Circular Reconfiguration for Reconfigurable Parallel Processor

    公开(公告)号:US20180267930A1

    公开(公告)日:2018-09-20

    申请号:US15919709

    申请日:2018-03-13

    申请人: Jianbin Zhu Yuan Li

    发明人: Jianbin Zhu Yuan Li

    IPC分类号: G06F15/80 G06F9/30 G06F9/38

    摘要: Processors, systems and methods are provided for thread level parallel processing. A processor may comprise a plurality of reconfigurable units that may include a plurality of processing elements (PEs) and a plurality of memory ports (MPs) for the plurality of PEs to access a memory unit. Each of the plurality of reconfigurable units may comprise a configuration buffer and a reconfiguration counter. The processor may further comprise a sequencer coupled to the configuration buffer of each of the plurality of reconfigurable units and configured to distribute a plurality of configurations to the plurality of reconfigurable units for the plurality of PEs and the plurality of MPs to execute a sequence of instructions.

    XPath evaluation in an XML repository
    22.
    发明授权
    XPath evaluation in an XML repository 有权
    XML存储库中的XPath评估

    公开(公告)号:US09135367B2

    公开(公告)日:2015-09-15

    申请号:US12890961

    申请日:2010-09-27

    IPC分类号: G06F17/30

    摘要: XPath evaluation in an XML data repository includes parsing an input XPath query using a simple path file to generate an execution tree about the XPath query, where the simple path file includes an XML file that is generated based on the hierarchical architecture of a plurality of XML files in the data repository, and the names of the nodes in the generated XML file are generated by recording the tag information of respective nodes in the plurality of XML files in the data repository. Execution of an execution tree for the data repository generates a final evaluation result.

    摘要翻译: XML数据存储库中的XPath评估包括使用简单的路径文件来解析输入的XPath查询,以生成关于XPath查询的执行树,其中简单路径文件包含基于多个XML的层次结构生成的XML文件 数据存储库中的文件以及生成的XML文件中的节点的名称是通过在数据存储库中记录多个XML文件中的相应节点的标签信息生成的。 执行数据存储库的执行树生成最终的评估结果。

    COMBINED TRANSMISSION OF MULTIPLE-PRIORITY NETWORK TRAFFIC
    23.
    发明申请
    COMBINED TRANSMISSION OF MULTIPLE-PRIORITY NETWORK TRAFFIC 审中-公开
    多重优先网络传输的组合传输

    公开(公告)号:US20150249546A1

    公开(公告)日:2015-09-03

    申请号:US14430195

    申请日:2012-10-29

    IPC分类号: H04L12/403 H04J3/16

    摘要: A method of transmitting data includes storing high-priority data blocks in a high-priority queue, storing low-priority data blocks in a low-priority queue, and generating a first data unit that includes one or more of the high-priority data blocks and one or more of the low-priority data blocks. Generating the first data unit includes arranging the one or more high-priority data blocks and the one or more low-priority data blocks in a sequence in which the one or more high-priority data blocks precede the one or more low-priority data blocks. Generating the first data unit further includes indexing the one or more high-priority data blocks and the one or more low-priority data blocks in accordance with the sequence. The first data unit is transmitted.

    摘要翻译: 一种发送数据的方法包括将高优先级数据块存储在高优先级队列中,将低优先级数据块存储在低优先级队列中,以及生成包括一个或多个高优先级数据块的第一数据单元 以及一个或多个低优先级数据块。 生成第一数据单元包括在一个或多个高优先级数据块在一个或多个低优先级数据块之前的序列中排列一个或多个高优先级数据块和一个或多个低优先级数据块 。 生成第一数据单元还包括根据序列索引一个或多个高优先级数据块和一个或多个低优先级数据块。 发送第一个数据单元。

    Systems and methods for matching visual object components
    24.
    发明授权
    Systems and methods for matching visual object components 有权
    用于匹配视觉对象组件的系统和方法

    公开(公告)号:US08625887B2

    公开(公告)日:2014-01-07

    申请号:US13182076

    申请日:2011-07-13

    申请人: Yuan Li Hartwig Adam

    发明人: Yuan Li Hartwig Adam

    IPC分类号: G06K9/62

    摘要: Systems and methods for modeling the occurrence of common image components (e.g., sub-regions) in order to improve visual object recognition are disclosed. In one example, a query image may be matched to a training image of an object. A matched region within the training image to which the query image matches may be determined and a determination may be made whether the matched region is located within an annotated image component of the training image. When the matched region matches only to the image component, an annotation associated with the component may be identified. In another example, sub-regions within a plurality of training image corpora may be annotated as common image components including associated information (e.g., metadata). Matching sub-regions appearing in many training images of objects may be down-weighted in the matching process to reduce possible false matches to query images including common image components.

    摘要翻译: 公开了用于对共同图像分量(例如,子区域)的发生进行建模以便改进视觉对象识别的系统和方法。 在一个示例中,查询图像可以与对象的训练图像匹配。 可以确定训练图像中与查询图像匹配的匹配区域,并且可以确定匹配区域是否位于训练图像的注释图像分量内。 当匹配区域仅与图像分量匹配时,可以识别与该组件相关联的注释。 在另一示例中,多个训练图像语料库中的子区域可以被注释为包括关联信息(例如,元数据)的公共图像分量。 出现在对象的许多训练图像中的匹配子区域可以在匹配处理中被加权,以减少对包括公共图像分量的查询图像的可能的假匹配。

    Systems and Methods for Matching Visual Object Components
    25.
    发明申请
    Systems and Methods for Matching Visual Object Components 有权
    用于匹配可视对象组件的系统和方法

    公开(公告)号:US20130016899A1

    公开(公告)日:2013-01-17

    申请号:US13182076

    申请日:2011-07-13

    申请人: Yuan Li Hartwig Adam

    发明人: Yuan Li Hartwig Adam

    IPC分类号: G06K9/62

    摘要: Systems and methods for modeling the occurrence of common image components (e.g., sub-regions) in order to improve visual object recognition are disclosed. In one example, a query image may be matched to a training image of an object. A matched region within the training image to which the query image matches may be determined and a determination may be made whether the matched region is located within an annotated image component of the training image. When the matched region matches only to the image component, an annotation associated with the component may be identified. In another example, sub-regions within a plurality of training image corpora may be annotated as common image components including associated information (e.g., metadata). Matching sub-regions appearing in many training images of objects may be down-weighted in the matching process to reduce possible false matches to query images including common image components.

    摘要翻译: 公开了用于对共同图像分量(例如,子区域)的发生进行建模以便改进视觉对象识别的系统和方法。 在一个示例中,查询图像可以与对象的训练图像匹配。 可以确定训练图像中与查询图像匹配的匹配区域,并且可以确定匹配区域是否位于训练图像的注释图像分量内。 当匹配区域仅与图像分量匹配时,可以识别与该组件相关联的注释。 在另一示例中,多个训练图像语料库中的子区域可以被注释为包括关联信息(例如,元数据)的公共图像分量。 出现在对象的许多训练图像中的匹配子区域可以在匹配处理中被加权,以减少对包括公共图像分量的查询图像的可能的假匹配。

    METHODS, SYSTEMS, AND APPARATUS FOR TAIL TERMINATION OF TURBO DECODING
    26.
    发明申请
    METHODS, SYSTEMS, AND APPARATUS FOR TAIL TERMINATION OF TURBO DECODING 有权
    方法,系统和装置TURBO解码的尾部终止

    公开(公告)号:US20130007555A1

    公开(公告)日:2013-01-03

    申请号:US13608905

    申请日:2012-09-10

    IPC分类号: H03M13/05 G06F11/10

    摘要: Methods and apparatus for tail termination are provided that include a decoder that includes a processor configured to a forward state metric and a backward state metric wherein each iteration of an initial state of the backward state metric is fetched from a memory and is pre-computed without feedback from a decoding iteration. Each decoding iteration is substantially identical, and the backward state metric that is pre-computed is used for a subsequent iteration. The decoder may include a turbo decoder or a radix-4 decoder.

    摘要翻译: 提供了用于尾端终止的方法和装置,其包括解码器,该解码器包括被配置为向前状态度量和后向状态量度的处理器,其中,从存储器取出后向状态量度的初始状态的每次迭代,并且在没有 来自解码迭代的反馈。 每个解码迭代基本相同,并且预先计算的向后状态度量用于后续迭代。 解码器可以包括turbo解码器或radix-4解码器。

    RATE MATCHING AND DE-RATE MATCHING ON DIGITAL SIGNAL PROCESSORS
    27.
    发明申请
    RATE MATCHING AND DE-RATE MATCHING ON DIGITAL SIGNAL PROCESSORS 有权
    数字信号处理器的速率匹配和速率匹配

    公开(公告)号:US20130007382A1

    公开(公告)日:2013-01-03

    申请号:US13609034

    申请日:2012-09-10

    IPC分类号: G06F12/00

    CPC分类号: H04L1/0068

    摘要: Provided are devices, systems and methods for rate matching and de-rate matching on digital signal processors. In one embodiment, a device for rate matching and de-rate matching, includes an interface for receiving a plurality of blocks of data and digital signal processor configured to pre-compute permutation parameters common to the plurality of blocks, wherein the plurality of blocks are subject to a set of given puncturing parameters and receive a set of pre-computed puncturing thresholds. For one or more blocks in the plurality of blocks, the DSP computes a block signature from the pre-computed puncturing thresholds; matches the block signature to one of a set of pre-computed zone signatures, derives a zone index corresponding to the one pre-computed zone signature, and applies pre-computed permutation and puncturing transformations corresponding to the zone index to the block.

    摘要翻译: 提供了用于数字信号处理器上的速率匹配和去速率匹配的设备,系统和方法。 在一个实施例中,用于速率匹配和去速率匹配的设备包括用于接收多个数据块的接口和被配置为预先计算多个块共同的置换参数的数字信号处理器,其中多个块是 受制于一组给定的打孔参数并接收一组预先计算的穿刺阈值。 对于多个块中的一个或多个块,DSP从预先计算的穿孔阈值计算块签名; 将块签名与一组预先计算的区域签名中的一个匹配,导出与一个预先计算的区域签名相对应的区域索引,并将对应于区域索引的预先计算的置换和删截变换应用于该块。

    Liquid crystal display having compensation circuit for reducing gate delay
    30.
    发明授权
    Liquid crystal display having compensation circuit for reducing gate delay 有权
    液晶显示器具有减小门延迟的补偿电路

    公开(公告)号:US07999776B2

    公开(公告)日:2011-08-16

    申请号:US11986380

    申请日:2007-11-21

    申请人: Yuan Li Xiao-Jin Qi

    发明人: Yuan Li Xiao-Jin Qi

    IPC分类号: G09G3/36

    CPC分类号: G09G3/3677 G09G2320/0223

    摘要: An exemplary liquid crystal display (400) includes a liquid crystal panel (430), a gate driving circuit (410), a data driving circuit (420), and a compensation circuit (440). The liquid crystal panel includes a plurality of gate lines (401) and a plurality of data lines (402) intersecting with the gate lines. The gate driving circuit is configured for providing a plurality of scanning signals to the gate lines in sequence. The data driving circuit is configured for providing a plurality of gray scale voltages to the data lines. The compensation circuit is configured for compensating the scanning signals. The compensation circuit is charged by alternate of the scanning signals, and discharges each such charge to provide a compensation signal to a gate line corresponding to a next scanning signal.

    摘要翻译: 示例性液晶显示器(400)包括液晶面板(430),栅极驱动电路(410),数据驱动电路(420)和补偿电路(440)。 液晶面板包括与栅极线相交的多条栅极线(401)和多条数据线(402)。 栅极驱动电路被配置为依次向栅极线提供多个扫描信号。 数据驱动电路被配置为向数据线提供多个灰度级电压。 补偿电路被配置为补偿扫描信号。 补偿电路通过扫描信号的交替进行充电,并且对每个这样的电荷进行放电,以向对应于下一个扫描信号的栅极线提供补偿信号。