Context-Aware Trust Prediction for Optimal Routing in Opportunistic IoT Systems.

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Social Opportunistic Internet of Things (SO-IoT) networks facilitate mobile, ad-hoc device communication through physical and social interactions.
• Effective routing in SO-IoT relies on selecting trustworthy intermediate nodes for secure and efficient data dissemination.
• Existing routing methods often fall short by focusing solely on reliability or social interest, neglecting multifaceted node trustworthiness.

Purpose of the Study:

• To propose a novel trust-based route optimization framework for SO-IoT environments.
• To integrate social interest and behavioral reliability for comprehensive node trustworthiness assessment.
• To enhance data dissemination security and efficiency in dynamic SO-IoT networks.

Main Methods:

• Developed a trust-based route optimization framework utilizing Bayesian inference and Jeffrey's conditioning.
• Computed a composite trust level for intermediate nodes based on integrated social and behavioral metrics.
• Conducted a two-phase simulation analysis using scenario-driven evaluations and large-scale NS-3 simulations.

Main Results:

• The proposed framework achieved a delivery ratio of up to 88.9%.
• Demonstrated minimal energy consumption compared to benchmark routing schemes.
• Achieved the highest trust accuracy at 86.5% in simulation results.

Conclusions:

• The proposed trust-based framework is robust and scalable for real-world SO-IoT applications.
• Integrating social interest and behavioral reliability significantly improves routing performance.
• The model offers a superior approach to node selection for secure and efficient data forwarding in SO-IoT.