What question did this study set out to answer?

The aim is to develop an efficient intrusion detection framework for unmanned aerial vehicle swarm networks.

May 13, 2026Scientific ReportsOpen Access

Residual-aware lightweight deep learning framework for high-fidelity intrusion detection in UAV swarm networks

View Full Paper

Ask AI

Bookmark

Authors

The novel framework enhances intrusion detection in unmanned aerial vehicle networks, suggesting enhanced scalability against cyber-attacks.

The aim is to develop an efficient intrusion detection framework for unmanned aerial vehicle swarm networks.
Developed a Residual-inspired One-Dimensional Convolutional Neural Network framework.
Integrated residual skip connections for improved gradient flow and feature extraction.
Applied dynamic class-weighting to address data imbalance in training processes.
Utilized memory-time profiling for efficiency in resource-constrained UAV systems.
Achieved 99.71% accuracy in detection performance on the UAVIDS-2025 dataset.
Obtained a macro F1-score of 0.9971 and a ROC-AUC of 0.9999.
Maintained a compact size with 474,629 trainable parameters and 1.81 MB footprint.
Demonstrated an inference time of approximately 0.0023 seconds per sample.

Wáng et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

The novel framework enhances intrusion detection in unmanned aerial vehicle networks, suggesting enhanced scalability against cyber-attacks.