Nov 16 2020
Fiber optic sensors utilized in crucial applications such as landslide prediction, leak detection in pipelines, and fire detection in tunnels are on the verge of turning much quicker and more precise.
Simon Zaslawski, Zhisheng Yan, and Prof. Luc Thévenaz. Image Credit: 2020 EPFL/Alain Herzog.
Engineers from École polytechnique fédérale de Lausanne (EPFL) have designed a modern encoding and decoding system that enables fiber optic sensors to transmit data up to 100 times quicker and over a broad area.
Unlike conventional sensors that take measurements at a given point, like thermometers, fiber optic sensors record data all along a fiber. But the technology has barely improved over the past few years.
Luc Thévenaz, Professor, School of Engineering, EPFL
Thévenaz is also the head of the Group for Fibre Optics at EPFL.
Used Widely in Safety Applications
In general, fiber optic sensors are employed in hazard detection systems, for example, to find deformations in civil engineering structures, identify cracks in pipelines, and detect possible landslides on mountain slopes.
The sensors are capable of taking temperature readings anywhere a fiber is placed, producing a constant heat diagram of a specific site—even if the site extends for dozens of kilometers. This offers a critical understanding of potential accidents before they can occur.
Improving Signal Quality
In collaboration with the Beijing University of Posts and Telecommunications, two GFO engineers—postdoc Zhisheng Yang and PhD student Simon Zaslawski—have designed a new system for encoding and decoding data sent together with the fibers.
With the help of their technique, sensors can receive signals of higher energy and decode them quickly, leading to faster measurements over a huge area. The study was recently published in the Nature Communications journal.
The team has explained that the system functions similarly to an echo. If one shouts a single word, that word is heard back. But if one sings out a song, what is heard back is a mix of sounds that are difficult to differentiate.
One would require a “key” to decode the sounds and make them intelligible. Fiber optic sensors function in a similar way, apart from the fact that an instrument transmits light pulses—and not sounds—along a fiber. Signals bouncing back up the fiber are decoded by a device, thereby transforming the signals into usable data.
Yang and Zaslawski made the sensors more effective by sorting the light pulses into sequences so that the signals bounce back with a higher intensity. But that did not resolve the “echo” issue—that is, discovering a key to make the signals readable.
Therefore, they devised a technique for encoding the data sent using a fiber; the technique involves using unique genetic optimization algorithms to deal with faults.
Other systems are either limited in scope or expensive. But with ours, you just have to add a software program to your existing equipment. No need to adapt your sensors or use complex devices.
Luc Thévenaz, Professor, School of Engineering, EPFL
Journal Reference:
Sun, X., et al. (2020) Genetic-optimised aperiodic code for distributed optical fibre sensors. Nature Communications. doi.org/10.1038/s41467-020-19201-1.
Source: https://www.epfl.ch/en/