Personalization Data Integration in Nonvolatile Memory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for introducing personalization data in nonvolatile memories of integrated circuits, such as eUICC, face challenges including the need for dedicated memory areas, lengthy first boot routines, simulator unreliability, and slow personalization times, particularly in Wafer Level Chip Scale Packaging (WLCSP) processes.
Innovation Solution
A method that generates test data images iteratively to identify invariant and dynamic memory areas, allowing for the creation of a static image and dynamic personalized image, which are then used to personalize integrated circuits without requiring a dedicated personalization data area or first boot routine, and can be performed on a test integrated circuit similar to the production batch, eliminating the need for simulators and reducing personalization time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If personalization data is introduced using conventional methods with dedicated memory areas and first boot routines, then personalization can be performed, but the process becomes slow and requires additional memory management complexity
Solution Approach 1:
The patent applies preliminary action by pre-organizing personalization data into a substitution table during the image generation phase, before the actual personalization process. The static image is prepared with placeholder positions for all personalization data fields, and the substitution table maps each field's position and length. This allows the personalization process to simply perform direct data replacement without complex memory management during operation, significantly speeding up the personalization process while reducing runtime complexity.
2Ease of manufacture
If simulators are used to generate test data images, then the process can be performed without physical test integrated circuits, but reliability decreases due to simulator unreliability
Solution Approach 1:
The patent uses real physical test integrated circuits from the production batch to generate test data images, creating accurate copies of the actual production conditions. Instead of using simulators that may not perfectly replicate hardware behavior, the method programs actual test chips with test personalization data and reads back the results. This ensures that the generated static images and substitution tables accurately reflect the true behavior of production integrated circuits, maintaining high reliability while still allowing iterative testing and validation.
3Ease of operation
If personalization data is stored in dedicated memory areas, then data organization is simplified, but additional memory space is consumed and first boot routines become necessary
Solution Approach 1:
The patent merges the personalization data storage with the existing operating system image structure. Instead of creating separate dedicated memory areas for personalization data, the method integrates personalization fields directly into the static image at specific positions within the existing memory layout. The substitution table tracks the positions and lengths of these embedded personalization fields, allowing them to be filled during the personalization process without requiring additional memory space or separate management routines. This approach maintains ease of operation through systematic organization while avoiding extra memory consumption.
Data Source
AI summary
Embodiments of the present disclosure relate to solutions for introducing personalization data in nonvolatile memories of a plurality of integrated circuits, comprising writing in the nonvolatile memory of a given integrated circuit a static data image, corresponding to an invariant part of nonvolatile memory common to the plurality of integrated circuits, and a personalization data image representing data specific to the given integrated circuit.


