Kernel Principal Subspace based Outlier Detection Method in Wireless Sensor Networks

An emerging class of Wireless Sensor Networks (WSNs) applications involves the acquisition of large amounts of sensory data from battery-powered, low computation and low memory wireless sensor nodes. The accuracy of sensor readings is without a doubt one of the most important measures to evaluate the quality of a sensor and its network. For this case, the task amounts to creating a useful model based on KPCA to recognize data as normal or outliers. Over the last few years, Kernel Principal Component Analysis (KPCA) has found several applications in outlier detection. Within this setting, we propose a new outlier detection method based on Kernel Principal Component Analysis (KPCA) using mahalanobis distance to implicitly calculate the mapping of the data points in the feature space so that we can separate outlier points from normal pattern of data distribution. The use of KPCA based mahalanobis kernel on real word data obtained from Intel Berkeley are reported showing that the proposed method performs better in finding outliers in wireless sensor networks.