Image Processing Reference
In-Depth Information
improvement came along with the “Cricket” system that has integrated additional hardware on the
badges whereby the lateration process can be distributed [PCB].
Another system is “RADAR,” which assumes a preinstalled Wi-Fi infrastructure [BP]. RADAR
offers two methods for the localization process. he first one is using signal strength measurements
and “signal to noise ratios” (SNRs) of Wi-Fi signals to estimate distances, which are basis for
lateration. he second method is comparing latest measurements with stored measurements from a
signal map, which was built in a preprocess. Systems that use this kind of pattern matching techniques
are “MotionStar” [RBSJ] and “Locus” [KH].
Since , the GPS has been a popular localization system used by military as well as civil
applications. It achieves a precision of some centimeters. GPS works very well for outdoor environ-
ments and is limited in indoor environments, depending on the number of satellites in direct sight.
Measuring the signal's ToF to geostationary satellites from a receiver is a basic requirement of GPS.
With distances to satellites and their positions, GPS receiver estimate their own position via lateration
[Gib]. The standard version of GPS has been extended by several improvements like differential
GPS, where fixed base stations on the ground allow error correction. Beside, the American GPS a
similar system is the Russian GLONASS, which was developed for the Russian Ministry of Defense.
Since years, the European Union has been working on a satellite navigation system with  satellites
that is planed to be ready in  [BCKM].
Loran-C works similar like the before mentioned systems except that base stations are placed along
the coast in a chain structure [lor]. his system is mainly used for marine navigation.
More systems based on lateration are SpotOn [HWB], a Bluetooth-based system developed by
Feldmann [Fel] and D-iD [WL]. Other systems that must be mentioned are LANDMARC,
which uses radio frequency identification (RFID) [NLLP] or Ubisense that combines lateration
and angulation with ultrawideband communication [ubi]. For more systems, [HB] gives a broad
overview.
All discussed positioning systems are not feasible on tiny and resource constrained sensor nodes.
Their energy consumption is too high and they assume a preinstalled infrastructure of base sta-
tions. Currently, only GPS is qualified, due to its global availability. he following problems are still
existent:
. Costs : he financial expense for a large network is not acceptable.
. Size : he size of nodes equipped with GPS is very high.
. Energy : GPS modules are energy intensive and require a current flow of milli amperes.
. Application : GPS requires direct line of sight to at minimum four satellites. hus, it does
not work in buildings.
For these reasons, numerous research groups develop special localization algorithms that are
adapted to the special sensor node's hardware and application demands.
6.3.2 Design Considerations
Resource-aware localization can only be achieved by carefully analyzing its behavior on all network
layers. For example from a sensor node's view, the distributed localization on every node is privileged
for robustness and reliability. But, a global network view avoids systematic errors and redundancy.
Generally, the following requirements should be fulfilled:
An estimated position must have a high precision, respectively a small localization error.
Position information must be provided as soon as possible, immediately after network
initialization, because many other protocols are based on this data.
 
Search WWH ::




Custom Search