Header

UZH-Logo

Maintenance Infos

Deferred decentralized movement pattern mining for geosensor networks


Laube, P; Duckham, M; Palaniswami, M (2011). Deferred decentralized movement pattern mining for geosensor networks. International Journal of Geographical Information Science, 25(2):273-292.

Abstract

This paper presents an algorithm for decentralized (in-network) data mining of the movement pattern flock amongst mobile geosensor nodes. The algorithm DDIG (Deferred Decentralized Information Grazing) allows roaming sensor nodes to ‘graze’ over time more information than they could access through their spatially limited perception range alone. The algorithm requires an intrinsic temporal deferral for pattern mining, as sensor nodes must be enabled to collect, memorize, exchange, and integrate their own and their neighbors’ most current movement history before reasoning about patterns. A first set of experiments with trajectories of simulated agents showed that the algorithm accuracy increases with growing deferral. A second set of experiments with trajectories of actual tracked livestock reveals some of the shortcomings of the conceptual flocking model underlying DDIG in the context of a smart farming application. Finally, the experiments underline the general conclusion that decentralization in spatial computing can result in imperfect, yet useful knowledge.

Abstract

This paper presents an algorithm for decentralized (in-network) data mining of the movement pattern flock amongst mobile geosensor nodes. The algorithm DDIG (Deferred Decentralized Information Grazing) allows roaming sensor nodes to ‘graze’ over time more information than they could access through their spatially limited perception range alone. The algorithm requires an intrinsic temporal deferral for pattern mining, as sensor nodes must be enabled to collect, memorize, exchange, and integrate their own and their neighbors’ most current movement history before reasoning about patterns. A first set of experiments with trajectories of simulated agents showed that the algorithm accuracy increases with growing deferral. A second set of experiments with trajectories of actual tracked livestock reveals some of the shortcomings of the conceptual flocking model underlying DDIG in the context of a smart farming application. Finally, the experiments underline the general conclusion that decentralization in spatial computing can result in imperfect, yet useful knowledge.

Statistics

Citations

4 citations in Web of Science®
6 citations in Scopus®
Google Scholar™

Altmetrics

Downloads

95 downloads since deposited on 23 Dec 2011
14 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Journal Article, refereed, original work
Communities & Collections:07 Faculty of Science > Institute of Geography
Dewey Decimal Classification:910 Geography & travel
Language:English
Date:2011
Deposited On:23 Dec 2011 08:15
Last Modified:05 Apr 2016 15:13
Publisher:Taylor & Francis
ISSN:1365-8816
Publisher DOI:https://doi.org/10.1080/13658810903296630

Download

Preview Icon on Download
Preview
Content: Accepted Version
Filetype: PDF
Size: 3MB
View at publisher