Abstract The fast propagation of Internet of Things (IoT) networks in smart cities have presented many advantages such as improving urban efficacy, automation, and sustainability. However, these unified methods also pose important cybersecurity tasks, such as unauthorized access, data breaches, and cyberattacks that can deal with crucial infrastructure. As smart cities trust deeply real-time data automation and exchange, certifying the integrity, security, and confidentiality of these methods is vital. Conventional security devices, like firewalls and traditional encryption models frequently drop short owing to the spread nature and resource restraints of IoT devices. To tackle these tasks, industry specialists and researchers have developed advanced cybersecurity tactics, with blockchain (BC) technology and artificial intelligence (AI)-based models. Recently, the AI-driven security solutions permit real-world anomaly detection (AD) by analyzing cyber-attack designs and mechanizing risk mitigation events. This study presents the Secure Cybersecurity Framework for Smart Infrastructure Using Feature Engineering and Quantum Deep Learning Model (SCFSI-FEQDLM) methodology. This paper aims to propose a robust cybersecurity model for protecting IoT-based smart infrastructure using BC technology. Initially, the BC technology-based IoT is applied to enhance security, transparency, and efficiency in smart infrastructure systems. Furthermore, the data standardization applies min-max normalization to convert input data into a suitable format. Moreover, the feature selection (FS) process is performed by correlation analysis model, mutual information method, and principal component analysis technique to select the most significant features from a dataset. For the classification process, the SCFSI-FEQDLM method implements quantum long short-term memory with stochastic gradient descent optimizer for improving the classification performance. The incorporatin of feature engineering process with hyperparameter tuned quantum classification model helps in the accomplishment of improved detection rate. The experimental evaluation of the SCFSI-FEQDLM model was examined on a benchmark dataset. The extensive results highlight the significant solution of the SCFSI-FEQDLM approach when compared to recent models.
Revathi et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: