The cellular network's signal-aware fingerprint-based positioning technique (SAFPPT)

Chung Ming Huang, Shih Yang Lin, Tsung Han Hsieh

Research output: Contribution to journalArticlepeer-review


The Fingerprint-based positioning technique provides an alternative choice for positioning. In order to meet different requirements for traveling, e.g.; positioning within a specific area or detection some approaching Point Of Interests (POIs), a Signal-Aware Fingerprint-based Positioning Technique (SAFPPT) is proposed in this paper. SAFPPT contains four positioning methods based on the signal/information of cellular network's base stations: (i) Positioning Method of Line (PMoL), (ii) Positioning Method of Plane (PMoP), (iii) Approaching Detection Method of Point (ADMoP), and (iv) Approaching Detection Method of Line (ADMoL). The basic idea is that SAFPPT uses user's cellular information to find the best match records in the pre-established Fingerprint database. SAFPPT can be used in different scenarios of the interested area: (i) Point of Interest (point), (ii) road (line), and (iii) region (plane). The experimental results show that (i) the positioning accuracy of PMoL and PMoP are higher than Google's "My Location", (ii) some parameters may affect the positioning accuracy of PMoL, e.g.; the moving speed of a user and the number of samples of the Fingerprint database, (iii) the stayed time length of a user may affect the positioning accuracy of PMoP, (iv) ADMoP and ADMoL have higher hit rates to determine the corresponding POIs that a user is approaching within the 150-m radius of the approaching range.

Original languageEnglish
Pages (from-to)233-260
Number of pages28
JournalWireless Personal Communications
Issue number1
Publication statusPublished - 2014 Mar

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Electrical and Electronic Engineering


Dive into the research topics of 'The cellular network's signal-aware fingerprint-based positioning technique (SAFPPT)'. Together they form a unique fingerprint.

Cite this