Research Article
Adaptive pruning of event decision trees for energy efficient collaboration in event-driven WSN
@INPROCEEDINGS{10.4108/ICST.MOBIQUITOUS2009.6856, author={Steffen Ortmann and Michael Maaser and Peter Langendoerfer}, title={Adaptive pruning of event decision trees for energy efficient collaboration in event-driven WSN}, proceedings={3rd International ICST Workshop on Information Fusion and Dissemination in Wireless Sensor Networks}, publisher={IEEE}, proceedings_a={SENSORFUSIONS}, year={2009}, month={11}, keywords={Analytical models Collaboration Costs Decision trees Energy efficiency Event detection Medical services Monitoring Terrorism Wireless sensor networks}, doi={10.4108/ICST.MOBIQUITOUS2009.6856} }- Steffen Ortmann
Michael Maaser
Peter Langendoerfer
Year: 2009
Adaptive pruning of event decision trees for energy efficient collaboration in event-driven WSN
SENSORFUSIONS
IEEE
DOI: 10.4108/ICST.MOBIQUITOUS2009.6856
Abstract
Wireless sensor networks (WSN) are considered to be the key-enabler for low cost highly distributed applications in the area of homeland security, healthcare, environmental monitoring etc. A necessary prerequisite is reliable and efficient event detection. This paper introduces a novel approach for event configuration and in network processing, called event decision trees (EDT). An EDT enables every node to self-divide event queries according to its resources. EDT autonomously adapt to the tasks assigned, even though it requires to organize collaboration between nodes to deliver expected results. The effort for maintain formal EDT is evaluated by analysis and simulations. Our results show that the proposed lease-based mechanism for maintaining producer/consumer pairs in an EDT outperforms even idealized acknowledgment-based approaches.