keenan
Uploaded by
26 SLIDES
408 VIEWS
260LIKES

Privacy-Preserving Data Aggregation: Balancing Smart Grids and User Privacy

DESCRIPTION

This work explores how to enable data analytics while preserving individual privacy in scenarios like smart grids and population surveys. We present a novel approach using homomorphic encryption and differential privacy to allow an untrusted data aggregator to perform meaningful analysis without compromising sensitive user information. Our methods support periodic aggregation and ensure no user interaction is needed post-upload. We discuss challenges, expressiveness of queries, and potential future improvements in privacy-preserving data analytics.

1 / 26

Download Presentation

Privacy-Preserving Data Aggregation: Balancing Smart Grids and User Privacy

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

Playing audio...

  1. Privacy-Preserving Stream Aggregation Elaine Shi (PARC/UC Berkeley), T-H. Hubert Chan (HKU), Eleanor Rieffel (FXPal), Richard Chow (PARC), Dawn Song (UC Berkeley)

  2. Privacy in Smart Grids Smart grid operator Time-series data

  3. Privacy in Population Surveys

  4. How can we allow a data aggregator to perform data analytics, while preserving individual privacy?

  5. Our Results – Privacy Notion

  6. Computing on Multiple Users’ Encrypted Data

  7. Homomorphic Encryption? (PK, SK)

  8. New Paradigm SK5 SK4 SK1 SK3 SK2

  9. New Paradigm Evaluate SK4 SK5 SK3 SK2 SK1

  10. Expressiveness: Summation Evaluate

  11. Expressiveness: Distributions Evaluate

  12. Aggregate Once: Simple Construction SK4 SK5 SK3 SK2 SK1

  13. Aggregate Once: Simple Construction … … SK4 SK5 SK3 SK2 SK1

  14. Aggregate Once: Simple Construction … … SK4 SK5 SK3 SK2 SK1

  15. Multiple Time Steps … … SK4 SK5 SK3 SK2 SK1

  16. Differential Privacy against an Untrusted Aggregator

  17. Differential Privacy [Dwork06] ? 0 0 1 1 1 1 0 1 1 1 8 neighboring vectors x and x’, 8 sets of transcripts S: Pr [π (x) є S] ≤ exp(ε) ∙ Pr [π (x’) є S]

  18. Naïve Scheme Error: v1+ r1 v2+ r2 v3+ r3 v4+ r4 v5+ r5

  19. Crypto + Differential Privacy Error: … … Enc(v1+ ρ1) Enc(v5+ ρ5)

  20. Open Problems and Future Work • More expressive queries • Larger plaintext space • Fault tolerance [CSS10] • Reduce privacy loss over multiple time steps [CSS10]

  21. Take-Home Messages • Differential Privacy against an Untrusted Aggregator • The Power of Combining Cryptography and Differential Privacy

  22. Thank you!

  23. Our Results – Property • Periodic aggregation • Non-interactive • No interactions among users • Users upload ciphertext to aggregator, and no more communication needed

  24. Power of Combining Crypto and Differential Privacy [CSS10]

  25. Privacy in Sensor Networks • Building monitoring • Employee sensing • Body sensor nets • …

  26. Privacy in Market Research

More Related