Format Preserving Encryption Padding Length Obfuscation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Format-preserving encryption (FPE) reveals information about the plaintext through the length of the ciphertext, making it vulnerable to inference attacks, especially when the encrypted data has a unique length, such as extremely short credit card numbers or names, which can compromise the security of specific fields in databases.
Innovation Solution
The technique involves adding an effectively random length of padding characters to the plaintext, ensuring the ciphertext maintains the same format but obscures the original length, using a sequence of characters that can be identified for decryption, thus preventing inference about the data type.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If format-preserving encryption is applied to encrypt plaintext, then the ciphertext maintains the same format as the plaintext, but the length of the ciphertext reveals information about the plaintext
Solution Approach 1:
The patent applies preliminary action by adding padding to the plaintext before encryption. The padding is determined in advance based on the plaintext length and a predefined padding scheme, ensuring that the ciphertext length does not reveal information about the original plaintext length. This preliminary padding step resolves the contradiction by preparing the data in advance to maintain format preservation while hiding length information.
Solution Approach 2:
The patent uses padding as an intermediary element between the plaintext and ciphertext. The padding acts as a mediator that bridges the gap between the variable-length plaintext and the fixed-length ciphertext requirement. By introducing this intermediary padding layer, the system maintains format preservation while preventing length information leakage, as the padding masks the relationship between plaintext and ciphertext lengths.
2Ease of manufacture
If the ciphertext length is kept the same as plaintext length, then format preservation is maintained, but attackers can infer data type and focus decryption attempts
Solution Approach 1:
The patent applies preliminary action by pre-determining the padding length based on the plaintext length before encryption. This ensures that regardless of the original plaintext length, the ciphertext will have a standardized length that does not reveal information about the data type. The preliminary padding step prevents attackers from inferring whether the data is numeric, alphabetic, or other types based on ciphertext length.
Solution Approach 2:
The patent changes the parameter of ciphertext length independence from plaintext length by introducing padding. Instead of maintaining a one-to-one relationship between plaintext and ciphertext lengths, the system transforms this relationship by adding padding, thereby changing the parameter space and preventing length-based inference attacks while maintaining format preservation.
3Object-affected harmful factors
If padding is added to obscure plaintext length, then security against inference attacks is improved, but the complexity of encryption and decryption processes increases
Solution Approach 1:
The patent applies local quality by implementing padding only where necessary - specifically when the plaintext length requires it to maintain format preservation. The padding is applied locally to the plaintext before encryption, and the same padding is removed locally during decryption. This localized approach minimizes the increase in process complexity while maintaining security benefits.
Solution Approach 2:
The patent uses discarding and recovering by adding padding during encryption and then removing the same padding during decryption. The padding is discarded (added) during the encryption phase to mask length information, and then recovered (removed) during the decryption phase to restore the original plaintext. This reversible process manages the complexity by introducing and then eliminating the padding, maintaining security while enabling correct decryption.
Data Source
AI summary
Techniques for using padding in format preserving encryption are provided. In one aspect, it may be determined if padding of a plaintext undergoing format preserving encryption is needed. A pseudo random padding length may be calculated when it is determined that padding is needed. The calculated length of padding may be added to the plaintext when it is determined that padding is needed. The plaintext and added padding may be encrypted using format preserving encryption to create a cipher text.


