Sensor Abstractions for Opportunistic Activity and Context Recognition Systems

Marc Kurz, Alois Ferscha

Research output: Chapter in Book/Report/Conference proceedingsConference contributionpeer-review

9 Citations (Scopus)

Abstract

Pervasive environments are inherently characterized to draw from sensor infrastructures in order to become situation aware. Very recent technological evolutions of sensor hardware (e.g. for geoposition, acceleration, orientation, noise, light, humidity, chemical properties, etc.) have fertilized an explosive growth of sensor infrastructures, introducing whole new challenges for sensor software architectures like heterogeneity, redundancy and replaceability, fault tolerance, mobility, massive deployment, but most of all frequency of change and spontaneous availability. In this paper we address the very fundamental issue of exploiting spontaneous sensor configurations by introducing mechanisms for sensor self-description, goal-oriented sensing missions and dynamic sensor ensemble management. The concept of "sensor abstractions" is introduced, making the use of physical as well as immaterial sensors transparent from any technical sensor properties. An opportunistic sensor software architecture has been implemented, reversing former architecture principles (e.g. fusion of all available sensors) into a spontaneous, selective, utility driven involvement of sensors based on their sensing mission contribution potentials. The framework is implemented in OSGi, and demonstrated for activity recognition missions.

Original languageEnglish
Title of host publicationSmart Sensing and Context - 5th European Conference, EuroSSC 2010, Proceedings
PublisherSpringer
Pages135-148
Number of pages14
ISBN (Print)3642169813, 9783642169816
DOIs
Publication statusPublished - 2010
Event5th European Conference on Smart Sensing and Context (EuroSSC2010) - Passau, Germany
Duration: 14 Nov 201016 Nov 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6446 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference5th European Conference on Smart Sensing and Context (EuroSSC2010)
CountryGermany
CityPassau
Period14.11.201016.11.2010

Fingerprint Dive into the research topics of 'Sensor Abstractions for Opportunistic Activity and Context Recognition Systems'. Together they form a unique fingerprint.

Cite this