Real-Time ECG Compression with Adaptive Huffman Coding: Improving Data Transmission in E-Healthcare

Abstract

E-healthcare monitoring systems are becoming increasingly common due to their widespread use in everyday life. Among these systems, the electrocardiogram (ECG) signal is the most reliable method for diagnosing cardiovascular diseases (CVDs). Processing ECG signals is crucial for monitoring the heart’s electrical activity and providing essential insights into CVDs. However, storing and transmitting sensitive ECG data presents a challenge, as traditional compression methods often cause issues in reconstructing the original data. Huffman coding processes input data as a stream and analyzes its frequency distribution. However, conventional Huffman coding requires prior knowledge of the data, which is not always available for all datasets. To address this limitation, this research proposes an adaptive, lossless ECG signal compression algorithm based on the Huffman coding technique for real-time e-health monitoring. The algorithm's performance is evaluated using both an evaluation and a validation dataset. When tested with the MIT-BIH database, the proposed algorithm achieves average compression ratios of 20.61, 24.41, and 30.96 for different dynamic and minimal window size variations, with corresponding percentage root mean square differences of 0.18 and 0.29.