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
    • 23. 发明公开
    • Error control on-demand
    • 按需差错控制
    • EP2194667A1
    • 2010-06-09
    • EP08291139.7
    • 2008-12-03
    • Alcatel Lucent
    • Laevens, KoenraadDe Vleeschauwer, DannyDegrande, Natalie
    • H04L1/00H04L1/12
    • H04L1/0057H04L1/12
    • The present invention relates to a new error control paradigm in a packet switched network that is particularly advantageous for multicast transmission.
      A transmission Status vector TSV (T) is updated at the sender side whenever a new outgoing data packet (Dj-1) is scheduled for transmission. A corresponding Reception Status Vector RSV (R) is similarly updated at the receive side whenever a new incoming and correctly indexed data packet (Dj-1) is validly received.
      As soon as a missing or corrupted data packet (Dj) is detected, the update of the RSV is suspended. A request is then sent to the sender to get the current TSV's value, or a part thereof. The erroneous data packet is then recovered from the current TSV'S value (sk,m), from the lastly updated RSV'S value (sj-1), and from otherwise validly received data packets (Dj+1, ..., Dk).
      The present invention more specifically relates to a transmitting unit (100) and a receiving unit (200) implementing that paradigm, and correspondingly to a method for protecting a flow of indexed data packets against data lost or data corruption, and a method for recovering a missing or corrupted data packet within a flow of indexed data packets.
    • 27. 发明公开
    • A NETWORKED APPLICATION COMPONENT WITH CREDIT BASED CONGESTION CONTROL, AND CORRESPONDING METHOD
    • 一种基于信誉的拥塞控制的网络应用组件及其响应方法
    • EP3306878A1
    • 2018-04-11
    • EP16290193.8
    • 2016-10-05
    • ALCATEL LUCENT
    • De Vleeschauwer, DannyVan Leekwijck, Werner Adriaan Josephine
    • H04L12/819H04L12/801
    • H04L47/39H04L47/193H04L47/2475
    • A computer-implemented networked application component (101; 102) comprises:
      - a need parameter manager (111; 121) adapted to regularly determine an instantaneous need (N(t)) of an application for throughput in said network;
      - a credit parameter manager (112; 122) adapted to regularly update a throughput credit (C(t)) built up as a function of throughput difference, i.e. difference between actual data throughput and fair share; and
      - a transport layer application programming interface (113; 123) adapted to alter one or more congestion control parameters (α(t), β(t)) in the transport layer (103, 104) of the network depending on the instantaneous need (N(t)) to temporarily receive a throughput above, respectively below, fair share in return for consuming, respectively building up, the throughput credit (C(t)).
    • 一种计算机实现的网络应用组件(101; 102),包括: - 需求参数管理器(111; 121),适用于定期确定应用在所述网络中的吞吐量的即时需求(N(t)); - 信用参数管理器(112; 122),适于定期更新作为吞吐量差异(即实际数据吞吐量与公平份额之间的差异)的函数而建立的吞吐量积分(C(t)); 以及 - 传输层应用程序编程接口(113; 123),其适于根据瞬时需求改变网络的传输层(103,104)中的一个或多个拥塞控制参数(α(t),β(t)) (N(t))临时接收吞吐量(C(t))以上的吞吐量,分别低于公平分享以消耗或分别构建吞吐量信用(C(t))。
    • 30. 发明公开
    • Recommendation method and apparatus for selecting content items for users
    • Empfehlungsverfahren und Vorrichtung zumAuswählenvon InhaltselementenfürBenutzer
    • EP2595066A1
    • 2013-05-22
    • EP11290532.8
    • 2011-11-21
    • Alcatel Lucent
    • Steck, HaraldDe Vriendt, Johan A.De Vleeschauwer, DannyVerhoeyen, Marc
    • G06F17/30
    • G06F17/30017G06F17/30244G06F17/3069G06F17/30749
    • A recommended method for selecting content items to be recommended to a user or users includes obtaining a data set of content items in which items are included irrespective of whether or not they have associated user feedback indicating relevance to users. User feedback associated with content items in the data set is used to estimate (a) a user-feature matrix Q of q_{u, k}, with K_u values of k (k = 1, ..., K_u) per user u, representing user profiles and acquired over a first time period prior to a time at which content items are to be recommended to the user or users; and (b) a content-feature matrix P of p_{i, l}, with L_i values of 1 (1=1, ..., L_i) per item i, representing content profiles and acquired over a second time period prior to the time at which content items arc to be recommended to the user or users. The user-feature matrix Q and the content-feature matrix P are used to build a matrix R of estimated rating values r_{i, u} where r_{i, u} = f (q_{u, k}, p_ {i, l}). Higher ranked content items of the matrix R are selected to be recommended to the user or users.
      The selected recommendations may be used to determine which content items are to be pre-loaded and /or cached for later access by the user or users.
    • 用于选择要向用户或用户推荐的内容项目的推荐方法包括获得其中包括项目的内容项目的数据集,而不管它们是否具有指示与用户的相关性的相关联的用户反馈。 使用与数据集中的内容项相关联的用户反馈来估计(a)每个用户u的k_u值为k(k = 1,...,K_u)的q_ {u,k}的用户特征矩阵Q ,表示用户简档并且在将内容项目推荐给用户或用户的时间之前的第一时间段中获取; 和(b)p_ {i,l}的内容特征矩阵P,每个项目i的L_i值为1(1 = 1,...,L_i),表示内容轮廓,并且在第 将内容项目推荐给用户或用户的时间。 用户特征矩阵Q和内容特征矩阵P用于构建估计的等级值r_ {i,u}的矩阵R,其中r_ {i,u} = f(q_ {u,k},p_ {i ,l})。 选择矩阵R的较高排名的内容项目被推荐给用户或用户。 所选择的建议可以用于确定哪些内容项目将被预加载和/或高速缓存以供用户或用户稍后访问。