Secured DDoS Attack Detection in SDN Using TS-RBDM With MDPP-Streebog Based User Authentication

Abstract

In a Distributed Denial of Service (DDoS) attack, the attacker aims to render a network resource unavailable to its intended users. A novel Software Defined Networking (SDN)-centered secured DDoS attack detection system is presented in this paper by utilizing TanhSoftmax-Restricted Boltzmann Dense Machines (TS-RBDM) with a Mean Difference of Public key and Private key based Streebog (MDPP-Streebog) user authentication algorithm. Primarily, in the registration phase, the users have registered their device details. The two-stage login process is performed after successful registration. Then, in the network layer, the nodes are initialized, and via the Gate/Router, the sensed data is transmitted to the SDN controller to enhance network energy efficiency. Later, by using the CIC DDoS 2019 dataset, the DDoS detection system is trained. This dataset undergoes preprocessing, and features are extracted from it. By employing the Adaptive Synthetic (ADASYN) technique, data balancing is achieved. Lastly, by using the TS-RBDM technique, the data is trained. The sensed data is categorized as either attacked or non-attacked data within this trained DDoS detection system. By employing the Entropy Binomial probability-based Shanon-Fano-Elias (EB-SFE) technique, the non-attacked data will be encoded and transmitted to the receiving terminal. Lastly, the experiential assessment illustrated that the proposed DDoS detection system attained 98% accuracy with 37 485 ms minimal training time, thus outperforming all state-of-the-art methods.