Novel Methods for Enhancing Statistical Randomness Characteristic Using Chaotic Systems

Abstract

This study explores various methods for enhancing the statistical randomness of sequences generated from chaotic systems. While the modulus function is commonly employed to improve randomness by addressing the clustering of chaotic outputs, it has certain limitations. To address these, we investigated alternative methods, including the fractional part method, threshold functions, trigonometric transformations, and linear transformations with scaling and shifting. The results of this study demonstrated that while the modulus function is effective, alternative methods like the fractional part method and threshold functions can further enhance the distribution of values, reducing bias and improving the overall randomness. These findings indicate that the proposed methods are valuable for applications requiring high-quality random sequences, including cryptographic systems, data science, and large-scale simulations. © 2025 IEEE.