In this paper, we have developed a privacy-preserving framework for the implementation of remote dynamical algorithms in the cloud. It is built on the synergy of random coding and system immersion tools from control theory to protect private information. This innovative approach ensures that while using cloud-based algorithms, user privacy remains uncompromised, enabling efficient and secure operations even with sensitive data.
The proposed immersion-based coding scheme provides the same utility as the original algorithm (i.e., when no coding is employed to protect against data inference), (practically) reveals no information about private data, can be applied to large-scale algorithms, is computationally efficient, and offers any desired level of differential privacy without degrading the algorithm utility. This highlights the efficiency and effectiveness of our approach in safeguarding user data.
Collection
[
|
...
]