Artificial intelligence-enhanced intrusion detection systems for drone security: a real-time evaluation of algorithmic efficacy in mitigating wireless vulnerabilities


Şentürk K., Gormus A. F., Gönen S., Barışkan M. A., Durmaz A. K.

Cluster Computing, cilt.28, sa.3, 2025 (SCI-Expanded, Scopus) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 28 Sayı: 3
  • Basım Tarihi: 2025
  • Doi Numarası: 10.1007/s10586-024-04911-8
  • Dergi Adı: Cluster Computing
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, PASCAL, Applied Science & Technology Source, Compendex, Computer & Applied Sciences, INSPEC
  • Anahtar Kelimeler: Artificial intelligence, Cyber security, Drone security, Drones, IoT
  • İstanbul Gelişim Üniversitesi Adresli: Evet

Özet

Advancements in science and technology have provided extensive opportunities and conveniences for mankind. One prime example of these advancements is wireless communication technology. This technology provides users with mobility during communication, initiating a paradigm shift. The convenience of wireless communication technology has initiated the production of versatile devices. Among these technologies developed in recent years for observation and detection purposes in various fields, drones have taken a leading role. Drones, with their versatile applications and access to real-time data, are being used in various operations. With such utilization, humans are increasingly interacting with these systems, leading to natural human-drone interaction. However, in these human-drone interactions, as is the case with many wireless devices, security often becomes an afterthought, leaving many drones vulnerable to cyber attacks. The most effective way to protect against these attackers is to conduct vulnerability analyses of the systems we use against emerging threats and address the detected vulnerabilities. This paper investigates the vulnerabilities of wireless communication regarding remote connectivity usage of a commercial drone, the DJI Ryze Tello, with the aim of examining its weaknesses. In this context, a test environment was created to reveal problems and threats in drone technology through attacks executed on the test environment (DEAUTH ATTACK, Port Scan DOS, DDoS, and MitM). Following the identification of these vulnerabilities, an artificial intelligence-based study was carried out to detect these attacks. In the study, the percentages of attack detection using different algorithms were verified with graphs.