PRNG Post-Processing Circuit for Secure Random Data Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating random numbers in digital semiconductor technology are limited by the need for linear transformations, which can be insecure and inefficient, particularly in ensuring computational unpredictability and robustness against environmental changes and physical attacks.
Innovation Solution
A method involving a PRNG operated in a non-autonomous mode, where the raw random binary sequence is continuously fed into the PRNG using XOR logic gates, ensuring the output is purely random and computationally unpredictable, even if the raw sequence is known.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If linear transformations are used for post-processing raw random sequences, then device complexity is reduced, but computational security and robustness deteriorate
Solution Approach 1:
The patent changes the fundamental parameter of the transformation type from linear to non-linear. The non-linear post-processing circuit uses non-linear Boolean functions to transform the raw random sequence, fundamentally altering the processing mechanism to achieve both security and efficiency. This parameter change resolves the contradiction by providing a solution that is neither purely linear (insecure) nor overly complex.
2Reliability
If non-linear transformations are used for post-processing, then computational security is improved, but device complexity increases
Solution Approach 1:
The patent segments the post-processing function into multiple non-linear stages or components. Rather than using a single complex non-linear transformation, the circuit divides the processing into sequential non-linear steps, each manageable in complexity. This segmentation allows the system to achieve high computational security through cumulative non-linear transformation while keeping individual circuit components relatively simple.
3Manufacturing precision
If raw random sequence is processed to eliminate statistical weaknesses, then randomness quality is improved, but generation speed is reduced
Solution Approach 1:
The patent replaces traditional mechanical or sequential processing methods with parallel digital logic operations. The non-linear post-processing circuit uses combinatorial logic gates that operate simultaneously on multiple bits of the raw random sequence, eliminating the need for sequential processing. This substitution of processing mechanics maintains high generation speed while achieving complete elimination of statistical weaknesses through parallel non-linear transformation.
Data Source
AI summary
An apparatus for generating random data includes a raw random sequence source adapted to generate a raw random sequence and a digital post processor adapted to process the raw random sequence to generate the random data, wherein the digital post-processor includes a synchronous finite state machine having at least one input adapted to repeatedly receive a current value of the raw random sequence and at least one output to provide a current output value depending on previous values of the raw random sequence.