Location Fingerprinting for Enhanced Positioning of Mobile Terminals in Cellular Networks

Abstract : In regards to the emerging interest for Location Based Services (LBS), this thesis was initiated with the goal of providing "low-cost" and "continuous" LBS to the end users. The Location Fingerprinting (LFP) method has been adopted as the main axis of our studies, where it is investigated in a machine learning perspective. As the first major contribution of this dissertation, we tackled the problem of radio database compression in LFP systems, by reducing the number of records. The cluster analysis in this work was proposed with the goal of reducing the computation and transmission loads, in order to decrease the terminal power consumption in mobile-based LFP systems. We proposed to perform the compression by applying a "clustering" process during the training phase. Once the standard clustering methods were examined, as a next step, a clustering algorithm well-tailored to the structure of the radio database was proposed. In the next part of the thesis, we tackled the problem of missing data in the RSS-based fingerprinting systems. A specific missing mechanism was proposed to describe the missingness occurring in RSS measurements, issued from the 3GPP-defined scanning process (as in 2G and 3G). Our modeled missing mechanism proceeds based on two parameters: the receiver minimum sensitivity for signal detection, and the maximum number of base stations to be measured in the radio measurements. Next, statistical methods were developed at two different levels, to deal with missing data.
Complete list of metadatas

Cited literature [180 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/pastel-00671865
Contributor : Azin Arya <>
Submitted on : Monday, February 27, 2012 - 12:49:53 PM
Last modification on : Monday, February 25, 2019 - 11:56:01 AM
Long-term archiving on : Wednesday, December 14, 2016 - 7:18:19 AM

Identifiers

  • HAL Id : pastel-00671865, version 1

Citation

Azin Arya. Location Fingerprinting for Enhanced Positioning of Mobile Terminals in Cellular Networks. Networking and Internet Architecture [cs.NI]. Télécom ParisTech, 2011. English. ⟨pastel-00671865⟩

Share

Metrics

Record views

592

Files downloads

843