Biometrics in a heartbeat
18 February 2013 17:47 GMT

Heart beat biometrics improving

Researchers in China have found a way of improving performance in a biometric system designed to recognise people based on the sound of their heart beat.

The work presented by researchers from HangZhou DianZi University presents a heart sound biometric system based on marginal spectrum analysis, which is a new feature extraction technique for identification purposes. The results indicate that the technique increases recognition rate to 94.4% compared with that of the traditional techniques (84.32%) based on a database of 280 heart sounds from 40 participants.

Heart sounds are a reflection of the mechanical movement of the heart and cardiovascular system. The feature contains physiological and pathological information about the heart and various parts of the body.

According to the research it has advantage over conventional biometric features:

  • high security because an individual’s heart sounds cannot be faked;
  • easy to process because the heart sound is a one-dimensional signal with frequency components that exist mainly in the low-frequency range, thus making the signal processing simple;
  • high universality because every person has a beating heart.

The team of researchers admit that heart sound identification technology remains at an early research stage, but say it has been receiving considerable attention in recent years. Nevertheless there are numerous unresolved issues including the poor robustness of the technology in a noisy environment and the impact of heart diseases on identification accuracy.