Order-Preserving Tokenization for Secure Cloud Data Retrieval
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cloud computing security is compromised due to the lack of enterprise-level controls in third-party cloud-based services, necessitating effective data encryption methods that preserve sort order for efficient data retrieval and storage.
Innovation Solution
The implementation of an order-preserving tokenization (OPT) method using a mapping table and a tail bounded probability distribution sampler to generate tokens that preserve the sort order of plaintext data, enabling efficient encryption and decryption while maintaining computational efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional encryption methods are used to secure cloud data, then data security is improved, but data retrieval efficiency deteriorates due to loss of sort order
Solution Approach 1:
The patent transforms the encryption approach by changing the parameter of ciphertext generation to preserve sort order. Instead of using traditional encryption that randomly maps plaintext to ciphertext, the system uses order-preserving encryption where the ciphertext values maintain the same relative order as plaintext values, enabling efficient range queries and sorting on encrypted data without decryption
Solution Approach 2:
The patent introduces an order-preserving encryption function as an intermediary between plaintext data and ciphertext storage. This intermediary transformation layer maintains the structural properties of the original data (sort order) while providing cryptographic protection, allowing cloud service providers to perform efficient operations on encrypted data without accessing the plaintext
2Productivity
If order-preserving encryption is implemented to maintain sort order, then data retrieval efficiency is improved, but computational intensity increases
Solution Approach 1:
The patent applies preliminary action by pre-processing the data through order-preserving encryption before cloud storage. The encryption transformation is performed once during data upload, and the resulting ciphertext maintains sort order properties that enable efficient querying without requiring complex computational operations during retrieval. The mapping relationships are established in advance, allowing O(log n) or O(1) query operations
Solution Approach 2:
The system changes the cryptographic approach from traditional high-computation encryption to order-preserving encryption with optimized mathematical properties. By using specific mathematical transformations that preserve order (such as deterministic encryption schemes with bounded output ranges), the computational complexity is reduced while maintaining both security and query efficiency
3Adaptability or versatility
If third-party cloud services are used for data storage, then accessibility and scalability are improved, but enterprise control over data security deteriorates
Solution Approach 1:
The patent introduces order-preserving encryption as an intermediary control mechanism that enables enterprises to maintain security control while using third-party cloud services. The enterprise controls the encryption keys and the encryption/decryption process, while the cloud provider only handles encrypted data. This intermediary cryptographic layer allows the cloud provider to perform efficient operations without compromising enterprise security requirements
Solution Approach 2:
The patent segments the data security control into two parts: the enterprise controls the encryption keys and sensitive operations, while the cloud provider controls the storage infrastructure and efficient query processing. This segmentation of control allows both parties to operate within their respective domains while maintaining overall system security and efficiency
Data Source
AI summary
An order-preserving tokenization (OPT) method receives a plaintext and generates a token as the ciphertext which preserves the sort order of the plaintext by using stored legacy plaintext-to-ciphertext mapping data and a tail bounded probability distribution sampler to sample a random order preserving function. More specifically, the OPT method uses a mapping table to store plaintext-to-ciphertext mapping for previously generated ciphertexts. The mapping table enables efficient searching of the ciphertext space as the OPT method only needs to look for ciphertext in the space that most likely will have the desired plaintext. In this manner, the order-preserving tokenization method of the present invention realizes a fast algorithm that is also more computational efficient. In one embodiment, the OPT method uses a tail bounded hypergeometric distribution sampler as the probability distribution sampler.