Abstract:The use of multicore optical fibers is now recognized as one of the most promising methods to implement the space-division multiplexing techniques required to overcome the impending capacity limit of conventional single-mode optical fibers. Nonetheless, new devices for networking operations compatible with these fibers will be required in order to implement the next-generation high-capacity optical networks. In this work, we develop a new architecture to build a high-speed core-selective switch, critical for efficiently distributing signals over the network. The device relies on multicore interference, and can change among outputs in less than 0.7 us, while achieving less than -18 dB of average inter-core crosstalk, making it compatible with a wide range of network switching tasks. The functionality of the device was demonstrated by routing a 1GBs optical signal and by successfully switching signals over a field-installed multicore fiber network. Our results demonstrate for the first time the operation of a multicore optical fiber switch functioning under real-world conditions, with switching speeds that are three orders of magnitude faster than current commercial devices. This new optical switch design is also fully compatible with standard multiplexing techniques and, thus, represents an important achievement towards the integration of high-capacity multicore telecommunication networks.
Abstract:A deep reinforcement learning approach is applied, for the first time, to solve the routing, modulation, spectrum and core allocation (RMSCA) problem in dynamic multicore fiber elastic optical networks (MCF-EONs). To do so, a new environment - compatible with OpenAI's Gym - was designed and implemented to emulate the operation of MCF-EONs. The new environment processes the agent actions (selection of route, core and spectrum slot) by considering the network state and physical-layer-related aspects. The latter includes the available modulation formats and their reach and the inter-core crosstalk (XT), an MCF-related impairment. If the resulting quality of the signal is acceptable, the environment allocates the resources selected by the agent. After processing the agent's action, the environment is configured to give the agent a numerical reward and information about the new network state. The blocking performance of four different agents was compared through simulation to 3 baseline heuristics used in MCF-EONs. Results obtained for the NSFNet and COST239 network topologies show that the best-performing agent achieves, on average, up to a four-times decrease in blocking probability concerning the best-performing baseline heuristic methods.
Abstract:In optical communications, space-division multiplexing is a promising strategy to augment the fiber network capacity. It relies on modern fiber designs that support the propagation of multiple spatial modes. One of these fibers, the ring-core fiber (RCF), is able to propagate modes that carry orbital angular momentum (OAM), and has been shown to enhance not only classical, but also quantum communication systems. Typically, the RCF spatial modes are used as orthogonal transmission channels for data streams that are coupled into the fiber using different Laguerre-Gaussian (LG) beams. Here, we study the optimal conditions to multiplex information into ring-core fibers in this scheme. We determine which are the most relevant LG beams to be considered, and how their coupling efficiency can be maximized by properly adjusting the beam width with respect to the fiber parameters. Our results show that the coupling efficiency depends upon the OAM value, and that this can limit the achievable transmission rates. In this regard, we show that LG beams are not the optimal choice to couple information into RCF. Rather, another class of OAM-carrying beam, the perfect vortex beam, allows for nearly perfect coupling efficiencies for all spatial modes supported by these fibers.