Masked Data Protection via Lookup Table Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for protecting binary data against side channel attacks, such as energy consumption monitoring, are ineffective in maintaining consistent processing times and energy consumption levels, allowing attackers to infer sensitive information.
Innovation Solution
The implementation of masked binary logic operations using lookup tables to consistently apply masked binary logic operations to masked data, ensuring uniform processing times and reduced energy consumption, thereby making side channel attacks more difficult.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If masked binary logic operations are implemented without lookup tables, then device complexity is reduced, but processing time varies and energy consumption becomes unpredictable, enabling side channel attacks
Solution Approach 1:
The patent pre-computes and stores all possible outcomes of masked binary logic operations in lookup tables during the design phase. During actual operation, the system simply retrieves pre-computed results based on input values, eliminating the need for complex real-time computations and ensuring consistent timing and energy consumption patterns that resist side channel attacks.
Solution Approach 2:
The lookup table acts as an intermediary between the masked input data and the output results. Instead of directly computing masked operations which may have variable execution paths, the system uses the lookup table to map inputs to outputs through a standardized retrieval process, thereby masking the actual computational behavior from side channel observers.
2Reliability
If consistent processing times are enforced to prevent side channel attacks, then security against timing analysis improves, but processing speed may be reduced due to uniform execution requirements
Solution Approach 1:
The system executes operations in a periodic and standardized manner by always following the same retrieval sequence from the lookup table, regardless of the actual data values. This periodic execution pattern ensures that timing information does not leak sensitive data while maintaining efficient access to pre-computed results.
Solution Approach 2:
Instead of performing actual computations that may have variable execution paths, the system copies pre-computed results from the lookup table. This copying operation has predictable and consistent timing characteristics, providing security against timing analysis while maintaining the functionality of the masked logic operations.
3Adaptability or versatility
If masked operations are performed without standardization, then operational flexibility is maintained, but energy consumption varies creating vulnerabilities to power analysis attacks
Solution Approach 1:
The lookup table is designed to handle multiple types of masked binary logic operations (AND, OR, XOR, etc.) and various data formats within a single unified structure. This universal approach ensures that all operations are executed through the same standardized retrieval process, eliminating energy consumption variations that could reveal operational type or data values to power analysis attacks.
Solution Approach 2:
The system changes the parameter of execution methodology from variable computation-based operations to fixed retrieval-based operations. By transforming the operational parameters to always use lookup table retrieval regardless of input data characteristics, the energy consumption profile becomes standardized and resistant to power analysis while maintaining full operational flexibility through the comprehensive lookup table design.
Data Source
AI summary
A device includes a memory and cryptographic processing circuitry coupled to the memory. The memory, in operation, stores one or more lookup tables. The cryptographic processing circuitry, in operation, processes masked data and protects the processing of masked data against side channel attacks. The protecting includes applying masked binary logic operations to masked data using lookup tables of the one or more lookup tables.