Anomaly Behavior Analysis for Fog Nodes Availability Assurance in IoT Applications

Abstract

The Internet of Things (IoT) is the new trend to make devices interact among themselves. The IoT will connect not only computers and mobile devices, but also wearable devices, smart buildings, smart cities, electrical grids, and automobiles just to name few. IoT will lead to the development of a wide range of advanced information services that need to be processed in real-time and require large storage and computational power than can be provided by Cloud and Fog Computing. The integration of IoT with Cloud and Fog Computing make IoT capable of processing large-scale geo-distributed information. In any IoT application, communications are critical to deliver the required information to the end user, device or application, for instance to take actions during crisis events. However, IoT communication elements such as gateways or Fog nodes, will introduce major security challenges as they contribute to increase the attack surface, preventing the IoT to deliver accurate information. In this paper, we propose a methodology to develop an Intrusion Detection System (IDS) based on Anomaly Behavior Analysis (ABA) to detect when a Fog node has been compromised. The preliminary experimental results show that our proposed approach accurately detects known and unknown anomalies due to misuses or cyber-attacks, with high detection rate and low false alarms.