Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

A federated learning-benchmarking framework for privacy-preserving UAV intrusion detection using adaptive aggregation

Ms Bithi^1,2, Tahani Alsubait³, Amani Ibraheem⁴

¹Skill Morph Research Lab., Skill Morph, Dhaka, Bangladesh.

Scientific Reports

|May 12, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

The relationship between smartphone usage and sleep quality among university students: a cross-sectional study.

Scientific reports·2026

Same author

A novel unified lightweight temporal-spatial transformer approach for intrusion detection in drone networks.

Scientific reports·2026

Same author

A new adaptive federated learning approach for privacy preserving UAV anomaly detection under non-IID distributions.

Scientific reports·2026

Same author

Teacher activities and student participation in university classrooms: a cross-sectional study in the technological era.

Scientific reports·2026

Same author

A novel deep learning approach for intrusion detection in maritime radar networks.

Scientific reports·2025

Same author

A novel deep learning framework with temporal attention convolutional networks for intrusion detection in IoT and IIoT networks.

Scientific reports·2025

The FedDrone-Shield framework enhances Unmanned Aerial Vehicle (UAV) network security by using federated learning for intrusion detection. Advanced aggregation algorithms like FedAdam and ClusterAvg achieve near-perfect accuracy, protecting against cyber-attacks.

Area of Science:

Cybersecurity
Artificial Intelligence
Network Security

Background:

Unmanned Aerial Vehicles (UAVs) face increasing cyber-attacks like Denial of Service (DoS) and identity impersonation.
Centralized Intrusion Detection Systems (IDS) in UAV networks present privacy risks and single points of failure.
Decentralized and privacy-preserving learning paradigms are crucial for robust UAV network security.

Purpose of the Study:

To introduce FedDrone-Shield, a federated learning architecture for detecting UAV intrusions.
To evaluate various aggregation algorithms (FedAvg, FedProx, FedAdam, FedMedian, ClusterAvg) for intrusion detection efficacy.
To assess the performance of FedDrone-Shield under Independent and Identically Distributed (IID) data scenarios.

Main Methods:

Keywords:

Adaptive federated learning Decentralized anomaly detection Lightweight neural networks Privacy-preserving machine learning UAV security

Related Experiment Videos

Implemented a federated learning framework (FedDrone-Shield) for distributed UAV intrusion detection.

Experimented with multiple aggregation algorithms including FedAdam, ClusterAvg, FedMedian, FedAvg, and FedProx.

Conducted experiments on a UAV anomaly detection dataset to evaluate algorithm performance.

Main Results:

FedAdam and ClusterAvg achieved superior performance with 99.98% test accuracy and 0.9999 F1-scores.
Low loss values (0.0009-0.0014) and high precision, recall, and F1-scores (0.9997-0.9999) were observed across all attack types.
FedMedian showed competitive results, while FedAvg and FedProx demonstrated slower convergence and slightly lower accuracy.

Conclusions:

FedDrone-Shield offers a robust, privacy-preserving intrusion detection model for UAV networks.
Adaptive aggregation strategies significantly improve detection accuracy, training efficiency, and data privacy.
The framework serves as a strong benchmark for federated intrusion detection in distributed UAV systems.