Researchers at the Technische Universität München (TUM or Technical University of Munich) have developed a way to prove location without revealing exact details, aiming to improve privacy while ensuring accuracy. Their method, called Zero-Knowledge Location Privacy (ZKLP), uses advanced mathematical proofs known as zero-knowledge proofs and floating-point numbers to let users confirm they’re in a specific area without exposing their exact coordinates.
Location tracking is common in mobile apps, often happening without users realizing it. This data can paint a picture of a person’s habits, work locations, and routines, sometimes leading to serious privacy risks. A 2019 New York Times report suggested how easily commercial location data could identify individuals, including a member of U.S. President Donald Trump’s team, revealing visits to sensitive places like Mar-a-Lago and the Pentagon. With concerns growing over how this data could be misused, the research team at TUM set out to find a way to verify location information while keeping personal data private.
Their approach, ZKLP, allows users to prove they are in a general area, such as a city or park, without giving away their exact position. It’s based on zero-knowledge proofs, which verify a statement without revealing the data behind it. To make this work in a practical way, the researchers introduced the first set of zero-knowledge proof circuits fully compliant with the IEEE 754 standard for floating-point arithmetic, ensuring precise calculations and avoiding errors common in older systems that relied on integer-based math.
The process is efficient, requiring only 64 constraints per operation for 2^15 single-precision floating-point multiplications, significantly reducing computational complexity compared to previous methods. Their optimized implementation uses 15.9 times fewer constraints when using single precision floating-point values and 12.2 times fewer when using double precision, making the system much more reliable and scalable.
One major use case is privacy-preserving peer-to-peer proximity testing. In this setup, two people can check if they are near each other without sharing exact locations. The system works fast—Bob can generate a proof in just 0.26 seconds, and Alice can verify proximity to about 470 peers per second. “Our method shows that location verification is possible and performant while preserving privacy,” said Prof. Sebastian Steinhorst, Professor of Embedded Systems and Internet of Things.
Beyond location verification, the technology developed in this study could have broader applications in cryptography. The floating-point circuits designed for zero-knowledge proofs could be useful in secure machine learning, digital healthcare, and mobility systems, allowing accurate verification while protecting user data. By combining precision and privacy, this research offers a promising way to safeguard location data in an era where tracking is becoming increasingly common.
Source: Technische Universität München, IEEE Computer Society
This article was generated with some help from AI and reviewed by an editor.
4 Comments - Add comment