Scholarlink

<p>Secure audio data transmission has become a critical challenge in the digital era due to increasing cyber threats and the sensitive nature of multimedia information. Traditional encryption techniques often struggle to maintain efficiency and robustness when applied to audio signals characterized by complex temporal and spectral patterns. To address this challenge, this study proposes a novel cryptographic framework that integrates S-box based transformations directly into audio signal processing. The proposed approach employs encryption and decryption mechanisms grounded in advanced cryptographic principles, where carefully designed substitution boxes ensure strong nonlinearity and secure data transformation within audio signals. Extensive experiments are conducted using various audio datasets, including standard WAV signals, classical music compositions, and complex chord progressions. The security and performance of the proposed framework are evaluated through multiple metrics such as correlation analysis, Log-Likelihood Ratio (LLR), Bit Error Rate (BER), NPCR, Unified Average Changing Intensity (UACI), histogram analysis, Signal-to-Noise Ratio (SNR), and Peak Signal-to-Noise Ratio (PSNR), along with both time-domain and frequency-domain evaluations. Furthermore, robustness against noise-based attacks demonstrates the stability of the proposed scheme. Machine learning techniques are incorporated for feature extraction, and SVM-based cryptanalysis achieves 99.15% accuracy, confirming strong resistance. The framework enhances confidentiality and has applications in cybersecurity and multimedia encryption.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A novel framework for audio signal encryption via S-box cryptography and feature extraction

  • 期刊论文

  • 发表日期：

    2026年5月15日

查看全文
  • Muhammad Waheed Rasheed,
  • Abid Mahboob,
  • Dur e Najaf,
  • Rabia Mehmood

摘要

Secure audio data transmission has become a critical challenge in the digital era due to increasing cyber threats and the sensitive nature of multimedia information. Traditional encryption techniques often struggle to maintain efficiency and robustness when applied to audio signals characterized by complex temporal and spectral patterns. To address this challenge, this study proposes a novel cryptographic framework that integrates S-box based transformations directly into audio signal processing. The proposed approach employs encryption and decryption mechanisms grounded in advanced cryptographic principles, where carefully designed substitution boxes ensure strong nonlinearity and secure data transformation within audio signals. Extensive experiments are conducted using various audio datasets, including standard WAV signals, classical music compositions, and complex chord progressions. The security and performance of the proposed framework are evaluated through multiple metrics such as correlation analysis, Log-Likelihood Ratio (LLR), Bit Error Rate (BER), NPCR, Unified Average Changing Intensity (UACI), histogram analysis, Signal-to-Noise Ratio (SNR), and Peak Signal-to-Noise Ratio (PSNR), along with both time-domain and frequency-domain evaluations. Furthermore, robustness against noise-based attacks demonstrates the stability of the proposed scheme. Machine learning techniques are incorporated for feature extraction, and SVM-based cryptanalysis achieves 99.15% accuracy, confirming strong resistance. The framework enhances confidentiality and has applications in cybersecurity and multimedia encryption.