Effective Intrusion Detection for UAV Communications using Autoencoder-based Feature Extraction and Machine Learning Approach
Published in 2024 International Symposium on Nonlinear Theory and Its Applications, 2024
Abstract
Unmanned Aerial Vehicles (UAVs) have seen widespread adoption across various domains, but their wireless communication systems remain vulnerable to cyber attacks. This paper addresses this critical security challenge by proposing a novel intrusion detection system specifically designed for UAV communication networks.
Our approach combines autoencoder-based feature extraction with advanced machine learning algorithms to detect malicious activities in UAV communication data. The autoencoder architecture effectively reduces dimensionality while preserving essential security-relevant information, allowing subsequent classification algorithms to identify attack patterns with higher accuracy and lower computational overhead.
We evaluate our system using a comprehensive UAV communication dataset that includes various attack scenarios. Experimental results demonstrate that our approach achieves superior detection accuracy compared to conventional methods, with particularly strong performance in identifying sophisticated attacks. The proposed system maintains high detection rates even under challenging conditions such as resource constraints and communication variability typical in UAV operations.
This research contributes to enhancing the security of UAV communications by providing an efficient and accurate intrusion detection mechanism that can be implemented within the limited computational resources available in UAV environments.
Keywords
UAV Security, Intrusion Detection, Autoencoder, Feature Extraction, Machine Learning, Cyber Security, Wireless Communication Security
Recommended citation: Vuong Tuan-Cuong, Cong Chi Nguyen, Pham Van-Cuong, Le Thi-Thanh-Huyen, Tran Xuan-Nam, Luong Thien Van. (2024). "Effective Intrusion Detection for UAV Communications using Autoencoder-based Feature Extraction and Machine Learning Approach." 2024 International Symposium on Nonlinear Theory and Its Applications, pp. 798-804.
Download Paper