Optimized detection of network defect exposure in network environment

    公开(公告)号:US11546227B2

    公开(公告)日:2023-01-03

    申请号:US16368735

    申请日:2019-03-28

    Abstract: Present technology is directed to preferred processing and the verification of diagnostic signatures for a plurality of network defect. The disclosed optimization process is based on expressing each Diagnostic Signature as a minimal sum of product Boolean function of associated network commands, followed by ranking of each command reference in the product terms of the Boolean expression as well as each Boolean product terms of the SOP Boolean expressions, and constructing a decision tree based on the provided rankings to thereby determine a minimum set of commands along with an preferred command dispatch sequence for evaluating a Diagnostic Signature. Further aspects include the translation of both the optimization computation (interpretation of network conditions associated with a network defect) and the computed workflow (dispatch of the command) into a series of declarative rules that can be processed by a machine reasoning engine to thereby automate the optimization process.

    REDUCING COMPLEXITY OF WORKFLOW GRAPHS THROUGH VERTEX GROUPING AND CONTRACTION

    公开(公告)号:US20210286644A1

    公开(公告)日:2021-09-16

    申请号:US16815980

    申请日:2020-03-11

    Abstract: Systems, methods, and computer-readable media for generating and presenting workflow graphs can include the following operations. A workflow graph including vertices is provided. An event is received to reorganize the vertices of the workflow graph. Each of the vertices is classified with a significance level, the significance level is based on at least one of a business rule and a vertex position of a vertex of the vertices. Vertices of the vertices having a low significance level are grouped together. The vertices in the group having the low significance level are combined into a new vertex, and the new vertex is expandable and collapsible to view the vertices in the group having the low significance level.

    Weight initialization for random neural network reinforcement learning

    公开(公告)号:US10257072B1

    公开(公告)日:2019-04-09

    申请号:US15718901

    申请日:2017-09-28

    Inventor: Samer Salam

    Abstract: A plurality of paths through a network are determined for transmitting a packet from a source device to a destination device. The paths are modelled as nodes in a Random Neural Network, each node corresponding to a path and a reward is calculated for each of the nodes. An excitatory weight and an inhibitory weight are determined for each of the nodes in the Random Neural Network. The excitatory weight is set directly proportional to the reward corresponding to the node for which the excitatory weight is being determined, and the inhibitory weight is set inversely proportional to the reward corresponding to the node for which the inhibitory weight is being determined. A potential is determined for each of the nodes based upon the excitatory and inhibitory weights. A path corresponding to the node with the highest potential is selected, and the packet is transmitted over the selected path.

    WEIGHT INITIALIZATION FOR RANDOM NEURAL NETWORK REINFORCEMENT LEARNING

    公开(公告)号:US20190097912A1

    公开(公告)日:2019-03-28

    申请号:US15718901

    申请日:2017-09-28

    Inventor: Samer Salam

    Abstract: A plurality of paths through a network are determined for transmitting a packet from a source device to a destination device. The paths are modelled as nodes in a Random Neural Network, each node corresponding to a path and a reward is calculated for each of the nodes. An excitatory weight and an inhibitory weight are determined for each of the nodes in the Random Neural Network. The excitatory weight is set directly proportional to the reward corresponding to the node for which the excitatory weight is being determined, and the inhibitory weight is set inversely proportional to the reward corresponding to the node for which the inhibitory weight is being determined. A potential is determined for each of the nodes based upon the excitatory and inhibitory weights. A path corresponding to the node with the highest potential is selected, and the packet is transmitted over the selected path.

    Protection against fading in a network ring

    公开(公告)号:US10039048B2

    公开(公告)日:2018-07-31

    申请号:US15178322

    申请日:2016-06-09

    CPC classification number: H04W40/34 H04L45/125 H04L45/22 H04W40/12

    Abstract: A method provided in one example embodiment includes detecting a first current bandwidth of a first link in a network ring, where the first current bandwidth indicates a signal degradation on the first link. The method also includes determining whether the first current bandwidth has degraded more than a second current bandwidth of a second link in the network ring, where the second current bandwidth indicates a signal degradation on the second link. The method further includes routing one or more network flows away from the first link if the first current bandwidth has degraded more than the second current bandwidth.

    Semantic data broker for dynamic association between devices and applications
    90.
    发明授权
    Semantic data broker for dynamic association between devices and applications 有权
    用于设备和应用程序之间的动态关联的语义数据代理

    公开(公告)号:US09553945B2

    公开(公告)日:2017-01-24

    申请号:US14174376

    申请日:2014-02-06

    CPC classification number: H04L67/2809 H04L41/00 H04L41/12

    Abstract: In one embodiment, a broker device receives device-identifying data to identify a device in a computer network. An ontology associated with the device-identifying data is then identified by the broker device and in response to identifying the ontology, interpretation instructions related to the identified ontology are determined. The broker device receives data from the identified device and interprets the received data based on the interpretation instructions.

    Abstract translation: 在一个实施例中,代理设备接收设备识别数据以识别计算机网络中的设备。 然后由代理设备识别与设备识别数据相关联的本体,并且响应于识别本体,确定与所识别的本体相关的解释指令。 代理设备从所识别的设备接收数据,并且基于解释指令来解释所接收的数据。

Patent Agency Ranking