TEPPIT: TEstbed for Privacy-PreservIng Technologies for Data Sharing and Analysis

Project Description

The project “TEPPIT: TEstbed for Privacy-PreservIng Technologies for Data Sharing and Analysis” will build and operate TEPPIT, a testbed to support assessment, comparative analysis, vulnerability and threat analysis, privacy risk assessments, and privacy-utility trade-off analysis of privacy-preserving technologies for data sharing and analysis (PPDSA) and of applications of those technologies.

TEPPIT is one of several projects funded through the NSF Privacy-Preserving Data Sharing in Practice (PDaSP) program, which advances practical, deployable approaches for privately sharing and analyzing data across a range of use cases and application domains. In addition to developing its own testbed capabilities, TEPPIT will provide an evaluation environment for privacy-preserving solutions developed by other PDaSP projects.

The project’s goal is to support assessment of both new PPDSA solutions, and to support assessment of how existing PPDSA solutions apply to real-world uses such as in AI applications. These assessments will improve the understanding of the trade-offs around privacy leakage, utility, performance, and scalability of PPDSA solutions.

TEPPIT builds on Security and Privacy Heterogeneous Environment for Reproducible Experimentation (SPHERE), the NSF-funded mid-scale research infrastructure for cybersecurity and privacy research. TEPPIT will extend SPHERE with diverse, rich, full scenarios for assessment of PPDSA solutions and applications, with new PPDSA-specific software tools and user interfaces, and with focused building of the PPDSA research community.

TEPPIT will improve evaluation of PPDSA in three ways:

(1) It will provide a rich, modular, extensible and composable evaluation scenario framework, with a sample of PPDSA technologies and applications to support to allow researchers reuse, combine, and extend evaluation workflow to explore their specific question.

(2) TEPPIT will provide PPDSA-relevant hardware systems and software tools, including virtual or bare-metal machines with different trusted compute technologies, servers with GPUs, and resource-constrained embedded CPUs and IoT devices, all connected by user-specified emulated network. These platforms will allow researchers to evaluate PPDSA approaches across diverse and realistic scenarios.

(3) TEPPIT will grow the research community for PPDSA and their applications, through workshops, tutorials, and BoFs at community events, as well as through support for easy research artifact storage and reuse. Our goal is to promote sharing, collaboration, and reproducible research.

TEPPIT’s broader impacts will be to improve privacy technologies and applications and accelerate research maturing and transition to practice. TEPPIT will also support community building and workforce education.

Support

TEPPIT is supported by NSF/TIP and NSF/CISE as proposal #2453092.

People

  • Jose Luis Ambite, co-PI on this project, principal scientist and research associate professor (USC/ISI)
  • David Balenson, sr. supervising computer scientist and director, networking and cybersecurity division (USC/ISI)
  • John Heidemann, co-PI on this project, project leader and professor (USC/ISI)
  • Jelena Mirkovic, PI on this project, project leader and assistant professor (USC/ISI)
  • Yuri Pradkin, researcher (USC/ISI)

Publications

  • Sulyab Thottungal Valapu, John Heidemann, Mattijs Jonker and Raffaele Sommese 2026. Zombies in Alternate Realities: The Afterlife of Domain Names in DNS Integrations. Technical Report arxiv:2605.06880 [cs.CR]. USC/Information Sciences Institute. [PDF] Details

For related publications, please see the ANT publications web page.

Software

See also the see the ANT distribution web page.

Datasets

We also make all datasets available through the ANT dataset page.