-
公开(公告)号：US11475282B2
公开(公告)日：2022-10-18
申请号：US16603950
申请日：2018-04-17
IPC分类号： G06N3/063 , G06N3/08 , G06N3/04 , G06F13/00 , G06F9/30 , H04L49/00 , G06F9/38 , H04L12/54 , G06F5/06 , G06F13/40 , H04L49/506 , G06F30/392
摘要： Techniques in advanced deep learning provide improvements in one or more of accuracy, performance, and energy efficiency. An array of compute elements and routers performs flow-based computations on wavelets of data. Some instructions are performed in iterations, such as one iteration per element of a fabric vector or FIFO. When sources for an iteration of an instruction are unavailable, and/or there is insufficient space to store results of the iteration, indicators associated with operands of the instruction are checked to determine whether other work can be performed. In some scenarios, other work cannot be performed and processing stalls. Alternatively, information about the instruction is saved, the other work is performed, and sometime after the sources become available and/or sufficient space to store the results becomes available, the iteration is performed using the saved information.
-
公开(公告)号：US11328208B2
公开(公告)日：2022-05-10
申请号：US17272141
申请日：2019-08-27
摘要： Techniques in advanced deep learning provide improvements in one or more of cost, accuracy, performance, and energy efficiency. The deep learning accelerator is implemented at least in part via wafer-scale integration. The wafer comprises a plurality of processor elements, each augmented with redundancy-enabling couplings. The redundancy-enabling couplings enable using redundant ones of the processor elements to replace defective ones of the processor elements. Defect information gathered at wafer test and/or in-situ, such as in a datacenter, is used to determine configuration information for the redundancy-enabling couplings.
-
公开(公告)号：US11062200B2
公开(公告)日：2021-07-13
申请号：US16603647
申请日：2018-04-16
IPC分类号： G06F3/06 , G06N3/063 , G06F9/30 , G06F17/16 , G06F9/455 , G06N3/08 , G06F13/00 , G06N3/04 , G06F9/32 , G06F9/38 , G06F9/48 , G06F9/52
摘要： Techniques in advanced deep learning provide improvements in one or more of accuracy, performance, and energy efficiency. An array of processing elements performs flow-based computations on wavelets of data. Each processing element has a compute element and a routing element. Each compute element has memory. Each router enables communication via wavelets with at least nearest neighbors in a 2D mesh. Routing is controlled by respective virtual channel specifiers in each wavelet and routing configuration information in each router. A compute element conditionally selects for task initiation a previously received wavelet specifying a particular one of the virtual channels. The conditional selecting excludes the previously received wavelet for selection until at least block/unblock state maintained for the particular virtual channel is in an unblock state. The compute element executes block/unblock instructions to modify the block/unblock state.
-
公开(公告)号：US10762418B2
公开(公告)日：2020-09-01
申请号：US16090049
申请日：2018-04-17
IPC分类号： G06N3/063 , G06N3/04 , G06F13/00 , G06N3/08 , G06F9/30 , H04L12/935 , G06F9/38 , H04L12/54 , H04L12/931
摘要： Techniques in advanced deep learning provide improvements in one or more of accuracy, performance, and energy efficiency. An array of processing elements performs flow based computations on wavelets of data. Each processing element has a compute element and a routing element. Each compute element has memory. Each router enables communication via wavelets with nearest neighbors in a 2D mesh. A compute element receives a wavelet. If a control specifier of the wavelet is a first value, then instructions are read from the memory of the compute element in accordance with an index specifier of the wavelet. If the control specifier is a second value, then instructions are read from the memory of the compute element in accordance with a virtual channel specifier of the wavelet. Then the compute element initiates execution of the instructions.
-
-
公开(公告)号：US09706439B2
公开(公告)日：2017-07-11
申请号：US14274704
申请日：2014-05-10
申请人： Firetide, Inc.
IPC分类号： H04W28/08 , H04L12/709 , H04L12/721 , H04L12/801 , H04L12/803 , H04W40/02 , H04W84/18
CPC分类号： H04W28/085 , H04L45/245 , H04L45/26 , H04L47/10 , H04L47/125 , H04L47/14 , H04W28/08 , H04W40/02 , H04W84/18
摘要： In a mesh network composed of multiple-radio nodes, we assign each radio to one of a plurality of channels, and treat a plurality of links between a pair of nodes as one logical link (bonded link). In some embodiments, the routing protocol is adapted to view each bonded link as one link having a combination of at least some of the properties of the constituent physical links. Traffic sent along a path is dynamically load balanced between the interfaces at each intermediate node based on the current utilization of each interface. In at least some embodiments, route discovery packets record the metrics of each component link of the bonded links leaving a node, but only one route discovery packet per pair of nodes is forwarded, reducing the route discovery packet traffic compared to if each route discovery packet were forwarded over each component link between the pair of nodes.
-
公开(公告)号：US09686792B2
公开(公告)日：2017-06-20
申请号：US14471248
申请日：2014-08-28
申请人： Firetide, Inc.
CPC分类号： H04W72/0453 , H04W28/085 , H04W72/0486 , H04W72/082 , H04W72/085 , H04W84/18
摘要： Techniques are described for automatically determining quasi-static per-link channel assignments for each radio in multiple-hop mesh networks having nodes with two or more radios and where only a small number of channels is available for use in the network. The method optimally assigns the channels to the radios of all of the nodes in the network so as to achieve the lowest interference among links and the highest possible bandwidth.
-
公开(公告)号：US09104931B2
公开(公告)日：2015-08-11
申请号：US14249233
申请日：2014-04-09
申请人： SOCKET MOBILE, INC.
CPC分类号： G06K7/1413 , G06K7/0004 , G06K7/10564 , G06K7/10584 , G06K7/10732 , G06K7/10881 , G06K7/10891
摘要： An accuracy-enhanced scanner provides (in response to a first user input) illumination of potential scan targets and scans (in response to a second user input) a selected scan target. The user uses the illumination to aim the scanner at the selected scan target in between providing the first and the second user inputs. The scanner has switches to communicate the user inputs, to specify an operating mode for the scanner, and/or to communicate information codes to a computing device. The scanner has one or more scan engines (such as a barcode reader or an RFID tag reader), and optionally communicates wirelessly with the computing device. A scanning system including the scanner optionally provides feedback to the user based on feedback from a host processor. The scanner is any of a Multi-Mode Ring Scanner (MMRS), a cordless hand scanner, or a Personal Digital Assistant (PDA) with an add-on scanner.
摘要翻译： 准确度增强的扫描器提供（对第一用户输入的响应）照射潜在的扫描目标和扫描（响应于第二用户输入）所选择的扫描目标。 在提供第一和第二用户输入之间，用户使用照明将扫描仪瞄准所选择的扫描目标。 扫描器具有用于传达用户输入，指定扫描仪的操作模式和/或将信息代码传送到计算设备的开关。 扫描器具有一个或多个扫描引擎（诸如条形码读取器或RFID标签读取器），并且可选地与计算设备进行无线通信。 包括扫描仪的扫描系统可以基于来自主机处理器的反馈来选择性地向用户提供反馈。 扫描仪是多模式环形扫描仪（MMRS），无绳手持扫描仪或具有附加扫描仪的个人数字助理（PDA）中的任何一种。
-
公开(公告)号：US08851255B2
公开(公告)日：2014-10-07
申请号：US13566046
申请日：2012-08-03
申请人： Vince Mitchell
发明人： Vince Mitchell
CPC分类号： F16G11/12 , B60P7/083 , Y10T24/2177 , Y10T24/2191
摘要： A ratchet assembly for a load binder includes a pawl bearing assembly biased between a surface of the pawl and a biased movable element whose movement is actuated by the insertion or withdrawal of the lever. In one position of the movable element, the bearing assembly is urged against the pawl to lock it. In the other position of the movable element, the bearing assembly is allowed to reduce its pressure on the pawl.
摘要翻译： 用于装载工具的棘轮组件包括偏置在棘爪的表面之间的棘爪轴承组件和偏置的可移动元件，其运动通过杆的插入或拉出来致动。 在可移动元件的一个位置中，轴承组件被推靠在棘爪上以将其锁定。 在可移动元件的另一位置，允许轴承组件减小其对棘爪的压力。
-
公开(公告)号：US08724468B2
公开(公告)日：2014-05-13
申请号：US13168988
申请日：2011-06-26
IPC分类号： G08C15/00
CPC分类号： H04W28/085 , H04L45/245 , H04L45/26 , H04L47/10 , H04L47/125 , H04L47/14 , H04W28/08 , H04W40/02 , H04W84/18
摘要： In a mesh network composed of multiple-radio nodes, we assign each radio to one of a plurality of channels, and treat a plurality of links between a pair of nodes as one logical link (bonded link). In some embodiments, the routing protocol is adapted to view each bonded link as one link having a combination of at least some of the properties of the constituent physical links. Traffic sent along a path is dynamically load balanced between the interfaces at each intermediate node based on the current utilization of each interface. In at least some embodiments, route discovery packets record the metrics of each component link of the bonded links leaving a node, but only one route discovery packet per pair of nodes is forwarded, reducing the route discovery packet traffic compared to if each route discovery packet were forwarded over each component link between the pair of nodes.
摘要翻译： 在由多个无线电节点组成的网状网络中，我们将每个无线电分配给多个信道中的一个，并且将一对节点之间的多个链路视为一个逻辑链路（绑定链路）。 在一些实施例中，路由协议适于将每个绑定链路视为具有组成物理链路的至少一些属性的组合的一个链路。 根据每个接口的当前利用率，沿着路径发送的流量在每个中间节点的接口之间动态负载平衡。 在至少一些实施例中，路由发现分组记录离开节点的绑定链路的每个分量链路的度量，但是每对节点仅转发一个路由发现分组，与每个路由发现分组相比减少了路由发现分组流量 被转发通过该对节点之间的每个组件链路。
-
-
-
-
-
-
-
-
-
搜索结果
169 条记录 (耗时 0.024 秒)