# Adversarial Robustness of Distilled and Pruned Deep Learning-based Wireless Classifiers

arXiv: https://arxiv.org/abs/2404.15344 · HTML: https://mlsanity.com/paper/2404.15344/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 22 references that are in the map.
- First posted on arXiv 11 Apr 2024.

## Builds on (most cited first, 11 of 11)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Adversarial Examples: Attacks and Defenses for Deep Learning](https://mlsanity.com/paper/1712.07107/) (2017)
- [Knowledge distillation: A good teacher is patient and consistent](https://mlsanity.com/paper/2106.05237/) (2021)
- [Adversarial Attacks on Deep-Learning Based Radio Signal Classification](https://mlsanity.com/paper/1808.07713/) (2018)
- [Deep Architectures for Modulation Recognition](https://mlsanity.com/paper/1703.09197/) (2017)
- [A Survey on Deep Neural Network Compression: Challenges, Overview, and Solutions](https://mlsanity.com/paper/2010.03954/) (2020)
- [Integrating Sensing and Communications for Ubiquitous IoT: Applications, Trends and Challenges](https://mlsanity.com/paper/2104.11457/) (2021)
- [Deep Learning Power Allocation in Massive MIMO](https://mlsanity.com/paper/1812.03640/) (2018)
- [Robust Adversarial Attacks Against DNN-Based Wireless Communication Systems](https://mlsanity.com/paper/2102.00918/) (2021)
- [SafeAMC: Adversarial training for robust modulation recognition models](https://mlsanity.com/paper/2105.13746/) (2021)
- [Sensing and Classification Using Massive MIMO: A Tensor Decomposition-Based Approach](https://mlsanity.com/paper/2109.00821/) (2021)

## Abstract (arXiv)

Data-driven deep learning (DL) techniques developed for automatic modulation classification (AMC) of wireless signals are vulnerable to adversarial attacks. This poses a severe security threat to the DL-based wireless systems, specifically for edge applications of AMC. In this work, we address the joint problem of developing optimized DL models that are also robust against adversarial attacks. This enables efficient and reliable deployment of DL-based AMC on edge devices. We first propose two optimized models using knowledge distillation and network pruning, followed by a computationally efficient adversarial training process to improve the robustness. Experimental results on five white-box attacks show that the proposed optimized and adversarially trained models can achieve better robustness than the standard (unoptimized) model. The two optimized models also achieve higher accuracy on clean (unattacked) samples, which is essential for the reliability of DL-based solutions at edge applications.