Circumventing Rank Attack on Routing Protocol for Low Power and Lossy Networks (RPL)

Abstract

Low power and Lossy Networks (LLNs) are used for data acquisition in industrial control & monitoring, home automation, security & military survei llance, health care etc. LLNs are categorized as a type of wireless personal area networks (WP AN). Routing Protocol for low power and Lossy (RPL) network is the underlying routing protocol of standard 6Lo WPAN. RPL is proposed by working group of Internet Engineering Task force (IETF) to be used in Internet of Things (loT). JETF has selected RPL as a standard protocol because it outperforms other wireless and ad hoc routing protocols in term of quality of service (device management and efficient energy saving performance) for LLNs. RPL creates a directed acyclic graph (DAG) of the network topology fo r routing. Research efforts are made to evaluate the performance of RPL in different network scenarios and topologies for LLNs. RPL provides some solutions for securing communication between devices however various internal attacks are recently highlighted. Rank attack is one of the most critical attacks that can be launched over RPL. It is a variation of sinkhole attack in ad hoc networks. Rank in RPL is the relative distance of a node from the DAG root (sink). Rank of node increases downstream & decreases upstream. Rank of a node in RPL serves multiple purposes apart from distance to the sink, it ensures route optimization, prevent loops, and manages control overhead. In a rank attack, a malicious node advertises false rank information (lower rank than neighboring nodes) to its possible child nodes. As result, neighboring nodes select the malicious as their prefeJTed parent for routing information to the sink node. This attack can have very adverse impact on the operation of network if it is launched near the sink. In this case, major portion of the networks traffic will converge to the malicious node and all the data will be routed through the malicious node. To the best of knowledge, recent studies have only highlighted the criticality of rank attacks in LLNs but no solution has been proposed to counter this attack. In this work, we have proposed a non-cryptographic neighbor based solution in which a node checks the advertised rank and trust level of the attacking nodes before selecting any node as its parent. Also, update of vulnerable activities (false control messages) is reported to the sink for authentication of malicious nodes. Simulations analysis has shown that the proposed solution is capable of identifying the attacking nodes with high level of accuracy.