An Accurate and Energy-Efficient Anomaly Detection in Edge-Cloud Networks

With the wide-adoption of edge-cloud networks in various domains, anomaly detection is a regularly performed task to guarantee the health of the Internet of Things (IoT) applications. Traditionally, sensory data are gathered at the network edge and completely routed to the cloud, where computational-heavy algorithms are mostly adopted to determine the locations of certain anomaly. Considering the occurrence infrequency of anomalies, this strategy may route relatively large-volume of sensory data, which reflects a healthy situation indeed, to the cloud. To reduce these anomaly-irrelevant sensory data transmitted in the network, this paper proposes an accurate and energy-efficient anomaly detection mechanism in three-Tier IoT-Edge-Cloud (3T-IEC) networks. Specifically, after gathering sensory data provided by IoT nodes in certain edge network, the edge node applies the Marching Squares algorithm to generate isopleths, where an isopleth may capture the boundary of anomaly. A filtering mechanism is conducted at the edge tier, such that anomaly-relevant sensory data are routed to the cloud. Thereafter, the boundary of anomaly is obtained, and the locations of candidate boundary nodes are derived by adopting the Kriging spatial interpolation algorithm at the cloud tier. These locations are traversed by mobile sensing nodes at edge networks, and their sensory data are gathered to refine the boundary. Extensive experiments are conducted, where an air quality hazardous gas dataset acquired from Towards Data Science is applied. Evaluation results show that our technique outperforms the state-of-the-art ones in boundary accuracy and energy-efficiency.