Performing lookup table operations on a single-instruction multiple data processor

    公开(公告)号:US10261939B2

    公开(公告)日:2019-04-16

    申请号:US14464134

    申请日:2014-08-20

    摘要: In accordance with at least one embodiment, a processor system is disclosed having a SIMD processor device that has a plurality of subsidiary processing elements that are controlled to process multiple data concurrently. In accordance with at least one embodiment, the SIMD processor is a vector processor (VPU) having a plurality of vector Arithmetic Units (AUs) as subsidiary processing elements, and the VPU executes an instruction to transfer table information from a global memory of the VPU to a plurality of local memories accessible by each AU. The VPU also executes an instruction that results in each processing element performing a table lookup from a table stored at its local memory. In response to the instruction, this table lookup uses a portion of a lookup value to access information from the table, and uses another portion of the lookup information to calculate an interpolated resultant based upon the accessed information.

    Reference signaling scheme using compressed feedforward codebooks for multi-user multiple input multiple output (MU-MIMO) systems
    2.
    发明授权
    Reference signaling scheme using compressed feedforward codebooks for multi-user multiple input multiple output (MU-MIMO) systems 有权
    使用多用户多输入多输出(MU-MIMO)系统的压缩前馈码本的参考信令方案

    公开(公告)号:US08509339B2

    公开(公告)日:2013-08-13

    申请号:US13100738

    申请日:2011-05-04

    IPC分类号: H04B7/02

    摘要: A multi-user multiple input multiple output (MIMO) downlink beamforming system with limited feed forward (200) is provided to enable precoding matrix information to be efficiently provided to a subset of user equipment devices (201.1), where zero-forcing transmit beamformers (wi) are computed at the base station (210) and assembled into a precoding matrix (W). The precoding matrix is encoded using a compact reference signal codebook (225, 207.1) for forward link signaling, either by sending bits indicating the index of the transmission matrix used, or by transmitting one or more precoded pilots or reference signals wherein the pilot signals are precoded using vectors uniquely representative of the transmission matrix used which includes candidate reference signal matrices which meet a predetermined condition number requirement, such as a condition number threshold.

    摘要翻译: 提供了具有有限的前馈(200)的多用户多输入多输出(MIMO)下行链路波束成形系统,以使得能够将预编码矩阵信息有效地提供给用户设备设备(201.1)的子集,其中迫零发射波束形成器 wi)在基站(210)处计算并组装成预编码矩阵(W)。 通过发送指示所使用的传输矩阵的索引的比特,或者通过发送一个或多个预编码的导频或参考信号,使用用于前向链路信令的紧凑参考信号码本(225,207.1)对预编码矩阵进行编码, 使用唯一地代表所使用的传输矩阵的向量进行预编码,所述传输矩阵包括符合预定条件号要求的候选参考信号矩阵,例如条件数阈值。

    Fourier transform processing and twiddle factor generation
    3.
    发明授权
    Fourier transform processing and twiddle factor generation 有权
    傅里叶变换处理和旋转因子生成

    公开(公告)号:US08386552B2

    公开(公告)日:2013-02-26

    申请号:US12211892

    申请日:2008-09-17

    IPC分类号: G06F15/00

    CPC分类号: G06F17/142

    摘要: In a data processing system, having a twiddle factor unit, a method for performing a mixed-radix discrete Fourier transform (DFT) having a block size, N, and a maximum block size, Nmax, wherein the maximum block size includes a radix that is not a power of 2 is provided. The method includes receiving a delta value at an input of the twiddle factor unit, the delta value representing a ratio of a modified maximum bock size to the block size, wherein the modified maximum block size is a power of 2. The method further includes using the delta value to obtain a step size for generating indices of a look-up table stored within the twiddle factor unit, wherein the look-up table stores real and imaginary components of twiddle factors corresponding to a set of block sizes of the DFT.

    摘要翻译: 在具有旋转因子单元的数据处理系统中,执行具有块大小N和最大块大小Nmax的混合离散傅里叶变换(DFT)的方法,其中最大块大小包括基数,其中, 不是提供2的力量。 该方法包括在旋转因子单元的输入处接收增量值,增量值表示修改的最大块大小与块大小的比率,其中修改的最大块大小是2的幂。该方法还包括使用 所述增量值用于获得用于产生存储在所述旋转因子单元内的查找表的索引的步长的步长,其中所述查找表存储与所述DFT的块大小的集合相对应的旋转因子的实部和虚部。

    TECHNIQUES FOR PERFORMING DISCRETE FOURIER TRANSFORMS ON RADIX-2 PLATFORMS
    4.
    发明申请
    TECHNIQUES FOR PERFORMING DISCRETE FOURIER TRANSFORMS ON RADIX-2 PLATFORMS 有权
    在RADIX-2平台上执行离散傅立叶变换的技术

    公开(公告)号:US20090313314A1

    公开(公告)日:2009-12-17

    申请号:US12140890

    申请日:2008-06-17

    IPC分类号: G06F17/14

    CPC分类号: G06F17/142

    摘要: A technique for performing a discrete Fourier transform (DFT) includes storing, in a single-port memory, multiple signal points. A first group of consecutive ones of the multiple signal points are fetched (from a first line of the single-port memory) to a first input register associated with a processor that includes multiple arithmetic units (AUs) that are each configured to perform multiply accumulate (MAC) operations. A second group of consecutive ones of the multiple signal points are then fetched (from a second line of the single-port memory) to a second input register associated with the processor. Selected pairs of the multiple signal points are then loaded (one from each of the first and second input registers for each pair) into the multiple arithmetic units during an initial butterfly stage. Radix-2 butterfly operations are then performed on the selected pairs of the multiple signal points (using the multiple AUs) to provide respective output elements.

    摘要翻译: 用于执行离散傅里叶变换(DFT)的技术包括在单端口存储器中存储多个信号点。 将多个信号点中的连续的多个信号点中的第一组(从单端口存储器的第一行)提取到与包括多个运算单元(AU)的处理器相关联的第一输入寄存器,每个运算单元被配置为执行乘法累积 (MAC)操作。 然后将多个信号点中的第二组连续的信号点从单端口存储器的第二行提取到与处理器相关联的第二输入寄存器。 然后,在初始蝴蝶阶段期间,将多个信号点的所选择的对加载(每对中的每个第一和第二输入寄存器中的一个)分配到多个运算单元中。 然后对所选择的多个信号点对(使用多个AU)执行基2蝶形运算,以提供相应的输出元件。

    SINGLE-INSTRUCTION MULTIPLE DATA PROCESSOR
    5.
    发明申请
    SINGLE-INSTRUCTION MULTIPLE DATA PROCESSOR 审中-公开
    单指令多数据处理器

    公开(公告)号:US20160054995A1

    公开(公告)日:2016-02-25

    申请号:US14464134

    申请日:2014-08-20

    IPC分类号: G06F9/30 G06F17/10

    摘要: In accordance with at least one embodiment, a processor system is disclosed having a SIMD processor device that has a plurality of subsidiary processing elements that are controlled to process multiple data concurrently. In accordance with at least one embodiment, the SIMD processor is a vector processor (VPU) having a plurality of vector Arithmetic Units (AUs) as subsidiary processing elements, and the VPU executes an instruction to transfer table information from a global memory of the VPU to a plurality of local memories accessible by each AU. The VPU also executes an instruction that results in each processing element performing a table lookup from a table stored at its local memory. In response to the instruction, this table lookup uses a portion of a lookup value to access information from the table, and uses another portion of the lookup information to calculate an interpolated resultant based upon the accessed information.

    摘要翻译: 根据至少一个实施例,公开了具有SIMD处理器设备的处理器系统,该处理器设备具有被控制以同时处理多个数据的多个辅助处理元件。 根据至少一个实施例,SIMD处理器是具有多个向量算术单元(AU)作为辅助处理单元的向量处理器(VPU),并且VPU执行从VPU的全局存储器传送表信息的指令 到由每个AU访问的多个本地存储器。 VPU还执行一个指令,导致每个处理元素从存储在其本地存储器中的表执行表查找。 响应于该指令,该表查找使用查找值的一部分来访问表中的信息,并且使用查找信息的另一部分基于所访问的信息来计算插值结果。

    Generalized reference signaling scheme for multi-user, multiple input, multiple output (MU-MIMO) using arbitrarily precoded reference signals
    7.
    发明授权
    Generalized reference signaling scheme for multi-user, multiple input, multiple output (MU-MIMO) using arbitrarily precoded reference signals 有权
    广泛参考信令方案,用于多用户,多输入,多输出(MU-MIMO)使用任意预编码的参考信号

    公开(公告)号:US08199846B2

    公开(公告)日:2012-06-12

    申请号:US12897417

    申请日:2010-10-04

    IPC分类号: H04B7/02

    摘要: A multi-user multiple-input multiple-output (MU-MIMO) downlink beamforming system (200) is provided to enable transmit beamforming vectors to be efficiently provided to a subset of user equipment devices (201.i), where spatial separation or zero-forcing transmit beamformers (wi) are computed at the base station (210) and used to generate precoded reference signals (216). The precoded reference signals (216) are fed forward to the user equipment devices (201.i) which apply one or more hypothesis tests (207.i, 208.i) to the precoded reference signals to extract the precoding matrix (W), including the specific transmit beamforming vector (WUE) designed for the user equipment, and this extracted information is used to generate receive beamformers (vi).

    摘要翻译: 提供多用户多输入多输出(MU-MIMO)下行链路波束形成系统(200)以使发送波束成形向量能够有效地提供给用户设备设备(201.i)的子集,其中空间分离或零 在基站(210)处计算强制发射波束形成器(wi)并用于产生预编码的参考信号(216)。 预编码的参考信号(216)被馈送到用户设备设备(201.i),其将一个或多个假设测试(207.i,208.i)应用于预编码的参考信号以提取预编码矩阵(W) 包括为用户设备设计的特定发射波束形成向量(WUE),并且该提取的信息用于生成接收波束形成器(vi)。

    Techniques for increasing decoding reliability in an adaptive minimum mean squared error with successive interference cancellation (MMSE/SIC) decoder
    8.
    发明授权
    Techniques for increasing decoding reliability in an adaptive minimum mean squared error with successive interference cancellation (MMSE/SIC) decoder 有权
    用于增加具有连续干扰消除(MMSE / SIC)解码器的自适应最小均方误差中的解码可靠性的技术

    公开(公告)号:US08184750B2

    公开(公告)日:2012-05-22

    申请号:US12358039

    申请日:2009-01-22

    IPC分类号: H04L27/06 H04L5/12 H03M13/03

    摘要: A technique for increasing decoding reliability in an adaptive minimum mean squared error with successive interference cancellation (MMSE/SIC) decoder in a channel-coded multiple-input multiple-output (MIMO) communication system. A code block selector evaluates reliability metrics and determines whether the reliability metric of the decoded symbols of a particular indexed code block of a first code word satisfies a quality threshold. Depending upon this determination, a composite second code word is formed at each indexed code block using a previously calculated MMSE-LLR output of a second code word or a SIC-LLR output using the indexed symbols of a first code word. Composite second code word is decoded with increased accuracy.

    摘要翻译: 一种用于在通道编码的多输入多输出(MIMO)通信系统中的连续干扰消除(MMSE / SIC)解码器的自适应最小均方误差中增加解码可靠性的技术。 代码块选择器评估可靠性度量,并确定第一代码字的特定索引代码块的解码符号的可靠性度量是否满足质量阈值。 根据该确定,使用先前计算的第二码字的MMSE-LLR输出或使用第一码字的索引符号的SIC-LLR输出,在每个索引码块处形成复合第二码字。 复合第二码字以更高的精度进行解码。

    Reference Signaling Scheme Using Compressed Feedforward Codebooks for MU-MIMO Systems
    9.
    发明申请
    Reference Signaling Scheme Using Compressed Feedforward Codebooks for MU-MIMO Systems 有权
    使用压缩前馈码本的MU-MIMO系统的参考信令方案

    公开(公告)号:US20120114064A1

    公开(公告)日:2012-05-10

    申请号:US13100738

    申请日:2011-05-04

    IPC分类号: H04L27/00

    摘要: A multi-user MIMO downlink beamforming system with limited feed forward (200) is provided to enable precoding matrix information to be efficiently provided to a subset of user equipment devices (201.i), where zero-forcing transmit beamformers (wi) are computed at the base station (210) and assembled into a precoding matrix (W). The precoding matrix is encoded using a compact reference signal codebook (225, 207.i) for forward link signaling, either by sending bits indicating the index of the transmission matrix used, or by transmitting one or more precoded pilots or reference signals wherein the pilot signals are precoded using vectors uniquely representative of the transmission matrix used which includes candidate reference signal matrices which meet a predetermined condition number requirement, such as a condition number threshold. The precoding matrix information (227) is extracted at the user equipment devices (201.i) using the compact reference signal codebook (207.i) and used by the MMSE receiver (209.i) to generate receive beamformers (vi).

    摘要翻译: 提供了具有有限的前馈(200)的多用户MIMO下行链路波束成形系统,以使预编码矩阵信息能够有效地提供给用户设备设备(201.i)的子集,其中计算迫零发射波束形成器(wi) 在基站(210)处并组装成预编码矩阵(W)。 预编码矩阵使用用于前向链路信令的紧凑参考信号码本(225,207.i)进行编码,通过发送指示所使用的传输矩阵的索引的比特,或者通过发送一个或多个预编码的导频或参考信号,其中导频 使用唯一地代表所使用的传输矩阵的矢量来对信号进行预编码,该传输矩阵包括符合预定条件数量要求的候选参考信号矩阵,例如条件数阈值。 使用紧凑参考信号码本(207.i)在用户设备设备(201.i)处提取预编码矩阵信息(227),并由MMSE接收机(209.i)使用以生成接收波束形成器(vi)。

    Generalized Reference Signaling Scheme for MU-MIMO Using Arbitrarily Precoded Reference Signals
    10.
    发明申请
    Generalized Reference Signaling Scheme for MU-MIMO Using Arbitrarily Precoded Reference Signals 有权
    使用任意预编码参考信号的MU-MIMO的广义参考信令方案

    公开(公告)号:US20110019631A1

    公开(公告)日:2011-01-27

    申请号:US12897417

    申请日:2010-10-04

    IPC分类号: H04W4/00

    摘要: A multi-user MIMO downlink beamforming system (200) is provided to enable transmit beamforming vectors to be efficiently provided to a subset of user equipment devices (201.i), where spatial separation or zero-forcing transmit beamformers (wi) are computed at the base station (210) and used to generate precoded reference signals (216). The precoded reference signals (216) are fed forward to the user equipment devices (201.i) which apply one or more hypothesis tests (207.i, 208.i) to the precoded reference signals to extract the precoding matrix (W), including the specific transmit beamforming vector (wUE) designed for the user equipment, and this extracted information is used to generate receive beamformers (vi).

    摘要翻译: 多用户MIMO下行链路波束形成系统(200)被提供以使发射波束成形向量能够有效地提供给用户设备设备(201.i)的子集,其中空间分离或迫零发射波束形成器(wi) 基站(210)并用于产生预编码的参考信号(216)。 预编码的参考信号(216)被馈送到用户设备设备(201.i),其将一个或多个假设测试(207.i,208.i)应用于预编码的参考信号以提取预编码矩阵(W) 包括为用户设备设计的特定发射波束形成向量(wUE),并且该提取的信息用于生成接收波束形成器(vi)。