Yao, H.; Yang, Y.; Fu, X., and Mi, C., 2018. An adaptive sliding-window strategy for outlier detection in wireless sensor networks for smart port construction. In: Ashraf, M.A. and Chowdhury, A.J.K. (eds.), Coastal Ecosystem Responses to Human and Climatic Changes throughout Asia.

With the rapid expansion of wireless sensor network in smart ports, the outlier detection in data streams produced by sensor has attracted a wide range of attention to meet the challenge of network security, fault diagnosis, and intrusion detection. In this paper, the focus is on the adaptive optimization strategy for sliding window–based outlier detection method, which has been involved in a few of the existing relevant literatures. A novel adaptive sliding-window method is introduced to dynamically optimize the sample set of data streams based on the local continuous fluctuations of data. The local outlier factor is calculated for data within the composite neighborhood in the sliding window to estimate its local ability to be an outlier. Then, data with the ability to be a continuously high outlier is detected as an outlier. Finally, the scalability, computational complexity and outlier-detection effect of the proposed method have been analyzed based on synthetic and real datasets. All results show that the proposed approach has superior outlier-detection performance compared with the fixed sliding window–based method.