Leveraging Blockchain With Chaotic Oppositional Barnacles Mating Optimizer-Based Deep Learning Model For Secure Iot Environment In Consumer Electronics

The security of IoT networks has become a major concern with the ubiquity of Internet of Things (IoT) technology. The concept of intrusion detection system (IDS) is a complex system to discover an intruder in the IoT platform, where the intruder can be a host that tries to gain some other nodes without authorization. Due to the complexity and resource constraints, classical IDS have their limitations in the context of IoT networks. There has been significant research in integrating both IDS and blockchain (BC) to detect existing and emerging cyberattacks and improve data privacy, correspondingly. In these approaches, learning-based ensemble algorithms can concurrently ensure data privacy and facilitate the detection of sophisticated malicious events. This study introduces a BC with fractal chaotic oppositional barnacles mating optimizer-based deep learning (BCOBMO-DL) Model for Secure IoT environment. The BCOBMO-DL technique exploits BC technology with DL-based intrusion detection to protect the IoT environment. In the BCOBMO-DL technique, the linear scaling normalization (LSN) approach can be used to scale the input data into a uniform format. In addition, the BCOBMO-DL technique designs the COBMO algorithm for electing the optimal subset of features. For intrusion detection, the self-attention bidirectional gated recurrent unit (SA-BiGRU) model is applied with fractal theory. Finally, the reptile search algorithm (RSA)-based hyperparameter tuning process is utilized for tuning the parameters based on the SA-BiGRU model. Furthermore, the BC technology is useful for achieving security in the IoT network. The experimental evaluation of the BCOBMO-DL method takes place on the benchmark NSLKDD dataset. The widespread experimental outcomes highlighted the enhanced detection outcomes of the BCOBMO-DL algorithm over other models.