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
    • 5. 发明申请
    • GENERAL PURPOSE EMBEDDED PROCESSOR
    • 一般用途嵌入式处理器
    • US20130185543A1
    • 2013-07-18
    • US13614011
    • 2012-09-13
    • Steven FrankShigeki Imai
    • Steven FrankShigeki Imai
    • G06F9/30
    • G06F9/3009G06F9/30047G06F9/30072G06F9/3013G06F9/3802G06F9/3814G06F9/383G06F9/3851G06F9/4812G06F9/542G06F2209/543Y02D10/24
    • The invention provides an embedded processor architecture comprising a plurality of virtual processing units that each execute processes or threads (collectively, “threads”). One or more execution units, which are shared by the processing units, execute instructions from the threads. An event delivery mechanism delivers events—such as, by way of non-limiting example, hardware interrupts, software-initiated signaling events (“software events”) and memory events—to respective threads without execution of instructions. Each event can, per aspects of the invention, be processed by the respective thread without execution of instructions outside that thread. The threads need not be constrained to execute on the same respective processing units during the lives of those threads—though, in some embodiments, they can be so constrained. The execution units execute instructions from the threads without needing to know what threads those instructions are from.
    • 本发明提供一种嵌入式处理器架构,其包括多个虚拟处理单元,每个虚拟处理单元执行处理或线程(统称为“线程”)。 由处理单元共享的一个或多个执行单元从线程执行指令。 事件传递机制通过非限制性的例子将事件(例如,硬件中断),软件发起的信令事件(“软件事件”)和存储器事件传送到相应的线程而不执行指令。 每个事件可以在本发明的各个方面由相应的线程处理,而不在该线程之外执行指令。 在这些线程的生命期间,线程不需要被限制为在相同的各个处理单元上执行 - 尽管在一些实施例中,它们可以被如此约束。 执行单元从线程执行指令,而不需要知道这些指令来自哪些线程。
    • 8. 发明授权
    • Dithered quantization using neighborhood mask array to approximate interpolate
    • 使用邻域掩码阵列的抖动量化来近似内插
    • US06809740B1
    • 2004-10-26
    • US09626709
    • 2000-07-26
    • Steven Frank Weed
    • Steven Frank Weed
    • G09G502
    • H04N1/6019G06F17/17
    • Methods and image forming systems for approximating the value of a function given specified values of input data using a sparse lookup table. Individual samples are quantized and rounded up or down to an adjacent lattice point of the lookup table. Rather than performing multiple lookup table accesses, which are required using conventional linear interpolation, the disclosed data processing techniques require as few as one lookup table access per sample. The quantized samples are obtained by truncating one or more least significant bits, designated as masked bits, such that the most significant, or index bits, remain. For each sample, the value of an individual one of the masked bits is examined by comparing it with a corresponding entry in a mask array to determine whether the index bits are to be incremented prior to being used as an index to the lookup table. The mask array defines a dithering process whereby different masked bits are examined for different samples, resulting in dithered quantized index bits for the samples, that when used to access the lookup table, yield an average result over a neighborhood of samples that approximates the result that could be obtained using linear interpolation.
    • 方法和图像形成系统,用于使用稀疏查找表近似给定输入数据的指定值的函数的值。 单个样本被量化并向上或向下舍入到查找表的相邻格点。 所公开的数据处理技术不是执行使用常规线性内插所需的多个查找表访问,而是每个样本需要少至一个查找表访问。 量化的样本是通过截断一个或多个被指定为掩蔽位的最低有效位,从而保留最高有效位或索引位来获得的。 对于每个样本,通过将其与掩码阵列中的相应条目进行比较来检查掩码位中的单个一个值的值,以确定在用作查找表的索引之前索引比特是否递增。 掩模阵列定义抖动过程,由此针对不同样本检查不同的掩蔽位,导致样本的抖动量化索引位,当用于访问查找表时,产生近似结果的结果的平均结果, 可以使用线性插值获得。