Event Driven Adaptive Security in Internet of Things

Abstract

With Internet of Things (IoT), new and improved personal, commercial and social opportunities can be explored and availed. However, with this extended network, the corresponding threat landscape will become more complex and much harder to control as vulnerabilities inherited by individual things will be multiplied. Conventional security controls, such as firewalls, intrusion detection systems (IDS) etc., may show some level of resistance to this self-organizing network but, as standalone mechanisms, are not sufficient to analyze the threat in a particular context. They fail to provide the essential context of a threat and yields false positives-negatives which can trigger pointless re-configurations, service unavailability and end user discomfort. Such unwanted events can be very catastrophic, for instance, in an IoT enabled eHealth services. We need to have an autonomous adaptive risk management solution for IoT, which can analyze an adverse situation in a distinct context and manage the risk involved intelligently so that the end user, service and security preferences are well-preserved. This paper details an event driven adaptive security model for IoT to approach the objective specified and explicates how it can be utilized in an eHealth scenario to protect against a threat faced at runtime.