EYES: Mitigating Forwarding Misbehavior in Energy Harvesting Motivated Networks

Abstract Energy harvesting motivated networks (EHNets) has been becoming increasingly popular in the presence of Internet-of-Things (IoT). Each self-sustainable node periodically harvests energy from an immediate environment but it is admittedly vulnerable to a Denial-of-Service (DoS) attack in the EHNets. In this paper, we propose a novel countermeasure, called EYES, to the forwarding misbehavior of multiple colluding malicious nodes in the realm of EHNets. Under the charge-and-spend harvesting policy, we first establish a set of adversarial scenarios, analyze its forwarding operations, and identify vulnerable cases. The EYES consists of two schemes, SlyDog and LazyDog , and cooperatively detects the forwarding misbehavior. In the SlyDog, each node actively disguises itself as an energy harvesting node and stealthily monitors the forwarding operations of adjacent nodes. In the LazyDog, however, each node periodically requests the number of overheard packets from its adjacent nodes and validates any prior uncertain forwarding operation. The combination of two schemes can efficiently detect the forwarding misbehaviors of colluding malicious nodes and quickly isolate them from the network. We also present a simple analytical model and its numerical result in terms of detection rate. We evaluate the proposed countermeasure through extensive simulation experiments using the OMNeT++ and compare its performance with two existing schemes, the hop-by-hop cooperative detection (HCD) and Watchdog. Simulation results show that the EYES provides 70–92% detection rate and achieves 23–60% lower detection latency compared to the HCD and Watchdog. The EYES also shows a competitive performance in packet delivery ratio.