Deep Learning Enabled Data Protection and Security (DPS) Techniques for Intrusion Mitigation, and Network Vulnerabilities Detection in the Internet of Things (IoTs)
DOI:
https://doi.org/10.63075/emt07360Keywords:
Machine Learning, Deep Neural Network, CNN, Prediction Models, Routing Attacks Detection, Deep Learning, Internet Of Things, Threat Detection, Deep Neural Network, Internet Of Things Networks, Wi-Fi security, wireless protocols, WEP, EncryptionAbstract
Deep learning operates as the main technology pillar of the present Industry 4.0. Various applications in healthcare along with visual recognition and text analytics besides cybersecurity functions have adopted Deep Learning implementation. Developing suitable DL models remains complicated because of real-world problems and data show constant changes and diverse patterns. A structured Deep Learning approach is proposed and discussed in this article that includes a DL taxonomy system. The DL-IDS framework examines different types of practical operations that include supervised or unsupervised protocols. This work proposes a novel deep learning technique for threat mitigation and a l73 accuracy rate stands at 71.73 listing real-life applications in which deep learning techniques serve practical purposes. This article works toward compiling IoT (Internet of Things) connected systems, applications, data storage, and services that may be a new gateway for cyberattacks as they continuously offer services in the organization. Currently, software piracy and malware attacks are high risks to compromise the security of IoT. The proposed DL-IDS system uses acquired information to determine if data needs to be transferred to the fog layer. The proposed approach demonstrates better functionality than available DL-IDS solutions operating on the RT-IoT2022 dataset. The accuracy rate stands at 71.73 when measuring the detection performance of Intrusion using the proposed IPS-DL system. Through an integration of Deep Learning DL-IDS based the Proposed IPS system achieves an anomaly identification with a precision of 70.63%, together with a recall of 96.30% and an F1-score at 92% for intrusion prevention tasks. 85% detection ratio coupled with a 0.99% ideal throughput and a 0.23% packet delivery ratio at a minimum energy usage of 0.11 joules with a bandwidth of 0.84 bps and the delay was measured at 0.21 ms using 100 nodes with 0.66% improved probability. Security issues within IoT networks is also addressed through quick response systems for intrusion detection in IoT networks.
