Modeling and implement of mobile phone user location discrimination based on heuristic strategy

With the all-pervading mobile devices and continuing advancement of big data technologies, mobile phone data research has been gaining widespread popularity in the past few years. Dealing with the implausible location caused by cell handover phenomenon in the communication system is one key problem...

Full description

Saved in:
Bibliographic Details
Published in:EURASIP journal on wireless communications and networking Vol. 2019; no. 1; pp. 1 - 9
Main Authors: Shan, Qingchao, Dong, Honghui, Jia, Limin, Yuan, Hua, Zhang, Hui
Format: Journal Article
Language:English
Published: Cham Springer International Publishing 02.09.2019
Springer Nature B.V
SpringerOpen
Subjects:
Algorithms
Big data mining
Cell oscillation phenomenon
Cell phones
Cellular telephones
Clustering
Communications Engineering
Communications systems
Discrimination
Distributed Big Data Processing in SCADA Communication Networks
Electronic devices
Engineering
Handover
Heuristic strategy
Information Systems Applications (incl.Internet)
Location discrimination
Mobile phone data
Modelling
Networks
Signal,Image and Speech Processing
Mobile phone data
Big data mining
Handover
Heuristic strategy
Cell oscillation phenomenon
Location discrimination
ISSN:1687-1499, 1687-1472, 1687-1499
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the all-pervading mobile devices and continuing advancement of big data technologies, mobile phone data research has been gaining widespread popularity in the past few years. Dealing with the implausible location caused by cell handover phenomenon in the communication system is one key problem of user mobility profile building based on mobile phone call detail records (CDRs) data. In this paper, we propose a location discrimination model aiming at CDRs data, where heuristic strategies for the characteristic of the oscillation phenomenon from practical CDRs and handover categories are added to distinguish the stay points, passing points, and oscillating points. A whole month of CDRs data from one communication operator is employed to select parameters and validate the model on the Spark platform. The experiment results betray that the proposed model can identify the false locations effectively. Compared with the threshold models, the result of the proposed model is more reasonable both in the population aggregate level and individual level. Besides, the model can retain more user’s trajectory points than clustering algorithm, so it can improve the quality of user mobility modeling.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1687-1499
1687-1472
1687-1499
DOI:10.1186/s13638-019-1535-9