Heart sound as a physiological biometric signature

This paper presents a system for using heart sound as a physiological biometric signature based on energy percentage in each wavelet coefficients. An electronic stethoscope designed and implemented by the authors is used to record more than 30 heart sound and for each recording a signature is calculated. The signature is calculated during one heart cycle by calculating the total energy for each wavelet coefficient after normalization, filtering and denoising of each recordings. A Euclidean distance is used as an identification tool between the signatures It was found that the distances for different recordings at different days for the same person are below a threshold value and above the threshold value as compared with the signatures of the other persons. A software based on Matlab is used for recordings, denoising, signature calculations And signature comparison to prove the implemented system.