The Project

The omnipresence of connected objects and the new habits of numerical services consumption have caused an ever-expanding need for high data rates. Today, most long-haul transmission systems and an increasing part of home Internet services rely on optical fibers. During the past decades, researchers have taken advantage of the diverse degrees of freedom offered by the fiber (time, wavelength, amplitude, phase and polarization of the optical field) to increase the achievable data rates. However, these systems are about to reach their theoretical limit of about 100 terabits per second while the capacity need undergoes a constant increase of 30 to 40% each year according to the industry players. Consequently, a breakthrough is required to meet the growing data rates needs. One type of degrees of freedom that has not yet been exploited involves the space dimension.

Optical fibers currently deployed for long-haul transmissions are single mode fibers: light can only follow one unique path. However, there exist multimode fibers in which light can take different trajectories. Taking advantage of the spatial degrees of freedom would allow increasing significantly the number of channels, similarly to MIMO (Multiple-Inputs/Multiple-Outputs) systems deployed for wireless communications.

The goal of our project is to develop original techniques, protocols and devices that take advantage of the spatial diversity of the multimode fibers in order to increase the data rates achievable for telecommunications and meet the traffic needs for the next decades. In this very active domain of research, most alternatives are either focused on the multiplexers, to inject light into independent channels, or on the numerical reconstruction of the signal, to compensate for the perturbations and the channel crosstalk. By considering the multimode optical system as a whole we intend to overcome the limitations of both approaches. To this end, we need to invent new hardware systems for modulation and multiplexing, to build new simulation models to study telecommunications channels and to develop new protocols for optical MIMO approaches.