Efficient quantum inspired blockchain-based cyber security framework in IoT using deep learning and huristic algorithms

Abstract

“Internet-of-Things (IoT)” systems and components are highly noticed by cybercriminals including the affiliated or the nation-state sponsored threat actors as become a united part of the linked ecosystem and the society. But, the difficulties in protecting the systems and the devices are combined of scale and multiple deployments, the speed-paced cyber threats landscape, and more parameters. With the enhanced internet services, cyber security grows one of the highest research issues of the latest digital world. It is very important to develop a cyber security model to identify the various types of attacks. To overcome these problems, a quantum-inspired blockchain-assisted cyber security model is obtained in the IoT platform. Firstly, the required information is obtained from quality online information resources. Then, the information is stored in the quantum-inspired blockchain with optimal key, where the key optimization is performed with the help of the Fitness-based Jellyfish Chameleon Swarm Algorithm (FJCSA). Then, the stored data are recovered and finally, fed to the intrusion detection stage to verify whether it is affected by any unauthorized entities. The intrusion detection is done with the support of “Adaptive Attention-based Long Short Term Memory (LSTM) with Adaboost (AALSTM-Ab)”, where the parameters are optimized by using the FJCSA. Furthermore, the experimental results of the developed model are validated by comparing the performance of various recently implemented blockchain-based cyber security approaches with respect to several positive and negative performance measures. From the result analysis, the accuracy and precision rate of the recommended model are 95.50% and 91.40%.