The wireless sensor networks (WSNs) suffer from several constraints such as harsh areas of deployment, the bad medium of communication and energy limitations. So, to guarantee the reliability and to enhance the quality of the measurements in this kind of networks, it is very crucial that the nodes of the network can detect and clean the sensed data which do not conform to an expected pattern. To reach this goal, we present in this paper a new algorithm, called Outlier Detection Algorithm based on Mahalanobis Distance (ODA-MD). In this algorithm, Mahalanobis theorem is used to compute and compare sensed data in a distributed fashion. The efficiency of this proposal that can detect outliers online even several data types exist in the network. The functionalities of the proposed scheme were performed by simulation using data sets from the Berkeley Intel Laboratory. The obtained results demonstrate the effectiveness of the solution compared to other algorithms in terms of energy consumption, false alarm rate and detection accuracy.
展开▼
