Signature Verification Offloading for Memory-Constrained IoT Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Memory-constrained electronic devices, such as IoT devices, face challenges in performing cryptographic verifications due to the need for arithmetic operations involving large input values, which exceed their storage capacity and processing capabilities.
Innovation Solution
The method involves offloading computational operations by obtaining and processing portions of input values from a sender device, using hash functions to verify integrity and correctness without storing the entire input or intermediate results, allowing for reduced storage requirements and increased processing bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If cryptographic verification is performed using large input values, then verification accuracy is improved, but storage capacity requirements increase beyond what memory-constrained devices can provide
Solution Approach 1:
The patent divides large input values into smaller portions or chunks that can be processed sequentially. Instead of loading entire large cryptographic values into memory at once, the system processes them in segmented portions, reducing peak memory requirements while maintaining verification accuracy.
Solution Approach 2:
The patent performs preliminary computations and preparations before the main verification process. Intermediate results are computed and stored in optimized formats that reduce memory requirements, and certain preprocessing steps are executed to minimize the storage burden during actual verification.
2Productivity
If large input values are stored for cryptographic operations, then computational completeness is improved, but processing bandwidth is reduced due to memory constraints
Solution Approach 1:
The patent extracts only the necessary portions of large input values that are needed for each computational step, rather than keeping entire values in memory. This extraction approach reduces memory access overhead and improves processing bandwidth by minimizing data movement between storage and computation units.
Solution Approach 2:
The patent changes the representation parameters of cryptographic values, using optimized data formats and encoding schemes that reduce memory footprint. By altering how values are stored and represented, the system achieves better memory efficiency without sacrificing computational completeness.
3Measurement precision
If intermediate results of computational operations are stored, then calculation accuracy is improved, but storage requirements exceed available memory capacity
Solution Approach 1:
The patent implements a strategy of discarding intermediate results that are no longer needed and recovering only those that are required for subsequent computation steps. This selective retention approach maintains calculation accuracy for necessary intermediates while discarding others to free up memory space.
Solution Approach 2:
The patent performs computations with slightly higher precision than strictly necessary for the final result, using partial precision arithmetic. This allows intermediate results to be stored in smaller formats while still achieving the required final accuracy, reducing storage requirements without sacrificing measurement precision.
Data Source
AI summary
In an aspect, an apparatus obtains at least a first input value and a second input value from a sender device. The apparatus performs a computational operation between portions of the first input value and portions of the second input value to obtain a plurality of partial results of the computational operation. The apparatus applies a hash function to each of the plurality of partial results of the computational operation to obtain a hash of a final result of the computational operation between the first input value and the second input value. The apparatus obtains the final result of the computational operation from the sender device. The apparatus verifies that the final result of the computational operation from the sender device is correct based on the hash of the final result of the computational operation.