A Model-Hardware Co-design Framework for Robust and Efficient CNN-Based SAR ATR

Convolutional Neural Networks (CNNs) have achieved state-of-the-art accuracy in Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR), but their high computational cost, latency, and memory usage make it challenging on resource-constrained platforms such as small satellites. While adversarial robustness is critical for real-world SAR ATR, it is often overlooked in system-level optimizations. Addressing both challenges requires more than model compression or accelerator design alone, demanding joint optimization in a unified framework.