Project – A network of sensors and algorithms

The project is completed.

Project start 01.01.2007. Expected completion 31.12.2009.

In and around us in our daily environment, for example in the health sector, we find a large amount of sensors that send information about our surroundings e.g. Fire alarms, theft alarms, thermometers, barometers and surveillance cameras. For certain individuals, such as home–bound patients or old age pensioners in care homes, this could in a form of observation for health asons, or possibly the jogger wishing to see the effects of their training. Whichever it may be, these sensors have a relatively high level of intelligence, the full potential of which is rarely used.

The purpose of the project is to gather know-how on technology, implementation techniques and use of sensor networks. The term ´sensor` should be understood in the wider sense i.e. everything from temperature gauges and barometers to advanced microphones, digital vision or heart rhythm monitors.

This opens up an exciting list of challenging projects, first and foremost interoperability, low energy consumption, large data quantity and complicated calculation.

Interoperability is challenging, as many communication protocols are designed for one specific purpose or are proprietary and therefore cannot be used for any purpose other than their originally intended use. Having said that, many systems can be used for other purposes, or can be combined to give many other useful results e.g. heart rhythm and skin temperature sensors as one combined entity would give much more information than a singular sensor for each specific purpose. The challenge with this is that there is often more than one manufacturer and many of the apparatus are not yet designed for everyday use. However, research in digital radio communication has resulted in problem solving solutions, which until only a few years ago would have been complete science fiction, but is possible today e.g. digital radio communication.

A low energy consumption is also important for many reasons, many sensors have to be implanted into ´Body Area Networks` where the energy supply obviously has to be minimal, or cast in buildings in which an energy supply is unavailable, or in which the sensor specific energy requirements are unable to be met (e.g. wireless fire alarms). Apart from these points, energy prices have, of course, an influence on the desire for lower energy consumption.

Many sensors in product technology, for example, generate large amounts of data, these must be handled locally, before being passed on, this could be because of poor resources in data communication e.g. networking broadband or a shortage in data storage.

Many of these tasks require complicated calculations, e.g. digital signal processing, these calculations can either be placed locally in the sensor or on a central station. Whatever the circumstance, consideration would have to be given to where these calculations take place most effectively, both with regard to cost and energy consumption.

Participants