中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

2521 records (took 0.047 seconds)

    Efficient beam search and data communication in millimeter-wave wireless networks
    151.
    发明授权
    Efficient beam search and data communication in millimeter-wave wireless networks 有权

    公开(公告)号：US10931352B2

    公开(公告)日：2021-02-23

    申请号：US16591070

    申请日：2019-10-02

    Applicant: NEC Laboratories America, Inc.

    Inventor： Shahram Shahsavari ,  Mohammad Khojastepour

    IPC: H04B7/06

    Abstract: A method for establishing communication links in a millimeter wave network is presented. The method includes determining an active communication link between first and second devices, the first device transmitting probing packets to the second device, employing a beam search technique to locate narrow beams by triggering the first device to adjust its transmission pattern to cover a fraction of an angular uncertainty region (AUR) at a beginning of a time-slot, and adjusting transmission coefficients of the first device based on a response received from the second device such that if the second device receives a probing packet, the second device sends an acknowledgment packet to the first device and the first device updates the AUR such that the AUR is set to a probed angular interval, and if the second device does not receive the probing packet, the first device updates the AUR to a complementary part of the probed interval.

    Sequence models for audio scene recognition
    152.
    发明授权
    Sequence models for audio scene recognition 有权

    公开(公告)号：US10930301B1

    公开(公告)日：2021-02-23

    申请号：US16997314

    申请日：2020-08-19

    Applicant: NEC Laboratories America, Inc.

    Inventor： Cristian Lumezanu ,  Yuncong Chen ,  Dongjin Song ,  Takehiko Mizuguchi ,  Haifeng Chen ,  Bo Dong

    IPC: G10L25/51 ,  G10L25/24 ,  G10L25/18 ,  G10L25/21

    Abstract: A method is provided. Intermediate audio features are generated from an input acoustic sequence. Using a nearest neighbor search, segments of the input acoustic sequence are classified based on the intermediate audio features to generate a final intermediate feature as a classification for the input acoustic sequence. Each segment corresponds to a respective different acoustic window. The generating step includes learning the intermediate audio features from Multi-Frequency Cepstral Component (MFCC) features extracted from the input acoustic sequence. The generating step includes dividing the same scene into the different acoustic windows having varying MFCC features. The generating step includes feeding the MFCC features of each of the different acoustic windows into respective LSTM units such that a hidden state of each respective LSTM unit is passed through an attention layer to identify feature correlations between hidden states at different time steps corresponding to different ones of the different acoustic windows.

    Recommender system for heterogeneous log pattern editing operation
    153.
    发明授权
    Recommender system for heterogeneous log pattern editing operation 有权

    公开(公告)号：US10929763B2

    公开(公告)日：2021-02-23

    申请号：US15684293

    申请日：2017-08-23

    Applicant: NEC Laboratories America, Inc.

    Inventor： Jianwu Xu ,  Biplob Debnath ,  Bo Zong ,  Hui Zhang ,  Guofei Jiang ,  Hancheng Ge

    IPC: G06N5/04 ,  G06F17/16 ,  G06N7/00 ,  G06F40/00 ,  G06F40/16 ,  G06F40/284

    Abstract: A heterogeneous log pattern editing recommendation system and computer-implemented method are provided. The system has a processor configured to identify, from heterogeneous logs, patterns including variable fields and constant fields. The processor is also configured to extract a category feature, a cardinality feature, and a before-after n-gram feature by tokenizing the variable fields in the identified patterns. The processor is additionally configured to generate target similarity scores between target fields to be potentially edited and other fields from among the variable fields in the heterogeneous logs using pattern editing operations based on the extracted category feature, the extracted cardinality feature, and the extracted before-after n-gram feature. The processor is further configured to recommend, to a user, log pattern edits for at least one of the target fields based on the target similarity scores between the target fields in the heterogeneous logs.

    SELF-SUPERVISED VISUAL ODOMETRY FRAMEWORK USING LONG-TERM MODELING AND INCREMENTAL LEARNING
    154.
    发明申请
    SELF-SUPERVISED VISUAL ODOMETRY FRAMEWORK USING LONG-TERM MODELING AND INCREMENTAL LEARNING 有权

    公开(公告)号：US20210042937A1

    公开(公告)日：2021-02-11

    申请号：US16939604

    申请日：2020-07-27

    Applicant: NEC Laboratories America, Inc.

    Inventor： Pan Ji ,  Quoc-Huy Tran ,  Manmohan Chandraker ,  Yuliang Zou

    IPC: G06T7/246 ,  G06T7/579

    Abstract: A computer-implemented method for implementing a self-supervised visual odometry framework using long-term modeling includes, within a pose network of the self-supervised visual odometry framework including a plurality of pose encoders, a convolution long short-term memory (ConvLSTM) module having a first-layer ConvLSTM and a second-layer ConvLSTM, and a pose prediction layer, performing a first stage of training over a first image sequence using photometric loss, depth smoothness loss and pose cycle consistency loss, and performing a second stage of training to finetune the second-layer ConvLSTM over a second image sequence longer than the first image sequence.

    System and method for detecting security risks in a computer system
    157.
    发明授权
    System and method for detecting security risks in a computer system 有权

    公开(公告)号：US10909242B2

    公开(公告)日：2021-02-02

    申请号：US16169081

    申请日：2018-10-24

    Applicant: NEC Laboratories America, Inc.

    Inventor： Ding Li ,  Xusheng Xiao ,  Zhichun Li ,  Guofei Jiang ,  Peng Gao

    IPC: G06F21/55 ,  G06F9/54 ,  G06F16/33 ,  G06F21/57 ,  G06F16/2455

    Abstract: A system and method are provided for identifying security risks in a computer system. The system includes an event stream generator configured to collect system event data from the computer system. The system further includes a query device configured to receive query requests that specify parameters of a query. Each query request includes at least one anomaly model. The query request and the anomaly model are included in a first syntax in which a system event is expressed as {subject-operation-object}. The system further includes a detection device configured to receive at least one query request from the query device and continuously compare the system event data to the anomaly models of the query requests to detect a system event that poses a security risk. The system also includes a reporting device configured to generate an alert for system events that pose a security risk detected by the detection device.

    WORD-OVERLAP-BASED CLUSTERING CROSS-MODAL RETRIEVAL
    158.
    发明申请
    WORD-OVERLAP-BASED CLUSTERING CROSS-MODAL RETRIEVAL 有权

    公开(公告)号：US20210027019A1

    公开(公告)日：2021-01-28

    申请号：US16918353

    申请日：2020-07-01

    Applicant: NEC Laboratories America, Inc. ,  NEC Corporation

    Inventor： Yuncong Chen ,  Hao Yuan ,  Dongjin Song ,  Cristian Lumezanu ,  Haifeng Chen ,  Takehiko Mizoguchi

    IPC: G06F40/30 ,  G06N3/08 ,  G06K9/62 ,  G06F17/16

    Abstract: A system for cross-modal data retrieval is provided that includes a neural network having a time series encoder and text encoder which are jointly trained using an unsupervised training method which is based on a loss function. The loss function jointly evaluates a similarity of feature vectors of training sets of two different modalities of time series and free-form text comments and a compatibility of the time series and the free-form text comments with a word-overlap-based spectral clustering method configured to compute pseudo labels for the unsupervised training method. The computer processing system further includes a database for storing the training sets with feature vectors extracted from encodings of the training sets. The encodings are obtained by encoding a training set of the time series using the time series encoder and encoding a training set of the free-form text comments using the text encoder.

    AERIAL FIBER OPTIC CABLE LOCALIZATION BY DISTRIBUTED ACOUSTIC SENSING
    160.
    发明申请
    AERIAL FIBER OPTIC CABLE LOCALIZATION BY DISTRIBUTED ACOUSTIC SENSING 审中-公开

    公开(公告)号：US20200319017A1

    公开(公告)日：2020-10-08

    申请号：US16838105

    申请日：2020-04-02

    Applicant: NEC Laboratories America, Inc.

    Inventor： Yue TIAN

    IPC: G01H9/00 ,  G02B6/44

    Abstract: Aspects of the present disclosure describe aerial fiber optical cable localization using distributed acoustic sensing (DAS) that advantageously may determine the locality of electrical transformers affixed to utility poles along with the aerial fiber optical cable as well as any length(s) of fiber optical cable between the poles. Further aspects employ survey manned or unmanned, aerial or terrestrial survey vehicles that acoustically excite locations along the fiber optical cable and associate those DAS excitations with global positioning location (GPS).

Patent Agency Ranking