Memory Sharing via Incompatible Data Transformation Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing memory sharing devices face security challenges, particularly with error injection attacks and unauthorized access, as components are physically accessible and Memory Protection Units (MPUs) do not adequately counter all types of attacks.
Innovation Solution
Implementing data transformation functions and incompatible inverse transformation functions for each functional entity to ensure data integrity and security, where each entity applies specific transformation functions before writing and inverse functions upon reading, with error status provision to prevent unauthorized access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If Memory Protection Units (MPUs) are used to protect memory areas, then access control between functional entities is improved, but error injection attacks on the address bus are not countered and device complexity increases
Solution Approach 1:
The memory space is segmented into multiple independent memory areas, each associated with a specific functional entity. Each memory area has its own transformation function and inverse transformation function, creating isolated security zones that prevent cross-entity attacks while maintaining overall system simplicity.
Solution Approach 2:
Data is transformed using entity-specific transformation functions before being written to memory areas. This preliminary transformation encodes the data in a way that only the intended functional entity can reverse, preventing unauthorized access and error injection attacks before they can affect the actual data.
2Adaptability or versatility
If functional entities share the same memory support, then device integration is improved, but physical accessibility enables attacks by malicious actors
Solution Approach 1:
Each memory area is given local quality through entity-specific transformation functions. The data in each memory area is encoded with characteristics unique to its intended functional entity, making it locally secure even though all memory areas physically share the same support structure.
Solution Approach 2:
Transformation functions act as intermediaries between functional entities and memory areas. These intermediary functions encode data with entity-specific characteristics, creating a protective layer that prevents direct physical attacks while allowing legitimate access through the intermediary transformation process.
3Reliability
If data transformation functions are assigned to each functional entity, then data integrity and security are improved, but device complexity increases
Solution Approach 1:
The transformation function and its inverse are merged into a unified entity-specific security mechanism. Each functional entity receives a pair of functions (transformation and inverse transformation) that work together as an integrated security solution, reducing overall system complexity compared to separate protection mechanisms.
Solution Approach 2:
The transformation function framework provides universal protection across all memory areas and functional entities. A single transformation approach can be applied to any memory area by assigning the appropriate entity-specific transformation function, making the security mechanism universally applicable while maintaining entity-specific security.
Data Source
AI summary
In a general aspect, a method for sharing a memory between two functional entities can include assigning, to the first functional entity, a first data transformation function and a first inverse transformation function, and assigning, to the second functional entity a second data transformation function and a second inverse transformation function. The second inverse data transformation function can be incompatible with the first data transformation function and the first inverse data transformation function can be incompatible with the second data transformation function.