Evaluating the impact of selective forwarding attacks on RPL-based 6LoWPAN in IoT: a comprehensive simulation study

• Published in: Computing Vol. 107; no. 3; p. 89
• Main Authors: Rajasekar, V. R., Rajkumar, S.
• Format: Journal Article
• Language: English
• Published: Vienna Springer Vienna 01.03.2025; Springer Nature B.V
• Subjects: Artificial Intelligence; Computer Appl. in Administrative Data Processing; Computer Communication Networks; Computer Science; Convergence; Energy consumption; Information Systems Applications (incl.Internet); Network latency; Nodes; Packet transmission; Radio range; Regular Paper; Software Engineering; Topology; Transmission rate (communications); 97-11; IPv6; Selective Forwarding attack; 6LoWPAN; IoT; RPL; LLN
• ISSN: 0010-485X; 1436-5057
• DOI: 10.1007/s00607-025-01440-z

RPL, an IETF-developed standard routing protocol for 6LoWPAN, is susceptible to Selective Forwarding Attacks (SFA), wherein a malicious node selectively forwards certain packets while discarding others. The detrimental impacts of SFA encompass increased overhead, latency, complexity, scalability issues, security risks, resource consumption, and compatibility challenges. A thorough analysis of the influence of SFA on RPL-based 6LoWPAN networks across various dimensions is essential for the research community to devise optimal solutions. In this work, we examined the adverse effects of SFA on RPL-based 6LoWPAN using a simulation-based experimental study, with six scenarios with 10%, 20%, 30%, and 50% of attacker nodes to investigate the detrimental impacts of SFA on control traffic overhead, duty cycling, latency, expected transmission rate, network convergence time, packet delivery rate, network overload, power consumption, and throughput. Our study shows that the CTO has climbed from 4.41 to 37.94%, resulting in negative network bandwidth usage. Rx and Tx duty cycles grew by 1670.71% and 341.66% across all the scenarios with attacker nodes, respectively. Latency increased by 111.58% and 108.27%, respectively, when selectively dropping control and data packets, and a maximum of 20.04% for both (control and data) packet dropping. The ETX is increased between 1.31 and 14.9% and remains the same as in the reference topology in the case of only data packet drop, indicating that the RPL’s self-healing mechanism does not affect data packet loss. In all scenarios, network convergence time increased by 46%, PDR decreased by 14.63%, network load increased by 8.14%, and the average energy consumed increased by 18.85% on average. This study further revealed that nodes positioned beyond the radio range of the sink node, which was not selected as the preferred parent, exhibited reduced energy consumption. The increase in attackers increases network throughput by 43.98% to transmit only control packets. As far as our knowledge extends, this pioneering research study thoroughly investigates the influence of RPL-specific SFA on 6LoWPAN networks across multiple dimensions, including variations in packet drop types (Data/Control), attacker deployment locations, and varying attacker quantities.