Managing Trust and Detecting Malicious Groups in Peer-to-Peer IoT Networks

Area of Science:

• Computer Science
• Network Security
• Artificial Intelligence

Background:

• Peer-to-peer (P2P) networking offers low-cost, low-latency advantages for Internet of Things (IoT) platforms.
• P2P networks face significant security vulnerabilities, including remote attacks, eavesdropping, and credential theft by malicious peers.
• Existing trust and reputation management systems struggle to detect sophisticated malicious peer behaviors, especially coordinated attacks.

Purpose of the Study:

• To propose an intelligent trust management system, Trutect, designed to address the security challenges in P2P IoT networks.
• To develop a system capable of identifying individual malicious peer models and detecting coordinated malicious groups.
• To improve the overall security and trustworthiness of P2P networks through advanced detection mechanisms.

Main Methods:

• Implementation of a novel intelligent trust management system named Trutect.
• Leveraging neural networks for analyzing peer behavior and assessing trustworthiness.
• Developing algorithms to identify individual peer models (good or malicious) and detect malicious collectives.

Main Results:

• Trutect demonstrates superior performance compared to existing trust management systems.
• The system significantly increases the success rates of legitimate peers.
• Trutect achieves accurate identification of individual peer models and detects malicious group activities with a lower running time.

Conclusions:

• Trutect effectively addresses the limitations of current trust management systems in P2P IoT environments.
• The proposed system enhances network security by accurately identifying and mitigating threats from individual malicious peers and coordinated groups.
• Neural network integration provides a robust and efficient solution for trust management in decentralized networks.