Scholarlink

<p>The integration of Unmanned Aerial Vehicles (UAVs) into 6G networks offers numerous benefits but also introduces significant security challenges, particularly concerning the confidentiality of sensitive information transmitted by UAVs. Physical layer security (PLS) techniques provide a promising approach to complement traditional cryptographic methods. We investigated the application of machine learning (ML) techniques for detecting pulse jamming attacks and classical energy detector (ED) for comparison, a critical threat to secure UAV communication. We propose a framework where a single UAV can leverage ML classifiers to perform hypothesis testing based on received signal characteristics. The performance of Support Vector Machine (SVM), Logistic Regression (LR), and K-Nearest Neighbors (KNN) classifiers is evaluated using simulated data under varying Signal-to-Noise Ratio (SNR) and Signal-to-Jamming Ratio (SJR) conditions. Receiver Operating Characteristic (ROC) curves are generated to analyze the detection capabilities of each classifier, providing insights into the trade-off between detection probability and false alarm rate. The results demonstrate the feasibility of using ML for real-time pulse jamming detection in 6G UAV scenario, paving the way for the activation of adaptive anti-jamming techniques and enhanced PLS.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Physical layer security in 6G UAVs with machine learning for pulse jamming detection: a baseline study

  • 期刊论文

  • 发表日期：

    2026年2月16日

查看全文
  • Ahmed Moumena

摘要

The integration of Unmanned Aerial Vehicles (UAVs) into 6G networks offers numerous benefits but also introduces significant security challenges, particularly concerning the confidentiality of sensitive information transmitted by UAVs. Physical layer security (PLS) techniques provide a promising approach to complement traditional cryptographic methods. We investigated the application of machine learning (ML) techniques for detecting pulse jamming attacks and classical energy detector (ED) for comparison, a critical threat to secure UAV communication. We propose a framework where a single UAV can leverage ML classifiers to perform hypothesis testing based on received signal characteristics. The performance of Support Vector Machine (SVM), Logistic Regression (LR), and K-Nearest Neighbors (KNN) classifiers is evaluated using simulated data under varying Signal-to-Noise Ratio (SNR) and Signal-to-Jamming Ratio (SJR) conditions. Receiver Operating Characteristic (ROC) curves are generated to analyze the detection capabilities of each classifier, providing insights into the trade-off between detection probability and false alarm rate. The results demonstrate the feasibility of using ML for real-time pulse jamming detection in 6G UAV scenario, paving the way for the activation of adaptive anti-jamming techniques and enhanced PLS.