Open Access

ARTICLE

An Efficient CSP-PDW Approach for ECG Signal Compression and Reconstruction for IoT-Based Healthcare

Hari Mohan Rai^1,#, Chandra Mukherjee^2,#, Joon Yoo¹, Hanaa A. Abdallah³, Saurabh Agarwal^4,*, Wooguil Pak^4,*

1 School of Computing, Gachon University, Seongnam-si, 13120, Republic of Korea
2 Department of Electronics and Communication Engineering, Indian Institute of Technology (IIT), Dhanbad, 826004, India
3 Department of Information Technology, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, P.O. Box 84428, Riyadh, 11671, Saudi Arabia
4 School of Computer Science and Engineering, Yeungnam University, Gyeongsan, 38541, Republic of Korea

* Corresponding Authors: Saurabh Agarwal. Email: ; Wooguil Pak. Email:
# These authors contributed equally to this work

(This article belongs to the Special Issue: Recent Advancements in Machine Learning and Data Analysis for Disease Detection)

Computers, Materials & Continua 2025, 85(3), 5723-5745. https://doi.org/10.32604/cmc.2025.070391

Received 15 July 2025; Accepted 10 September 2025; Issue published 23 October 2025

Abstract

A hybrid Compressed Sensing and Primal-Dual Wavelet (CSP-PDW) technique is proposed for the compression and reconstruction of ECG signals. The compression and reconstruction algorithms are implemented using four key concepts: Sparsifying Basis, Restricted Isometry Principle, Gaussian Random Matrix, and Convex Minimization. In addition to the conventional compression sensing reconstruction approach, wavelet-based processing is employed to enhance reconstruction efficiency. A mathematical model of the proposed algorithm is derived analytically to obtain the essential parameters of compression sensing, including the sparsifying basis, measurement matrix size, and number of iterations required for reconstructing the original signal and determining the type and level of wavelet processing. The low time complexity of the proposed algorithm makes it an ideal candidate for ECG monitoring systems in IoT-based e-healthcare applications. A feature extraction algorithm is also developed to show that the important ECG peaks remain unaltered after reconstruction. The clinical relevance of the reconstructed signal and the efficiency of the developed algorithm are evaluated using four validation parameters at three different compression ratios.

---

Keywords

---