Secure Text Retrieval via Client-Side Key Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data security systems face challenges in balancing security and efficiency, particularly when encrypting documents for storage, as they often require decrypting documents to search them, which can compromise security and efficiency, especially in scenarios where providers have access to both encrypted documents and keys, making them vulnerable to breaches.
Innovation Solution
A secure text retrieval system where the encryption key is secured locally with the client device, allowing encrypted search terms to be generated for searching encrypted documents without decrypting them, using a combination of word and sentence-level encryption, and storing the encrypted documents and indexers in separate locations to maintain security and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the provider encrypts each word in a document to maintain higher security, then the security level is improved, but the system efficiency deteriorates due to the need to decrypt each word for searching
Solution Approach 1:
The patent segments the encryption approach by applying encryption at different levels (word-level and sentence-level) rather than uniformly encrypting every word. This allows the system to maintain security while reducing the computational overhead associated with decrypting and searching every individual word, thus resolving the contradiction between security and efficiency.
Solution Approach 2:
The patent changes the encryption parameter from encrypting every single word to encrypting at selective levels (word-level or sentence-level based on security requirements). This parameter change allows the system to adjust the balance between security and search efficiency dynamically, improving productivity without significantly compromising reliability.
2Ease of operation
If the provider stores both the encrypted document and the encryption key in the provider's storage networks, then the ease of operation is improved, but the reliability deteriorates due to security risks from potential breaches
Solution Approach 1:
The patent segments the storage location of cryptographic materials by storing the encryption key separately from the encrypted document. The key is stored in a secure key management system while the encrypted document is stored in cloud storage, creating spatial separation that enhances security while maintaining operational convenience through automated key management.
Solution Approach 2:
The patent introduces a key management system as an intermediary between the user and the encrypted document. This intermediary securely manages the encryption key, allowing the provider to store encrypted documents without having access to the decryption key, thus maintaining both ease of operation and enhanced security.
3Ease of operation
If the provider decrypts the document to search for content, then the ease of operation is improved, but the reliability deteriorates as the encryption protection is compromised
Solution Approach 1:
The patent replaces the mechanical decryption process with a cryptographic search mechanism. Instead of decrypting the document to search, the system uses encrypted search techniques where search queries are processed in encrypted form, allowing search capability to be maintained while the encryption protection remains intact during the search process.
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Methods and systems described herein may perform a word-level encryption and a sentence-level encryption of one or more documents. The word-level encryption and the sentence-level encryption may be performed with an encryption key generated by a client device. A document indexer is stored in the one or more storage networks. The document indexer includes encrypted word frequencies and encrypted word position identifiers based on the encrypted words of the one or more encrypted documents. The client device receives search terms and encrypts the search terms with the encryption key. The one or more encrypted documents are identified in the one or more storage networks based on searching with the encrypted search terms and at least one of the encrypted word frequencies and/or the encrypted word position identifiers.