IoTGuard: Enhancing Trust Management in IoT with Long Short-Term Memory and Multi-criteria Decision-Making Techniques

IoTGuard: Enhancing Trust Management in IoT with Long Short-Term Memory and Multi-criteria Decision-Making Techniques

연구 분야: Safety

학회: SN Computer Science

The growing adoption of IoT devices has introduced significant security challenges, including trust management, which is essential to ensure secure and reliable network operations. Traditional trust models often fall short in adapting to the dynamic and heterogeneous nature of IoT environments. This paper addresses these challenges by proposing a trust management model that integrates Long Short-Term Memory (LSTM) networks with Multi-Criteria Decision-Making (MCDM) techniques for a dynamic and accurate evaluation of IoT device trustworthiness. The study focuses on key security aspects, including predicting trust scores and mitigating potential risks associated with compromised devices. Our approach involves thorough data collection, preprocessing, and feature extraction to predict trust scores using the LSTM-based model, which is then aggregated via MCDM methods to provide a comprehensive evaluation. The proposed model is validated using the TON_IoT dataset, designed for testing and validating various cybersecurity applications, including intrusion detection and threat intelligence. Results demonstrate superior performance, achieving lower Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) alongside higher R-squared (R2) values, confirming the model's accuracy and reliability. By combining deep learning and decision-making techniques, this framework offers a robust solution to the trust management challenges in IoT networks, addressing critical security issues such as privacy preservation, adversarial machine learning, and threat detection. This study underscores the importance of integrating advanced AI techniques to enhance trust management and ensure safer, more secure IoT network operations.