Brain signal's low-frequency fits the continuous authentication

Yasuo Matsuyama, Michitaro Shozawa, Ryota Yokote

    Research output: Contribution to journalArticle

    7 Citations (Scopus)

    Abstract

    In this paper, we propose a method to utilize low-frequency brain signals for continuous authentication of users. During such monitoring, the users to be authenticated can work without interruption. This style of authentication is expected to complement traditional methods based on passwords, which can be easily forgotten or stolen. For brain signal-based continuous authentication, we measured oxyhemoglobin changes in the brain through near-infrared spectroscopy (NIRS). There are two cases of NIRS measurement: a rest case, and a keyboard typing task case. In both cases, the brain signals were found to show specific patterns in the range around 1.5. Hz. Identified personality was used to prevent impersonators. For the detection of impostors, we first carried out a principal component analysis (PCA) of the logarithmic power spectra of the NIRS signals. Small eigenvalues were discarded so that excessive learning of system parameters can be avoided. The processed spectral data were utilized to obtain an average weight vector for support vector machines (SVMs). The average weight vector was applied to the spectral data to emphasize characteristic patterns in low-frequency regions. This process generated separable clusters for each subject's NIRS signals. In the test phase, unknown subject's NIRS signals were measured and pre-processed. Following this, we carried out continuous authentication by computing the Mahalanobis distance to the registered cluster set. For both the rest and task cases of the NIRS, the authentication accuracy of our proposed method was greater than 99% at the equal error rate (EER). Dynamic authentication of this sort using brain signals can offer a viable method for reducing excessive dependence on traditional password-based methods.

    Original languageEnglish
    Pages (from-to)137-143
    Number of pages7
    JournalNeurocomputing
    Volume164
    DOIs
    Publication statusPublished - 2015 Sep 21

    Fingerprint

    Near infrared spectroscopy
    Near-Infrared Spectroscopy
    Authentication
    Brain
    Weights and Measures
    Oxyhemoglobins
    Power spectrum
    Principal Component Analysis
    Principal component analysis
    Support vector machines
    Personality
    Learning
    Monitoring

    Keywords

    • Brain signal
    • Continuous authentication
    • Near-infrared spectroscopy (NIRS)
    • Principal component analysis (PCA)
    • Support vector machine (SVM)

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Computer Science Applications
    • Cognitive Neuroscience

    Cite this

    Brain signal's low-frequency fits the continuous authentication. / Matsuyama, Yasuo; Shozawa, Michitaro; Yokote, Ryota.

    In: Neurocomputing, Vol. 164, 21.09.2015, p. 137-143.

    Research output: Contribution to journalArticle

    Matsuyama, Yasuo ; Shozawa, Michitaro ; Yokote, Ryota. / Brain signal's low-frequency fits the continuous authentication. In: Neurocomputing. 2015 ; Vol. 164. pp. 137-143.
    @article{0fbe19e4c4c94085bd3e866e822792af,
    title = "Brain signal's low-frequency fits the continuous authentication",
    abstract = "In this paper, we propose a method to utilize low-frequency brain signals for continuous authentication of users. During such monitoring, the users to be authenticated can work without interruption. This style of authentication is expected to complement traditional methods based on passwords, which can be easily forgotten or stolen. For brain signal-based continuous authentication, we measured oxyhemoglobin changes in the brain through near-infrared spectroscopy (NIRS). There are two cases of NIRS measurement: a rest case, and a keyboard typing task case. In both cases, the brain signals were found to show specific patterns in the range around 1.5. Hz. Identified personality was used to prevent impersonators. For the detection of impostors, we first carried out a principal component analysis (PCA) of the logarithmic power spectra of the NIRS signals. Small eigenvalues were discarded so that excessive learning of system parameters can be avoided. The processed spectral data were utilized to obtain an average weight vector for support vector machines (SVMs). The average weight vector was applied to the spectral data to emphasize characteristic patterns in low-frequency regions. This process generated separable clusters for each subject's NIRS signals. In the test phase, unknown subject's NIRS signals were measured and pre-processed. Following this, we carried out continuous authentication by computing the Mahalanobis distance to the registered cluster set. For both the rest and task cases of the NIRS, the authentication accuracy of our proposed method was greater than 99{\%} at the equal error rate (EER). Dynamic authentication of this sort using brain signals can offer a viable method for reducing excessive dependence on traditional password-based methods.",
    keywords = "Brain signal, Continuous authentication, Near-infrared spectroscopy (NIRS), Principal component analysis (PCA), Support vector machine (SVM)",
    author = "Yasuo Matsuyama and Michitaro Shozawa and Ryota Yokote",
    year = "2015",
    month = "9",
    day = "21",
    doi = "10.1016/j.neucom.2014.08.084",
    language = "English",
    volume = "164",
    pages = "137--143",
    journal = "Neurocomputing",
    issn = "0925-2312",
    publisher = "Elsevier",

    }

    TY - JOUR

    T1 - Brain signal's low-frequency fits the continuous authentication

    AU - Matsuyama, Yasuo

    AU - Shozawa, Michitaro

    AU - Yokote, Ryota

    PY - 2015/9/21

    Y1 - 2015/9/21

    N2 - In this paper, we propose a method to utilize low-frequency brain signals for continuous authentication of users. During such monitoring, the users to be authenticated can work without interruption. This style of authentication is expected to complement traditional methods based on passwords, which can be easily forgotten or stolen. For brain signal-based continuous authentication, we measured oxyhemoglobin changes in the brain through near-infrared spectroscopy (NIRS). There are two cases of NIRS measurement: a rest case, and a keyboard typing task case. In both cases, the brain signals were found to show specific patterns in the range around 1.5. Hz. Identified personality was used to prevent impersonators. For the detection of impostors, we first carried out a principal component analysis (PCA) of the logarithmic power spectra of the NIRS signals. Small eigenvalues were discarded so that excessive learning of system parameters can be avoided. The processed spectral data were utilized to obtain an average weight vector for support vector machines (SVMs). The average weight vector was applied to the spectral data to emphasize characteristic patterns in low-frequency regions. This process generated separable clusters for each subject's NIRS signals. In the test phase, unknown subject's NIRS signals were measured and pre-processed. Following this, we carried out continuous authentication by computing the Mahalanobis distance to the registered cluster set. For both the rest and task cases of the NIRS, the authentication accuracy of our proposed method was greater than 99% at the equal error rate (EER). Dynamic authentication of this sort using brain signals can offer a viable method for reducing excessive dependence on traditional password-based methods.

    AB - In this paper, we propose a method to utilize low-frequency brain signals for continuous authentication of users. During such monitoring, the users to be authenticated can work without interruption. This style of authentication is expected to complement traditional methods based on passwords, which can be easily forgotten or stolen. For brain signal-based continuous authentication, we measured oxyhemoglobin changes in the brain through near-infrared spectroscopy (NIRS). There are two cases of NIRS measurement: a rest case, and a keyboard typing task case. In both cases, the brain signals were found to show specific patterns in the range around 1.5. Hz. Identified personality was used to prevent impersonators. For the detection of impostors, we first carried out a principal component analysis (PCA) of the logarithmic power spectra of the NIRS signals. Small eigenvalues were discarded so that excessive learning of system parameters can be avoided. The processed spectral data were utilized to obtain an average weight vector for support vector machines (SVMs). The average weight vector was applied to the spectral data to emphasize characteristic patterns in low-frequency regions. This process generated separable clusters for each subject's NIRS signals. In the test phase, unknown subject's NIRS signals were measured and pre-processed. Following this, we carried out continuous authentication by computing the Mahalanobis distance to the registered cluster set. For both the rest and task cases of the NIRS, the authentication accuracy of our proposed method was greater than 99% at the equal error rate (EER). Dynamic authentication of this sort using brain signals can offer a viable method for reducing excessive dependence on traditional password-based methods.

    KW - Brain signal

    KW - Continuous authentication

    KW - Near-infrared spectroscopy (NIRS)

    KW - Principal component analysis (PCA)

    KW - Support vector machine (SVM)

    UR - http://www.scopus.com/inward/record.url?scp=84935585478&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84935585478&partnerID=8YFLogxK

    U2 - 10.1016/j.neucom.2014.08.084

    DO - 10.1016/j.neucom.2014.08.084

    M3 - Article

    VL - 164

    SP - 137

    EP - 143

    JO - Neurocomputing

    JF - Neurocomputing

    SN - 0925-2312

    ER -