Novel hyperchaotic system: Implementation to audio encryption

To overcome the limitations of existing low-dimensional chaotic systems, particularly their vulnerability to degradation, this study introduces a novel family of discrete hyper-chaotic systems, designed using a one-dimensional quadratic map. The dynamic behavior of the systems is analysed using Lyapunov exponents and sample entropy to evaluate their complexity and robustness. The results demonstrate that the proposed systems exhibit higher ergodicity, greater Lyapunov exponents and better randomness compared to existing chaotic systems. Exploiting these systems, a novel fractal K-means audio encryption (FKM-AE) algorithm is proposed, integrating fractal algorithms with the K-means grouping approach. Simulations reveal that the proposed method effectively reduces the correlation of audio messages across adjacent time intervals and robustly resists various attacks, demonstrating its high performance.