TY  - EJOU
AU  - Emeto, Ifeanyi C. 
AU  - Galadima, Adamu A. 
AU  - Ezeh, Ikechukwu H. 
AU  - Atomatofa, Emmanuel O. 
AU  - Okoloegbo, Christiana A. 

TI  - Optimizing Network Security at the Control Plane through Software Defined Networking (SDN)
T2  - Journal of Cyber Security

PY  - 2026
VL  - 8
IS  - 1
SN  - 2579-0064

AB  - Software-Defined Networking (SDN) introduces centralised network control and programmability, but its centralised control plane creates significant security vulnerabilities that can be exploited by cyber attackers. This study proposes and evaluates a hybrid security framework that integrates machine learning (ML)-based anomaly detection with blockchain-based authentication to enhance the security of the SDN control plane. The study aimed to analyse vulnerabilities in SDN architectures, develop an ML-driven attack detection model, secure API communications using blockchain-based authentication, and evaluate the performance and scalability of the proposed framework. Vulnerability assessment was conducted using Nmap and STRIDE threat modelling, while an LSTM-based anomaly detection model was implemented using TensorFlow. Hyperledger Fabric was employed for secure API authentication and key management, while Mininet was used to emulate the SDN environment and generate network traffic and attack scenarios. The CICIDS2017 dataset, developed by the Canadian Institute for Cybersecurity at the University of New Brunswick, was used to evaluate the machine learning-based intrusion detection component. The dataset provides labelled benign and malicious network traffic representing common attack categories and network-flow features suitable for machine learning-based intrusion detection. The proposed framework integrates detected anomalies with blockchain-based authentication and SDN policy enforcement to support automated threat response. The performance of the proposed framework was evaluated based on detection accuracy, false-positive rate, mitigation latency, throughput, and scalability. The framework is designed to provide improved security, responsiveness, and scalability for SDN control planes and can support security applications in modern network environments, including 5G, IoT, and cloud networks.
KW  - Software-defined networking (SDN); machine learning (ML); anomaly detection; blockchain security; network vulnerabilities

DO  - 10.32604/jcs.2026.084424