Image Processing Reference
In-Depth Information
30
Register value
R
TX
R
TX
+ σ
SW
R
TX
- σ
SW
Regression
R
TX
=
25
1,3
d
20
15
10
5
0
5
50
100
150
200
250
300
350
400
Distance
d
between sender and receiver (cm)
FIGURE .
Register value
R
TX
of minimal transmission power
P
TX
,min
depending on the distance (ESB,
measurements).
S
FIGURE .
Schematic of the Active Bat system. (From Blumenthal, J., Resource aware and decentral localization of
autonomous sensor nodes in sensor networks, PhD thesis, University of Rostock, Rostock, Germany, , submitted.
With permission.)
All positioning systems are based on one of the before explained distance measurement techniques
(Section .) or a combination of them. So far, these positioning systems are not applicable in large
sensor networks, but sufficient for ad hoc networks with only some hundreds nodes.
One of the first systems was “Active Badge,” which was developed by the Olivetti Research Lab-
oratory [Wan]. It uses badges, mounted at persons moving within a building. hese badges send
every s an infrared signal, which allows to identify the specific person. hese signals are received
by sensors, installed at the ceiling of the room and then, the signals are forwarded to a central server.
The server stores incoming signals with the sensor's location and thus the person's last position.
Infrared communication is only useful for short distances, because external light sources interfere
thesignal.ForthatreasonActiveBadgewasimprovedbydistancemeasurementswithultrasound.
This new “Active Bat” system estimates the position of persons by lateration with distances and
known sensor positions, which is demonstrated in Figure .. As a result, the precision of the older
Active Badge system could be improved to about cm in % of all cases [HHS
+
]. Finally, another
