1. Note: Gaussian mixture model for event recognition in optical time-domain reflectometry based sensing systems
- Author
-
Valery E. Karasik, Aleksey Fedorov, M.N. Anufriev, Dmitry Namiot, Alexey B. Pnev, E. T. Nesterov, A. A. Zhirnov, and K. V. Stepanov
- Subjects
Physics - Instrumentation and Detectors ,Computer science ,FOS: Physical sciences ,02 engineering and technology ,01 natural sciences ,010309 optics ,symbols.namesake ,020210 optoelectronics & photonics ,0103 physical sciences ,0202 electrical engineering, electronic engineering, information engineering ,Cluster (physics) ,Time domain ,Reflectometry ,Instrumentation ,business.industry ,Event recognition ,Probability and statistics ,Pattern recognition ,Instrumentation and Detectors (physics.ins-det) ,Mixture model ,Gaussian noise ,Physics - Data Analysis, Statistics and Probability ,symbols ,Artificial intelligence ,business ,Sensing system ,Data Analysis, Statistics and Probability (physics.data-an) ,Physics - Optics ,Optics (physics.optics) - Abstract
We propose a novel approach to the recognition of particular classes of non-conventional events in signals from phase-sensitive optical time-domain-reflectometry-based sensors. Our algorithmic solution has two main features: filtering aimed at the de-nosing of signals and a Gaussian mixture model to cluster them. We test the proposed algorithm using experimentally measured signals. The results show that two classes of events can be distinguished with the best-case recognition probability close to 0.9 at sufficient numbers of training samples., Comment: 4 pages; published version
- Published
- 2016