DeepStrike: Remotely-Guided Fault Injection Attacks on DNN Accelerator in Cloud-FPGA
| Autor: | Yunsi Fei, Yukui Luo, Cheng Gongye, Xiaolin Xu |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Cryptography and Security business.industry Computer science Cloud computing Fault injection Fault (power engineering) Virtualization computer.software_genre Embedded system Logic gate Electronic design automation business Field-programmable gate array Cryptography and Security (cs.CR) computer Digital signal processing |
| Zdroj: | DAC |
| Popis: | As Field-programmable gate arrays (FPGAs) are widely adopted in clouds to accelerate Deep Neural Networks (DNN), such virtualization environments have posed many new security issues. This work investigates the integrity of DNN FPGA accelerators in clouds. It proposes DeepStrike, a remotely-guided attack based on power glitching fault injections targeting DNN execution. We characterize the vulnerabilities of different DNN layers against fault injections on FPGAs and leverage time-to-digital converter (TDC) sensors to precisely control the timing of fault injections. Experimental results show that our proposed attack can successfully disrupt the FPGA DSP kernel and misclassify the target victim DNN application. 6 pages, 6 figures |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|