An efficient detection of Sinkhole attacks using machine learning: Impact on energy and security

Abstract

In the realm of Wireless Sensor Networks (WSNs), the detection and mitigation of sinkhole attacks remain pivotal for ensuring network integrity and efficiency. This paper introduces SFlexCrypt, an innovative approach tailored to address these security challenges while optimizing energy consumption in WSNs. SFlexCrypt stands out by seamlessly integrating advanced machine learning algorithms to achieve high-precision detection and effective mitigation of sinkhole attacks. Employing a dataset from Contiki-Cooja, SFlexCrypt has been rigorously tested, demonstrating a detection accuracy of 100% and a mitigation rate of 97.31%. This remarkable performance not only bolsters network security but also significantly extends network longevity and reduces energy expenditure, crucial factors in the sustainability of WSNs. The study contributes substantially to the field of IoT security, offering a comprehensive and efficient framework for implementing Internet-based security strategies. The results affirm that SFlexCrypt is a robust solution, capable of enhancing the resilience of WSNs against sinkhole attacks while maintaining optimal energy efficiency.