Evolving Lightweight Intrusion Detection Systems for RPL-Based Internet of Things.

With the integration of efficient computation and communication technologies into sensory devices, the Internet of Things (IoT) applications have increased tremendously in recent decades. While these applications provide numerous benefits to our daily lives, they also pose a great potential risk in terms of security. One of the reasons for this is that devices in IoT-based networks are highly resource constrained and interconnected over lossy links that can be exposed by attackers. The Routing Protocol for Low-Power and Lossy Network (RPL) is the standard routing protocol for such lossy networks. Despite the efficient routing built by RPL, this protocol is susceptible to insider attacks. Therefore, researchers have been working on developing effective intrusion detection systems for RPL-based IoT. However, most of these studies consume excessive resources (e.g., energy, memory, communication, etc.) and do not consider the constrained characteristics of the network. Hence, they might not be suitable for some devices/networks. Therefore, in this study, we aim to develop an intrusion detection system (IDS) that is both effective and efficient in terms of the cost consumed by intrusion detection (ID) nodes. For this multiple-objective problem, we investigate the use of evolutionary computation-based algorithms and show the performance of evolved intrusion detection algorithms against various RPL-specific attacks.