Decentralized Consensus Blockchain and IPFS-Based Data Aggregation for Efficient Data Storage Scheme

By the development and advancement of blockchain technique, Internet of Things (IoT) proliferation driven devices and the application of blockchain-enabled IoT alter the view and operating infrastructure of the smart networks. The blockchain is responsible for supporting decentralized systems and offers secured means of authentication, management, and access to IoT system thereby deploying smart contracts offered by Ethereum. The increasing demand and the blockchain expansion generate huge volume of sensitive data. The growing demand and expansion of blockchain-IoT systems is generating large volume of sensitive data. Furthermore, distributed denial-of-service (DDoS) attacks are regarded as the most promising threats for smart contracts in the blockchain-based systems. Therefore, there is a need to detect and classify the attack type and the data should be stored in server more securely with the use of blockchain and data aggregation method. For this purpose, this presented technique aims at introducing decentralized consensus blockchain and Interplanetary file system (IPFS) based data aggregation for effective classification and data storage. The attack is detected using meta-hyperparameter random forest (MHP-RF) classifier. Once the attack is detected, the transaction information is stored in server securely by means of smart contract-based blockchain system. The transaction handling stage classifies the transaction type as normal or abnormal one which then followed by execution of business logic by smart contract thereby appending the transaction of blockchain in the network cloud. The consensus blockchain technique is employed with the use of PoW-enabled scheme integrated with Elgamal-based data aggregation. Therefore, the system security is improved and the intrusion is prevented greatly. The performance analysis of the system is analyzed in terms of accuracy, precision, recall, F-score, Encryption time, decryption time, execution time, and space complexity. The attained outcomes are compared with traditional approaches to prove the effectiveness of proposed strategy. The proposed system is said to be effective in time consumption, classifier performance, and in overcoming space complexity issues.