Unified Data Path for Flexible Cryptographic Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cryptographic systems are inefficient and costly due to the need for hardware accelerators designed for a fixed number of shares, which are underutilized for most operations, and struggle to balance protection against side-channel attacks while handling both protected and unprotected data efficiently.
Innovation Solution
Implementing a unified data path that allows for both protected and unprotected nonlinear computations using a low number of hardware circuits, enabling flexible protection with different numbers of shares and reducing the complexity and cost of cryptographic accelerators by performing computations sequentially.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If hardware accelerators are designed for a fixed number of shares to protect against side-channel attacks, then security protection is improved, but device complexity and cost increase
Solution Approach 1:
The patent implements a unified data path that can operate in multiple modes: unprotected mode for non-sensitive data and protected mode with configurable number of shares for sensitive data. This single hardware accelerator can adapt to different security requirements without requiring separate dedicated circuits for each share configuration, thereby reducing overall device complexity while maintaining security protection when needed.
Solution Approach 2:
The system dynamically configures the number of shares based on the security sensitivity of the data being processed. The control logic determines whether to operate in unprotected mode or protected mode with a specific number of shares, allowing the hardware accelerator to adapt its complexity level to the actual security requirements of each operation, thus avoiding unnecessary complexity for non-sensitive operations.
2Reliability
If hardware accelerators are designed for a fixed number of shares, then protection against side-channel attacks is improved, but productivity decreases due to underutilization
Solution Approach 1:
The unified data path serves dual purposes: it processes both protected and unprotected data through the same hardware circuits. When processing unprotected data or data with lower security requirements, the system operates in a more efficient mode without the overhead of multiple share computations, thereby improving overall productivity while maintaining the capability to provide strong protection when needed.
Solution Approach 2:
The system changes the operational parameters (number of shares) based on the security requirements of the data being processed. For non-sensitive data, it uses fewer shares or no sharing at all, reducing computational overhead and improving processing efficiency. For sensitive data, it increases the number of shares to provide stronger protection, thus optimizing the balance between security and productivity dynamically.
3Reliability
If separate hardware circuits are used for protected and unprotected data, then reliability is improved, but device complexity and cost increase
Solution Approach 1:
The patent merges the processing paths for protected and unprotected data into a single unified data path. The same hardware circuits perform both protected nonlinear computations (with share splitting and recombination) and unprotected computations, controlled by control logic that routes operations appropriately. This eliminates the need for separate dedicated hardware circuits for each type of data processing, reducing device complexity and cost while maintaining reliability through proper isolation of sensitive operations.
Solution Approach 2:
The unified data path is designed to be universal, handling both protected and unprotected data through the same physical circuits. This multi-functional approach reduces the total number of hardware components required while ensuring that protected data operations maintain their security properties through the use of share splitting and recombination logic integrated into the same processing path.
Data Source
AI summary
Disclosed systems and techniques include a cryptographic processor for processing of both unprotected data and protected data using an unprotected data path. In one implementation, the cryptographic processor includes a processing unit, and a control unit to selectively cause the processing unit to operate in a public mode or a secure mode. In the public mode, the processing unit performs a computational operation to compute a nonlinear function of a public data. In the secure mode, the processing unit computes, over a plurality of iterations, a plurality of shares of the nonlinear function of a secure data. At each iteration, the processing unit performs multiple instances of the computational operation to compute a respective share of the nonlinear function of the secure data.