Lossless ECG Encoder with Multi-Channel Adaptive Fuzzy Predictor and Enhanced Entropy Encoding in Wireless Body Sensor Networks

Abstract

With the advancement of wireless technology, Wireless Body Sensor Networks, such as Electrocardiograms (ECGs) will serve as state-of-the-art method for electronic-healthcare systems and applications. Like other digital communications, however, ECGs highlight power consumption as the main design constraint and bottleneck as it affects device lifespan and data accuracy. Hence, power reduction and power management techniques and schemes have been developed to eliminate this constraint such as hardware optimization, source and channel coding, signal conditioning, and resolution control. This paper proposes a lossless ECG encoder that combines existing data compression techniques specifically the adaptive fuzzy predictor based on fuzzy decision making and an enhanced entropy encoding that utilizes algorithms in both run length encoding (RLE) and incremental prefix encoding. Simulation results that the proposed scheme outperforms the entropy encoding using Huffman, RLE, and predictive encoding schemes in compression ratio (CR), with the enhanced entropy encoding leading the pre-existing compression techniques by 24.0907 on RLE and 23.6580 on Huffman.