Intrusion Detection for Smart Environmental Drones

Citation

Huzaifa, Malik and Attaullah, Hafiz Muhammad and Nawaz, Menaa and Rahman, Hameedur (2025) Intrusion Detection for Smart Environmental Drones. Artificial General Intelligence-Based Drones for Climate Change. pp. 153-182. ISSN 2327-0411

Full text not available from this repository.

Abstract

AGI-enabled drones are vital yet vulnerable systems, especially in climate monitoring and autonomous aerial missions, making cybersecurity a top priority. Dynamic, decentralized UAVs face cyber threats like DoS, spoofing, replay, and FDI attacks. This research proposes and evaluates lightweight ML-based Intrusion Detection Systems (IDS) tailored for UAVs. Five supervised models—Naïve Bayes, Decision Tree, K-Nearest Neighbors, Random Forest, and LightGBM—were tested across three datasets: a CTGAN-augmented mix of CIC-IDS2017 and UNSW-NB15, a physical-cyber UAV testbed dataset, and a T-ITS behavioral dataset. The experimental design included preprocessing, normalization, class balancing, and performance evaluation using accuracy, precision, recall, F1-score, and confusion matrix. LightGBM achieved the highest accuracy and generalization, while Decision Tree and Naïve Bayes offered optimal efficiency and speed, making them ideal for onboard IDS in AGI-driven drones.

Item Type: Article
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7800-8360 Electronics > TK7871 Electronics--Materials
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Suzilawati Abu Samah
Date Deposited: 30 Jun 2025 04:56
Last Modified: 30 Jun 2025 04:56
URII: http://shdl.mmu.edu.my/id/eprint/14168

Downloads

Downloads per month over past year

View ItemEdit (login required)