Computing perception from sensor data

This paper describes a framework for perception creation from sensor data. We propose using data abstraction techniques, in particular Symbolic Aggregate Approximation (SAX), to analyse and create patterns from sensor data. The created patterns are then linked to semantic descriptions that define th...

Full description

Saved in:
Bibliographic Details
Main Authors: Barnaghi, P., Ganz, F., Henson, C., Sheth, A.
Format: Conference Proceeding
Language:English
Subjects:
Cognition
Data mining
Humans
Internet
Ontologies
Semantics
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper describes a framework for perception creation from sensor data. We propose using data abstraction techniques, in particular Symbolic Aggregate Approximation (SAX), to analyse and create patterns from sensor data. The created patterns are then linked to semantic descriptions that define thematic, spatial and temporal features, providing highly granular abstract representation of the raw sensor data. This helps to reduce the size of the data that needs to be communicated from the sensor nodes to the gateways or highlevel processing components. We then discuss a method that uses abstract patterns created by SAX method and occurrences of different observations in a knowledge-based model to create perceptions from sensor data.
ISSN:1930-0395
2168-9229
DOI:10.1109/ICSENS.2012.6411505