Abstarct: Long-term recording of Phonocardiograph (PCG) signals involves large amount of data which indicates the importance of using a proper compression method. By using a compression method we aim to remove redundant information which has no or negligible effect on clinical and/or diagnosis applications and at the same time, preserve memory storage and bandwidth requirements. In this thesis, a new method for compression and reconstruction of PCG signals baxsed on two- dimensionalization and wavelet transform is proposed. The compression phase includes five stages: preprocessing, decimation, sound segment extraction, two-dimensionalization and wavelet transform/encoding to generate the output bitstream. The evaluation measures we used includes total compression ratio (TotalCR), Percentage root mean square difference (PRD), Bit per pixel (bpp), and Peak signal to noise ratio (PSNR). As one of obtained results, the STW and GBL_MMC_H, wavelet encoding methods had the highest compression performance with respect to TotalCR, PRD, PSNR and bpp criteria respectively.