NETWORK DIMENSIONING AND ROUTING OPTIMIZATION - Ethernet is being investigated as a promising technology for the Access and Metro network segments of public operators and its spanning tree routing paradigm is different of previous technologies; traffic engineering has many challenging issues like user fairness, routing survivability and demand uncertainty that require not only efficient optimization techniques, not yet available, but also tight monitoring and reconfiguration techniques that will be studied and validated on the existing Metro Ethernet lab infrastructure existent on IT. Triple play is already being offered by service and network providers; in the near future, the users required traffic will impose critical network design issues on networks since there is a trade-off in terms of cost between the number and location of servers and the transmission facilities to be installed; these issues can be formulated as non-linear optimization problems and require efficient solving techniques that will be addressed.
TRAFFIC STATISTICAL CHARACTERIZATION, MODELING AND MEASUREMENTS - We will continue our work on network modeling for QoS prediction, now investing in a Markovian framework suitable for this purpose; one step in this direction will be the research of multi-dimensional Markov models that jointly characterize the inbound/outbound traffic. We will increase our focus on advanced statistical methodologies, in particular, in the fast and real time identification of (eventually ciphered) Internet applications, in the use of multi-scaling traffic analysis for the purpose of traffic classification and identification, and in classification methodologies for multimedia content and users' preferences that enable personalized multimedia content distribution over p2p systems. We will research on the automation of traffic management tasks, while integrating some of the above procedures in our traffic monitoring tool.
NETWORK ARCHITECTURES AND PROTOCOLS FOR NOVEL NETWORKS - We will continue the research on new paradigms for networks, with special focus on mobile environments, and keeping the emphasis on user-centric and heterogeneous networks for mobile environments. This includes new security paradigms, addressing the integration of mobility, sharing and rewarding, and identity-based access control. We will also address management architectures and monitoring tools, handling such diverse aspects as intrusion detection, charging and rewarding records, or QoS conformance measures. Mobility will remain a center concern in all these activities, and its integration with QoS, security and multimedia, will still be considered, especially in post-IP networks, including heterogeneous networks, such as mesh and ad-hoc networks. Abstractions on the communication mode (unicast, multicast, anycast) will be pursued, as well as intelligent mechanisms for service selection in mobile environments. Key aspects for network structuring, and its relation with administrative domains, will be also researched in this context.
ADVANCED SERVICE PLATFORMS - We will research new service platforms bridging multicast delivery with network QoS, in context-aware networks, and integrating entertainment delivery with sensor networks. We will research new service platforms integrating pervasive environments in a telecom infrastructure, including context awareness, in a full mobile environment.
EXPERIMENTAL INFRASTRUCTURES - We will continue to evolve our experimental infrastructures, now with more emphasis on multiple wireless technologies and multiple service platforms, and in defining benchmarking methodologies for network evaluation.