Authors:
Paul Harvey
1
;
Alan Dearle
2
;
Jonathan Lewis
2
and
Joseph Sventek
1
Affiliations:
1
University of Glasgow, United Kingdom
;
2
University of St Andrews, United Kingdom
Keyword(s):
Wireless Sensor Network, Insense, InceOS, nesC, TinyOS, Operating System, Programming Language.
Related
Ontology
Subjects/Areas/Topics:
Platforms and Operating Systems
;
Programming and Middleware
;
Scheduling, Tasking and Control
;
Sensor Networks
;
Software and Architectures
Abstract:
To support the programming of Wireless Sensor Networks, a number of unconventional programming models have evolved, in particular the event-based model. These models are non-intuitive to programmers due to the introduction of unnecessary, non-intrinsic complexity. Component-based languages like Insense can eliminate much of this unnecessary complexity via the use of active components and synchronous channels. However, simply layering an Insense implementation over an existing event-based system, like TinyOS, while proving efficacy, is insufficiently space and time efficient for production use. The design and implementation of a new language-specific OS, InceOS, enables both space and time efficient programming of sensor networks using component-based languages like Insense.