Published on: 2026-09-21
Source: Novosibirsk State University –
An important disclaimer is at the bottom of this article.
Employees of Novosibirsk State University, together with the company “T8,” conducted an experiment using an existing fiber optic communication line to monitor objects on the university campus. As part of the experiment, specialists tested whether the laid fiber optic cable could be used as a distributed vibration sensor and recognized events in the environment using artificial intelligence methods. In the future, the technology could be applied to monitor industrial and urban infrastructure.
The project uses the hardware and software complex of the “Danube” fiber-optic distributed system. Its key element is an interrogator, a device that connects to the fiber-optic line and analyzes the backscattering of light within the fiber. Vibrations and deformations of the fiber, caused by events occurring nearby, alter the characteristics of this scattered signal. By analyzing these changes, the system can determine the location and nature of the impact on the fiber-optic line.
The pilot project involves a 4.3 km long service fiber optic line laid on the territory of the Novosibirsk Scientific Center between NSU and “Akadempark.” When registering vibrations with a spatial step of about 10 meters, it effectively turns into approximately 400 distributed observation points. Unlike traditional monitoring systems, where a separate seismometer or vibration sensor is installed at each observation point, here the sensory function is performed by an existing optical fiber.
— It is possible to connect a special device to an existing fiber optic line and use the fiber itself as a distributed sensor without laying a new sensor line, — said a senior researcher at NSU, a researcher at the Institute of Oil and Gas Geology and Geophysics of the Siberian Branch of the Russian Academy of Sciences Pyotr Dergach.
The received data are sent to the server of the NSU Center for Artificial Intelligence, where machine learning algorithms classify the recorded vibration signals. The system allows recognizing various events, such as a car passing by, a person walking, or earthworks being carried out. For each detected event, the time and approximate location of its occurrence are determined, after which the information is displayed in the monitoring system interface.
Depending on the task, such a system can be used for perimeter security, monitoring construction work and road traffic, as well as observing the condition of buildings and engineering structures. Changes in the vibration background and the nature of oscillations may indicate unusual loads, changes in operating conditions, or equipment operating in an abnormal mode. This opens up opportunities to use a fiber optic line as a tool for early anomaly detection and additional infrastructure condition monitoring.
A separate application scenario for the technology is seismic activity monitoring. A distributed fiber optic sensor can record vibrations associated with the passage of seismic waves and determine their characteristics and the location along the line where they are registered. At the same time, the fiber optic system does not necessarily have to fully replace traditional seismometers: its data can complement the readings of individual sensors and, thanks to distributed measurement, expand the observation area.
— The system receives a continuous stream of vibration signals, among which artificial intelligence recognizes characteristic features of various events and allows determining what happened and roughly where it happened., — explained the lead research scientist of the Artificial Intelligence Center, head of the laboratory of photonics and machine learning technologies for sensory systems at NSU, Alexey Redyuk.
The project involves specialists from NSU, the Institute of Oil and Gas Geology and Geophysics (INGG) of the Siberian Branch of the Russian Academy of Sciences, NSU Artificial Intelligence Center and the company “T8.” The company provides the interrogator and participates in connecting it to the line and data collection, while the university team is responsible for the server infrastructure, software, visualization, and data processing algorithms. To correlate the results of the distributed fiber optic sensor with the readings of traditional single sensors, data collection is conducted jointly with the company “Industrial Monitoring.” All equipment involved in the experiment, including the hardware and software complex “Danube,” was integrated into the existing vibration monitoring system of the NSU campus, developed by staff and students of the NSU Advanced Engineering School.
— We are testing the concept of reusing existing fiber optic infrastructure. For monitoring systems, special sensor lines can be laid, but in urban environments, there are already many fiber optic communication lines that can potentially also be used as distributed sensors., — noted Pyotr Dergach.
Distributed acoustic sensing is based on analyzing the backscattering of light in an optical fiber. The interrogator sends short laser pulses into the fiber, and a small portion of the light is scattered back due to the natural inhomogeneities of the optical fiber. Vibrations and deformations of the fiber change the characteristics of the scattered signal. Analyzing these changes allows obtaining information about the dynamic deformation of the fiber at various sections of the line and determining the location of the impact. This approach is already being studied worldwide for monitoring roads, bridges, pipelines, railways, and other extensive structures.
The technology has limitations. In some parameters, traditional seismometers can provide higher sensitivity, whereas the characteristics of a fiber-optic sensor largely depend on the way the line is laid and its mechanical connection to the surrounding environment. For example, weak contact with the ground, sagging fiber, or the cable passing through wells can reduce vibration detection efficiency. However, the distributed nature of the measurements and the ability to use existing infrastructure allow these limitations to be partially compensated.
The initial data has already been collected, and now the developers are verifying how accurately the algorithms can recognize various events through the vibrations of the optical fiber. The next stage is to expand the dataset and test the system under different conditions. If the results confirm the approach’s viability, the team plans to proceed to new trials at urban and industrial sites, where the existing fiber optic infrastructure can be used simultaneously for data transmission and environmental monitoring.
Please note; This information is raw content received directly from the source. It is an accurate report of what the source claims and does not necessarily reflect the position of MIL-OSI or its clients.