A Robust Positioning Method using WiMAX Fingerprints
Date Issued
2012
Date
2012
Author(s)
Hsu, Wei-Ting
Abstract
Due to the rising amount of mobile devices and location-based services (LBSs), the accuracy and efficiency of localization become a more and more important issue. Global Positioning System (GPS) is the most well-known positioning techniques; it can provide highly accurate location results in open sky environments, but its accuracy and availability declines in outdoor urban canyons. Beside GPS, radio signals are used for localization both in practice and in research; these signals include Global System for Mobile Communication (GSM), Wi-Fi and Worldwide interoperability for Microwave Access (WiMAX).
Related work of wireless positioning mostly rely on signal strength, but we know that signal strength differs in different weathers; reflection and signal decay when passing through obstacles are also problems. Thus, we proposed a robust positioning method using the hybrid of a deterministic method K nearest neighbor(s) (KNN) and two novel techniques including Grid average Signal Strength (SS) Matching and Signal Difference Matching for location estimation, which puts weather conditions into concerns. Also, we performed a complete experiment and done accuracy and robustness evaluation of our positioning method along with other deterministic and probabilistic techniques. Our positioning method reaches a median error 52 meters, which is much superior than the accuracy requirement (latitude and longitude of emergency callers within 300 meters) of Wireless enhanced 911 from the U.S. Federal Communications Commission (FCC), and shows better accuracy and robustness than others; it doesn’t require any pre-knowledge of base stations and can be used on GPS-free mobile device or supporting GPS.
Subjects
Positioning
Positioning Accuracy
Robust
Weather
Signal Delta
Signal Difference
Location
Localization
Fingerprinting
Deterministic
Probabilistic
LBS
Mobile Device
Transportation
SDGs
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-101-R99725052-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):75239d8840971a67084e999cd48537c1