Efficient event prewarning for sensor networks with multi microenvironments

  • Authors:
  • Yinglong Li;Hong Chen;Suyun Zhao;Shangfeng Mo

  • Affiliations:
  • Key Laboratory of Data Engineering and Knowledge Engineering, MOE, China,School of Information, Renmin University of China, Beijing, China,Hunan University of Science and Technology, Xiangtan, Chi ...;Key Laboratory of Data Engineering and Knowledge Engineering, MOE, China,School of Information, Renmin University of China, Beijing, China;Key Laboratory of Data Engineering and Knowledge Engineering, MOE, China;Key Laboratory of Data Engineering and Knowledge Engineering, MOE, China,School of Information, Renmin University of China, Beijing, China,Hunan University of Science and Technology, Xiangtan, Chi ...

  • Venue:
  • Euro-Par'13 Proceedings of the 19th international conference on Parallel Processing
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Early detecting the approaching events is the primary way of minimizing their damages in the sensor-based systems. The majority of existing approaches of event description and detection rely on using crisp raw sensory data, which requires large amount of data transmission as well as is memory-consuming, moreover, these approaches are only applicable to homogeneous sensor networks. This paper describes a novel efficient framework for event prewarning in sensor networks with multi microenvironments, which mainly includes a simple and practical data preprocessing method, Node-level Noteworthy Event (NNE) detection algorithm, event probability encodings of NNEs and two distributed Node-level Alert Event (NAE) detection algorithms. We demonstrate our algorithms by experimentally evaluating their performance in various scenarios using real and synthetic data. Our NAE detection algorithm by leveraging spatial correlation only requires a small amount of data transmission and can detect over 90% of NAEs with few false negatives.