A hybrid scheme for energy-efficient object tracking in sensor networks

Ming Hua Hsieh, Kawuu W. Lin, S. Tseng

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

Energy saving is a critical issue in many sensor-network-based applications. Among the existing sensor-network-based applications, the surveillance application has attracted extensive attention. Object tracking in sensor networks (OTSNs) is a typical surveillance application. Previous studies on energy saving for OTSNs can be divided into two main approaches: (1) improvements in hardware design to lower the energy consumption of attached components and (2) improvements in software to predict the movement of objects. In this paper, we propose a novel scheme, namely hybrid tracking scheme (HTS), for tracking objects with energy efficiency. The scheme consists of the two parts: (1) adaptive schedule monitoring and (2) a recovery mechanism integrated with seamless temporal movement patterns and seeding-based flooding to relocate missing objects with the purpose of saving energy. Furthermore, we also propose a frequently visited periods mining algorithm, which discovers the corresponding frequently visited periods for adaptive schedule monitoring efficiently from the visitation information of sensor nodes. To decrease the number of sensor nodes activated in flooding, a seeding-based flooding mechanism is first proposed in our work. Empirical evaluations of various simulation conditions and real datasets show that the proposed HTS delivers excellent performance in terms of energy efficiency and low missing rates.

Original languageEnglish
Pages (from-to)359-384
Number of pages26
JournalKnowledge and Information Systems
Volume36
Issue number2
DOIs
StatePublished - 1 Aug 2013

Keywords

  • Data mining
  • Object tracking sensor networks
  • Recovery mechanism
  • Schedule monitoring
  • Temporal movement pattern

Fingerprint Dive into the research topics of 'A hybrid scheme for energy-efficient object tracking in sensor networks'. Together they form a unique fingerprint.

Cite this