Data compression of ECG's by high-degree polynomial approximation

Abstract
A method for the compression of ECG data is presented. The method is based on high-degree polynomial expansions. Data rates of about 350 bits per second are achievable at an acceptable signal quality. The high compression is obtained by a carefully selected subdivision of the ECG signal into intervals that make optimal use of the special properties of the polynomial base functions. Each interval corresponds to one ECG period. The method is compared to the discrete cosine transform and is found to yield a significantly higher data compression for a given signal quality (quantified by mean squared error and peak error).

This publication has 16 references indexed in Scilit: