Virtual Ruler: Mobile Beacon Based Distance Measurements for Indoor Sensor Localization

1
Virtual Ruler Mobile Beacon Based Distance
Measurements for Indoor Sensor Localization
Department of Computer Science
EngineeringCollege of Engineering
Yong Ding, Chen Wang Advisor Li Xiao
Excluding Outliers We let the virtual ruler
measure each pairwise distance multiple times
from different perspectives, and select the
distance value with the highest frequency. We
also assign a confidence value for each distance
based on some statistical features. The distance
with higher confidence is given higher priority
in the localization algorithm. Therefore, a
subset of high confidence distances is used by
the localization algorithm to localize each
sensor.
Introduction In sensor localization, ultrasound
based distance measurements will have large
errors if line-of-sight paths are blocked between
pairwise sensors. Because these outliers, once
mixed together with other correct distance
measurements, are difficult for localization
algorithms to identify, we propose to exclude the
outliers in the first step of distance
measurement. We use mobile beacons to measure the
distance between pairwise sensors from multiple
perspectives and filter incorrect values through
a statistical approach. Our performance
evaluation shows that the proposed algorithm can
achieve better localization results than previous
approaches in an indoor environment where
multipath effects cannot be avoided.
Virtual Ruler Approach We attach two beacons at a
fixed distance away on a small vehicle which
randomly moves around in the deployed area and
measures distances between pairwise sensors.
The relative positions of sensors n can be
inferred by their measured distances from the
mobile beacons m. Thus, distances between sensors
can be calculated.
Performance Evaluation We perform the simulation
in an obstructed environment where 50 sensors are
randomly deployed.
Accuracy We perform some experiments to study the
accuracy that virtual ruler can achieve in the
line-of-sight condition.
The virtual ruler randomly moves around the area
for 100 steps, and measures distances between
pairwise sensors.
Ranging Methods RSS (Radio Signal Strength)
method estimates the distance between two nodes
by assuming a known rate of signal attenuation
over distance and measuring the received radio
signal strength. Ultrasound TOF (Time of Flight)
method estimates distance by assuming a constant
speed of ultrasound and measuring the time it
takes for an ultrasound signal to travel from
transmitter to receiver.
The accuracy of virtual ruler is related with the
length of vehicle. When the length of vehicle is
0.8m, the relative error of virtual ruler is
within 3.
By using confidence ranking, 201 distances are
selected which have only 2 incorrect distance
measurements (plotted in blue lines).
Motivation Although Ultrasound ToF is a more
accurate and robust distance measurement approach
than RSS in the indoor environment, It must
address two challenges the short measurable
range and the outliers due to the multipath
effect of ultrasound in obstructed environments.
Multipath Effect When the line-of-sight between
mobile beacons and sensors is blocked, virtual
ruler may obtain incorrect distance measurement.
The Iterative Least Square Fitting approach
filters outliers in the localization algorithm
layer, while our approach filters from distance
measurement layer. Therefore, our approach can
achieve better localization accuracy.
The sensors incorrect position, estimated from
the distance measurement along a reflected path
instead of a straight line, is mirrored to the
sensors correct position.
The incorrect distance is the virtual distance
between the image n1 and real location n2 of
sensors.
04/18/2006