In optical fibers, sensing is traditionally carried out using a single mode operation, technically denoted as Single Mode Fibers (SMF). This operation greatly reduces the complexity of sensor signal processing, but at the same time, significantly limits the capabilities of fiber sensors and sensing systems. In this project, adaptive optics methodologies are employed in conjunction with mode division multiplexing techniques to greatly expand the overall sensing capacity of fiber optics sensor networks and to reduce their per-sensor-cost by a factor of 10 to 100. The novel mode division multiplexing, or MDM, sensing technology investigated in this project can impact a wide array of critical sensing applications and data communication, and will enable large-scale low-cost infrastructure monitoring as well as flexible and ultra-high-data-rate internet communications.
Technically, this project aims to develop a first-of-its-kind MDM sensor network for distributed sensing. The most critical element of this sensor network is an adaptive optical component that can: 1) achieve highly selective mode excitation in a multi mode fiber, and 2) dynamically route the interrogation signals towards different sensor sub-networks. Specifically, two multi modal sensor networks will be demonstrated: a quasi-distributed one for temperature/strain sensing, and a fully distributed one for optical time-domain reflectometry sensing. The goal is to demonstrate that it is possible to integrate multiple effective sensors in a single fiber by exploiting the multi mode operation. A secondary goal is to confirm the full compatibility of mode division multiplexing with state-of-the-art multiplexing methods. Accomplishing these two goals will suggest that by employing fiber modes utilizing M modes, one can increase the overall sensing capacity of any existing fiber sensor network by a factor of M. While the project will be limited to two-mode operation (M=2), which will lead to the doubling of the sensors density available on a given fiber, it is fully expected that the developed principles will be applicable to cases with M as large as 10 to 100, corresponding to ten-fold or hundred-fold increase in sensor densities.
