RUNES
Reconfigurable Ubiquitous Networked Embedded Systems

In order to set the context for the consideration of the ways of realising the technical objectives we have set ourselves, let us first consider the application scenarios, which will be used as case studies, architectural requirements starting points and reference points for our technology trials and integration. Note that these scenarios are realistic and have been drawn from real case studies.

Healthcare – The quality and cost of healthcare continue to be significant issues for every nation in the EU and perhaps in the world. One of the emerging concepts to address both of these issues is to enable the provision of healthcare services closer to the patient, even into their homes, where it is often found that recovery is faster given proper care can be provided. A scenario of interest for RUNES, is the ability to provide continuous health information which usually derived from some form of sensor –enabled device from the hospital or clinical environment through the local surgery and into the home. The patient monitoring devices are a network of sensors which provide the same information regardless of location and transition automatically from one network such as that found in a clinic to that found in the patient home with the appropriate network hand-offs, associations, security and privacy that is critically important in the health industry. The RUNES platform and middleware support would enable the controlled, transient distribution across this scenario so that high quality care is possible at any level of healthcare provision. We have a specific demonstrator planned for this area.

Emergency Services – Natural and man-made disasters are recognised as a continual threat to the welfare of individuals, communities and entire regions. Hurricanes, floods and earthquakes are ever-present in our world as are the smaller, but potentially devastating events, such as chemical spills, fires and bombings. Information can be a key element in reducing the loss of life and destruction that such events cause. Coordinated information through the use of wide area networks of sensors and embedded devices can provide the means to insure that fire fighters have accurate area maps, that emergency officials are instantly alerted to new indicators of seismic dangers regardless of their location, and that necessary multimedia communications including satellite imagery can be provided to those responding to such events. The RUNES technology would need to enable the distribution of the sensor-based information in a controlled manner combined with the interpreted information to the most appropriate node or location with stringent service guarantees.

Factory Automation – Manufacturing operations are pressed to produce at lower and lower costs forcing more efficient processes in order to compete in the global marketplace. Coupled with this is the need for flexibility, to be able to attract new business. Therefore, flexibility is the third key driver to be considered in automating a factory. The move is towards shorter product lifecycles and greater customisation. This means manufacturing facilities have to be more flexible with the corresponding knock-on effect for the control and information systems. Being able to reconfigure and adapt control and information systems is something, which will be directly facilitated by RUNES. Process manufacturing, for example, must continually research and implement new ways to reduce waste, reduce time of operations and improve volumes. Sensor-based process control has been one technological way to enable this. Through the use of networked sensors, and process control algorithms, a manufacturing operation can quickly and efficiently alter its operational parameters to improve performance or to enable the addition of a new flow entirely. Improving performance can in some cases, mean millions of euros are saved in reduction of waste or higher production. The issue with these systems is that they often need to adapt to an existing environment, be retrofitting into existing large capital assets, or need to cover vast industrial sites with heterogeneous network engineering. The RUNES technology must deliver the ability to enable distributed process control with ad-hoc and highly robust networking in a very challenging environment. For example, manufacturing plants can be particularly difficult to network in a cost-effective manner and may have specific RF interference issues, or time dependent process operations. To provide a generic architecture, which supports such diversity will provide a worthwhile challenge.

Retail Settings – Sensor and embedded systems are already found in great numbers within the retail environment. Inventory tags, price checking and other consumer retail tasks are currently enabled by such devices. The next generation of retail environment will be the automated shop where sensor-embedded tags of the shelves report their own inventory to centralised purchasing systems, which place orders for goods through the Internet. In addition, already there are large grocery chains in the US, which are automating the check out process through the use of smart tags or sensors on goods. A network of embedded sensors provides the means to control and manage the lowest level of the supply chain, which is often the most costly. The RUNES technology will need to scale from single and highly specialised sensor input to the high volume, highly distributed yet simplistic sensor technology. The control and management of an adaptive sensor network may pose significantly different challenges in the simple, high volume sensor case as it does in the low volume, high specialised sensor case. And wherever commerce and hence pricing is involved, security and accuracy are of paramount importance, these will be critical to success of any generic architecture supporting this usage.

In-home safety and security – On a smaller, but no less important scale is the safety and security of the individual. The home is often mentioned as the single most dangerous location for many reasons. Security in the home has already become a networked system using the public network to connect the home to private security companies for monitoring and emergency service. These companies are moving to wireless networks and providing access to the public Internet. There are other aspects of safety and security that will be enabled by network embedded systems technology. For example, carbon monoxide sensors to detect unsafe levels emitted from home heating systems are already available. Enabling this sensor information across the in-home network and on to an emergency number would enable a request for help to be made on behalf of the inhabitants who may be unable to do so due to carbon monoxide poisoning. Other sensors or “sensor” devices can also join a home cluster segment of the RUNES architecture. Sensors such as mobile phones for on-demand surveillance across the Internet are feasible allowing a person to capture an image or video and view it online from their home at a very low cost. Sensors in personal vehicles can also connect to the home middleware and control network, so that maps and information acquired inside the home can be easily distributed to an in-car sensor network. Important information like route plans, insurance data, vehicle maintenance records and current status can be freely communicated between the sensors in the home and those in vehicle.

The project will concentrate on the design and development of a platform independent model of these key scenarios, which will provide the basis for understanding the architectures, design attributes and other engineering parameters required for such systems. Non-functional requirements such as scalability, security, and heterogeneity will be drawn from these applications and will condition the architectural decisions on the framework. The model will be accessible and reusable by other entities, i.e. businesses and academic institutions, which might be interested in creating RUNES products and systems in the future.