Abstract: Systems and methods for predicting performance of a modulation system are provided. A neural network model is trained using performance information of a source system. The neural network model is modified with transferable knowledge about a target system to be evaluated. The neural network model is tuned using specific characteristics of the target system to create a source-based target model. The target system performance is evaluated using the source-based target model to predict system performance of the target system.

Abstract: This application discloses a resource pushing method performed by a computer device. The method includes: obtaining a target recommendation model and a preference feature and a candidate resource set corresponding to a target object, the preference feature including at least a channel preference feature and a content preference feature; obtaining at least one target resource from the candidate resource set based on the target recommendation model and the preference feature; and pushing the at least one target resource to the target object. Such a resource pushing process integrates preferences of the target object in different dimensions, so that the target resource pushed to the target object not only conforms to channel preferences of the target object, but also conforms to content references of the target object, which is beneficial to improving the resource pushing effect, and further increasing the click-through rates (CTRs) of the pushed resources.

Abstract: Aspects of the present disclosure describe a method for digital coherent transmission systems that advantageously provides low-complexity, single-step nonlinearity compensation based on artificial intelligence (AI) implemented in a deep neuron network (DNN).

Abstract: Aspects of the present disclosure describe systems, methods. and structures in which a hybrid neural network combining a CNN and several ANNs are shown useful for predicting G-ONSR for Ps-256QAM raw data in deployed SSMF metro networks with 0.27 dB RMSE. As demonstrated, the CNN classifier is trained with 80.96% testing accuracy to identify channel shaping factor. Several ANN regression models are trained to estimate G-OSNR with 0.2 dB for channels with various constellation shaping. Further aspects include the tuning of existing optical networks and the characterization of retrofit/upgraded optical networks to estimate capacity—both aspects employing our inventive hybrid neural network methodology.

Abstract: Aspects of the present disclosure describe systems, methods, and structures that advantageously amplify optical signals through the effect of optical pump signals generated by a multicore laser diode and multicore rare-earth doped optical fiber in optical communication with a 3D waveguide structure and a multicore input signal fiber providing a plurality of optical signals for amplification.

Abstract: Aspects of the present disclosure describe systems, methods. and structures in which a hybrid neural network combining a CNN and several ANNs are shown useful for predicting G-ONSR for Ps-256QAM raw data in deployed SSMF metro networks with 0.27 dB RMSE. As demonstrated, the CNN classifier is trained with 80.96% testing accuracy to identify channel shaping factor. Several ANN regression models are trained to estimate G-OSNR with 0.2 dB for channels with various constellation shaping. Further aspects include the tuning of existing optical networks and the characterization of retrofit/upgraded optical networks to estimate capacity—both aspects employing our inventive hybrid neural network methodology.

Abstract: Aspects of the present disclosure describe systems, methods and structures in which a hybrid neural network combining a CNN and several ANNs are shown useful for predicting G-ONSR for Ps-256QAM raw data in deployed SSMF metro networks with 0.27 dB RMSE. As demonstrated, the CNN classifier is trained with 80.96% testing accuracy to identify channel shaping factor. Several ANN regression models are trained to estimate G-OSNR with 0.2 dB for channels with various constellation shaping.

Abstract: Aspects of the present disclosure describe systems, methods and structures for optical fiber nonlinearity compensation using neural networks that advantageously employ machine learning (ML) algorithms for nonlinearity compensation (NLC) that advantageously provide a system-agnostic model independent of link parameters, and yet still achieve a similar or better performance at a lower complexity as compared with prior-art methods. Systems, methods, and structures according to aspects of the present disclosure include a data-driven model using the neural network (NN) to predict received signal nonlinearity without prior knowledge of the link parameters. Operationally, the NN is provided with intra-channel cross-phase modulation (IXPM) and intra-channel four-wave mixing (IFWM) triplets that advantageously provide a more direct pathway to underlying nonlinear interactions.

Abstract: An unrepeatered transmission system includes a receiver coupled to a receive span; a transmitter coupled to the receive span; and a plurality of cascaded amplifiers in the receive span with dedicated fiber cores to supply one or more optical pumps from the receiver to each amplifier, wherein the plurality of cascaded amplifiers increase system reach by increasing the length of a back span in an unrepeatered link.

Abstract: Systems and methods for transmission filtering are provided. A receiver includes an input coupled to a transmission line to receive distorted optical symbols. A distortion filter is coupled to the input to replace the distorted optical symbols with predicted symbols using a trained neural network. A decoder is coupled to the distortion filter to decode the predicted symbols.

Abstract: Aspects of the present disclosure describe systems, methods and structures in which a hybrid neural network combining a CNN and several ANNs are shown useful for predicting G-ONSR for Ps-256QAM raw data in deployed SSMF metro networks with 0.27 dB RMSE. As demonstrated, the CNN classifier is trained with 80.96% testing accuracy to identify channel shaping factor. Several ANN regression models are trained to estimate G-OSNR with 0.2 dB for channels with various constellation shaping.

Abstract: An optical communication substrate includes a plurality of cores to communicate optical signals; a rectangular input delivering a pump laser, and a shaped portion to combine the optical signals and the pump laser into a ring geometry at an output.

Abstract: Systems and methods for orbital angular momentum (OAM)-based multidimensional wireless communication. The OAM-based multidimensional wireless communication is preformed with a transmitter for generating an RF modulated signal carrying a data sequence. Further included is an OAM antenna array including OAM antenna elements, each of which includes an azimuthal phase shifter and an antenna element. The azimuthal phase shifter shifts an azimuthal phase term of a wavefront generated by the antenna element such that the OAM antenna element imposes the multidimensional modulated signal on a pre-determined OAM mode of a carrier signal corresponding to the azimuthal phase term.

Abstract: Systems and methods for quantum key distribution using orbital angular momentum (OAM)-modes-enabled secure multidimensional coded modulation. The quantum key distribution including symbols of a raw key across subcarriers and multiplexing the subcarriers to form an electrical carrier. Then optically modulating the electrical carrier with an electro-optical modulator using an optical signal. The optically modulated electrical carrier is then imposed on a pre-determined OAM mode using an optical OAM multiplexer, attenuated, and transmitted across a quantum channel. A receiver then receives the transmitted signal and demultiplexes it with an optical OAM demultiplexer to extract projections. The projections then undergo optical-to-electrical conversion using a coherent optical detector with an optical signal from a local oscillator. A resulting estimated electrical carrier is then demultiplexed into subcarriers.

Abstract: Aspects of the present disclosure describe systems, methods and structures for optical fiber nonlinearity compensation using neural networks that advantageously employ machine learning (ML) algorithms for nonlinearity compensation (NLC) that advantageously provide a system-agnostic model independent of link parameters, and yet still achieve a similar or better performance at a lower complexity as compared with prior-art methods. Systems, methods, and structures according to aspects of the present disclosure include a data-driven model using the neural network (NN) to predict received signal nonlinearity without prior knowledge of the link parameters. Operationally, the NN is provided with intra-channel cross-phase modulation (IXPM) and intra-channel four-wave mixing (IFWM) triplets that advantageously provide a more direct pathway to underlying nonlinear interactions.

Abstract: Aspects of the present disclosure describe systems, methods, and structures that advantageously provide hybrid free-space optical (FSO)-radio frequency (RF) communication links (HFRCLs) that enable building integrated software-defined network (SDN) infrastructure capable of integrating ultra-high-throughput satellite networks, composite wireless infrastructures, heterogeneous networks (HetNets), hybrid networks, satellite networks, and fiber-optics networks—among others.

Abstract: Systems and methods for optical data transport, including controlling data transport across an optical transmission medium by generating two-dimensional (2D) distribution matchers (DMs) based on probabilistic fold shaping (PFS) and arbitrary probabilistic shaping (APS). The 2D PFS-based DM is can encode any N-fold rotationally symmetrical Quadrature Amplitude Modulation (QAM) format by applying the 2D PFS-based DM only to symbols in one quadrant based on a target entropy. A fold index yield uniform distribution is determined, and is utilized to carry generated uniform distributed parity check bits across the optical transmission medium. The 2D APS-based DM can encode any arbitrary modulation formats by encoding uniform binary data to generate non-uniform target symbols, and generating a probability distribution for the target symbols by indirectly applying the 2D APS-based DM based on a target probability distribution and a detected code rate of generated FEC code.

Abstract: A system and method are provided for distributed acoustic sensing in a multimode optical fiber. The system includes a transmitter for simultaneously propagating a sequence of M light pulses through the multimode optical fiber using a spatial mode selected from a set of N spatial modes provided by a spatial mode selector for the transmitter that is coupled to an input to the multimode optical fiber, with M and N being respective integers greater than one. The system further includes a receiver for receiving the sequence of M light pulses at an output of the multimode optical fiber and detecting an environmental perturbation in the multimode optical fiber based on an evaluation of a propagation of the sequence of M light pulses through the multimode optical fiber.

Abstract: Aspects of the present disclosure describe fiber nonlinearity induced transmission penalties are reduced both in fibers with large polarization-mode dispersion, and in coupled-core multicore fibers (CC-MCF). In the case of coupled multi-core fibers, the requirement for modal delay is less.

Abstract: Aspects of the present disclosure describe a method for digital coherent transmission systems that advantageously provides low-complexity, single-step nonlinearity compensation based on artificial intelligence (AI) implemented in a deep neuron network (DNN).