Scholarlink

Unmanned Aerial Vehicles (UAVs) and their Ground Control Stations (GCS) are increasingly being utilized across various industries. However, their vulnerability to cyber threats, particularly malware, presents significant security challenges. Traditional defenses, such as digital signatures, are becoming less effective against sophisticated attacks. In this paper, we propose a machine learning-based method for classifying and predicting malicious software within UAV and GCS systems based on API call sequences. We evaluated several models, including Naive Bayes, Logistic Regression, Random Forest, and Support Vector Machines (SVM). Our results indicate that, Random Forest and SVM offers the highest accuracy and efficiency for real-time malware detection in both UAVs and GCS. This study highlights the importance of incorporating advanced machine learning techniques into UAV and GCS cybersecurity frameworks to enhance protection against emerging threats.

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

Enhancing Malware Detection for UAV Ground Control Stations: A Comparative Study of Machine Learning Models

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Soltane Merzoug,
  • Abdeljalil Gattal,
  • Mohamed Larbi Gasri,
  • Amina Kemache,
  • Abdellatif Gahmousse

摘要

Unmanned Aerial Vehicles (UAVs) and their Ground Control Stations (GCS) are increasingly being utilized across various industries. However, their vulnerability to cyber threats, particularly malware, presents significant security challenges. Traditional defenses, such as digital signatures, are becoming less effective against sophisticated attacks. In this paper, we propose a machine learning-based method for classifying and predicting malicious software within UAV and GCS systems based on API call sequences. We evaluated several models, including Naive Bayes, Logistic Regression, Random Forest, and Support Vector Machines (SVM). Our results indicate that, Random Forest and SVM offers the highest accuracy and efficiency for real-time malware detection in both UAVs and GCS. This study highlights the importance of incorporating advanced machine learning techniques into UAV and GCS cybersecurity frameworks to enhance protection against emerging threats.