Announcement of Report: A.V. Makarenko, Deep Learning Algorithms for Signal Recognition in Long Perimeter Monitoring Distributed Fiber Optic Sensors
In this paper, we show an approach to build deep learning algorithms for recognizing signals in distributed fiber optic monitoring and security systems for long perimeters. One of the types of fiber optic systems operates based on time-domain detection of backward Rayleigh scattered light of short laser pulse injected into the cable. Synthesizing such detection algorithms poses a non-trivial research and development challenge, because these systems face stringent error (type I and II) requirements and operate in difficult signal-jamming environments, with intensive signal-like jamming and a variety of changing possible signal portraits of possible recognized events. To address these issues, we have developed a two-level event detection architecture, where the primary classifier is based on an ensemble of deep convolutional networks, can recognize 7 classes of signals and receives time-space data frames as input. Using real-life data, we have shown that the applied methods result in efficient and robust multiclass detection algorithms that have a high degree of adaptability.
The report was accepted for presentation at 26th IEEE International Workshop on Machine Learning for Signal Processing, Italy, Salerno, on September 13–16, 2016. The conference is held under auspices of the IEEE Signal Processing Society. Read more: mlsp2016.conwiz.dk.
06 July 2016.
News Source: Own information.
