To obtain access to full text of journal and articles you must register!
- Article name
- Early indirect leakage probability in data enclaves
- Authors
- Galmanov P. A., , galmanov.pa@phystech.edu, Moscow Institute of Physics and Technology (National Research University); SC "OKB SAPR"; JSC "Sberbank-Technology", Moscow Region, Dolgoprudny, Russia; Moscow, Russia
- Keywords
- indirect leaks / security / IT pipeline / blind data processing / Markov model / leak threshold
- Year
- 2026 Issue 1 Pages 40 - 48
- Code EDN
- VEEZUS
- Code DOI
- 10.52190/2073-2600_2026_1_40
- Abstract
- A probabilistic formulation of the problem of early indirect leakage in blind data processing systems is proposed. Early indirect leakage is defined as compromise occurring before the detectability threshold is reached or under a prescribed limit on the number of queries. Two characteristics are introduced: the probability pA(N) of successful recovery of a protected attribute under a query budget N, and the distribution of the time to compromise t. For the linear-Gaussian model, analytical expressions for pA(N) are obtained, and a recursive method for computing the distribution of t is developed based on a Markov representation of the error trajectory. A normalization with respect to the success-probability threshold is introduced, and a protocol is proposed for estimating the effective noise level from repeated queries in order to construct risk curves and choose limits on the number of queries. It is shown that limiting the number of queries below the detectability threshold does not eliminate the risk, but only reduces it gradually.
- Text
- BUY for read the full text of article
- Buy
- 500.00 rub
