D.V. Lutsenko - Ph. D. (Eng.), Assistant Professor L.A. Maitakova - Ph. D. (Eng.), Assistant Professor A.N. Tolmachev - Ph. D. (Eng.), Assistant Professor R.A. Mikhailov

The paper provides a rationale hardware false alarms and meteorological classifier. Under hardware false alarmmeans detection is a false alarmrefers to funds due to physical principle laid down, particularly of constructive execution and installation.To separate them from the general stream of false alarmson the stage field testsuse meteorological classifier. His work is based on the hypothesis that in the space of the first three principal component patterns of meteorological factors observations corresponding to the mode, linearly separable from the patterns corresponding to false alarms. Errors of the second kind, when false alarms are meteorological classifierto standby mode, identify hardware false alarms. At the stage of field testsmeteorological classifierapplication allows us to estimate the average period of repetition and intensity of hardware false alarms.

 

1. Ajjvazjan S.A. i dr.Klassifikacija i snizhenie razmernosti: Spravochnoe izdanie. M.: Finansy i statistika. 1989. 607 s.
2. Mangasarian O.L., Musicant D.R. Lagrangian Support Vector Machines // Journal of Machine Learning Research. 2001. № 1. P. 161−177.
3. Dvojjris L.I., Lucenko D.V., Mikhajjlov R.A., Zavjalov V.S. Rezultaty poligonnykh ispytanijj po issledovaniju ustojjchivosti sovremennykh sredstv obnaruzhenija k vozdejjstviju meteorologicheskikh pomekh // Radiotekhnika. 2013. № 2. S. 95−97.