SaTC: CORE: Small: Leveraging Physical Side-Channel Information to Build Detection-Based Rowhammer Defenses

Preventing a process from modifying a memory it does not own is a cornerstone of security. However, this essential protection has recently come under threat due to the discovery of a vulnerability, known as the Rowhammer for dynamic random-access memory (DRAM). It has been found that this vulnerability commonly exists in DRAM chips and can be exploited to form a wide range of powerful attacks.

Defending against Rowhammer attack remains a challenging problem. This project aims to solve this problem by developing novel detection-based defense techniques.

The project has 3 major thrusts. Thrust 1 will analyze certain inevitable physical side effects of computation (e.g., electromagnetic radiation, acoustic emanation, and power consumption) to discover malicious hammering. Thrust 2 will design and implement deployable defense techniques based on the discovered physical side-channel information which can robustly detect and promptly thwart any potential Rowhammer attacks, including elusive ones hidden inside trusted execution environments.

Thrust 3 concerns experiments and evaluation to quantify and demonstrate the significance of the project.

The techniques developed in this project may impact industry and help practitioners design more secure and dependable systems.

Code, data, publications, presentations, will be released publicly. A project repository will be maintained on CPS-VO (https://cps-vo.org/) for at least 5-years.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.