Scalar to Tau-Adic Conversion Algorithm for Cryptographic Circuits

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for converting scalars to τ-adic representation in elliptic curve cryptography are complex and time-consuming, especially when dealing with high scalar values that require multiple reduction operations.

Innovation Solution

A cryptographic method and device that perform scalar conversion by executing an algorithm applied to a scalar to generate an output vector, where each iteration updates the scalar value and sets digits of the output vector based on intermediate data values calculated modulo a complex number τ, with the number of iterations determined by a stop criterion.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If traditional scalar conversion methods are used, then the conversion process is straightforward, but it becomes complex and time-consuming for high scalar values requiring multiple reduction operations

Engineering Contradiction:
Improvescalar conversion speedVSAvoidconversion algorithm complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent segments the scalar conversion process into iterative steps that process the scalar in chunks rather than requiring complete reduction. The algorithm divides the conversion into manageable iterations where each iteration processes a portion of the scalar and produces intermediate results, avoiding the need for multiple complete reduction operations on high scalars.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary actions by pre-computing and storing reduction results for common cases. The algorithm prepares intermediate values and reduction constants in advance, so that during actual scalar conversion, these pre-computed values can be quickly applied without performing full reduction operations from scratch.

Inventive Principle:
Principle #10Preliminary action

2Difficulty of detecting and measuring

If the scalar value is kept visible during conversion, then the conversion process is simple to implement, but the scalar value becomes easily detectable during cryptographic operations

Engineering Contradiction:
Improvescalar value detectabilityVSAvoidconversion implementation ease
Core Design Contradiction:
Difficulty of detecting and measuringVSEase of operation

Solution Approach 1:

The patent introduces intermediate representations as mediators between the original scalar and the final τ-adic form. Instead of directly converting and exposing the scalar, the algorithm uses intermediate values that obscure the original scalar while maintaining mathematical equivalence. These intermediates serve as protective layers that make it difficult to detect the original scalar value during conversion operations.

Inventive Principle:
Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent changes parameters during conversion by transforming the scalar through multiple representation forms with different properties. The algorithm dynamically adjusts the representation parameters (such as the base and digit constraints) throughout the conversion process, making it difficult to trace back to the original scalar while maintaining the mathematical relationship needed for correct conversion.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS12284279B2Reduction and conversion of a scalar to a τ-ADIC representation
Publication Date: 2025.04.22 STMICROELECTRONICS (ROUSSET) SAS
  • US12284279B2 patent drawing
  • US12284279B2 patent drawing
  • US12284279B2 patent drawing

AI summary

The present disclosure relates to a cryptographic method including the execution, by a cryptographic circuit, of an algorithm applied to a scalar in order to generate an output vector, of length L+n, which digits are d0, . . . , dL+n−1, the algorithm comprising iterations i, each iteration i taking an input data value, initially equal to said scalar and an input vector of length c, which digits are d′i, . . . , d′i+c−1, where for each j∈{i, . . . , i+c−1}, the digit d′j is such that:dj′={dj⁢ if⁢ j<Ldj-m⁢ otherwise.