Distributed Computing Anonymity via Trusted Execution Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distributed computing systems face challenges in collecting data from numerous devices while preserving anonymity, as existing techniques like differential privacy often compromise data accuracy due to bandwidth and latency requirements.
Innovation Solution
A distributed computing system utilizing trusted execution environments (TEEs) to receive and decrypt encrypted data, determine trust between nodes, and generate anonymized computation results, ensuring data is anonymized and secure through multiple stages of computation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If differential privacy techniques are used to preserve anonymity, then anonymity is improved, but data accuracy deteriorates
Solution Approach 1:
The patent divides the distributed computing system into multiple nodes, each performing partial computations on encrypted data. This segmentation allows the system to maintain anonymity through encryption while improving data accuracy by distributing and aggregating computational results across multiple nodes, avoiding the need for differential privacy noise addition.
Solution Approach 2:
The patent introduces trusted execution environments (TEEs) as intermediary secure computing modules within each node. These TEEs act as mediators that can process encrypted data and generate accurate computational results while preserving anonymity, resolving the contradiction between anonymity and accuracy by providing a trusted intermediate layer.
2Measurement precision
If large amounts of data are collected to improve accuracy, then data accuracy is improved, but bandwidth and latency requirements worsen
Solution Approach 1:
The patent enables each distributed node to perform local computations on encrypted data within its own trusted execution environment. This local processing approach improves data accuracy through distributed computation while reducing bandwidth requirements, as only encrypted data and computational results need to be transmitted rather than processing large volumes of raw data centrally.
Solution Approach 2:
The patent performs encryption and computational processing in advance within trusted execution environments before data transmission. This preliminary action allows accurate computations to be completed on encrypted data, reducing the need for subsequent data transmission and processing, thereby lowering bandwidth and latency requirements.
3Measurement precision
If data is decrypted for computation, then computation accuracy is improved, but security deteriorates
Solution Approach 1:
The patent introduces trusted execution environments as secure intermediary modules that handle data decryption and computation. These TEEs act as protected intermediaries between encrypted data input and computational processing, ensuring that decryption occurs only within secure boundaries, thus maintaining both computation accuracy and security simultaneously.
Solution Approach 2:
The patent creates a secure, isolated computational environment within trusted execution environments where data decryption and processing occur. This inert secure environment protects decrypted data from external access or interference, allowing accurate computations on decrypted data while maintaining security through isolation from untrusted system components.
Data Source
Figure 1
Figure 2~5
Figure 3
AI summary
According to an example aspect of the present invention, there is provided receiving first encrypted data from at least one data source. First decrypted data is obtained by decrypting the first encrypted data using a trusted execution environment (TEE) of a node in a distributed computing system. An intermediate result is produced by determining a trust to a subsequent node, generating a computation result based on the first decrypted data, obtaining an encrypted computation result by encrypting the computation result for the subsequent node, and transmitting the encrypted computation result to the subsequent node if the subsequent node is deemed trusted. Producing a final result comprises determining a trust to a previous node, and controlling, on the basis of the previous node being deemed trusted, at least one of generating an anonymized computation result, and transmitting the anonymized computation result to a data receiver.