Intruder recognition using ECG signal

The electrocardiogram (ECG) is becoming a promising technology for biometrichuman identi?cation. Usually ECG is used for health measurements and this isuseful for biometric applications to state that the subject under analysis is live.But an individual identi?cation shouldn't require a classical ECG clinical analysiswhere several contacts are applied to the person to be identi?ed. In literature,ECG biometric recognition is usually studied for the recognition of a subjectwithin a group of known subjects. The aim of the designed embedded wearablecontroller is to authorize a subject or to reject him, labeling as an intruderunknown to the system. The research uses 40 healthy subjects: 2 authorized and38 intruders. A one-lead ECG trace has been recorded from the wrists of subjects,features have been extracted using a combination of Autocorrelation andDiscrete Cosine Transform (AC/DCT) and then classi?ed using a MultilayerPerceptron. Results show that intruder recognition can be performed with asuccess rate equal to 100%.


ERC Sector:

  • PE7_11 Components and systems for applications
  • PE6_11 Machine learning, statistical data processing and applications using signal processing (e.g. speech, image, video)


  • Biometrics
  • Support vector machines
  • Electrocardiogram
  • Intruder detection

Gruppi di ricerca

Persone di riferimento