Yixin Sun
University of Virginia, VA·Computer Science
Records of internet traffic help researchers build better networks and spot security threats. But these logs contain sensitive personal information, so usually only a few trusted researchers can see them.
- Fields
Computer Science
Engineering
- Topics
network traffic logs
internet measurement
data anonymization
- Grants
- 1 known grant, through Sep 2027
What they work on
- Build an extensible model that automatically detects personally identifiable information in structured and semi-structured campus network logs, fine-tune it on historical logs, and turn it into a tunable service for live feeds and archives.
- Design tiered levels of data access matched to research needs, using differential privacy (adding calibrated statistical noise) to give provable privacy guarantees for restricted access levels.
- Feed the privacy-preserving access framework back into anonymization so that the system anonymizes logs dynamically based on usage patterns.
- Deploy and evaluate the anonymization and sharing system directly on the University of Virginia and Virginia Tech traffic collection pipelines.
More about the research
That makes results hard to reproduce and limits who can use the data. This collaborative project builds a systematic way to share campus network traffic logs from the University of Virginia (UVA) and Virginia Tech (VT) with the wider research community while protecting user privacy.
Why it matters
Safer sharing of real network traffic data could let many more researchers test networking and cybersecurity ideas on realistic data and reproduce one another's results, without exposing users' personal information. The UVA–VT system could also serve as a template for other campuses and networks that want to share internet measurement data responsibly.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Jelena Mirkovic
University of Southern California, CA·Computer Science
Build SPHERE, a shared research infrastructure with diverse hardware, configurable networks and safe network policies, for reproducible cybersecurity and privacy experiments on IoT, edge computing and programmable networks.
Rose Qingyang Hu
Virginia Polytechnic Institute and State University, VA·Engineering
Reconstruct how wireless signal energy varies with position and direction by combining sparse channel state information measurements with 3D environment layouts rebuilt from images, using radiance-field-inspired models and array signal processing.
Zeyu Ding
Binghamton University, NY·Computer Science
Build automated tools that analyze an SQL database schema and help users interactively define a customizable privacy model specifying which data elements need differential-privacy protection, and which variant.
The school
University of Virginia: 43 labs on PI Pages, $814M a year on research, 413 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.