Feather-light distributed systems (TES.6388)

Research
The ubiquitous computing vision has inspired early research into computing systems and applications over the last 10 years. They become pervasively embedded in our everyday environments, and bring the unique flexibility of digital technology to the activities around which our lives evolve. Components of such a smart environment will be small nodes with sensing and wireless communications capabilities, able to organize flexibly into a network for data collection and delivery. Caused by rapid progress in technology, this early research tended to focus on experimental prototypes of infrastructure, devices, and applications.
As the field progresses, another, more fundamental, challenge emerges, which is the focus of this project. There is an increasing need for distributed systems support that fit mobile ambient systems. This project's ambition is to offer a new stable framework that will allow easy implementation of the algorithms needed to make the system function properly. The system will offer easy mechanisms for efficiently supporting distributed algorithms with timeconstraints, which will be a clear step towards future ubiquitous systems. Although nodes will be equipped with a power supply (battery) and embedded processor that makes them autonomous and self-aware, their functionality and capabilities will be very limited. Therefore, efficient operation and collaboration between nodes will show emergent behaviour and is essential to deliver smart services in a ubiquitous setting.
In this project we investigate new feather-light distributed mechanisms for networking and distributed collaboration, and evaluate their feasibility through experimentation. These mechanisms can operate in a challenging environment of self-organizing collaborative ambient systems where nodes move, fail, and energy is a scarce resource. The main thrust of the work is therefore towards the development of new distributed timely system support, taking into account those specific features. In particular, schemes, which are able to work efficiently and dependably, in the presence of limited energy, processing power and memory, will be developed.

Utilisation
Ambient Systems are networked embedded systems, which are not only used for dedicated tasks in environment control, but also to support users in their daily activities. They will be based on an unbounded set of hardware artefacts and software entities, embedded in everyday objects or realized as new types of applications in nature environments, on work, at home, in the house, or on or around the body. The proposed new feather-light protocols, algorithms, and firmware for networking and distributed collaboration will be a basis for building self-organizing, dependable and collaborative ambient systems. We will evaluate their feasibility not only through theoretical analysis but also through experimentation in order to keep the gap to further industrial development as small as possible. The delivered projects results will allow systems integrators to extend their product lines with tiny wireless sensors that can be placed in locations never possible before. It will enable them to build better ambient systems that are easily extendible and adaptable to user requirements and it will open the possibility to developed and build much faster selected user applications, which is a considerable market advantage. Their charm is based on the right combination of distributed operating systems, protocols and hardware and software components. Their exploitation is expected to have immediate utility in an abundance of industrial (NEDAP, TNO, Thales), medical (NEDAP, Philips), civil (Philips) and military (Thales, TNO) applications.
Our industrial partners fully support the program as described in this proposal, and have clearly identified the need for research in this area. Their support consists of substantial contribution in equipment, laboratories, advisory, and hosting the PhD students in their research groups. The area of Ambient Systems is identified by many industries and business people as an emerging field, that might become the 'next big thing' [BusinessWeek August 25, 2003].

For results, see http://wwwhome.cs.utwente.nl/~jansen/featherlight/default.htm

Er zijn nog geen gebruikers geïdentificeerd, interesse?