Networked Systems
Architecture and Engineering of Networked Service and Multimedia Systems
The research focus of Networked Systems is broadly described by the elements of Telematics system and discipline dimensions. The research focus of the various faculty members that constitutes Networked Systems is described as follows:
Adaptable and Autonomic Service Systems Finn Arve Aagesen
Adaptable service systems are service systems that are able to adapt dynamically to changes in time and position related to Users, Nodes, Capabilities, Status, and changed Service Requirements and Policies. Autonomic systems are adaptable systems based on a distributed autonomous architecture. Present research is related to Mobility management, Capability and status management, Configuration management, and Service management. The research is related to the
TAPAS architecture and platform. (TAPAS = Telematics Architecture for Play-based Adaptable Systems)
Model Driven Service Engineering Rolv Bræk
The goal is rapid, modular and incremental development of real-time services with controlled quality. Part of the activity is related to the
PATS program and the
SARDAS project. Three main topics are covered: 1)
Modeling languages and methods; We are currently investigating UML2 collaborations for modular service specification and service components with semantic interfaces that enable dynamic service discovery and composition with compatibility guarantees 2)
Tools; A suite of experimental tool components RAMSES, for modular and incremental validation and consistency checking as well as flexible UML2 to Java code generation targeted for Ericssons
ActorFrame 3)
Architecture and platforms; Agents as system components and roles as service components supporting dynamic deployment and dynamic role binding. It has been partially implemented in the ActorFrame platform of Ericsson and has been used to develop experimental services. (See
PATS - Hall of fame)
This activity comprises adaptive 3D scene & traffic control architectures in "Distributed Multimedia Plays", that is, virtual collaborations in games, business, education, concerts, opera, theatre and future television. (See
http://www.item.ntnu.no/~leifarne)
Mobile services to day are closed. Mobile services to day are mostly deployed, operated and managed totally by the mobile operators. Open mobile services are services that are open in use, design, implementation, deployment, operation and management. Future open mobile services should have an architecture which is distributed in multiple networks, devices, domains, governed by multiple players (user, mobile operator, third parties, enterprises), ubiquitously available continuous across devices, domains and networks, adapted to the context, and personalized, customised and presentable to the user. Three aspects of open mobile services are focused: 1) universal personalised services, 2) ambient intelligence and 3) identity management. (See
http://www.item.ntnu.no/~thanhvan/openmobileservices.htm)
Multiprovider Open Realtime Services Lill Kristiansen
New technology enables the convergence of services. One example is multimedia telephony combined with features from web, instant messaging, presence and context. The services will be distributed in multiple networks and in competing and/or cooperating domains. There are, however, economical, human and organisational issues that intertwine with the technical issues. The research is considering: 1) technical issues such as architecture and system development, and 2) human and organisational issues. Cases from health care domain are focused for both issues. (See
http://www.item.ntnu.no/~lillk/)
Correctness-Preserving Design of Distributed Systems Peter Herrmann
Methods for development of distributed systems by composition of heterogeneous software components need formal methods to guarantee that a system fulfils certain correctness properties. In particular, compositional modelling techniques simplifying the design and verification of system specifications are well-suited for component-based software. For instance, one can extend the interface description of a component by a formal model specifying aspects of the component behaviour and compose these specifications to prove correctness properties of the overall system. Three types of system constraints are focused: 1) functional correctness, 2) security properties, 3) trust building behaviour. (See
http://www.item.ntnu.no/~herrmann/)
