A Dense Neural Network Approach for Detecting Clone ID Attacks on the RPL Protocol of the IoT.

Area of Science:

• Cybersecurity and Network Engineering
• Artificial Intelligence in IoT Security
• Wireless Sensor Networks

Background:

• The proliferation of Internet of Things (IoT) devices necessitates advanced security measures.
• Existing security systems like IDS, IPS, and SIEM struggle with sophisticated IoT threats, particularly identity-based attacks.
• The Routing Protocol for Low Power and Lossy Networks (RPL) is vulnerable to Clone ID attacks, a critical threat in IoT mesh networks.

Purpose of the Study:

• To address the inadequacy of traditional security systems against novel IoT cyber-attacks.
• To propose and evaluate an Artificial Intelligence (AI)-based protection framework specifically designed to counter Clone ID attacks in RPL networks.
• To enhance the detection accuracy of identity impersonation and counterfeiting attempts in IoT environments.

Main Methods:

• Utilizing unsupervised pre-training techniques to extract salient features from RPL network data.
• Training a Dense Neural Network (DNN) model to leverage deep feature engineering for improved classification.
• Developing a robust AI framework to identify and mitigate Clone ID attacks.

Main Results:

• The proposed AI framework demonstrates superior performance in detecting identity impersonation attacks compared to conventional methods.
• Deep feature engineering via DNN training enhances the accuracy of identifying malicious counterfeiting attempts.
• The system effectively addresses the limitations of signature-based detection for novel IoT threats.

Conclusions:

• AI-based security frameworks are essential for protecting IoT networks against sophisticated attacks like Clone ID.
• The developed DNN model offers a promising solution for enhancing RPL network security and preventing identity theft.
• Continuous adaptation of detection mechanisms using AI is crucial for safeguarding evolving IoT ecosystems.