tion, speed and so forth into bits and bytes. Increas-
ingly, sensors will bridge the physical and the digital
world, and together with actuators form `context
agents' in ambient intelligence networks. Particular-
ly smart are sensors that can be adapted to new func-
tions and conditions during the operating phase or
even have self-calibration, self-adaptation and self-
Sensors on semiconductor basis and fibre-optic
sensors (e.g. for use in LANs) are far from being new
developments; rather, they have represented enor-
mous growth markets for many years.
troMechanicalSystems (MEMS) technologies play a
key role, making it possible to miniaturise sensors and
integrate their sensor elements with microelectron-
ic functions in minimal space. A benchmark for the
miniaturisation of sensors is the vision of `smart dust',
an autonomous sensing, computing and communica-
tion system called `mote' packaged within one cubic
millimetre (1 mm
There are already a number of ambient intelli-
gence prototypes that incorporate sensor and relat-
ed technologies, such as the `smart-its' developed in
the European project of the same title. Objects fitted
with `smart-its' can recognise their local environment
and share their observations with other objects with-
in the environment.
Another example is SoapBox
(Sensing, Operating and Activating Peripherial Box
for intelligent environments), developed by the Finn-
ish technology centre VTT.
Also important to watch will be the developments
in smart tags. It is expected that the costs for active
Radio Frequency Identification (RFID) tags might
well be brought down from 1 Euro to a few Euro
cents per unit. Coupled with a growth in their capa-
bility, this would lead to an explosion in the applica-
tion of smart tags. Low-cost levels may be reached by
new fabrication methods for silicon-based solutions,
Example: Intelligent carpets
Infineon Technologies AG (Munich, Germany)
develops `Technology Lifestyle Solutions', appli-
cations for everyday life into which electronic
functions are integrated. For example, together
with Vorwerk Teppichwerke (Hamlin, Germany),
Infinion's technology lab developed a `Thinking
Carpet', a networked carpet with a self-organising
network of microchips. Depending on the desi-
red dispersal of detected signals, the carpet carries
from one to 25 sensors per square metre in its bac-
king. Self-organising means: If a sensor fails for
some reason, the neighbouring processors can look
for a new connecting route, which circumvents
the defective region and maintains functionality.
This also makes it possible to cut the carpet, repla-
ce segments or add new ones at any time. The car-
pet allows for simple and space-saving placement
of sensors for a wide variety of functions, e.g. alarm
and climate control. Also guidance systems are pos-
sible through light-emitting diodes in the carpet.
The prototype was presented at the international
Orgatec 2004 trade fair in October 2004 in Colog-
ne, Germany. It is estimated that two years of fur-
ther development will be necessary before the
`thinking carpet' attains market maturity.
Imagine a future museum in which the flooring
(though not a carpet) would sense the visitors, gui-
de them, and also redirect their flow if some exhi-
bition space is already overcrowded.
Sources: "Future requires Visions: Vorwerk Teppich-
werke and Infineon are revolutionising the wall-
to-wall carpet..." (29 October 2004). http://www.
infineon.com (see news archive; the detailed article
also contains several images).
Cf. Intechno: "World
Report: Sensor Markets
2008. Worldwide analysis
and forecasts of sensor
markets until 2008", Basle,
May 1999. http://www.
pdf (gives a concise over-
view of the results).
Cf. "Desirable Dust",
Economist.com, 31 January
id=949030; "Smart Dust.
Mighty motes for medicine,
manufacturing, the military
24 March 2003. http://
VTT SoapBox. http://
06.12.2004 8:37:14 Uhr