Blackhole and Sinkhole Mitigation in Wireless Sensor Networks

Abstract

Wireless sensor networks (WSNs), as a technological newlinebreakthrough; have demonstrated considerable potential in many areas, newlinesuch as industrial process management, healthcare, military surveillance, newlineand environmental monitoring. Unfortunately, WSNs are prone to a range newlineof threats due to their intrinsic limitations, which include limited energy, newlineprocessing, and communication capabilities. Data availability, newlineconfidentiality, and integrity are all susceptible to compromise on WSNs newlinedue to malicious attacks. Assuring the safety and reliability of WSNs newlinerequires the implementation of attack detection and mitigation newlinetechniques. To prolong the lifespan of WSNs, energy-efficient attack newlinedetection and mitigation techniques are therefore required. newlineThis study presents trust-based energy-efficient attack detection newlineand mitigation techniques. The effectiveness of these approaches is newlineassessed depending on the following metrics: PDR, delay, throughput, newlineNRL, and scalability with 40, 70, and 100 nodes. A 1.25% FPR and a newline95% DR were achieved with blackhole mitigation based on trust-based newlinemethodology. On the other hand, the backbone node method yielded a newline2.5% FPR and a 90% DR. Using an agent-based strategy, sinkhole attack newlinemitigation produced a 1.25% FPR and a 95% DR, a scheme based on trust newlineachieved a 2% FPR and a 90% DR. Hybrid anomaly detection scheme newlinelocates sinkhole attacker nodes and blackhole attacker nodes by carrying newlineout the processes of both the offline and online phases. The simulation newlineresult showed that the FPR is 1.2% and the DR is 98.6%. newlinevi newlineM-LEACH performed better than LEACH and B-LEACH in newlineterms of throughput, energy efficiency, scalability, and network life. This newlineis because M-LEACH combines several techniques, such as data newlineaggregation, dynamic cluster building, cluster-head selection, and multihop newlinecommunication, to increase the overall network life, energy newlineefficiency, and scalability. Based on the results it was found that the Qlearning newlinebased energy efficient routing protocol outperformed LEACH, newlineM-LEACH, and B-LEACH.