A Transform Domain Approach for the Compression of Fetal Phonocardiogram Signal

Abstract

In this letter, a transform-based methodology is presented for the compression of fetal phonocardiogram (fPCG) signal. Transform techniques, such as discrete cosine transform (DCT), discrete wavelet transform (DWT), and fast Walsh-Hadamard transform (FWHT), are applied for the decomposition of fPCG signal and a comparative study on the performance of different transforms is made in terms of measures, such as compression ratio (CPR), percent root-mean-square difference (PRD), mean square error (MSE), quality score (QS), and signal-to-noise ratio (SNR). The simulation results reveal the effectiveness of the transform techniques in fPCG signal compression. In comparison, it is observed that FWHT gives a better CPR, whereas DWT yields good fidelity parameters with comparable CPR.