An Efficient DDoS Attack Detection using Attention based Hybrid Model in Blockchain based SDN-IoT

Due to the advancement of IoT (Internet of Things) has created a new technology and enhanced modelling capabilities, contributing to the evolution of modern standards of living. The prevalence of insecure and portable devices in the IoT platform has led to a significant rise in cyberattacks. Various research endeavors have been undertaken to identify potential security threats. However, these efforts have encountered challenges such as limitations in storage capacity, high computation costs, system failures, and increased latency. The fundamental programming and centralized features integrated into the SDN (Software Defined Networking) serve as a solution to streamline network management, make abstraction of the network, facilitate the smooth evolution of networks, and potentially address the intricate challenges linked with IoT environment. Despite these advantages, security concerns continue to present a noteworthy impact dilemma for IoT. Notably, the attack DDoS (Distributed Denial of Service) emerge as one of the most prominent security threats within IoT systems. This paper delves into a comprehensive attack examination in both IoT and SDN environments. To fortify SDN-IoT technology security, the combination of BC (Blockchain) is explored. Additionally, the Attention based Convolutional long short term memory (At-C-L) is employed to enhance the detection. The proposed attack detection model differentiated the normal and DDoS attacks on the InSDN database and achieved a better accuracy of 98.3%.