The precise geolocation of underground fiber optic telecom cables is crucial for operator maintenance, to avoid construction-related damage, and to minimize costs during network upgrades. Our innovative calibration technology provides non-instrusive high-accuracy localization of subsurface fiber optic cables, using only video inputs and ambient urban under any noise conditions. In addition, the same data are processed to yield an initial estimation of subsurface structure, useful for in-situ geotechnical decisions.
UNMET NEED
The telecom, construction, and infrastructure industries face significant challenges in accurately locating underground fiber optic cables, as existing mapping is often nonexistent or inaccurate. Traditional location methods are often imprecise and time-consuming, leading to potential damage during construction and inefficient access to infrastructure. There is a critical need for a reliable, precise, and efficient solution to address these challenges.
OUR SOLUTION
Our solution involves the calibration of underground fiber optic sensors using advanced algorithms and sensor technology. Key features include:
1. High-Accuracy Localization: Determining the 3D location of fiber optic cables with high precision.
2. Subsurface models: Actionable geotechnical properties of the subsurface below the fiber.
3. Full automation: no need for active sources or human intervention.
APPLICATIONS
1. Construction and Excavation: Accurate localization of underground fiber optic cables to avoid damage and ensure safe excavation.
2. Geotechnical Analysis: Providing detailed subsurface models under fiber locations, replacing costly and time-consuming surveys.
3. Urban Planning: Assisting city planners in designing and maintaining smart city infrastructure with precise underground mapping.
4. Utility Management: Enhancing the management of utilities by providing accurate location data for underground assets.
STATUS
Full proof of concept validated by successful pilot projects demonstrating its effectiveness in real-world conditions. Ongoing field tests continue to validate the accuracy and reliability of our calibration technology.
INTELLECTUAL PROPERTY
US provisional patent application, Serial No. 63/663,951, filed June 25, 2024
“METHOD FOR MONITORING TRAFFIC USING CONVOLUTIONAL NEURAL NETWORKS WITH VIDEO SIGNALS OVER DISTRIBUTED FIBER-OPTIC SENSING”
Papers (Under preparation)
Khen Cohen, Liav Hen, and Ariel Lellouch, Training a Distributed Acoustic Sensing traffic analysis network with video inputs.
Khen Cohen, Netanel Nisan, Ofir Nissan, and Ariel Lellouch, Self-calibrating optical fibers using visual guidance.