Huawei Technologies Co., Limited

Ciena Corporation

Future Wei Technologies, Inc.

Cienna Corporation

Petar Djukic

Todd D Morris

Jianglei Ma

Hang Zhang

Kelvin Kar Kin Au

Hosein Nikopour

Liqing Zhang

Xu Li

Aaron Callard

Alireza Bayesteh

Mohammadhadi Baligh

Zhihang Yi

David Jordan Krauss

Romualdas Armolavicius

Peiying Zhu

Alex Stephenne

Nimal Gamini Senarath

Peter J Ashwood-Smith

Vipul Anil Desai

Ngoc Dung Dao

Hamidreza Farmanbar

Philippe Leroux

David Côté

Amine Maaref

Sheng Sun

Stephen Daniel Shew

Emil Janulewicz

Mohamed Adel Salem

David Miedema

Kaniz A Mahdi

Halim Yanikomeroglu

Thomas Triplet

Ngoc-Dung Dao

Maryam Amiri

Firouzeh Golaghazadeh

Kaveh Shafiee

Christopher Janz

James P'ford't Carnes, Iii

Sid Ryan

Mitchell Howard Auster

James P Carnes, Iii

Jietao Zhang

Yinqing Pei

Paul Littlewood

John Wade Cherrington

Balaji Subramaniam

Burak Kantarci

William Kaufmann

Murat Simsek

Jinxin Liu

Marco Naveda

Kelvin Au

Yu Shen

Mehran Bagheri

Dmitri Fedorov

Zhiyan Chen

Joseph Alexander

Ivan Radovic

Nelson Chao

Wentao Cui

Salim Tariq

Paul Gosse

Dana Dennis

Patrick Premont

Alireza Sharifian

Kelvin Kar Au

Systems and methods for detecting patterns in data from a time-series and for detecting outliers in network data in an unsupervised manner are provided. In one implementation, a method includes the steps of obtaining network data from a network to be monitored and creating a window from the obtained network data. The method also includes the step of detecting outliers of the obtained data with respect to the window using an unsupervised deep learning process (e.g., using a Generalized Adversarial Network (GAN) learning technique and/or a Bidirectional GAN (BiGAN) learning technique) for enabling the learning of a data distribution. The unsupervised process, for example, does not require manual intervention.

Unsupervised outlier detection in time-series data

Systems and methods for embedding concealed meta-data into DNNs. Specifically, the system and method presented consists of receiving a trained neural network that includes one or more layers each having weights. The disclosed process includes transforming at least one layer to a transformed domain, adding information to the layer(s) in the transformed domain, and performing an inverse domain transform on at least one layer such that the layer(s) has new weights with an embedded watermark.

Embedding concealed meta-data into deep neural networks (DNNs)

Systems and methods for monitoring a network slice are provided. A method, according to one implementation, include extracting information from network traffic received from one or more User Plane Function (UPF) components of a network slice; examining the extracted information using Machine Learning (ML), and, in response to detecting of one or more malicious threats based on the examined extracted information by the ML, causing one or more actions to isolate the network traffic to protect at least the network slice from the one or more malicious threats.

Detecting malicious threats in a 5G network slice

Systems and methods for detecting intrusions, attacks, and sub-attacks launched against a network under observations are provided. A method, according to one implementation, includes obtaining network traffic information regarding data traffic in a network under observation and obtaining system log information regarding operations of the network under observation. The method further includes the step of inserting the network traffic information and system log information into one or more analysis procedures, where each analysis procedure is configured to detect a respective sub-attack of a multi-stage attack to which the network under observation is susceptible. Also, the method includes the step of combining the outputs of the one or more analysis procedures to detect whether one or more sub-attacks have been launched against the network under observation. In response to detecting that one or more sub-attacks have been launched, the methods include the step of determining the type of the one or more sub-attacks.

Machine learning detection of network attacks using traffic and log information

Systems and methods include monitoring packets, by a network edge device, from one or more endpoint devices where the packets are destined for corresponding application services in a network; classifying the one or more endpoint devices based on the monitoring into a corresponding trust level of a plurality of trust levels; and, responsive to a first endpoint device of the one or more endpoint devices being untrusted, steering the packets from the first endpoint device into a restricted zone.

Network architecture providing device identification and redirection using whitelisting traffic classification

The present disclosure relates to systems and methods for embedding concealed meta-data into DNNs. Specifically, the system and method presented consists of receiving a trained neural network that includes one or more layers each having weights. The disclosed process includes transforming at least one layer to a transformed domain, adding information to the layer(s) in the transformed domain, and performing an inverse domain transform on at least one layer such that the layer(s) has new weights with an embedded watermark.

Systems, methods, and computer-readable medium for forecasting a time-series are provided. In one implementation, a method is configured to include a step of providing a time-series to a neural network including one or more branches for processing one or more portions of the time-series. In each of the one or more branches, the method includes separating the respective portion of the time-series into individual portions and applying each portion to a respective sub-branch of a plurality of sub-branches of the one or more branches. The method also includes generating forecasting coefficients for each output time point in each of the respective sub-branches and providing a forecast of the time-series based at least on the forecasting coefficients.

Forecasting time-series data in a network environment

A method embodiment includes implementing, by a base station (BS), a grant-free uplink transmission scheme. The grant-free uplink transmission scheme defines a first contention transmission unit (CTU) access region in a time-frequency domain, defines a plurality of CTUs, defines a default CTU mapping scheme by mapping at least some of the plurality of CTUs to the first CTU access region, and defines a default user equipment (UE) mapping scheme by defining rules for mapping a plurality of UEs to the plurality of CTUs.

System and method for uplink grant-free transmission scheme

Systems and methods are provided for utilizing polarization parameters obtained from an optical network to determine vibrations in optical fibers using coherent optics equipment and machine learning techniques. A method, according to one implementation, includes the step of obtaining a time-series dataset that includes measurements of polarization characteristics of light traversing an optical fiber of an optical network. The method also includes the step of detecting vibration characteristics of the optical fiber based on the time-series dataset. In some implementations, the time-series dataset may be a multi-variate dataset and the polarization characteristics may be related to transients in a State of Polarization (SOP). The SOP, for example, may be represented by an amplitude and a phase of an electric field vector and may be defined as having one of a linear polarization, elliptical polarization, and circular polarization.

Utilizing polarization characteristics to detect vibrations in optical fibers

Deep Neural Networks (DNNs) for forecasting future data are provided. In one embodiment, a non-transitory computer-readable medium is configured to store computer logic having instructions that, when executed, cause one or more processing devices to receive, at each of a plurality of Deep Neural Network (DNN) forecasters, an input corresponding to a time-series dataset of a plurality of input time-series datasets. The instructions further cause the one or more processing devices to produce, from each of the plurality of DNN forecasters, a forecast output and provide the forecast output from each of the plurality of DNN forecasters to a DNN mixer for combining the forecast outputs to produce one or more output time-series datasets.

Forecasting routines utilizing a mixer to combine deep neural network (DNN) forecasts of multi-variate time-series datasets

Systems and methods include receiving network topology information of a network including a plurality of routers; receiving link measurements defining bandwidth on links in the network; determining routes in the network based on the network topology information; and utilizing the routes and the link measurements to determine an estimate of an initial traffic matrix that includes the bandwidth between origin routers and destination routers.

Estimating a traffic matrix of a communication network using network topology features

Systems and methods for detecting patterns in data from a time-series are provided. In one implementation, a method for pattern detection includes obtaining data in a time-series and creating one-dimensional or multi-dimensional windows from the time-series data. The one-dimensional or multi-dimensional windows are created either independently or jointly with the time-series. The method also includes training a deep neural network with the one-dimensional or multi-dimensional windows utilizing historical and/or simulated data to provide a neural network model. Also, the method includes processing ongoing data with the neural network model to detect one or more patterns of a particular category in the ongoing data, and localizing the one or more patterns in time.

Pattern detection in time-series data

An optical network element includes a connection to an optical fiber in an optical line system including a coherent receiver; a microphone configured to detect sound; and circuitry connected to the microphone and configured to cause transmission of information related to sounds detected by the microphone to a receiver at an end of the optical line system, wherein the transmission is over the optical fiber in the optical line system to the coherent receiver. The optical network element can include a polarization controlling device connected to the circuitry and configured to modulate a state-of-polarization (SOP) envelope for the transmission.

Detecting and localizing acoustic signals with an optical network

A network edge device includes switching circuitry configured to switch traffic from one or more endpoint devices to corresponding application services over a network; and processing circuitry configured to monitor the traffic from the one or more endpoint devices, compare the monitored traffic to classify the one or more endpoint devices into a corresponding trust level of a plurality of trust levels, and route the traffic from each of the one or more endpoint devices based on its corresponding trust level. The network edge element is configured to provide network connectivity to the one or more endpoint devices.

Systems, methods, and computer-executable programs are provided for estimating margin in a network path. A method, according to one implementation, includes the step of executing a section-based analysis by estimating a margin parameter for each section of a plurality of sections forming a path through a network. The margin parameter is related to at least an available Optical Signal-to-Noise Ratio (OSNR) parameter and a Required OSNR (ROSNR) parameter. The method further includes the step of combining the estimated margin parameters for the plurality of sections to obtain a section-based estimate. In response to determining that the section-based estimate falls outside an acceptable confidence range, the method includes executing an additional path-based analysis to modify the estimate of the margin parameter.

Estimating performance margin of a multi-section path using optical signal-to-noise ratio (OSNR)

Growing community of inventors

Ottawa, Canada

Petar Djukic