会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 3. 发明申请
    • Peer-to-Peer Network Address Translator (NAT) Traversal Techniques
    • 对等网络地址转换器(NAT)穿越技术
    • US20100146126A1
    • 2010-06-10
    • US12328214
    • 2008-12-04
    • Qingwei LinJiang LiJian-guang LouYusuo HuFan Li
    • Qingwei LinJiang LiJian-guang LouYusuo HuFan Li
    • G06F15/16
    • H04L67/104H04L29/12509H04L29/12528H04L61/2567H04L61/2575H04L61/2589H04L63/029H04L67/1093
    • Techniques described herein perform network address translator (NAT) traversal using a peer of a peer-to-peer architecture in lieu of a server. A first peer maintains a list of public domain peers that are not hidden behind a NAT or a firewall and have previously been connected with the first peer. The first peer distributes the list to other peers via a gossip-based protocol. When a second peer desires to connect with the first peer, the second peer initiates a connection with the first peer and sends a request to a public domain peer of the list, requesting that the public domain peer instruct the first peer to initiate a connection with the second peer. By leveraging a public domain peer to establish a connection between the first and second peers, these techniques lessen the burden on the server of the architecture. Furthermore, because different peers may make requests to different public domain peers, the techniques allow for better scalability of the architecture.
    • 本文描述的技术使用对等体系结构的对等体代替服务器来执行网络地址转换器(NAT)遍历。 第一个对等体维护一个未被隐藏在NAT或防火墙后面并且之前已经与第一个对等体连接的公共域对等体的列表。 第一个对等体通过基于八卦的协议将列表分发给其他对等体。 当第二对等体希望与第一对等体连接时,第二对等体发起与第一对等体的连接,并向列表的公共域对等体发送请求,请求公共域对等体指示第一对等体发起与 第二个同伴 通过利用公共域对等体建立第一和第二对等体之间的连接,这些技术减轻了体系结构的服务器负担。 此外,由于不同的对等体可以向不同的公共域对等体发出请求,所以这些技术允许架构的更好的可扩展性。
    • 7. 发明授权
    • Lecture capture and broadcast system
    • 讲座和播放系统
    • US08239905B2
    • 2012-08-07
    • US12357627
    • 2009-01-22
    • Fan LiJian-guang LouYusuo HuQingwei LinJiang Li
    • Fan LiJian-guang LouYusuo HuQingwei LinJiang Li
    • H04N7/173
    • H04N21/2187G06Q10/00G09B5/06G09B5/065H04N7/15H04N21/4788
    • An end-to-end lecture broadcast system (LBS) is described. The LBS system typically comprises three modules: a schedule module, a capture module and a transmission module. The schedule module is comprised of a scheduler that is implemented as a software add-on on the lecture host's computing device, a schedule server, and a schedule agent component in the uploading peer to provide for channel based management. The capture module is comprised of a video camera for capturing video content, one or more microphones for capturing audio content, and a VGA signal capture for capturing presentation materials. The transmission module is comprised of an uploading peer and a web server. The uploading peer is responsible for transmitting video/audio content to the Internet/intranet using peer-to-peer (P2P) streaming technology. The web server is responsible for the delivery of captured presentation materials and for ensuring the synchronous play back of video/audio content and presentation materials.
    • 描述了端到端讲座广播系统(LBS)。 LBS系统通常包括三个模块:调度模块,捕获模块和传输模块。 该调度模块包括一个调度器,该调度程序被实现为演讲主机的计算设备上的软件附件,调度服务器和上传对等体中的调度代理组件,以提供基于频道的管理。 捕获模块包括用于捕获视频内容的摄像机,用于捕获音频内容的一个或多个麦克风,以及用于捕获演示材料的VGA信号捕获。 传输模块由上传对等体和Web服务器组成。 上传对等体负责使用点对点(P2P)流技术将视频/音频内容传输到互联网/内部网。 Web服务器负责传送捕获的演示资料,并确保视频/音频内容和演示资料的同步播放。
    • 8. 发明申请
    • Context-Aware User Input Prediction
    • 上下文感知用户输入预测
    • US20120102417A1
    • 2012-04-26
    • US12912395
    • 2010-10-26
    • Fan LiQingwei LinJiang Li
    • Fan LiQingwei LinJiang Li
    • G06F3/01
    • G06N99/005G06K9/72
    • Predicted input data is obtained by initially acquiring a set of operations that are performed on non-editable user interface elements of a user interface prior to a subsequent data input into an editable user interface of the user interface. The set of operations is then classified into a pattern class based on the operations in the set and the subsequent data input that corresponds to the set. Subsequently, a pattern type for the pattern class is generated that describes a relationship between the set of operations and the subsequent data input. Accordingly, when a new set of operations is received, input data is predicted for the new set by applying the pattern type to the new set.
    • 通过在随后的数据输入到用户界面的可编辑用户界面之前,首先获取在用户界面的不可编辑的用户界面元素上执行的一组操作来获得预测的输入数据。 然后,基于集合中的操作和对应于该集合的后续数据输入将该组操作分类为模式类。 随后,生成用于模式类的模式类型,其描述操作集与后续数据输入之间的关系。 因此,当接收到一组新的操作时,通过将模式类型应用于新集合来预测新集合的输入数据。