Secure Database Aggregation via Static Shifting and Homomorphic Encryption
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data encryption methods, such as AES and FHE, hinder efficient data aggregation in databases by preventing servers from processing encrypted data, leading to increased bandwidth usage, processing resources, and potential data leakage, while fully homomorphic encryption is too inefficient for practical applications.
Innovation Solution
The system employs static shifting and additive homomorphic encryption to encode group membership into encrypted data, allowing servers to perform aggregation operations without decrypting the data, thereby reducing processing power and preventing data leakage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data is encrypted using standard database encryption techniques before uploading to a shared server, then data security is improved, but the server is unable to perform grouping or data aggregation functionality on the encrypted data, leading to increased bandwidth usage and processing resources
Solution Approach 1:
The patent transforms the encryption approach by changing the mathematical parameters of the encryption scheme to enable aggregation. Specifically, it uses additive homomorphic encryption where encrypted values can be added together to produce an encryption of the sum, allowing aggregation operations on encrypted data without decryption.
Solution Approach 2:
The patent introduces an intermediary encoding scheme that maps original data values to transformed encrypted representations. This intermediary layer allows the server to perform aggregation operations on the transformed data while maintaining security, as the transformation enables mathematical operations that reveal aggregate statistics without exposing individual values.
2Reliability
If data is encrypted to prevent data leakage, then data security is improved, but individual devices must individually aggregate received non-aggregated datasets, increasing processing resources required
Solution Approach 1:
The patent enables the encrypted data itself to serve the aggregation function directly on the server side. The homomorphic encryption properties allow the server to perform aggregation operations autonomously on the encrypted data without requiring client devices to download and process the complete datasets, thus reducing client processing resources.
3Reliability
If standard encryption methods are used to protect sensitive data, then data security is improved, but bandwidth usage and transmission overhead increase due to inability to aggregate data on the server
Solution Approach 1:
The patent changes the encryption parameters to homomorphic encryption schemes that preserve mathematical structures necessary for aggregation. This allows the server to compute aggregates of encrypted values and return only the aggregated results, dramatically reducing the bandwidth required compared to transmitting and processing complete encrypted datasets at client devices.
Data Source
AI summary
Disclosed herein are system, method, and computer program product embodiments for secure data aggregation in databases. An embodiment operates by identifying a value column and a group column of a plurality of columns of a dataset. Two distinct group values of the group column are identified. An offset value corresponding to the first group value is determined. One or more of the plurality of records including the first group value are identified. A value of the value column of each of the identified one or more plurality of records is encoded with the offset value. Values of the encoded value column are encrypted. The encrypted values are uploaded to a server.