Artificial Intelligence-Based Technique for Detection of Selective Forwarding Attack in RPL-Based Internet of Things Networks

Abstract

Internet of Things (IoT) is an upcoming technology in computing networks, which uses the novel and constrained technologies like IPv6 over Low-powered wireless personal area networks (6LoWPAN), Routing protocol for low power and Lossy networks (RPL), etc. The 6LoWPAN is a compressed version of the IPv6 protocol with the integration of the wireless personal area network, which plays a major role to connect the things (any objects around us such as chair, refrigerator, light, etc.) in the IoT-based network. The RPL is a distance-based network routing protocol that used to establish a route to send packets in the 6LoWPAN-based IoT networks. As the 6LoWPAN and RPL protocols are compressed/constrained versions, will leads to many attacks in IoT networks. The few attacks in the RPL protocol are wormhole, black hole, sink hole, sybil, rank, selective forwarding, various denial of service attacks, etc. This paper focused on the detection technique, for the selective forwarding attack, which occurs in the 6LoWPAN based RPL protocol. This attack will disrupt the routing path in IoT network by selectively forwarding the few packets and dropping remaining all the packets between the nodes in the network. To mitigate this attack, a novel artificial intelligence-based detection technique was proposed in this paper. This proposed technique was implemented in the ContikiCooja simulator with the Sky motes and the results shown the efficiency of the proposed system in terms of the delay, packets delivery ratio, and in detection of selective forwarding attack.