Scholarlink

<p>Electrocardiogram (ECG) signals have emerged as a promising biometric modality owing to their inherent uniqueness, temporal stability, and intrinsic liveness characteristics. However, the increasing use of biometric data in authentication systems raises critical concerns regarding privacy, secure storage, revocability, and the ethical handling of sensitive physiological information. To address these challenges, this paper proposes a privacy-preserving ECG-based biometric authentication method based on cancelable biometric template generation. The proposed approach employs the Lorenz chaotic system, a well-established nonlinear dynamic model, to mask the original ECG features and generate protected templates. Subject-specific morphological features are extracted from ECG segments centered around the <i>R</i>-peaks within the QRS complex and are subsequently transformed using a Lorenz-based chaotic masking mechanism. This transformation enhances template privacy and revocability while preserving the discriminative characteristics required for reliable authentication. Experimental results obtained on two widely used ECG datasets, ECG-ID and MIT-BIH, demonstrate that the proposed method effectively protects user privacy without degrading recognition performance, achieving an accuracy of up to 99.5%. The framework also supports template revocation and replacement by regenerating protected templates using updated transformation parameters, thereby strengthening long-term biometric data protection. Overall, the findings confirm the potential of chaotic systems as an effective mechanism for enhancing privacy, security, and revocability in ECG-based biometric authentication systems.</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

Enhancing privacy in electrocardiogram biometric authentication: a cancelable template approach using a Lorenz chaotic systems

  • 期刊论文

  • 发表日期：

    2026年7月23日

查看全文
  • Samia A. El-Moneim Kabel,
  • Walid El-Shafai

摘要

Electrocardiogram (ECG) signals have emerged as a promising biometric modality owing to their inherent uniqueness, temporal stability, and intrinsic liveness characteristics. However, the increasing use of biometric data in authentication systems raises critical concerns regarding privacy, secure storage, revocability, and the ethical handling of sensitive physiological information. To address these challenges, this paper proposes a privacy-preserving ECG-based biometric authentication method based on cancelable biometric template generation. The proposed approach employs the Lorenz chaotic system, a well-established nonlinear dynamic model, to mask the original ECG features and generate protected templates. Subject-specific morphological features are extracted from ECG segments centered around the R-peaks within the QRS complex and are subsequently transformed using a Lorenz-based chaotic masking mechanism. This transformation enhances template privacy and revocability while preserving the discriminative characteristics required for reliable authentication. Experimental results obtained on two widely used ECG datasets, ECG-ID and MIT-BIH, demonstrate that the proposed method effectively protects user privacy without degrading recognition performance, achieving an accuracy of up to 99.5%. The framework also supports template revocation and replacement by regenerating protected templates using updated transformation parameters, thereby strengthening long-term biometric data protection. Overall, the findings confirm the potential of chaotic systems as an effective mechanism for enhancing privacy, security, and revocability in ECG-based biometric authentication systems.