Deterministic Block Cipher Encryption Without Initialization Vectors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing encryption schemes that use initialization vectors face security issues due to non-trivial communication and storage costs, potential compromise of security if vectors are not unique or random, and the need for identical ciphertexts for identical plaintexts in applications like deduplication and searchable encryption.
Innovation Solution
A deterministic block cipher chaining scheme that segments data into equal length blocks, swaps subsets of data between blocks, and encrypts them separately without using an initialization vector, ensuring secure encryption without leaking common parts of distinct plaintexts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If initialization vectors are used in encryption schemes, then semantic security is improved, but communication and storage costs increase
Solution Approach 1:
The patent extracts and removes the initialization vector component from the encryption scheme, replacing it with a deterministic processing approach that operates directly on the plaintext blocks without requiring additional vector data for security
Solution Approach 2:
The patent segments the plaintext into fixed-size blocks and processes each block deterministically through the encryption function, allowing secure encryption without initialization vectors by treating each block independently through repeated encryption rounds
2Reliability
If initialization vectors are used in encryption schemes, then security is improved, but security compromise risk increases if vectors are not unique or random
Solution Approach 1:
The encryption scheme becomes self-sufficient by using the plaintext itself and deterministic processing to generate secure ciphertext, eliminating the external dependency on initialization vectors that could introduce security vulnerabilities if not properly managed
Solution Approach 2:
The patent introduces deterministic processing rounds as an intermediary mechanism between plaintext and ciphertext, replacing the initialization vector intermediary with a controlled computational process that eliminates security risks associated with vector management
3Reliability
If initialization vectors are used in encryption schemes, then semantic security is improved, but identical ciphertexts cannot be produced for identical plaintexts
Solution Approach 1:
The patent inverts the conventional approach by making encryption deterministic rather than probabilistic, allowing identical plaintexts to produce identical ciphertexts while maintaining security through repeated encryption rounds and block processing
Data Source
AI summary
In some examples, a non-transitory machine readable storage medium has machine readable instructions to cause a computer processor to segment a datastream into a plurality of equal length blocks each of which has a fixed length, separately encrypt each equal length block using a first encryption key, swap a subset of bits of a first encrypted equal length block with a subset of bits of a second encrypted equal length block such that both of the blocks each have a length equal to the fixed length, and separately encrypt each block using a second encryption key.