DocumentCode :
3517398
Title :
Toward easily analyzable sensor networks via structuring of time-triggered tasks
Author :
Kim, Kane ; Li, Yuqing
Author_Institution :
DreamLab, UCI, Irvine, CA, USA
fYear :
2003
fDate :
28-30 May 2003
Firstpage :
344
Lastpage :
351
Abstract :
Software techniques which enable efficient construction of sensor networks that are easily analyzable and can rarely run into a chaotic situation are in high demands. With the motivation for developing such software techniques, a skeleton of a software architecture for sensor networks which uses time-triggered functions (TTFs) extensively has been formulated. The TTF is a fundamental building-block for software which realizes global time based coordination of distributed actions (TCoDA). TDMA communication is adopted and harmonious joining of a sensor node into an existing cooperating group of sensor nodes is facilitated. Approaches for supporting preemptive TTFs by use of a multi-threading OS kernel as well as the approaches for supporting non-preemptive TTFs by use of application-level components are also presented. The latter approaches were devised for the situations where incentives for avoiding the use of a multi-threading OS would exist.
Keywords :
multi-threading; operating system kernels; real-time systems; time division multiple access; wireless sensor networks; OS kernel; TCoDA; TDMA; multi-threading; real time system; sensor network; software architecture; time-triggered task; wireless network; Chaotic communication; Communication channels; Costs; Intelligent sensors; Kernel; Sensor systems and applications; Skeleton; Software architecture; Time division multiple access; Wireless sensor networks;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Distributed Computing Systems, 2003. FTDCS 2003. Proceedings. The Ninth IEEE Workshop on Future Trends of
ISSN :
1071-0485
Print_ISBN :
0-7695-1910-5
Type :
conf
DOI :
10.1109/FTDCS.2003.1204357
Filename :
1204357
Link To Document :
بازگشت