Intelligent Deep Learning-aided Malicious Node Detection and Encrypted Routing Optimization in Software-Defined Networking
Indian Journal Of Science And Technology · 15.05.2026 · 19 (18) · 1169-1181
DOI: 10.17485/ijst/v19i18.367 ↗Kurz gefasst
Fachartikel in der Zeitschrift Indian Journal Of Science And Technology, online erschienen am 15.05.2026. Verfasst von V Mangaiyarkarasi und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Objectives
This study aims to develop a secure and efficient routing framework for Software-Defined Networking (SDN) environments that can dynamically adapt to network changes while mitigating malicious activities.
Method
A deep learning (DL)-assisted secure routing system is proposed, where a network emulator is employed to simulate real-time network operations with virtual hosts, OpenFlow switches, and communication links. A centralized controller manages traffic flow and installs forwarding rules dynamically. Flow-level statistics are collected from both normal and attack traffic. A Gated Convolution Kolmogorov-Arnold Network (GCKAN) is utilized for detecting malicious nodes through traffic pattern classification. For secure communication, a hybrid Elliptic Curve ElGamal Diffie-Hellman (ECE-DH) scheme is applied for key exchange. Furthermore, the Draco Lizard Optimization (DLO) algorithm is adopted to determine optimal and secure routing paths.
Findings
The proposed framework demonstrates superior performance in terms of security and efficiency. It achieves over 95% accuracy in malicious node detection, reduces energy consumption by 30%, and decreases communication distance by 45%. Comparative analysis with existing methods confirms the effectiveness of the model in enhancing secure routing within SDN-IoT environments. Novelty: The study integrates GCKAN-based malicious node detection, ECE-DH secure key exchange, and DLO-based optimal routing within a unified SDN framework. This combined approach ensures enhanced security, efficient routing, and adaptability to dynamic network conditions. Keywords: Software-Defined Networking (SDN), Malicious Node Detection, Data Encryption, Secure Optimal Routing Framework, Deep Learning (DL), Draco Lizard Optimization (DLO)
Zitieren
APA
Mangaiyarkarasi, V., & Malathi, S. (2026). Intelligent Deep Learning-aided Malicious Node Detection and Encrypted Routing Optimization in Software-Defined Networking. Indian Journal Of Science And Technology, 19(18), 1169-1181. https://doi.org/10.17485/ijst/v19i18.367
Vancouver
Mangaiyarkarasi V, Malathi S. Intelligent Deep Learning-aided Malicious Node Detection and Encrypted Routing Optimization in Software-Defined Networking. Indian Journal Of Science And Technology. 2026;19(18):1169-1181. doi:10.17485/ijst/v19i18.367
BibTeX
@article{Mangaiyarkarasi2026,
author = {Mangaiyarkarasi, V and Malathi, S},
title = {Intelligent Deep Learning-aided Malicious Node Detection and Encrypted Routing Optimization in Software-Defined Networking},
journal = {Indian Journal Of Science And Technology},
year = {2026},
volume = {19},
number = {18},
pages = {1169--1181},
doi = {10.17485/ijst/v19i18.367}
}
EUVIMED