Privacy-Preserving Querying on Semi-Trusted Cloud
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for querying encrypted data on semi-trusted networks fail to preserve the privacy of both the data and the entity requesting the query, as they either rely on untrusted cloud service providers for decryption or limit data sharing and querying features.
Innovation Solution
A method and system utilizing an asymmetric cryptosystem, such as the Goldwasser-Micali cryptosystem, combined with a private comparison protocol like Fischlin's protocol, where a survey server manages queries by receiving encrypted queries, performing private comparisons with encrypted data, and returning randomized results to ensure secure and private querying without revealing sensitive information to untrusted entities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the cloud service provider is allowed to decrypt the sensitive data stored in the cloud in order to run the requested query, then the querying capability is improved, but the data security and privacy are compromised
Solution Approach 1:
The private key is segmented into multiple shares distributed to different entities (data owner and cloud service provider). Neither entity alone can decrypt the data, but together they can perform queries. This segmentation resolves the contradiction by enabling querying functionality while preventing any single entity from accessing the plaintext data.
Solution Approach 2:
The patent introduces an intermediary cryptographic protocol that mediates between the data owner and cloud service provider. The protocol allows the cloud provider to perform comparisons on encrypted data without seeing the plaintext, acting as a trusted intermediary that enables querying while preserving security.
2Object-affected harmful factors
If users keep the decryption keys to ensure access is not given to the cloud service providers, then the data security is improved, but the data sharing and querying features are limited
Solution Approach 1:
The decryption capability is segmented into multiple key shares held by different parties. The data owner holds one share and the cloud provider holds another. This allows the cloud provider to participate in querying operations without having full decryption capability, thus maintaining security while enabling data sharing and querying features.
Solution Approach 2:
The segmented key system provides multi-functionality: it maintains security by preventing unauthorized access, enables data sharing among authorized parties, and supports querying operations on encrypted data. This universal approach resolves the contradiction by achieving multiple goals simultaneously.
3Object-affected harmful factors
If Attribute Based Encryption (ABE) is used to encrypt and store PHRs on semi-trusted servers, then the patient privacy is preserved, but the ability of health organizations to query PHRs is prevented
Solution Approach 1:
The patent segments the decryption capability into key shares, allowing the cloud provider to perform querying operations without having the complete private key. This enables health organizations to query PHRs while patient privacy is preserved, as no single entity can access the plaintext data.
Solution Approach 2:
The patent replaces the traditional ABE mechanism with a segmented key system that uses homomorphic encryption and comparison protocols. This substitution allows querying operations to be performed on encrypted data without requiring the data to be accessible in plaintext, thus maintaining privacy while enabling querying.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for a survey server for managing query communications between at least a requester server and at least one data holding entity is provided. The requester server holds a first share of a private key and the survey server holds a second share of the private key. An encrypted query is received and then forwarded to at least one data holding entity. A plurality of comparison responses is received. Each comparison response is generated by a private comparison protocol that compares the encrypted query with encrypted data. The encrypted data having been encrypted using the public key. Each comparison response having been partially decrypted with the first share of the private key and placed in the array in a randomized order. The array is decrypted using the second share of the private key. At least one result of the query is determined.