Blockchain and IoT Based Infrastructure for Secure Smart City Using Deep Learning Algorithm with Dingo Optimization

【Author】 Mishra, Sourav; Chaurasiya, Vijay Kumar

【Source】WIRELESS PERSONAL COMMUNICATIONS

【影响因子】2.017

【Abstract】Infrastructure for the Internet of Things is being created in smart cities with long-term viability for a range of manufacturing uses, including smart manufacturing as well as smart industries. However in a smart city environment, security measures may not be effective and the existing system also have several drawbacks such as latency, privacy, scalability and security. A block chain-based IoT framework is being developed to address these problems. Initially, the raw data's are collected from the smart cities through various IoT devices. Then the data's are pre-processed using Adaptive Data Cleaning, it also contain a prediction method called denoising auto encoder which is used to converting the data from low into high quality form. Then the pre-processed data is given to the block chain based distributed network. Block construction, Request + transaction, transmit block to other nodes in the network, and verification are the four block chain functionalities in this network. For verifying the transaction Modified Deep Neural Network is used. The verifier must select between two possibilities after the transaction has been verified: yes or no. If the state is correct, the transaction is performed and a block is generated in the blockchain. Or else, the process will be halted if any attack is identified in the transaction. The simulation analysis shows that the proposed method obtain 96% accuracy, 0.04% error, precision is 95% so on. This demonstrates that the proposed strategy outperforms other methods currently being used. Based on this proposed method the transaction is executed securely to provide secure smart city environment.

【Keywords】Adaptive data cleaning; Denoising auto encoder; Block and IoT based infrastructure; Modified DNN; Dingo optimization

【发表时间】2023 2023 AUG 7

【收录时间】2023-08-24

【文献类型】

【主题类别】

--