Search Results

Understanding the movements of travelers is essential for sustainable city planning, and unique identifiers from wireless network access points or smart card check-ins provide the necessary information to count and track individuals as they move between locations. Nevertheless, it is challenging to deal with such uniquely identifying data in a way that does not violate the privacy of individuals. Even though several protection techniques have been proposed, the data they produce can often still be used to track down specific individuals when combined with other external information. To address this issue, we use a novel method based on encrypted Bloom filters. These probabilistic data structures are used to represent sets while preserving privacy under strong cryptographic guarantees. In our setup, encrypted Bloom filters offer statistical counts of travelers as the only accessible information. However, the probabilistic nature of Bloom filters may lead to undercounting or overcounting of travelers, affecting accuracy. We explain our privacy-preserving method and examine the accuracy of counting the number of travelers as they move between locations. To accomplish this, we used a simulated subway dataset. The results indicate that it is possible to achieve highly accurate counting while ensuring that data cannot be used to trace and identify an individual.