An intelligent blockchain technology for securing an IoT-based agriculture monitoring system

Nowadays, securing the sensed data in the cloud server is one of the significant concerns in blockchain technology. Although different Machine Learning (ML) based security frameworks are developed, they face specific issues in confidentiality, time consumption, if the dataset is large, processing the data in an existing security system isn't easy, etc. Thus, a novel hybrid Recurrent Neural Elliptical Curve Blockchain (RNECB) was designed to securely store the sensed agricultural data in the cloud server. The dataset was gathered from a standard website. This model filters the input dataset in the pre-processing phase and enters it into the field monitoring module. The monitoring mechanism in the presented approach provides continuous monitoring and extracts meaningful features. In addition, crypto analysis was carried out to hide the extracted features from third parties. These encrypted data were then stored in the cloud server. Furthermore, security analysis was performed by launching attacks on the cloud server, and the results are estimated in two cases before and after the attack. The presented model was implemented in python software, and the accuracy attained about 97.7%, the confidential rate about 97.98%, encryption, decryption, and execution time taken were about 2.7 ms, 2.6 ms, and 11 ms, respectively. And also, the proposed model attained a lower error rate of about 0.0227%. The calculated results were compared with the existing security approaches. The comparative assessment verifies that the designed model earned better results than others.