Blockchain-enabled Architectures for Reliable Sensor Data and Trust Evaluation in IoT: A Comprehensive Survey

Abstract

The rapid expansion of the Internet of Things (IoT) and autonomous systems has increased the need for reliable, transparent, and secure sensor data management. Existing research explores diverse trust management strategies ranging from statistical models such as Markov chains for dynamic trust prediction to deep learning frameworks like DeepTrust and ensemble methods using XGBoost and AdaBoost for anomaly detection. Recent works also highlight blockchain-enabled trust architectures for decentralized verification and immutable audit trails. In autonomous vehicle studies, sensor reliability is often enhanced through fusion, redundancy, and self-calibration, yet challenges remain regarding fault tolerance, scalability, and tamper-resistant validation. This literature survey consolidates these developments to emphasize the necessity of integrating trust computation at the edge with blockchain-based transparency. The reviewed models reveal that while AI and reputation-based frameworks improve detection accuracy, they often depend on centralized or resource-intensive computation. Therefore, this work identifies a research gap in developing lightweight, edge-optimized trust frameworks that can evaluate sensor reliability in real time and record trust events immutably. The proposed direction bridges IoT reliability, edge computing, and blockchain auditability, laying a foundation for secure and verifiable sensor trust management with TrustScore in smart transportation and industrial IoT ecosystems.