Using Provenance to Detect Selective Forwarding Attack in RPL-Based Internet of Things

In the Internet of Things (IoT), resource-constrained things can connect to the Internet via IPv6 and 6LoWPAN networks. The Routing Protocol for Low-Power and Lossy Networks (RPL) has enabled such interconnection. However, the data transportation using RPL is vulnerable to various attacks due to the interaction between unattended things with the unreliable Internet. For instance, the data generated by sensors are vulnerable to attacks (for instance, selective forwarding attack). Therefore, error-free and reliable information cannot be assured in the decision-making process. During data transmission from source to destination, provenance can be used to track data acquisition and data traversal. In this paper, we use provenance to evaluate the network performance by computing the packet delivery ratio (PDR) at each forwarding node in the packet path. Furthermore, to identify the faulty nodes, we counted the packets received from the respective child nodes in the routing table at each parent node participating in the network. We have evaluated the proposed approach for RPL-based IoT in terms of provenance size, provenance generation time, and memory consumption.