Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
12 tokens/sec
GPT-4o
12 tokens/sec
Gemini 2.5 Pro Pro
41 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
37 tokens/sec
DeepSeek R1 via Azure Pro
33 tokens/sec
2000 character limit reached

A Novel Federated Learning-Based IDS for Enhancing UAVs Privacy and Security (2312.04135v3)

Published 7 Dec 2023 in cs.CR and cs.LG

Abstract: Unmanned aerial vehicles (UAVs) operating within Flying Ad-hoc Networks (FANETs) encounter security challenges due to the dynamic and distributed nature of these networks. Previous studies focused predominantly on centralized intrusion detection, assuming a central entity responsible for storing and analyzing data from all devices. However, these approaches face challenges including computation and storage costs, along with a single point of failure risk, threatening data privacy and availability. The widespread dispersion of data across interconnected devices underscores the need for decentralized approaches. This paper introduces the Federated Learning-based Intrusion Detection System (FL-IDS), addressing challenges encountered by centralized systems in FANETs. FL-IDS reduces computation and storage costs for both clients and the central server, which is crucial for resource-constrained UAVs. Operating in a decentralized manner, FL-IDS enables UAVs to collaboratively train a global intrusion detection model without sharing raw data, thus avoiding delay in decisions based on collected data, as is often the case with traditional methods. Experimental results demonstrate FL-IDS's competitive performance with Central IDS (C-IDS) while mitigating privacy concerns, with the Bias Towards Specific Clients (BTSC) method further enhancing FL-IDS performance even at lower attacker ratios. Comparative analysis with traditional intrusion detection methods, including Local IDS (L-IDS), sheds light on the strengths of FL-IDS. This study significantly contributes to UAV security by introducing a privacy-aware, decentralized intrusion detection approach tailored to UAV networks. Moreover, by introducing a realistic dataset for FANETs and federated learning, our approach differs from others lacking high dynamism and 3D node movements or accurate federated data federations.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (53)
  1. I. Bekmezci, O. K. Sahingoz, and Ş. Temel, “Flying ad-hoc networks (fanets): A survey,” Ad Hoc Networks, vol. 11, no. 3, pp. 1254–1270, 2013.
  2. A. D. Boursianis, M. S. Papadopoulou, P. Diamantoulakis, A. Liopa-Tsakalidi, P. Barouchas, G. Salahas, G. Karagiannidis, S. Wan, and S. K. Goudos, “Internet of things (iot) and agricultural unmanned aerial vehicles (uavs) in smart farming: A comprehensive review,” Internet of Things, vol. 18, p. 100187, 2022.
  3. G. A. Kakamoukas, P. G. Sarigiannidis, and A. A. Economides, “Fanets in agriculture-a routing protocol survey,” Internet of Things, vol. 18, p. 100183, 2022.
  4. J.-P. Yaacoub, H. Noura, O. Salman, and A. Chehab, “Security analysis of drones systems: Attacks, limitations, and recommendations,” Internet of Things, vol. 11, p. 100218, 2020.
  5. J. Sharma and P. S. Mehra, “Secure communication in iot-based uav networks: A systematic survey,” Internet of Things, p. 100883, 2023.
  6. K.-Y. Tsao, T. Girdler, and V. G. Vassilakis, “A survey of cyber security threats and solutions for uav communications and flying ad-hoc networks,” Ad Hoc Networks, vol. 133, p. 102894, 2022.
  7. A. Chriki, H. Touati, H. Snoussi, and F. Kamoun, “Fanet: Communication, mobility models and security issues,” Computer Networks, vol. 163, p. 106877, 2019.
  8. G. Amponis, T. Lagkas, P. Sarigiannidis, V. Vitsas, P. Fouliras, and S. Wan, “A survey on fanet routing from a cross-layer design perspective,” Journal of Systems Architecture, vol. 120, p. 102281, 2021.
  9. O. Ceviz, P. Sadioglu, and S. Sen, “Analysis of routing attacks in fanets,” in Ad Hoc Networks and Tools for IT.   Springer, 2021, pp. 3–17.
  10. M. Nayfeh, Y. Li, K. Al Shamaileh, V. Devabhaktuni, and N. Kaabouch, “Machine learning modeling of gps features with applications to uav location spoofing detection and classification,” Computers & Security, vol. 126, p. 103085, 2023.
  11. S. Ouiazzane, F. Barramou, and M. Addou, “Towards a Multi-Agent based Network Intrusion Detection System for a Fleet of Drones,” Int. J. Adv. Comput. Sci. Appl., vol. 11, no. 10, pp. 351–362, 2020.
  12. C. Zhang, Y. Xie, H. Bai, B. Yu, W. Li, and Y. Gao, “A survey on federated learning,” Knowledge-Based Systems, vol. 216, p. 106775, 2021.
  13. Z. Baig, N. Syed, and N. Mohammad, “Securing the smart city airspace: Drone cyber attack detection through machine learning,” Future Internet, vol. 14, no. 7, p. 205, 2022.
  14. Labs, “V. drone forensics,” 2020. [Online]. Available: https://www.vtolabs.com/drone-forensics(accessedon15April2022)
  15. L. M. da Silva, H. B. d. B. Menezes, M. d. S. Luccas, C. Mailer, A. S. R. Pinto, A. Boava, M. Rodrigues, I. G. Ferrão, J. C. Estrella, and K. R. L. J. C. Branco, “Development of an efficiency platform based on mqtt for uav controlling and dos attack detection,” Sensors, vol. 22, no. 17, p. 6567, 2022.
  16. C. Kolias, G. Kambourakis, A. Stavrou, and S. Gritzalis, “Intrusion detection in 802.11 networks: Empirical evaluation of threats and a public dataset,” IEEE Communications Surveys & Tutorials, vol. 18, no. 1, pp. 184–208, 2015.
  17. R. Shrestha, A. Omidkar, S. A. Roudi, R. Abbas, and S. Kim, “Machine-learning-enabled intrusion detection system for cellular connected uav networks,” Electronics, vol. 10, no. 13, p. 1549, 2021.
  18. D. Chulerttiyawong and A. Jamalipour, “Sybil attack detection in internet of flying things-ioft: A machine learning approach,” IEEE Internet of Things Journal, 2023.
  19. A. Varga, “Omnet++,” in Modeling and tools for network simulation.   Springer, 2010, pp. 35–59.
  20. W. Zhai, L. Liu, Y. Ding, S. Sun, and Y. Gu, “Etd: An efficient time delay attack detection framework for uav networks,” IEEE Transactions on Information Forensics and Security, 2023.
  21. A. Keränen, J. Ott, and T. Kärkkäinen, “The one simulator for dtn protocol evaluation,” in Proceedings of the 2nd international conference on simulation tools and techniques, 2009, pp. 1–10.
  22. Q. Abu Al-Haija and A. Al Badawi, “High-performance intrusion detection system for networked uavs via deep learning,” Neural Computing and Applications, vol. 34, no. 13, pp. 10 885–10 900, 2022.
  23. L. Zhao, A. Alipour-Fanid, M. Slawski, and K. Zeng, “Prediction-time efficient classification using feature computational dependencies,” in Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2018, pp. 2787–2796.
  24. R. A. Ramadan, A. H. Emara, M. Al-Sarem, and M. Elhamahmy, “Internet of drones intrusion detection using deep learning,” Electron., vol. 10, no. 21, 2021.
  25. KDD Cup 1999, “Kdd cup 1999,” http://kdd.ics.uci.edu/databases/kddcup99/kddcup99.html, October 2007.
  26. M. Tavallaee, E. Bagheri, W. Lu, and A. A. Ghorbani, “A detailed analysis of the kdd cup 99 data set,” in 2009 IEEE Symposium on Computational Intelligence for Security and Defense Applications, 2009, pp. 1–6.
  27. O. Bouhamed, O. Bouachir, M. Aloqaily, and I. Al Ridhawi, “Lightweight ids for uav networks: A periodic deep reinforcement learning-based approach,” in 2021 IFIP/IEEE International Symposium on Integrated Network Management (IM).   IEEE, 2021, pp. 1032–1037.
  28. R. Panigrahi and S. Borah, “A detailed analysis of cicids2017 dataset for designing intrusion detection systems,” International Journal of Engineering & Technology, vol. 7, no. 3.24, pp. 479–482, 2018.
  29. M. P. Arthur, “Detecting signal spoofing and jamming attacks in uav networks using a lightweight ids,” in 2019 international conference on computer, information and telecommunication systems (CITS).   IEEE, 2019, pp. 1–5.
  30. H. Sedjelmaci, S. M. Senouci, and N. Ansari, “A Hierarchical Detection and Response System to Enhance Security Against Lethal Cyber-Attacks in UAV Networks,” IEEE Trans. Syst. Man, Cybern. Syst., vol. 48, no. 9, pp. 1594–1606, 2018.
  31. N. I. Mowla, N. H. Tran, I. Doh, and K. Chae, “Federated learning-based cognitive detection of jamming attack in flying ad-hoc network,” IEEE Access, vol. 8, pp. 4338–4350, 2019.
  32. ——, “Afrl: Adaptive federated reinforcement learning for intelligent jamming defense in fanet,” Journal of Communications and Networks, vol. 22, no. 3, pp. 244–258, 2020.
  33. L. M. Da Silva, I. G. Ferrão, C. Dezan, D. Espes, and K. R. Branco, “Anomaly-based intrusion detection system for in-flight and network security in uav swarm,” in 2023 International Conference on Unmanned Aircraft Systems (ICUAS).   IEEE, 2023, pp. 812–819.
  34. “The ns-3 network simulator,” http://www.nsnam.org/.
  35. O. Puñal, C. Pereira, A. Aguiar, and J. Gross, “Crawdad dataset uportorwthaachen/vanetjamming2014 (v. 2014-05-12),” 2014.
  36. J. Whelan, T. Sangarapillai, O. Minawi, A. Almehmadi, and K. El-Khatib, “Novelty-based intrusion detection of sensor attacks on unmanned aerial vehicles,” in Proceedings of the 16th ACM symposium on QoS and security for wireless and mobile networks, 2020, pp. 23–28.
  37. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Artificial intelligence and statistics.   PMLR, 2017, pp. 1273–1282.
  38. S. Reddi, Z. Charles, M. Zaheer, Z. Garrett, K. Rush, J. Konečnỳ, S. Kumar, and H. B. McMahan, “Adaptive federated optimization,” arXiv preprint arXiv:2003.00295, 2020.
  39. M. A. Ferrag, O. Friha, L. Maglaras, H. Janicke, and L. Shu, “Federated deep learning for cyber security in the internet of things: Concepts, applications, and experimental analysis,” IEEE Access, vol. 9, pp. 138 509–138 542, 2021.
  40. S. A. Rahman, H. Tout, C. Talhi, and A. Mourad, “Internet of things intrusion detection: Centralized, on-device, or federated learning?” IEEE Network, vol. 34, no. 6, pp. 310–317, 2020.
  41. I. Sharafaldin, A. H. Lashkari, and A. A. Ghorbani, “Toward generating a new intrusion detection dataset and intrusion traffic characterization.” ICISSp, vol. 1, pp. 108–116, 2018.
  42. N. Moustafa and J. Slay, “Unsw-nb15: a comprehensive data set for network intrusion detection systems (unsw-nb15 network data set),” in 2015 military communications and information systems conference (MilCIS).   IEEE, 2015, pp. 1–6.
  43. I. Almomani, B. Al-Kasasbeh, M. Al-Akhras et al., “Wsn-ds: A dataset for intrusion detection systems in wireless sensor networks,” Journal of Sensors, vol. 2016, 2016.
  44. N. Moustafa, “New generations of internet of things datasets for cybersecurity applications based machine learning: Ton_iot datasets,” in Proceedings of the eResearch Australasia Conference, Brisbane, Australia, 2019, pp. 21–25.
  45. N. Koroniotis, N. Moustafa, E. Sitnikova, and B. Turnbull, “Towards the development of realistic botnet dataset in the internet of things for network forensic analytics: Bot-iot dataset,” Future Generation Computer Systems, vol. 100, pp. 779–796, 2019.
  46. I. Vaccari, G. Chiola, M. Aiello, M. Mongelli, and E. Cambiaso, “Mqttset, a new dataset for machine learning techniques on mqtt,” Sensors, vol. 20, no. 22, p. 6578, 2020.
  47. S. Agrawal, S. Sarkar, O. Aouedi, G. Yenduri, K. Piamrat, M. Alazab, S. Bhattacharya, P. K. R. Maddikunta, and T. R. Gadekallu, “Federated learning for intrusion detection system: Concepts, challenges and future directions,” Computer Communications, 2022.
  48. H. Ullah, M. Abu-Tair, S. McClean, P. Nixon, G. Parr, and C. Luo, “An unmanned aerial vehicle based wireless network for bridging communication,” in Proc. - 14th Int. Symp. Pervasive Syst. Algorithms Networks, I-SPAN 2017, 11th Int. Conf. Front. Comput. Sci. Technol. FCST 2017 3rd Int. Symp. Creat. Comput. ISCC 2017, 2017.
  49. L. Che, J. Wang, Y. Zhou, and F. Ma, “Multimodal federated learning: A survey,” Sensors, vol. 23, no. 15, p. 6986, 2023.
  50. A. Daniel, “A Survey on Detection of Sinkhole Attack in Wireless Sensor Networks,” vol. 91, no. 7, pp. 48–52, 2014.
  51. O. Ceviz, P. Sadioglu, and S. Sen, “A survey of security in uavs and fanets: Issues, threats, analysis of attacks, and solutions,” arXiv preprint arXiv:2306.14281, 2023.
  52. S. Sen and J. A. Clark, “Evolutionary computation techniques for intrusion detection in mobile ad hoc networks,” Computer Networks, vol. 55, no. 15, pp. 3441–3457, 2011.
  53. E. Canbalaban and S. Sen, “A cross-layer intrusion detection system for rpl-based internet of things,” in Ad-Hoc, Mobile, and Wireless Networks: 19th International Conference on Ad-Hoc Networks and Wireless, ADHOC-NOW 2020, Bari, Italy, October 19–21, 2020, Proceedings 19.   Springer, 2020, pp. 214–227.

Summary

  • The paper introduces a federated learning-based IDS that decentralizes intrusion detection to enhance UAV privacy and reduce computational load.
  • It utilizes local model weight aggregation to maintain data privacy in FANETs while effectively countering sinkhole, blackhole, and flooding attacks.
  • Experimental results indicate that the FL-IDS achieves similar performance to centralized systems, highlighting its scalability and potential for broader IoT security applications.

Federated Learning-Based Intrusion Detection for UAV Privacy and Security

The paper under review discusses an innovative approach to Intrusion Detection Systems (IDS) specifically designed for Unmanned Aerial Vehicles (UAVs) operating within Flying Ad-hoc Networks (FANETs). UAVs have become integral to various critical applications such as surveillance and disaster management. However, the dynamic and distributed nature of FANETs poses significant security challenges, particularly when using traditional centralized intrusion detection approaches. Significant computational and storage burdens, alongside potential privacy violations and a single point of failure, necessitate exploring decentralized methodologies. This is where the Federated Learning-based IDS (FL-IDS) proposed in the paper becomes relevant, offering a decentralized and privacy-conscious solution to these challenges.

Methodology

FL-IDS operates on the principle of federated learning (FL), a decentralized machine learning approach that allows for privacy-preserving training by enabling UAVs to collaboratively build a global intrusion detection model without sharing raw data. This model addresses the unique challenges of FANETs by reducing computation and storage load on resource-constrained UAVs and eliminating the possible latency caused by data transmission to centralized servers. By aggregating locally updated model weights rather than raw data, FL-IDS maintains data privacy and effectively mitigates risks associated with unauthorized data access.

Experimental Results and Analysis

The paper utilizes realistic FANET scenarios to simulate sinkhole, blackhole, and flooding attacks against the Ad-Hoc On-Demand Distance Vector (AODV) routing protocol using the NS-3 simulator. These attacks were chosen due to their pertinence in exposing vulnerabilities in existing systems. The results demonstrate that FL-IDS achieves competitive performance relative to the traditional centralized IDS (C-IDS), particularly at higher attacker ratios, which signals a robust tendency to infer the anomaly patterns collaboratively even in challenging scenarios.

The quantitative results suggest that at higher attacker ratios, FL-IDS approaches the efficacy levels of C-IDS, particularly when employing methods such as Bias Towards Specific Clients (BTSC) to prioritize client contributions that enhance model accuracy. This strategy further bolsters FL-IDS's standing, marking a significant advancement over Local IDS (L-IDS), which demonstrated lower accuracy due to the lack of collaboration among individual nodes.

Implications and Future Directions

The implications of adopting FL-IDS extend beyond enhanced UAV security; they promote the integration of AI-driven security solutions into broader IoT and autonomous systems, where data privacy and distributed processing resources present similar challenges. The research opens avenues for employing additional federated learning methodologies, such as differential privacy or secure multi-party computations, to further enhance the security protocols in deployment environments that prioritize confidentiality alongside performance efficiency.

Looking forward, future iterations of FL-IDS can benefit from reducing communication costs further and optimizing the balance between model complexity and computation resource constraints. Exploring different federated learning algorithms and assessing their impact on broader network topographies would also be beneficial. Moreover, the creation of even more comprehensive datasets reflecting diverse attack vectors in variable node density environments could greatly fortify the empirical base from which FL-IDS solutions are derived.

In conclusion, the FL-IDS presented constitutes an important stride towards addressing the significant privacy and security challenges intrinsic to UAV operations in FANETs. Its ability to harness and improve upon decentralized data for intrusion detection while maintaining robustness in the face of increasing attack ratios underpins its applicability as a viable, scalable security framework in evolving network paradigms.

File Document Download Save Streamline Icon: https://streamlinehq.com PDF File Document Download Save Streamline Icon: https://streamlinehq.com Markdown
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets