A Trust Management Model for IoT Devices and Services Based on the Multi-Criteria Decision-Making Approach and Deep...

Area of Science:

• Computer Science
• Cybersecurity
• Artificial Intelligence

Background:

• Internet of Things (IoT) technology is increasingly integrated into daily life, offering smart services across various sectors.
• The heterogeneity of IoT devices makes them vulnerable to attacks, compromising service reliability and user trust.
• Existing trust management techniques struggle with large datasets and dynamic behavioral changes in IoT environments.

Purpose of the Study:

• To propose an advanced trust management model for IoT devices and services.
• To address the limitations of current trust management methods in handling complex IoT ecosystems.
• To enhance the security and reliability of IoT smart services.

Main Methods:

• Development of a hybrid model combining the Simple Multi-Attribute Rating Technique (SMART) for trust value calculation.
• Integration of the Long Short-Term Memory (LSTM) algorithm for detecting behavioral anomalies based on trust thresholds.
• Evaluation of the model's performance using metrics like accuracy, loss rate, precision, recall, and F-measure.

Main Results:

• The proposed model demonstrated superior performance compared to existing machine learning and deep learning approaches.
• Achieved high accuracy (99.87%) and F-measure (99.76%) with 100 iterations.
• Effectively identified behavioral changes and calculated trust values across diverse data samples.

Conclusions:

• The SMART-LSTM model offers a robust solution for trust management in IoT.
• The proposed approach significantly improves the security and trustworthiness of IoT services.
• This model provides a scalable and effective method for securing heterogeneous IoT environments.