Performances of disorder detection procedure for autoregressive process with unknown noise distribution

S. I. Vorobejchikov, V. V. Konev

    Research output: Contribution to journalArticle


    The sequential procedure of detection of the jump change of autoregressive process parameters with unknown noise distribution is considered. It is supposed that the disorder occurs in unknown moment in time. Proposed disorder detection procedures uses cumulative sums with quantization of statistics. Asymptotic formulae for mean delay of disorder detection and mean time between false alarms are established. The accuracy of obtained performances is illustrated with the help of Monte Carlo method for Gaussian autoregressive first-order process with four quantization levels.

    Original languageEnglish
    Pages (from-to)68-75
    Number of pages8
    JournalAvtomatika i Telemekhanika
    Issue number2
    Publication statusPublished - Feb 1992


    ASJC Scopus subject areas

    • Control and Systems Engineering

    Cite this