中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

106 records (took 0.033 seconds)

    ENCODING AND CONCEALING INFORMATION USING DEEP LEARNING
    31.
    发明申请
    ENCODING AND CONCEALING INFORMATION USING DEEP LEARNING 有权

    公开(公告)号：US20220382901A1

    公开(公告)日：2022-12-01

    申请号：US17886046

    申请日：2022-08-11

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Wei Wang ,  Mikhail Istomin

    IPC: G06F21/62 ,  G06N3/04 ,  G06N3/08 ,  G06T5/50 ,  G06F17/16 ,  G06Q50/00

    Abstract: Aspects of the subject disclosure may include, for example, a method for training a deep learning model that includes encoding a content item; generating a blended image by combining a background image and the encoded content; decoding the blended image to generate decoded content corresponding to the content item; and defining or specifying a loss function related to the deep learning model. The method also includes determining values of training parameters for the deep learning model to minimize the loss function, thereby obtaining a trained deep learning model. The method also includes an information concealing procedure using the trained deep learning model to conceal user content by encoding the user content and blending the encoded user content with a user-selected image; the information concealing procedure is substantially independent of the user-selected image. Other embodiments are disclosed.

    ENCODING AND CONCEALING INFORMATION USING DEEP LEARNING
    32.
    发明申请
    ENCODING AND CONCEALING INFORMATION USING DEEP LEARNING 有权

    公开(公告)号：US20210124837A1

    公开(公告)日：2021-04-29

    申请号：US16662779

    申请日：2019-10-24

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Wei Wang ,  Mikhail Istomin

    IPC: G06F21/62 ,  G06N3/04 ,  G06N3/08 ,  G06T5/50 ,  G06Q50/00 ,  G06F17/16

    Abstract: Aspects of the subject disclosure may include, for example, a method for training a deep learning model that includes encoding a content item; generating a blended image by combining a background image and the encoded content; decoding the blended image to generate decoded content corresponding to the content item; and defining or specifying a loss function related to the deep learning model. The method also includes determining values of training parameters for the deep learning model to minimize the loss function, thereby obtaining a trained deep learning model. The method also includes an information concealing procedure using the trained deep learning model to conceal user content by encoding the user content and blending the encoded user content with a user-selected image; the information concealing procedure is substantially independent of the user-selected image. Other embodiments are disclosed.

    Recovery of subscriber data in mobile communication networks
    35.
    发明授权
    Recovery of subscriber data in mobile communication networks 有权

    公开(公告)号：US10127116B2

    公开(公告)日：2018-11-13

    申请号：US14811362

    申请日：2015-07-28

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Wei Wang ,  Cristina Serban

    IPC: G06F11/00 ,  G06F11/14 ,  H04W8/30 ,  H04W24/02

    Abstract: Methods, apparatus, systems and articles of manufacture (e.g., physical storage media) to recover subscriber data in mobile communication networks are disclosed. Example subscriber data recovery methods disclosed herein include communicating with mobile devices operating in a mobile network to configure the mobile devices to (1) locally store backup subscriber data corresponding to subscriber data maintained at a first network element to manage access to the mobile network and (2) transmit the backup subscriber data to a destination based on configuration data. Disclosed example methods also include queueing respective backup subscriber data received from respective ones of the mobile devices for transmission to at least one of the first network element and/or a second network element to restore the subscriber data maintained at a first network element.

    Devices, Systems, and Methods for Detecting Proximity-Based Mobile Malware Propagation
    38.
    发明申请
    Devices, Systems, and Methods for Detecting Proximity-Based Mobile Malware Propagation 有权
    用于检测基于邻近的移动恶意软件传播的设备，系统和方法

    公开(公告)号：US20160044050A1

    公开(公告)日：2016-02-11

    申请号：US14886421

    申请日：2015-10-19

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Wei Wang ,  Gang Xu ,  Gustavo de los Reyes

    IPC: H04L29/06 ,  G06F21/56

    CPC classification number: H04L63/145 ,  G06F21/00 ,  G06F21/564 ,  G06F2221/034 ,  G06F2221/2111 ,  H04L63/1416

    Abstract: Devices, systems, and methods are disclosed. An agent resides in a mobile communication device. The agent detects Proximity-based Mobile Malware Propagation. The agent injects one or more trigger network connections in the candidate connection list. These connections appear as legitimate networks and devices, but instead trigger connection to an agent server on a service provider's network. By attempting to connect through the trigger network connection, the malware reveals itself. The system helps collect the malware signature within a short period of time after the malware outbreak in local areas, though such attacks typically bypass network based security inspection in the network.

    Abstract translation: 公开了设备，系统和方法。 代理驻留在移动通信设备中。 代理检测基于邻近的移动恶意软件传播。 代理在候选连接列表中注入一个或多个触发网络连接。 这些连接显示为合法的网络和设备，而是触发与服务提供商网络上的代理服务器的连接。 通过尝试通过触发网络连接进行连接，恶意软件显示本身。 该系统在局部恶意软件爆发后，可以在短时间内收集恶意软件签名，但这种攻击通常会绕过网络中的网络安全检查。

    Method and apparatus for providing subscriber identity module-based data encryption and remote management of portable storage devices
    39.
    发明授权
    Method and apparatus for providing subscriber identity module-based data encryption and remote management of portable storage devices 有权
    用于提供基于用户身份模块的数据加密和便携式存储设备的远程管理的方法和装置

    公开(公告)号：US08972728B2

    公开(公告)日：2015-03-03

    申请号：US13652369

    申请日：2012-10-15

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Wei Wang ,  Mikhail Istomin

    IPC: G06F21/00

    CPC classification number: G06F21/44 ,  G06F21/32 ,  G06F21/79 ,  H04L63/0876 ,  H04L63/102 ,  H04W12/06 ,  H04W12/08 ,  H04W52/0261 ,  H04W88/02 ,  Y02D70/00 ,  Y02D70/1224 ,  Y02D70/142 ,  Y02D70/144 ,  Y02D70/162 ,  Y02D70/164 ,  Y02D70/166 ,  Y02D70/23

    Abstract: Portable storage devices and methods for remotely managing such portable storage devices are disclosed. For example, a method receives a request from an endpoint device to send a command to a portable storage device. The method then authenticates the endpoint device that has sent the request. The method then transmits the command wirelessly to the portable storage device. Similarly, a portable storage device includes a processor and a computer-readable medium in communication with the processor, the computer-readable medium to store instructions. The instructions, when executed by the processor, cause the processor to perform operations that include: wirelessly receiving a command related to an access of a memory of the portable storage device, verifying an authenticity of the command and executing the command when the authenticity of the command is verified.

    Abstract translation: 公开了用于远程管理这种便携式存储设备的便携式存储设备和方法。 例如，一种方法从端点设备接收到向便携式存储设备发送命令的请求。 该方法然后验证发送请求的端点设备。 该方法然后将无线命令发送到便携式存储设备。 类似地，便携式存储设备包括处理器和与处理器通信的计算机可读介质，该计算机可读介质存储指令。 所述指令在由所述处理器执行时使得所述处理器执行以下操作：所述操作包括：无线地接收与所述便携式存储设备的存储器的访问相关的命令，验证所述命令的真实性并在执行所述命令的真实性时执行所述命令 命令被验证。

    Multilayered Deception for Intrusion Detection and Prevention
    40.
    发明申请
    Multilayered Deception for Intrusion Detection and Prevention 有权
    入侵检测和预防的多层欺骗

    公开(公告)号：US20140259172A1

    公开(公告)日：2014-09-11

    申请号：US14283248

    申请日：2014-05-21

    Applicant: AT&T Intellectual Property I, L.P.

    Inventor： Wei Wang ,  Qi Shen ,  Andrea Forte ,  Jeffrey Bickford

    IPC: H04L29/06

    CPC classification number: H04L63/14 ,  G06F21/552 ,  G06F21/566 ,  H04L63/1491

    Abstract: Concepts and technologies are disclosed herein for multilayered deception for intrusion detection. According to various embodiments of the concepts and technologies disclosed herein, a multilayer deception system includes honey servers, honey files and folders, honey databases, and/or honey computers. A multilayer deception system controller generates honey activity between the honey entities and exposes a honey profile with contact information associated with a honey user. Contact directed at the honey user and/or activity at any of the honey entities can trigger alarms and/or indicate an attack, and can be analyzed to prevent future attacks.

    Abstract translation: 这里公开了用于入侵检测的多层欺骗的概念和技术。 根据本文公开的概念和技术的各种实施例，多层欺骗系统包括蜂蜜服务器，蜂蜜文件和文件夹，蜂蜜数据库和/或蜂蜜计算机。 多层欺骗系统控制器在蜂蜜实体之间产生蜂蜜活动，并且露出具有与蜂蜜用户相关联的联系信息的蜂蜜简档。 任何蜂蜜实体的蜂蜜用户和/或活动的联系可以触发报警和/或指示攻击，并且可以进行分析以防止将来的攻击。

Patent Agency Ranking