Ruzica Piskac
Yale University, CT·Computer Science
Computers leak information through their physical behavior. How much power a chip draws, how long an operation takes, or what electromagnetic field it emits can all depend on the secret data being processed, and an attacker who measures those "side channels" can recover keys without ever breaking the math of the encryption.
- Fields
Computer Science
Engineering
Physics
- Topics
side-channel attacks
power analysis
electromagnetic leakage
- Grants
- 2 known grants, through Aug 2028
What they work on
- Automatically locate power and electromagnetic side-channel vulnerabilities in low-level code by combining dynamic taint tracking and random testing with symbolic execution, satisfiability modulo theories solving and model counting.
- Develop register analysis techniques based on Hamming weight and Hamming distance leakage models, which tie power consumption to bit flips in registers holding sensitive data, to formally describe leakage at each intermediate variable.
- Repair vulnerable code locations using syntax-guided synthesis, accepting or rejecting candidate fixes with test-vector leakage assessment measured on real hardware.
- Build prototypes of a quantum trusted execution environment that keeps a user's computation confidential on shared quantum hardware.
- Develop a systematic understanding of physical attacks, including timing and power-based side channels, on noisy intermediate-scale and fault-tolerant quantum computing architectures, and design remediation techniques for them.
More about the research
Two projects attack this problem from different ends: one makes ordinary software provably free of power and electromagnetic leaks using automated program analysis, and the other extends side-channel thinking to quantum computers, which are expensive, shared among many users, and have barely been examined for security.
Why it matters
Side-channel attacks can extract cryptographic keys and private data from devices that are mathematically secure, and writing leak-free code by hand is error-prone. Automated detection and repair tools could let ordinary developers ship software that protects user data. On the quantum side, the work could allow expensive quantum machines to be shared among many users without one user stealing another's algorithms or data, which matters as valuable intellectual property moves onto these systems.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Himanshu Thapliyal
Southern Methodist University, TX·Engineering·Early-career
Investigate how secret information leaks through the power traces of energy recovery circuits, and propose design methods that reduce leakage against differential power analysis attacks.
Yuyan Bao
Augusta University, GA·Computer Science
Develop verification methodologies for reasoning about function values — functions handled as data — in industrial programming languages, using first-order assertions and auxiliary logical variables suited to automated reasoning.
Todd Brun
University of Southern California, CA·Physics
Develop weakly fault-tolerant quantum computing using error-detecting codes that store n–2 logical qubits in n physical qubits, detecting single faults and discarding runs that show errors, for use on near-term processors.
The school
Yale University: 73 labs on PI Pages, $1.5B a year on research, 438 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.