Blockchain-based cryptographic framework for secure data transmission in IoT edge environments using ECaps-Net.

In the evolving landscape of Internet of Things (IoT), the integration of interconnected devices and cloud computing has revolutionized data collection and processing. However, this connectivity poses numerous security challenges about data privacy, integrity, and security. Traditional cloud-based security approaches inadequate for managing the distributed and dynamic nature of IoT ecosystems. The emergence of the edge computing paradigm allowed for the transfer of data processing and storage closer to local edge devices, but introduces new vulnerabilities at the edges. Thus, an Intrusion Detection System (IDS) is required in this situation. IDS built at the edge can quickly detect and mitigate possible attacks by continually monitoring network traffic, device interactions, and real-time anomalies. Therefore, in this study, we propose an Enhanced Deep Learning (DL)-based IDS integrated with a Blockchain-Based Cryptographic-Algorithm to ensure secure data transmission in an IoT edge computing environment. Initially, the intrusion dataset undergoes preprocessing step to enhance its quality by eliminating unnecessary data and normalizing the dataset. then, the pre-processed data is classified using an Enhanced Capsule Network (ECaps-Net), which incorporates a Squeeze and Excitation (SE) block to highlight important features and surpasses less important ones. After classification, the classified normal data is converted into blocks using Blockchain technology. Every block is hashed using the Merkle-Damgard cryptographic algorithm to ensure data integrity and confidentiality. The proposed framework outperformed existing methods with a maximum accuracy of 98.90% and 98.78% on the KDD Cup-99 and UNSW-NB 15 datasets, respectively. The proposed mechanism protects cloud server and edge devices from malicious access, offering a reliable and efficient solution for secure data transmission in IoT edge environments.