Abstract:
The importance of services, based on current location of objects is swiftly increasing.
Mostly, commerce and business activities are conducted inside buildings. Global
Navigation Satellite System (GNSS) cannot provide object position inside buildings,
since for object position determination GNSS not only requires direct line of sight to
the object, but also GNSS signals are badly affected by multipath. In near future, the
Location Based Services (LBS) will be tremendously dependent on Indoor
Positioning System (IPS). In order to develop a 3D IPS based on Wi-Fi Received
Signal Strength (RSS) in a heterogeneous environment, the following approach was
used. Parallel RPROP and greedy algorithm were combined for development of IPS
using Received Signal Strength (RSS) in heterogeneous environment, the
environment comprised of human activity, walls' material, cupboards, and various
type of surveying machines and equipment. The propagation of Wi-Fi signal varies
directionally; therefore, to cope with direction changes in signals; our proposed model
produces three sets of weights, which could be considered best for easting, northing
and height, respectively. Our model was trained with 75% of collected data and tested
on remaining 25% data. Distance error was used for accuracy assessment. Through
our experiments, we achieved maximum accuracy of 0.87m and we came to know
that median error was less than mean error. Median error were about 3.32m and the
mean error was about 4.62m. Based on our results, we recommend the use of parallel
RPROP and greedy algorithm for 3D position determination in heterogeneous
environment.
