Ensuring data integrity in deep learning-assisted IoT-Cloud environments: Blockchain-assisted data edge verification with consensus algorithms

Ensuring the reliability and trustworthiness of massive IoT-generated data processed in cloud-based systems is paramount for data integrity in IoT-Cloud platforms. The integration of Blockchain (BC) technology, particularly through BC-assisted data Edge Verification combined with a consensus system, utilizes BC's decentralized and immutable nature to secure data at the IoT network's edge. BC has garnered attention across diverse domains like smart agriculture, intellectual property, and finance, where its security features complement technologies such as SDN, AI, and IoT. The choice of a consensus algorithm in BC plays a crucial role and significantly impacts the overall effectiveness of BC solutions, with considerations including PBFT, PoW, PoS, and Ripple in recent years. In this study, I developed a Football Game Algorithm with Deep learning-based Data Edge Verification with a Consensus Approach (FGADL-DEVCA) for BC assisted IoT-cloud platforms. The major drive of the FGADL-DEVCA algorithm was to incorporate BC technology to enable security in the IoT cloud environment, and the DL model could be applied for fault detection efficiently. In the FGADL-DEVCA technique, the IoT devices encompassed considerable decentralized decision-making abilities for reaching an agreement based on the performance of the intrablock transactions. Besides, the FGADL-DEVCA technique exploited deep autoencoder (DAE) for the recognition and classification of faults in the IoT-cloud platform. To boost the fault detection performance of the DAE approach, the FGADL-DEVCA technique applied FGA-based hyperparameter tuning. The experimental result analysis of the FGADL-DEVCA technique was performed concerning distinct metrics. The experimental values demonstrated the betterment of the FGADL-DEVCA approach with other existing methods concerning various aspects.