Fan Sang (Georgia Institute of Technology), Jaehyuk Lee (Georgia Institute of Technology), Xiaokuan Zhang (George Mason University), Meng Xu (University of Waterloo), Scott Constable (Intel), Yuan Xiao (Intel), Michael Steiner (Intel), Mona Vij (Intel), Taesoo Kim (Georgia Institute of Technology)

Effectively mitigating side-channel attacks (SCAs) in Trusted Execution Environments (TEEs) remains challenging despite advances in existing defenses. Current detection-based defenses hinge on observing abnormal victim performance characteristics but struggle to detect attacks leaking smaller portions of the secret across multiple executions. Limitations of existing detection-based defenses stem from various factors, including the absence of a trusted microarchitectural data source in TEEs, low-quality available data, inflexibility of victim responses, and platform-specific constraints. We contend that the primary obstacles to effective detection techniques can be attributed to the lack of direct access to precise microarchitectural information within TEEs.

We propose SENSE, a solution that actively exposes underlying microarchitectural information to userspace TEEs. SENSE enables userspace software in TEEs to subscribe to fine-grained microarchitectural events and utilize the events as a means to contextualize the ongoing microarchitectural states. We initially demonstrate SENSE’s capability by applying it to defeat the state-of-the-art cache-based side-channel attacks. We conduct a comprehensive security analysis to ensure that SENSE does not leak more information than a system without it does. We prototype SENSE on a gem5-based emulator, and our evaluation shows that SENSE is secure, can effectively defeats cache SCAs, and incurs negligible performance overhead (1.2%) under benign situations.

View More Papers

AnonPSI: An Anonymity Assessment Framework for PSI

Bo Jiang (TikTok Inc.), Jian Du (TikTok Inc.), Qiang Yan (TikTok Inc.)

Read More

Make your IoT environments robust against adversarial machine learning...

Hamed Haddadpajouh (University of Guelph), Ali Dehghantanha (University of Guelph)

Read More

CrowdGuard: Federated Backdoor Detection in Federated Learning

Phillip Rieger (Technical University of Darmstadt), Torsten Krauß (University of Würzburg), Markus Miettinen (Technical University of Darmstadt), Alexandra Dmitrienko (University of Würzburg), Ahmad-Reza Sadeghi (Technical University of Darmstadt)

Read More

MASTERKEY: Automated Jailbreaking of Large Language Model Chatbots

Gelei Deng (Nanyang Technological University), Yi Liu (Nanyang Technological University), Yuekang Li (University of New South Wales), Kailong Wang (Huazhong University of Science and Technology), Ying Zhang (Virginia Tech), Zefeng Li (Nanyang Technological University), Haoyu Wang (Huazhong University of Science and Technology), Tianwei Zhang (Nanyang Technological University), Yang Liu (Nanyang Technological University)

Read More