Abstract: A method may include publishing metadata to a blockchain, the metadata describing a training task associated with a global machine-learning model and computational resource requirements for performing the training task. The method may include receiving a request to participate in training the global machine-learning model from one or more clients based on a relevance of a respective local dataset of each of the clients to the training task and a suitability of the clients to the computational resource requirements for performing the training task. The method may include obtaining local model updates in which each respective local model corresponds to a respective client and each respective local model update is generated based on training the global machine-learning model with each of the local datasets. The method may include aggregating the plurality of local model updates and generating an updated global machine-learning model based on the aggregated local model update.

Abstract: A method may include publishing metadata to a blockchain, the metadata describing a training task associated with a global machine-learning model and computational resource requirements for performing the training task. The method may include receiving a request to participate in training the global machine-learning model from one or more clients based on a relevance of a respective local dataset of each of the clients to the training task and a suitability of the clients to the computational resource requirements for performing the training task. The method may include obtaining local model updates in which each respective local model corresponds to a respective client and each respective local model update is generated based on training the global machine-learning model with each of the local datasets. The method may include aggregating the plurality of local model updates and generating an updated global machine-learning model based on the aggregated local model update.

Abstract: An apparatus serves as a node device included in a distributed data sharing network, and shares, by using a blockchain, a piece of event information related to an event generated in a terminal, among node devices included in the distributed data sharing network, where the blockchain is a continuously growing list of blocks which are linked and secured using cryptography. The apparatus generates a piece of event data including the piece of event information related to the event, and causes the generated piece of event data to be stored in one of the node devices included in the distributed data sharing network.

Abstract: An edge server disposed on an edge of a cloud network, includes: a processor, wherein when encryption key information relating to a terminal which requests a connection to the edge server and the encryption key information generated before the request, is included in shared information shared between a cloud server and another edge server in the cloud network, the processor is configured to start encryption communication with the terminal using the encryption key information of the shared information.

Abstract: Methods and systems enable amplifying optical signals using a Bragg reflection waveguide (BRW) having second order optical nonlinearity to generate an optical pump by injection locking. The BRW may also be used for parametric amplification of optical signals using the optical pump. Feedback phase-power control may be performed to maximize output power.

Abstract: A method and system for implementing a network service computation system uses distributed graph processing at a plurality of network controllers corresponding to a plurality of network domains. Each network controller may manage and maintain a network graph for its respective network domain. Each network controller many communicate with nodes (or vertices) in its respective network domain, while the network controllers may communicate with each other for path discovery and computation purposes.

Abstract: Methods and systems enable amplifying optical signals using a Bragg reflection waveguide (BRW) having second order optical nonlinearity to generate an optical pump by injection locking. The BRW may also be used for parametric amplification of optical signals using the optical pump. Feedback phase-power control may be performed to maximize output power.

Abstract: An apparatus serves as a node device included in a distributed data sharing network, and shares, by using a blockchain, a piece of event information related to an event generated in a terminal, among node devices included in the distributed data sharing network, where the blockchain is a continuously growing list of blocks which are linked and secured using cryptography. The apparatus generates a piece of event data including the piece of event information related to the event, and causes the generated piece of event data to be stored in one of the node devices included in the distributed data sharing network.

Abstract: An edge server disposed on an edge of a cloud network, includes: a processor, wherein when encryption key information relating to a terminal which requests a connection to the edge server and the encryption key information generated before the request, is included in shared information shared between a cloud server and another edge server in the cloud network, the processor is configured to start encryption communication with the terminal using the encryption key information of the shared information.

Abstract: Methods and systems enable amplifying optical signals using a Bragg reflection waveguide (BRW) having second order optical nonlinearity to generate an optical pump by injection locking. The BRW may also be used for parametric amplification of optical signals using the optical pump. Feedback phase-power control may be performed to maximize output power.

Abstract: Methods and systems may estimate nonlinear penalties for optical paths using spectral inversion in optical transport networks. Certain values of nonlinear transfer functions for nonlinear penalty estimation may be pre-calculated for optical paths between given nodes. When an optical path computation for using spectral inversion between a given source node and a given destination node is desired, the pre-calculated values may be concatenated for improved computational efficiency.

Abstract: Methods and systems for amplifying optical signals include generating idler signals for input signals using an optical pump at a first Bragg reflection waveguide (BRW) having second order optical nonlinearity. Phase and amplitude regulation is performed using the output from the first BRW. Optical power monitoring of the input signals may be used for power equalization. The phase-sensitive amplified signal is generated at a second BRW using the optical pump. Optical power monitoring of the input signals may be used for power equalization.

Abstract: Object-oriented network virtualization may involve creating and operating virtual network objects (VNO) using a software-programmed networking operating system (SPN OS). A VNO may be a complete representation of a virtual network service provided under the SPN OS. A VNO may have a unique identity and properties, along with an internal set of methods for executing functionality encapsulated by the VNO. A VNO may exhibit persistence and autonomous control to enable improved virtual network services.

Abstract: A method and system for amplifying optical signals includes generating idler signals for input signals using an optical pump at a first non-linear element (NLE). An intermediate stage including a Raman amplifier performs pump amplification using the output from the first NLE along a single optical path. Optical power monitoring of the input signals may be used for power equalization. The intermediate stage may include a wavelength selective switch for a certain degree of phase modulation. The phase-sensitive amplified signal is generated at a second NLE using the optical pump. Optical power monitoring of the input signals may be used for power equalization and other control functions to achieve low-noise operation.

Abstract: Methods and systems for superchannel power pre-emphasis may adjust power levels of selected subcarriers of the superchannel. The power pre-emphasis may be performed at a transmission stage using a laser source, a variable optical attenuator, or a wavelength selective switch. The power pre-emphasis may be performed in-line at a reconfigurable optical add-drop multiplexer node. The power pre-emphasis may be performed using feedback control based on a receiver output. The power pre-emphasis may be performed using feedforward control based on optical path computations.

Abstract: Methods and systems for transmitting superchannels with mixed baud rate subcarriers include modifying baud rates for certain subcarriers in order to improve or equalize optical signal-to-noise ratio penalties incurred during transmission. Additionally frequency shifts may be applied to individual subcarriers. The baud rate modification and frequency shifts may be symmetrical for spectral positions of subcarriers within the superchannel.

Abstract: Methods and systems for superchannel subcarrier monitoring using amplitude modulated (AM) tones include modulating an optical subcarrier of a superchannel with a first frequency that is chosen to be substantially smaller than a baud rate for data modulated onto the optical subcarrier. Certain pairs of subcarriers in the superchannel may be modulated with AM tones having a common frequency with complementary phase for power equalization. Then, downstream detection and monitoring of the optical subcarrier may be performed based on demodulation of the first frequency. Each optical subcarrier in the superchannel may be modulated using a respectively unique combination of AM tone and phase.

Abstract: Methods and systems for superchannel subcarrier monitoring using frequency modulated (FM) tones includes frequency modulating an optical subcarrier of a superchannel with a first frequency that is unique to the optical subcarrier and is chosen to be smaller than an optical transmission frequency for carrier data modulated onto the optical subcarrier. Then, downstream detection and monitoring of the optical subcarrier may be performed based on demodulation of the first frequency. Each optical subcarrier in the superchannel may be modulated using a respectively unique FM tone.

Abstract: Optical transceiver sharing methods may be based on different ROADM node architectures for shared restoration in flexible grid optical networks. A ROADM node architecture with a pool of transceivers may improve transceiver utilization for backup optical paths, compared to a conventional ROADM node architecture. Sharing of transceivers in the pool for working and backup optical paths may further improve transceiver utilization. The methods disclosed herein may be used for multiple bit rates and different modulation formats.

Abstract: Methods and systems are provided for optical signal transmission using span-wise spectrum management. The method includes transmitting a first optical signal at a first wavelength by a control system. The first optical signal has a first optical path distance. The method further includes determining a first guard band requirement for the first optical signal, and transmitting a second optical signal by the control system. The second optical signal has a second optical path distance. The method includes determining a second guard band requirement for the second optical signal, and placing the second optical signal spectrally adjacent to the first optical signal by selecting a second wavelength. The second wavelength is selected to satisfy the first and second guard band requirements.