Abstract—The rapid growth in sensors, low-power integrated circuits, and wireless communication standards has enabled a new generation of applications based on ultra-low powered wireless sensor networks. These are employed in many environments including health-care, industrial automation, environmental monitoring and intelligent transportation. Furthermore, a significant portion of low powered data requires a certain type of security that offers higher availability, confidentiality and data integrity. The objective of this work is to investigate the feasibility of using attributes of IEEE 802.15.4e/TSCH and machine learning techniques to determine traffic anomalies in ultra-low powered wireless networks. Several factors including the sample size, noise influence, classification algorithm and model aging process are investigated against prediction accuracy and other performance indicators. The experiments have demonstrated that machine learning models trained using carefully selected input features and adequate training data are able to detect traffic anomalies of low powered wireless networks with remarkable accuracy (over 95 percent), while keeping the false positive and negative rates to minimum.
 

Index Terms—LoWSN, LoWPAN, low powered sensor networks, IEEE 802.15.4e/TSCH, IoTs, wireless security, anomaly detection

Cite: Sajeeva Salgadoe and Fletcher Lu, "An Anomaly Detection Model for Ultra Low Powered Wireless Sensor Networks Utilizing Attributes of IEEE 802.15.4e/TSCH," Journal of Communications, vol. 14, no. 5, pp. 335-341, 2019. Doi: 10.12720/jcm.14.5.335-341