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

Area of Science:

• Computer Science
• Network Security
• Machine Learning

Background:

• Wireless Sensor Networks (WSNs) are vulnerable to security threats, particularly sinkhole attacks.
• Ensuring network integrity and efficiency in WSNs requires robust security mechanisms.
• Energy consumption is a critical factor for the sustainability of WSNs.

Purpose of the Study:

• To introduce SFlexCrypt, a novel approach for detecting and mitigating sinkhole attacks in WSNs.
• To optimize energy consumption in WSNs while enhancing security.
• To provide an efficient framework for Internet-based security strategies in IoT.

Main Methods:

• Integration of advanced machine learning algorithms for attack detection and mitigation.
• Utilizing a dataset from Contiki-Cooja for rigorous testing and validation.
• Developing a framework for implementing Internet-based security strategies.

Main Results:

• Achieved 100% detection accuracy for sinkhole attacks.
• Demonstrated a 97.31% mitigation rate for sinkhole attacks.
• Significantly extended network longevity and reduced energy expenditure.

Conclusions:

• SFlexCrypt is a robust solution for enhancing WSN resilience against sinkhole attacks.
• The approach maintains optimal energy efficiency, crucial for WSN sustainability.
• SFlexCrypt contributes significantly to the field of IoT security.