Secure Processor Comparison Masking Side Channel Attacks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing secure comparison methods in electronic components are vulnerable to 'side channel' attacks, where attackers can infer secret data by analyzing behavioral differences during comparison operations, and previous solutions are difficult to implement or easily detectable.
Innovation Solution
A secure comparative processing method where a processor compares test data with both main secret data and additional secret data, generating variations in behavior that mask the comparison process, making it impossible for external observers to distinguish between the two, by implementing additional operations that create similar electromagnetic leaks and current variations for all input values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional comparison methods are used, then the comparison operation is simple and fast, but the system is vulnerable to side channel attacks
Solution Approach 1:
The patent changes the operational parameters by introducing additional secret data and modifying the comparison program to include extra comparison operations. This transforms the original simple comparison into a more complex procedure that executes the same number of operations regardless of whether the test data matches the main secret data, thereby masking power consumption patterns and resisting side channel attacks
Solution Approach 2:
The patent introduces additional secret data as an intermediary element that mediates the comparison process. By comparing test data against both main secret data and additional secret data, the system creates a layered comparison structure where the additional secret data acts as a protective intermediary, obscuring the true comparison target from side channel analysis
2Reliability
If dummy manipulations are implemented to mask behavior differences, then security against side channel attacks improves, but the implementation becomes very difficult
Solution Approach 1:
The patent creates a universal comparison program structure that handles both matching and non-matching cases through the same sequence of operations. The comparison logic is designed to universally execute all comparison steps regardless of intermediate results, making the program structure uniform and easier to implement while maintaining security against side channel attacks
3Reliability
If random execution times are generated to mask comparison results, then security against side channel attacks improves, but the solution becomes easily detectable by attackers
Solution Approach 1:
Instead of changing the time parameter through random delays, the patent changes the operational structure by introducing additional comparison operations. This maintains consistent execution timing while altering the computational workload, making it harder for attackers to detect security measures without introducing obvious artificial delays
4Reliability
If symmetric programs are implemented to equalize power consumption profiles, then security against side channel attacks improves, but the implementation becomes very difficult
Solution Approach 1:
The patent segments the comparison operation into distinct phases: initial comparison between test data and main secret data, and subsequent comparison with additional secret data. This segmentation allows each phase to be independently structured to contribute to uniform power consumption patterns, simplifying the overall implementation of symmetric program behavior
Data Source
AI summary
Method of secure comparative processing of the type in which a processor of an electronic component compares proof data (EPR, AUTH) that it receives as input with main secret data (CTRL, PIN) stored in said electronic component, characterized in that the processor implements, in addition to the comparison with the secret data (CTRL, PIN), complementary operations on the proof data (EPR, AUTH) which generate on the electronic component a variation of behaviour which is dependent on the proof data (EPR, AUTH) which it receives as input and which is compounded with the behavioural variation related to the comparison with the main secret data (CTRL, PIN).