD traffic causes high delays in the networks. Hence, traffic burstiness in IP networks cannot be averaged out, thus posing significant problems in dimensioning link capacities to fulfil given QoS requirements.

Finally, suitable design rules and planning guidelines will be defined for ad-hoc wireless systems in order to achieve important features of 4G, such as: connectivity, re-configurability, security and support of integrated traffics. In this context, planning does not exist due to the spontaneous and dynamic nature while centralised dimensioning is also not feasible due to potential topology volatility. Distributed approaches to dimensioning for QoS could be investigated.

Service aspects

The need of future MWNs will be largely dependent on the pervasive universal growth of the use of existing and new, narrowband to broadband, applications & services and their diffusion out to mobile users. It is expected that many of the new applications will be bandwidth-intensive to provide the users with multimedia contents, high degree of interactivity and real-time traffics, as those achievable from current high-speed wireline access systems (e.g., Asynchronous Digital Subscriber Line, ADSL, access).

Expected killer applications for future MWNs can be as follows:

• Mobile Internet services;
• E-commerce and e-government applications
• Location-aware information services;
• Broadcasting of news, sports etc.;
• High-capacity Web browsing;
• Instant messaging;
• Voice over IP;
• Personal Communications Services (PCS);
• Real-time video-on-demand services;
• Real-time interactive games;
• Peer-to-peer networking;
• Multimedia conferencing;
• Mobile multicast streaming data (audio-visual) data transfer;
• Download of audio- and video-clips;
• Video-telephony.

The implementation of these services poses new problems related to service billing, user privacy, user profiling, middleware architecture introduction for service and QoS adaptability (depending on user preferences, access characteristics, history of interaction), security, transactional interactivity, network reconfigurability, interoperability and interworking of networks (intra-terrestrial, intra-satellite, inter terrestrial and satellite), network engineering, traffic engineering and so on. Each service has specific requirements that influence on both the overall network design and the detailed information carried within protocols. Therefore, for a proper design of both networks and protocols, it is necessary to understand the services to be supported. The real-time aspects of the service can be described in terms of the transport level (such as transmission delay or packet jitter) and the session level (such as time to establish the session). Note that some QoS requirements understood in this phase of the study can be useful in all the previous WGs.

This Action aims at focusing on these service aspects by increasing the knowledge on new mobile applications and by providing adequate analysis on service provision aspects (i.e., enabling technologies and available design approaches) and characteristics. Other important issues to be addressed by this Action will be the relationship between service and pricing policy. This is a novel field of research activity that has a significant relevance for future mobile communication systems. It is motivated by the deployment of new services and hence new traffic types to be managed by the system, particularly in the context of the sophisticated evolution of MWN beyond 3G and encompassing the evolution of all forms of MWNs. Such aspects are of vital importance, because a right charging policy entails revenue and the possibility to enrich the service offer, thus allowing a positive feedback on the network deployment and traffic itself.