| Abstrakt: |
The Internet of Things (IoT) links different gadgets together and enables data transmission, device tracking, and device monitoring. Blockchainbased solutions are being developed to guarantee security in IoT devices, but they have a number of drawbacks. To increase data security and prevent inaccurate data detection, we propose a novel model named Markov’s Fusion-based MM Ring Toss Blockchain. In this novel model, an innovative Fusion-Based Deep Multimodal with Radial Basis Function Neural Network (FBMR-BFNN) is utilized in which Multilayer Perceptron (MLP) with Global Max Pooling Layer eliminates overfitting of data to increase execution and packing time. In order to prevent false data detection, the corrupted input data is rebuilt by a Denoising Auto Encoder (DAE), which is then supplied to a Radial Basis Function Neural Network (RBNN). The absence of identifying capabilities, the inability to determine whether the same user or a hacker is operating the system for an extended period of time, and insufficient authorization methods made the existing systems vulnerable to inadequate security. In order to prevent any unauthorized data usage and achieve a successful second-stage authorization mechanism without data loss, a unique Markov Ring Toss VGBO technique is developed. This proposed method have superior security, efficient authorization and verification system, reduced packaging and execution times, and greater hash rates. [ABSTRACT FROM AUTHOR] |
