ECS Intranet:
Pervasive Computing and Networks

Themes

The Pervasive Computing and Networks theme is the successful collaboration of those within the Intelligence, Agents, Multimedia Research Group with specific interest in future pervasive computing and networking technology. We aim to both develop and research new technologies, as exemplified by our work on Distributed Systems and IPv6, and to take existing technology and apply it to an academic environment, as undertaken by our JISC technology transfer projects.

Outlined below is a collection of some of the current research and deployment activities that are on-going within the theme. Further information about the rest of the IAM Research Group can be found at the group's information site. For further information on the activities within the Pervasive Computing and Networks theme, please contact the Senior Academic Lead, Professor David De Roure.

Where Physical meets Digital

We are exploring how digital technology could be deployed to radically affect the homes, schools, theatres and even streets of the future-changing and evolving the relationship between the physical and digital worlds. Our principal aim is to determine how people will live with the technology of the future, investigating new ways of using digital technology in everyday items such as signposts, furniture and even clothes.

As part of a large-scale interdisciplinary research colloboration, the local Southampton team concentrate on the design of software systems. As digital technologies have matured, they have begun to move beyond the workplace to other areas of our everyday lives, our homes and neighbourhoods, as well what we wear and carry with us. The phenomenal spread of the Internet has also enabled the public to participate in a variety of new online experiences, such as email, distributed hypermedia and virtual reality. Despite developments there are still many everyday activities where the boundary between physical environments and digital space is often over-complex and poorly designed. This work looks at ways of improving the integration of the physical with the digital.

Further information: Equator@Soton or contact David De Roure

IPv6 Wireless Initiative

6WINIT, an initiative co-ordinated by major European Telecom companies, equipment manufacturers, solutions/software providers, research laboratories and end-user hospitals - are being supported by the European Commission to develop and demonstrate future internet technologies. The multi-million Euro project also includes partners from North America and the Pacific Rim. It intends to lead to the provision of a production IPv6-2.5/3G transit service that facilitates high-quality, high-performance, operationally robust and secure Mobile Wireless 2.5/3G-IPv6 networks to facilitate the wider deployment of European M/E-commerce and convergence. It should enable advanced mobile wireless technology to identify the main problems associated with connecting in a seamless way through wireless mobile and fixed IPv6 infrastructures.

The objectives of the 6WINIT project are to validate the introduction of the new mobile wireless internet in Europe - based on a combination of the new Internet Protocol version 6 (IPv6) and the new wireless protocols (GPRS and UMTS/3GPP). Such an offering will solve the current problems of the dual scarcity in the IP and wireless world of the following: IP address limitation, quality of service and security from the IP side and lack of bandwidth from the wireless side. By piloting some important mobile applications in both healthcare and business environments, we expect to take a major step towards the eEurope vision.

Further information: 6WINIT Project or contact Tim Chown

Linking smart spaces

Incorporating work from many of the projects outlined here, we are currently researching issues that develop the state of the art in collaborative mediated spaces. An example of this research is our work towards enabling distributed e-Science collaboration through the novel application of knowledge technologies. In particular, we harness our work on ontologies to enhance media-rich annotations of group problem solving.

This work will develop and deploy our work on planning and knowledge-based task support to enhance issue-based process/activity discussions. We will utilise work on scholarly discourse and argumentation to enhance collaborative meeting structures, and our work on presence and visualisation to enhance group peripheral awareness at a distance.

Further information: contact David De Roure

Bringing Enhanced Routing to Multimedia Users in Distributed Academia

Internet Protocol Version 6, or IPv6, is the successor to the current version of the Internet Protocol (IPv4) which has grown with the Internet since the 1970's. IPv4 was not designed with the current rapid expansion and pervasiveness of IP in mind. While IP has outlived most other network protocols, it too is now showing signs of strain; IPv4 does not scale to deliver the Internet of tomorrow.

In development since the early 1990's, IPv6 as now reached the stage where the core specifications have been finalised, and where initial vendor implementations exist. With initial trial deployments happening around the world, UKERNA is seeking to gain some early deployment experience to assist in drawing up a roadmap for IPv6 on UK academic networks.

The success of IPv4 cannot be underestimated, yet it also poses a major obstacle for the roll-out of IPv6. The issue of IPv6 deployment, and in particular IPv6 transition and integration, is an important area of study. Bermuda 2 considers these issues both from the perspective of the UK academic network backbone and of sites on the network. How should a University plan adoption of IPv6? What are the implications of the services to be enabled by the backbone provider? These questions form the core issues of the project's research.

Further information: Bermuda2 Project or contact Tim Chown

Location-sensitive mobile information

The aim of the LinkMe project is to look at the issues surrounding delivery of media and hypertext data to mobile devices within a framework of Quality of Service and location awareness. Specificially, via our colaboration with Ordnance Survey, we look at the delivery of appropriate map and data in a timely manner as an example application. By applying a model of context to the user's requests we're able to tailor the data sent to maximise the usefulness of a low bandwidth connection.

Some of the infrastructure built for LinkMe allows us to explore how we can meld mobile devices, location and ad hoc networking into a rich information environment. In this experiment we have in mind a world where relevant information fragments exist on devices everywhere and the challenge is to locate, assemble and present the information to the user.

Further information: contact Danius Michaelides

Pan-European IPv6 Deployment

The group is also participating in two large (~15M Euro) three-year European projects due to start by 1st January 2002. In the first, 6NET, the group will be leading the studies of IPv6 transition methods, including those that can be run by end users and sites to enable interoperability with IPv4 services. In the second, Euro6IX, the group will be developing IPv6 messaging systems, targeted at end devices that can leverage the huge IPv6 address space for peer-to-peer messaging.

Managing intrusiveness

The FEEL project - Non-intrusive services to support Focussed, Efficient and Enjoyable Local activities - deals with the core problem of intrusiveness of today's mobile technology and how work in local environments can be enhanced by introducing the idea of non-intrusive services realised partly by disappearing computer environments.

The case under consideration is the conflict between activities in the local physical and social context and parallel activities in distributed contexts. When technologies for distributed activities develop, in particular the rapidly developing mobile technology, a larger part of our time will be spent connected to our various distributed contexts. When we meet physically we bring technology, both artefacts and services, that enable us to participate in these non-local contexts, e.g. cellular phones, laptops, PDAs, etc., and the tendency now is that they intrude more and more on the local situations.

Further information: Principal project page or contact David De Roure

Towards Knowledgeable Devices

In conjunction with the Advanced Knowledge Technologies interdisciplinary research collaboration, we are pioneering the application of Knowledge Engineering techniques to wearable and pervasive computing devices. So-called 'knowledgeable devices' exploit in-built reasoning ability and an agent architecture to acquire, and reason over, ontologically structured 'information in context'. They gain knowledge both by explicit information exchanges, and sniffing public information exchanges between neighbouring devices. The ambient wireless information environment also forms a contextual cue, which, combined with conventional physical contextual inputs, will allow knowledgeable devices to construct rich situation profiles.

Our initial research is concerned with the role such devices have in supporting ad-hoc meetings. We are developing a test-bed PDA application, i-See!, which will use situation profiling to deliver relevant resources to meeting participants. We envisage that future generation knowledgeable devices will take a more proactive role, co-operating to, for example, increase the occurrence of useful meetings by subtle manipulation of their physical and information environments.

Related information: AKT IRC site or contact Richard Beales

Systems security

Within the PCaN theme we have participated in a number of JISC-funded security-related projects. These include the deployment of firewalls in academic environments, and customised anti-virus systems. As the Internet paradigm shifts back towards its original always-on mode, but in a massively larger scale, the security and integrity of systems and the data they hold becomes an increasingly important area. Another JISC project covers authentication systems, in which we consider both certificate technologies and methods to distribute such tokens securley.

Temporal linking

Temporal media streams, as exemplified by Internet telephony, radio and TV, and interactive TV, are set to become a significant part of the multimedia information space. We are developing a new set of formal models that enable the application of open hypermedia principles and technologies to such media.

Models exist in other research that have been applied successfully to multimedia documents including stored temporal media, but there has been little work with streamed, or live media. This project addresses the link storage, transmission and resolution architecture for both point-to-point streams and multicast, and it explores the quality of service issues associated with branching media. We also take a longer-term view by considering connectionless networks, addressing navigation in the context of a pervasive computing environment.

Related information: contact Don Cruickshank

Asynchronous link services

We are interested in using distributed hypermedia link services as Contextual Information Systems that provide context-relevant messages to mobile users. An example scenario that we are working with is where the systems uses the geographical location of a mobile user, the capabilities of the user's devices, the time, and user preferences as contextual cues that scope the level of detail, format and manner of presentation of information.

Given the dislocated nature of the devices and services being studies, we are investigating the use of asynchronous message queues (e.g. IBM's MQseries product) to store these results until the user is in a network accessible environment.

Further information: contact David De Roure

Enabling webs on the fly

The last few years have seen the emergence of suitable infrastructure technologies that enable pervasive computing. For example, Next generation Internet Protocol (IPv6) supports many aspects of pervasive computing, including the address space for large numbers of devices, and there is ongoing activity within the IETF (Internet Engineering Task Force) in support of mobility, ad hoc networking and automatic configuration. There has also been a huge increase in the availability, diversity and richness of digital content, which compounds the difficulties of content delivery and, more importantly, navigation in the increasingly pervasive setting.

We are exploring mechanisms that enable navigation of hypermedia data in an ad hoc multimedia information space. In essence, the work centralises around the concept of automatic generation of hypermedia links between documents and other information resources brought together on-the-fly and the services infrastructure required to enable this.

Further information: contact Mark Thompson

Ultra-large scale computing

When looking at large scale pervasive computing systems, there can be many, possibly inumerable, interacting computational devices. There are currently few paradigms for programming distributed systems of such a large scale. Through simulations, new paradigms can be created and explored without the prohibitive cost of manufacture and configuration of physical devices.

The challenge of this ultra-large scale computing is to reliably obtain a desired behaviour by engineering the cooperation of many parts, without assuming any precision interconnect or precision arrangement of the parts, and - through simulation - invent the computational substrate that can support this kind of engineering.

Further information: contact John Revill

Internet conferencing

The UKERNA VIP Project saw the group deploy and report on security issues in IP-based (H.323) videoconferencing systems. As a follow-up, we have also participated in the UKERNA H.323 Architecture working group, which is defining an H.323 service for the UK academic community. In parallel with this activity, we have participated in UKERNA's Quality of Service (QoS) Think Tank, assisting UKERNA to determine appropriate QoS methods to deploy on the UK academic network. QoS requirements also filter right down to the local network level, where in some cases (for example in 802.11b wireless LANs), the bandwidth restrictions may have a dramatic effect on application performance.

Continuous metadata

As enabling technologies become available there is an increasing use of temporal media streams, such as audio and video, within a hypertext context. In this research theme we explore the rationale and requirements for delivering continuous metadata alongside the media stream, and focus on linking as our case study.

We consider the mechanism for delivery of the metadata across a distributed system, the format and content of the metadata flow itself, and the presentation of the media and augmenting metadata to the user. Two initial proof of concept applications have been developed to demonstrate these concepts, and are in use in other research activities. Part of our on-going work is the formalisation of a framework for highly distributed delivery and processing of multicast continuous metadata, as a part of the infrastructure necessary to provide a more complete multimedia environment for hypermedia systems.

Further information: contact Kevin Page

Augmented Reality and Hypermedia

Applying Adaptive Hypermedia techniques to Augmented Reality museum tour guide applications promises great advances in presenting material on museum exhibits. Information selection and presentation can be adapted according to the visitor's goals, preferences, knowledge, and interests and this information can be overlaid over the real object and its features. Existing mobile AR tour guides tend to present information using hand held devices or display hard coded, predefined labels over the object's features. Our initial work has been on developing a system that can provide dynamic, adaptable information overlaid on objects.

Further information: contact Patrick Sinclair

ECS Intranet:Pervasive Computing and Networks