Scholarlink

<p>Electrocardiogram (ECG) signals, crucial for monitoring cardiovascular health, often lead to large data volumes, necessitating effective compression techniques. Recent advancements in biomedical signal processing have underscored the need for more efficient ECG signal compression techniques. This paper introduces a novel ECG signal compression method using Compressed Sensing (CS), which capitalizes on the inherent sparsity of ECG signals. By employing the Discrete Wavelet Transform (DWT), the signal is transformed into a sparse domain, allowing for minimal sampling using a deterministic sensing matrix. The Iteratively Reweighted Least Squares (IRLS) algorithm is then utilized for accurate signal reconstruction. This approach significantly reduces the number of measurements and computational complexity while ensuring high-quality signal recovery. Comprehensive evaluations using the MIT-BIH arrhythmia database demonstrate improvements in both compression efficiency and reconstruction accuracy. It consistently achieved a Compression Ratio (CR) of up to 12.52 and a Percentage Root-mean-square Difference (PRD) as low as 1.02, outperforming recent approaches in efficiency and reconstruction accuracy.</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

ECG Signal Compression Using Compressed Sensing and Deterministic Matrices: A Wavelet-Based Approach

  • 期刊论文

  • 发表日期：

    2026年2月4日

查看全文
  • Haroon Yousuf Mir,
  • Omkar Singh

摘要

Electrocardiogram (ECG) signals, crucial for monitoring cardiovascular health, often lead to large data volumes, necessitating effective compression techniques. Recent advancements in biomedical signal processing have underscored the need for more efficient ECG signal compression techniques. This paper introduces a novel ECG signal compression method using Compressed Sensing (CS), which capitalizes on the inherent sparsity of ECG signals. By employing the Discrete Wavelet Transform (DWT), the signal is transformed into a sparse domain, allowing for minimal sampling using a deterministic sensing matrix. The Iteratively Reweighted Least Squares (IRLS) algorithm is then utilized for accurate signal reconstruction. This approach significantly reduces the number of measurements and computational complexity while ensuring high-quality signal recovery. Comprehensive evaluations using the MIT-BIH arrhythmia database demonstrate improvements in both compression efficiency and reconstruction accuracy. It consistently achieved a Compression Ratio (CR) of up to 12.52 and a Percentage Root-mean-square Difference (PRD) as low as 1.02, outperforming recent approaches in efficiency and reconstruction accuracy.