Distributed algorithm implementation and interaction in Wireless Sensor Networks
L. L. Shum, I. Wokoma, Toks Adebutu, A Djafari Marbini, L. Sacks, Matthew Britton · Adelaide Research & Scholarship (AR&S) (University of Adelaide) · 2005
The SECOAS project is exploring the use of wireless sensor networks as an alternative to traditional oceanographic data acquisition methods for monitoring harsh and inaccessible environment. In this paper we present our ap-proach to the development of the distributed applications that runs on the sensor nodes to give collaborative processing, with an emphasis on creating clusters of nodes with similar spatial attributes. The use of distributed algorithms fulfil the requirements of a self-organising, self-managing and self-optimising network, and provide the extra benefit of the simple addition, removal and alteration of functionality. Our approach is illustrated with a description of clustering and auto-positioning algorithms to show self-organisation and emergence of global order by applying simple local rules. Dissemination of operational parameters between algorithms are realised by the gossip protocol. The distributed algorithms are supported by the light-weight operating system in SECOAS which is called kOS. The way this platform supports the algorithms as well as collaboration between them is discussed in the paper.