International Business Machines Corporation

Alcatel Lucent

Nokia Solutions and Networks Oy

Nicolas Dupuis

Benjamin G Lee

Alexander V Rylyakov

Clint Lee Schow

Christopher Richard Doerr

Daniel Michael Kuchta

Jonathan E Proesel

Mehmet Soyuer

Benoit Drooghaag

Tam N Huynh

Renato Rimolo-Donadio

Long L Chen

Axel Van Damme

Claude Basso

Fuad Elias Doany

Frank Robert Libsch

Bengi Karacali-Akyamac

Laurent Schares

Marc A Taubenblatt

Jeonghwan Song

Philippe Dierickx

Thomas James Christopher Ward

Md Abdul Alim

Stijn Meersman

Pavlos Maniotis

Embodiments for dynamic bandwidth allocation in cloud network switches in a cloud computing environment are provided. Quality of service (QoS) policies may be dynamically changed in one or more cloud network switches based on dynamically estimating expected traffic demands for each of a plurality of traffic classes, wherein bandwidth is dynamically allocated among queues based on changing the QoS policies.

Dynamic bandwidth allocation in cloud network switches based on traffic demand prediction

An optical switch fabric comprises two or more optical switch elements. The optical switch elements are configured in a topology. A switch control has a plurality of bias control signals. The switch control can address one or more of the optical switch elements and can apply one of the bias control signals to bias of the addressed optical switch element to establish a switch setting. The topology and switch settings determine how each of one of the inputs is connected to each of one of the outputs of the optical switch fabric. The switch settings are determined by a machine learning process which includes a model creation. The model can be made to adapt dynamically during optical switch fabric operation.

Configuration of an optical switch fabric using machine learning

Embodiments relate to an apparatus for monitoring a telecommunication network including one or more telecommunication channels. The apparatus includes a processor configured after executing computer code to design a convolutional neural network configured for determining an impairment type of a telecommunication channel as a function of a channel frequency response of the telecommunication channel, by selecting at least one of a number of convolutional layers, a number of filters for respective convolutional layers, and/or a size of filters for respective convolutional layers; and train the convolutional neural network based on training data specifying, for respective telecommunication channels, a channel frequency response and an associated impairment type.

Method and apparatus for monitoring a telecommunication network

Systems, computer-implemented methods, and computer program products to facilitate rotated polarization detection and adjustment are provided. According to an embodiment, a system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise an optical component that can comprise a polarization monitor component that can detect a rotated polarization state of an optical signal. The computer executable components can further comprise a second optical component that can comprise a polarization controller component that can control a rotation polarization state of the second optical component. The computer executable components can further comprise a feedback loop component that can couple the polarization monitor component to the polarization controller component.

Polarization-insensitive optical link

Techniques that facilitate control of dual phase tuners are provided. In one example, a system includes a first phase tuner and a second phase tuner. The first phase tuner is driven to a first phase bias that is different than a desired phase bias for an interferometer. The second phase tuner is driven to a second phase bias. A combination of the first phase bias and the second phase bias corresponds to the desired phase bias.

Control of dual phase tuners

The apparatus includes a memory configured to store executable code; and a processor configured to execute the executable code and cause the apparatus to perform the operations of generating and training. The generating generates a dataset specifying, for a plurality of communication lines, i) a channel frequency response of a communication line having one or two bridged taps, and ii) a set of M lengths of bridged taps associated with the communication line, with M greater than one. The training trains, based on the dataset, a machine learning model, the machine learning model configured for determining, based on the channel frequency response of a communication line, a set of M lengths of bridged taps associated with the communication line.

Method and apparatus for determining a bridged tap length

Embodiments relate to a method and an apparatus for predicting the bitrate of a repaired communication channel. The method may include, generating a dataset specifying, for a plurality of communication channels, a channel frequency response of a communication channel affected by an impairment, and a channel frequency response of the communication channel non-affected by the impairment, training, based on the dataset, a machine learning model configured to predict a channel frequency response of the repaired communication channel based on the channel frequency response of the communication channel affected by an impairment.

Method and apparatus for predicting the bitrate of a repaired communication channel

Optical switches and methods of switching include a first hybrid coupler configured to accept an input and to provide two branches. A phase tuner on a first branch includes a Mach-Zehnder phase shifter configured to shift a signal on the first branch by a selected phase. A loss compensator on a second branch is configured to match a loss incurred on the first branch. A second hybrid coupler is configured to recombine the two branches such that the phase shift generated by the phase tuner determines which output of the second hybrid coupler is used.

Low-crosstalk electro-optical Mach-Zehnder switch

Methods and systems for bias control in an optical switch fabric include monitoring optical power at outputs of a plurality of switch elements in an N&#xd7;N switch fabric that has N inputs, N outputs, and M&#x2265;2 stages, each switch element having a respective unique radio frequency tone. A bias control of each of the plurality of switch elements is adjusted. It is determined whether the optical power at the outputs of each respective switch element after bias control adjustment conform more closely to a predetermined criterion relative to the monitored optical power at the outputs of the respective switch element prior to adjustment.

Optical switch fabric with bias control

Optical switches include a phase shifter on a first branch, a first heater on the first branch, and a second heater on a second branch. A hybrid coupler combines the first branch and the second branch. A first photodetector and a second photodetector are at outputs of the second hybrid coupler to measure crosstalk between the outputs of the second hybrid coupler. A controller is configured to activate the first heater or the second heater to reduce the measured crosstalk.

Low-crosstalk electro-optical mach-zehnder switch

A device and method of optical equalization using an optical modulator is provided. An electrical modulation signal is split into a first modulation signal and a second modulation signal. The second modulation signal is delayed relative to the first modulation signal. An amplitude of the second modulation signal is attenuated relative to the first modulation signal. The first modulation signal is applied to a first waveguide segment of the optical modulator. The second modulation signal that is delayed and attenuated relative to the first modulation signal is applied to a second waveguide segment of the optical modulator. Both the applied first and second modulation signals generate a feed-forward equalized optical signal that is recombined in the optical domain.

Feed-forward optical equalization using an electro-optic modulator with a multi-segment electrode and distributed drivers

Techniques for using planar photonic switch fabrics with reduced waveguide crossings are described. In one embodiment, a system is provided that comprises a memory that stores computer-executable components and a processor that executes computer-executable components stored in the memory. In one implementation, the computer-executable components comprise an arrangement component that arranges a first planar switch fabric topology. The computer-executable components further comprise a transformation component that interleaves a plurality of inputs of the first planar switch fabric topology and a plurality of outputs of the first planar switch fabric topology to form a second planar switch fabric topology, the second planar switch fabric topology having a lower number of waveguide crossings than the first planar switch fabric topology.

Planar photonic switch fabrics with reduced waveguide crossings

Optical switches include a first hybrid coupler configured to accept an input and to provide two branches. A phase tuner on a first branch includes a phase shifter. A first heater is positioned on the first branch and a second heater is positioned on the second branch, each configured to compensate for phase error. A second hybrid coupler configured to recombine the two branches. A first photodetector and a second photodetector are positioned at outputs of the second hybrid coupler and are configured to measure crosstalk between the outputs of the second hybrid coupler. A controller is configured to activate the first heater or the second heater to reduce the measured crosstalk.

Growing community of inventors

New York, NY, United States of America

Nicolas Dupuis