Table 2.

Comparison of LDP and global privacy GDP

AspectLDPGDP
Data collectionEach participant perturbs their data before sending it to the aggregatorData collector introduces noise into the aggregated data before releasing it to a third party
Privacy guaranteeProvides better privacy guarantee by perturbing individual data points before aggregationPrivacy protection relies on noise introduced by the data collector, ensuring privacy from third-party attacks
UtilitySignificant utility loss due to perturbation of individual data points, especially with a large number of participantsImproved privacy/utility trade-offs compared to LDP, particularly in scenarios with a large number of participants
PerformanceMay face convergence difficulties with a limited number of participantsRequires a large number of participants for satisfactory performance, making it unsuitable for applications with a limited number of participants
Trust assumptionDoes not require a trusted aggregator; suitable for scenarios with untrusted data collectorsAssumes a trusted aggregator, which may not be practical in distributed contexts and introduces a single point of failure
Source (s): Authors’ own work

or Create an Account

Close Modal
Close Modal