Intelligent decision-making framework for selecting optimized substitution boxes in modern block cipher design.

Area of Science:

• Cryptography
• Information Security
• Applied Mathematics

Background:

• Substitution boxes (S-boxes) are critical for block cipher security, but selecting optimal S-boxes is challenging due to competing cryptographic requirements.
• Evaluating and comparing S-boxes involves complex multi-criteria optimization tasks.

Purpose of the Study:

• To introduce a novel hybrid Multi-Criteria Decision-Making (MCDM) approach for systematically ranking S-boxes.
• To provide a data-oriented evaluation of key cryptographic parameters for S-box assessment.

Main Methods:

• Combines the Evaluation based on Distance from Average Solution (EDAS) technique with an Entropy-weighting framework.
• Evaluates S-boxes based on nonlinearity, strict avalanche criterion, bit independence, differential approximation probability, and linear approximation probability.

Main Results:

• The proposed hybrid MCDM model effectively ranks S-boxes based on their cryptographic strengths.
• Identified S-boxes with superior linear and differential attack resistance.

Conclusions:

• The developed approach offers evidence-based design directions for creating secure and efficient block ciphers.
• Future work will explore AI-generated and optimization-algorithm driven S-box designs for enhanced digital confidentiality.