CRCauligi Raghavendra
Cauligi Raghavendra
*This is an algorithmically generated author record based on publicly available information.
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Cauligi Raghavendra has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Sachini Wickramasinghe
University of Southern California, USA
2 shared publications - Tian Ye
University of Southern California, Los Angeles, USA
2 shared publications - Viktor Prasanna
University of Southern California
2 shared publications
Latest Publications
- Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Reconfigurable Technology and Systems
ARMOR: Robust and Efficient CNN-Based S AR ATR through Mo del-Ha r dware Co-Design
Convolutional Neural Networks (CNNs) have achieved state-of-the-art accuracy in Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR). However, their high computational cost, latency, and memory footprint make deployment 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. Achieving both robustness and inference efficiency requires a unified framework that considers adversarially trained models together with hardware constraints. In this paper, we present a model-hardware co-design framework for CNN-based SAR ATR that integrates robustness-preserving model compression with FPGA accelerator design. The compression stage includes hardware-guided structured pruning, where a hardware performance model derived from the FPGA design predicts the pruning impact on latency and resource usage. This enables the generation of Pareto-optimal models that improve hardware efficiency under user-defined objectives while maintaining adversarial robustness within a predefined tolerance. We design an FPGA accelerator with channel-aware processing element (PE) allocation that supports both fully pipelined streaming and temporal resource-reuse architectures. An automated design generation flow efficiently maps the compressed models to optimized FPGA implementations. Experiments on the widely used MSTAR and FUSAR-Ship datasets across three CNN architectures show that our framework produces models up to 18.3 \(\times\) smaller with 3.1 \(\times\) fewer MACs while preserving robustness. Our FPGA implementation achieves up to 68.1 \(\times\) (6.4 \(\times\) ) lower inference latency and up to 169.7 \(\times\) (33.2 \(\times\) ) better energy efficiency compared to CPU (GPU) baselines, demonstrating the effectiveness of the proposed co-design framework for robust and efficient SAR ATR on FPGA platforms.- 11 August 2026
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Reconfigurable Technology and Systems
Open AccessARMOR: Robust and Efficient CNN-Based S AR ATR through Mo del-Ha r dware Co-Design
Convolutional Neural Networks (CNNs) have achieved state-of-the-art accuracy in Synthetic Aperture Radar (SAR) Automatic Target Recognition (ATR). However, their high computational cost, latency, and memory footprint make deployment 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. Achieving both robustness and inference efficiency requires a unified framework that considers adversarially trained models together with hardware constraints. In this paper, we present a model-hardware co-design framework for CNN-based SAR ATR that integrates robustness-preserving model compression with FPGA accelerator design. The compression stage includes hardware-guided structured pruning, where a hardware performance model derived from the FPGA design predicts the pruning impact on latency and resource usage. This enables the generation of Pareto-optimal models that improve hardware efficiency under user-defined objectives while maintaining adversarial robustness within a predefined tolerance. We design an FPGA accelerator with channel-aware processing element (PE) allocation that supports both fully pipelined streaming and temporal resource-reuse architectures. An automated design generation flow efficiently maps the compressed models to optimized FPGA implementations. Experiments on the widely used MSTAR and FUSAR-Ship datasets across three CNN architectures show that our framework produces models up to 18.3 \(\times\) smaller with 3.1 \(\times\) fewer MACs while preserving robustness. Our FPGA implementation achieves up to 68.1 \(\times\) (6.4 \(\times\) ) lower inference latency and up to 169.7 \(\times\) (33.2 \(\times\) ) better energy efficiency compared to CPU (GPU) baselines, demonstrating the effectiveness of the proposed co-design framework for robust and efficient SAR ATR on FPGA platforms. - Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 39
A Model-Hardware Co-design Framework for Robust and Efficient CNN-Based SAR ATR
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 39
Open AccessA Model-Hardware Co-design Framework for Robust and Efficient CNN-Based SAR ATR