An optimized hybrid deep neural network architecture for intrusion detection in real-time IoT networks

The Internet-of-Things (IoT) refers to the interconnection of things to the physical network that is embedded with software, sensors, and other devices to exchange information from one device to the other. The interconnection of devices means there is the possibility of challenges like security, trustworthiness, reliability, confidentiality, and so on. To address those issues, we have proposed a novel GTBSS-HDNN approach which hybridization of Group theory (GT), Binary Spring search (BSS) algorithm, and Hybrid deep neural network (HDNN). The proposed GTBSS-HDNN approach effectively detects the intrusion in the IoT nodes. Initially, the privacy-preserving technology was implemented using a Blockchain-based methodology. Our proposed privacy-preserving methods are divided into two parts. The first stage utilizes blockchain and the second stage involves Modified Independent Component Algorithm (MICA) to prevent intrusion attacks. The authentication of data is performed by blockchain-based Enhanced Proof of Work (EPoW) and achieves better authentication. Furthermore, the experimental study is carried out using the ToN-IoT dataset, which is used to evaluate the performance of our proposed work. To analyze the performance we have taken the performance metrics such as F1-measure, Detection Rate, Precision, and Accuracy. The performance analyzes depict that the proposed method effectively preserves the accuracy and thereby avert the intrusion attacks. The proposed model achieved 95.3% accuracy, 96.54% precision, 95.23% recall, and 95.67% F-score values on the ToN-IoT dataset and 96.23% accuracy, 95.94% precision, 97.03% recall, and 96.70% F-score results on the BoT-IoT dataset.