-
公开(公告)号：US07433820B2
公开(公告)日：2008-10-07
申请号：US10844093
申请日：2004-05-12
IPC分类号： G10L15/14
CPC分类号： G10L15/144 , G06K9/6297
摘要： A system, method and program storage device implementing a method for modeling a data generating process, wherein the modeling comprises observing a data sequence comprising irregularly sampled data, obtaining an observation sequence based on the observed data sequence, assigning a time index sequence to the data sequence, obtaining a hidden state sequence of the data sequence, and decoding the data sequence based on a combination of the time index sequence and the hidden state sequence to model the data sequence. The method further comprises assigning a probability distribution over time stamp values of the observation sequence, wherein the decoding comprises using a Hidden Markov Model. The method further comprises using an expectation maximization methodology to learn the Hidden Markov Model.
摘要翻译： 一种实现数据生成过程建模方法的系统，方法和程序存储设备，其中所述建模包括观察包含不规则采样数据的数据序列，基于观察到的数据序列获得观测序列，向数据分配时间索引序列 获取数据序列的隐藏状态序列，并且基于时间索引序列和隐藏状态序列的组合对数据序列进行解码以建模数据序列。 该方法还包括分配观测序列的时间戳值的概率分布，其中解码包括使用隐马尔科夫模型。 该方法还包括使用期望最大化方法来学习隐马尔可夫模型。
-
公开(公告)号：US07430717B1
公开(公告)日：2008-09-30
申请号：US09669680
申请日：2000-09-26
IPC分类号： G06N3/00
CPC分类号： G06F17/3071 , G06K9/6223
摘要： A method and structure for clustering documents in datasets which include clustering first documents and a first dataset to produce first document classes, creating centroid seeds based on the first document classes, and clustering second documents in a second dataset using the centroid seeds, wherein the first dataset and the second dataset are related. The clustering of the first documents in the first dataset forms a first dictionary of most common words in the first dataset and generates a first vector space model by counting, for each word in the first dictionary, a number of the first documents in which the word occurs, and clusters the first documents in the first dataset based on the first vector space model, and further generates a second vector space model by counting, for each word in the first dictionary, a number of the second documents in which the word occurs. Creation of the centroid seeds includes classifying second vector space model using the first document classes to produce a classified second vector space model and determining a mean of vectors in each class in the classified second vector space model, the mean includes the centroid seeds.
摘要翻译： 一种用于在数据集中聚类文档的方法和结构，所述方法和结构包括聚集第一文档和第一数据集以产生第一文档类，基于第一文档类创建质心种子，以及使用质心种子将第二文档聚类在第二数据集中，其中， 数据集和第二个数据集相关。 第一数据集中的第一文档的聚类形成第一数据集中最常用词的第一字典，并通过对第一字典中的每个单词进行计数，生成第一个向量空间模型，其中该单词的第一个文档的数目 发生，并且基于第一向量空间模型对第一数据集中的第一文档进行聚类，并且还通过对第一字典中的每个单词计数出现该单词的第二文档的数量来生成第二向量空间模型。 重心种子的创建包括使用第一文档类分类第二矢量空间模型以产生分类的第二矢量空间模型并且确定分类的第二矢量空间模型中每个类中的矢量的平均值，平均值包括质心种子。
-
63.发明授权Method of generating wiring routes with matching delay in the presence of process variation 有权在存在过程变化的情况下生成具有匹配延迟的布线路线的方法公开(公告)号：US07418689B2
公开(公告)日：2008-08-26
申请号：US10908102
申请日：2005-04-27
IPC分类号： G06F17/50
CPC分类号： G06F17/5077
摘要： A method and service of balancing delay in a circuit design begins with nodes that are to be connected together by a wiring design, or by being supplied with an initial wiring design that is to be altered. The wiring design will have many wiring paths, such as a first wiring path, a second wiring path, etc. Two or more of the wiring paths are designed to have matching timing, such that the time needed for a signal to travel along the first wiring path is about the same time needed for a signal to travel along the second wiring path, the third path, etc. The method/service designs one or all of the wiring paths to make the paths traverse wire segments of about the same length and orientation, within each wiring level that the first wiring path and the second wiring path traverse. Also, this process makes the first wiring path and the second wiring path traverse the wire segments in the same order, within each wiring level that the first wiring path and the second wiring path traverse.
摘要翻译： 电路设计中的平衡延迟的方法和服务从通过布线设计连接在一起的节点开始，或通过提供要被改变的初始布线设计。 布线设计将具有许多布线路径，例如第一布线路径，第二布线路径等。两条或多条布线路径被设计成具有匹配的定时，使得信号沿着第一布线路径行进所需的时间 信号沿着第二布线路径，第三路径等移动所需的大致相同的时间。该方法/服务设计一个或所有布线路径，以使路径穿过大约相同长度的线段，并且 在第一布线路径和第二布线路径横越的各布线层内。 此外，该处理使得第一布线路径和第二布线路径在第一布线路径和第二布线路径横越的各布线层内以相同的顺序横穿线段。
-
64.发明授权公开(公告)号：US07415695B2
公开(公告)日：2008-08-19
申请号：US11748575
申请日：2007-05-15
申请人： Bette L. Bergman Reuter , David L. DeMaris , Mark A. Lavin , William C. Leipold , Daniel N. Maynard , Maharaj Mukherjee
发明人： Bette L. Bergman Reuter , David L. DeMaris , Mark A. Lavin , William C. Leipold , Daniel N. Maynard , Maharaj Mukherjee
IPC分类号： G06F17/50
CPC分类号： G06T7/001 , G06T2207/30148
摘要： Disclosed is a method of locating systematic defects in integrated circuits. The invention first performs a preliminary extracting and index processing of the circuit design and then performs feature searching. When performing the preliminary extracting and index processing the invention establishes a window grid for the circuit design and merges basis patterns with shapes in the circuit design within each window of the window grid. The invention transforms shapes in a each window into feature vectors by finding intersections between the basis patterns and the shapes in the windows. Then, the invention clusters the feature vectors to produce an index of feature vectors. After performing the extracting and index processing, the invention performs the process of feature searching by first identifying a defect region window of the circuit layout and similarly merging basis patterns with shapes in the defect region window. This merging process can include rotating and mirroring the shapes in the defect region. The invention similarly transforms shapes in the defect region window into defect vectors by finding intersections between basis patterns and the shapes in the defect region. Then, the invention can easily find feature vectors that are similar to the defect vector using, for example, representative feature vectors from the index of feature vectors. Then, the similarities and differences between the defect vectors and the feature vectors can be analyzed.
摘要翻译： 公开了一种定位集成电路系统缺陷的方法。 本发明首先进行电路设计的初步提取和索引处理，然后执行特征搜索。 当执行初步提取和索引处理时，本发明建立了用于电路设计的窗口网格，并且将窗体网格的每个窗口内的电路设计中的形状与基本图案合并。 本发明通过在窗口中找到基本图案和形状之间的交点来将每个窗口中的形状转换为特征向量。 然后，本发明聚集特征向量以产生特征向量的索引。 在执行提取和索引处理之后，本发明通过首先识别电路布局的缺陷区域窗口并且将基本模式与缺陷区域窗口中的形状类似地合并来执行特征搜索的处理。 该合并过程可以包括旋转和镜像缺陷区域中的形状。 本发明类似地通过在缺陷区域中找到基础图案和形状之间的交点来将缺陷区域窗口中的形状转换为缺陷向量。 然后，本发明可以使用例如来自特征向量的索引的代表性特征向量容易地找到与缺陷向量相似的特征向量。 然后，可以分析缺陷向量和特征向量之间的相似性和差异。
-
65.发明授权公开(公告)号：US07395187B2
公开(公告)日：2008-07-01
申请号：US11348010
申请日：2006-02-06
IPC分类号： G06F11/30
CPC分类号： H04L43/16 , G06F11/3409 , G06F11/3452 , G06F11/3495 , G06F2201/81 , H04L41/046 , H04L41/5003 , H04L43/0817 , H04L43/0829 , H04L43/0852 , H04L43/0888 , H04L43/0894 , H04L67/1097
摘要： Disclosed is an autonomic abnormality detection device having a plurality of agents, a server with a one or more processors, a data storage device and a corrective actions engine. The device is adapted to detect and diagnose abnormalities in system components. Particularly, the device uses agents to track performance/workload measurements of system components and dynamically compiles a history of those performance/workload measurements for each component. In order to detect abnormalities a processor compares current performance/workload measurements for a component to the compiled histories for that component and for other components. The processor can further be adapted to determine possible causes of a detected abnormality and to report the abnormality, including the possible causes, to a corrective actions engine.
摘要翻译： 公开了具有多个代理的自主异常检测装置，具有一个或多个处理器的服务器，数据存储装置和纠正动作引擎。 该设备适用于检测和诊断系统组件中的异常。 特别地，该设备使用代理跟踪系统组件的性能/工作负载测量，并动态编译每个组件的性能/工作负载测量的历史记录。 为了检测异常，处理器将组件的当前性能/工作负载测量结果与该组件和其他组件的已编译历史进行比较。 处理器还可以适于确定检测到的异常的可能原因，并将包括可能的原因的异常报告给校正动作引擎。
-
66.发明授权Metal oxide field effect transistor with a sharp halo and a method of forming the transistor 有权具有尖锐光晕的金属氧化物场效应晶体管和形成晶体管的方法公开(公告)号：US07384835B2
公开(公告)日：2008-06-10
申请号：US11420318
申请日：2006-05-25
申请人： Huajie Chen , Judson R Holt , Rangarajan Jagannathan , Wesley C Natzle , Michael R Sievers , Richard S Wise
发明人： Huajie Chen , Judson R Holt , Rangarajan Jagannathan , Wesley C Natzle , Michael R Sievers , Richard S Wise
IPC分类号： H01L21/335
CPC分类号： H01L29/66636 , H01L29/6659 , H01L29/7833
摘要： Disclosed are embodiments of a MOSFET with defined halos that are bound to defined source/drain extensions and a method of forming the MOSFET. A semiconductor layer is etched to form recesses that undercut a gate dielectric layer. A low energy implant forms halos. Then, a COR pre-clean is performed and the recesses are filled by epitaxial deposition. The epi can be in-situ doped or subsequently implanted to form source/drain extensions. Alternatively, the etch is immediately followed by the COR pre-clean, which is followed by epitaxial deposition to fill the recesses. During the epitaxial deposition process, the deposited material is doped to form in-situ doped halos and, then, the dopant is switched to form in-situ doped source/drain extensions adjacent to the halos. Alternatively, after the in-situ doped halos are formed the deposition process is performed without dopants and an implant is used to form source/drain extensions.
摘要翻译： 公开了具有限定的限定的卤素的MOSFET的实施例，其限定的源极/漏极扩展部分以及形成MOSFET的方法。 蚀刻半导体层以形成切割栅极介电层的凹部。 低能量植入物形成晕轮。 然后，执行COR预清洁，并且通过外延沉积填充凹部。 外延可以被原位掺杂或随后植入以形成源/漏扩展。 或者，蚀刻之后紧接着是COR预清洁，随后进行外延沉积以填充凹部。 在外延沉积工艺期间，沉积的材料被掺杂以形成原位掺杂的光晕，然后切换掺杂剂以形成邻近光晕的原位掺杂的源极/漏极延伸。 或者，在形成原位掺杂的光晕之后，进行沉积工艺而没有掺杂剂，并且使用注入来形成源极/漏极延伸部。
-
公开(公告)号：US07383550B2
公开(公告)日：2008-06-03
申请号：US10328255
申请日：2002-12-23
IPC分类号： G06F9/54 , G06F9/46 , G06F9/44 , G06F15/163
CPC分类号： G06F9/5072 , G06F9/5027
摘要： The invention provides a system/method of processing client requests over a computer network of hosts, that includes creating persistent containers in the hosts, creating objects in the containers, grouping the containers into grid container arrays, grouping objects within containers that are within a single grid container array into grid object arrays, creating one micro scheduler for each grid container array, dividing each of the client requests into a plurality of tasks, and assigning groups of the tasks to the microschedulers, wherein the microschedulers assign individual tasks to the objects. The invention assigns the microschedulers additional groups of tasks as the microschedulers return groups of completed tasks. The method can also include passing the client requests through gateways to multiple grid services schedulers.
摘要翻译： 本发明提供了一种通过主机的计算机网络处理客户端请求的系统/方法，其包括在主机中创建持久容器，在容器中创建对象，将容器分组为网格容器阵列，将容器内的对象分组在单个 将网格容器阵列转换为网格对象阵列，为每个网格容器阵列创建一个微调度器，将每个客户端请求划分为多个任务，以及将任务组分配给微计划器，其中微计划器将各个任务分配给对象。 本发明在微计划器返回已完成任务的组时，分配另外的任务组。 该方法还可以包括将客户端请求通过网关传递到多个网格服务调度器。
-
公开(公告)号：US07383337B2
公开(公告)日：2008-06-03
申请号：US10707972
申请日：2004-01-29
申请人： Brian T. Denton , Robert J. Milne , Chi-Tai Wang
发明人： Brian T. Denton , Robert J. Milne , Chi-Tai Wang
IPC分类号： G06F15/173
CPC分类号： G06Q10/06 , G06Q10/06315
摘要： The invention disclosed comprises a method for encouraging fair sharing of limited material inventory and capacity between multiple customers when creating a production plan. the invention allocates resources among competing demands in a linear programming production planning system by first classifying the demands into fair share sets, wherein all demands within each set have the same priority, calculating the cumulative demand for each resource within each set, and then allocating the resources to the demands in order of fair share set priority. If, during the allocating process, the supply of a given resource cannot satisfy a given cumulative demand of a given set, the given resource is allocated proportionally (e.g., evenly or according to usage proportions) among all demands that contribute to the given cumulative demand within the given set. This is an advanced planning system for optimizing established planning objectives (e.g., customer service, short lead times, low inventory, and prioritized allocation of supply and capacity) to compute a feasible production plan for the enterprise.
摘要翻译： 所公开的发明包括在创建生产计划时鼓励在多个客户之间公平共享有限材料库存和容量的方法。 本发明通过首先将需求分类为公平共享集合，在线性规划生产计划系统中的竞争需求中分配资源，其中每个集合内的所有需求具有相同的优先级，计算每个集合内的每个资源的累积需求，然后分配 资源按要求按照公平分配优先顺序排列。 如果在分配过程中，给定资源的供给不能满足给定集合的给定累积需求，则给定资源按比例（例如，均匀地或根据使用比例）在有助于给定累积需求的所有需求中分配 在给定的集合。 这是一个先进的规划系统，用于优化既定的规划目标（例如，客户服务，交货期短，库存低，供应和容量优先分配），以计算企业的可行生产计划。
-
公开(公告)号：US07382036B2
公开(公告)日：2008-06-03
申请号：US11161320
申请日：2005-07-29
IPC分类号： H01L27/148
CPC分类号： H01L27/10 , H01L23/5256 , H01L2924/0002 , H01L2924/3011 , H01L2924/00
摘要： An eFuse begins with a single crystal silicon-on-insulator (SOI) structure that has a single crystal silicon layer on a first insulator layer. The single crystal silicon layer is patterned into a strip. Before or after the patterning, the single crystal silicon layer is doped with one or more impurities. At least an upper portion of the single crystal silicon layer is then silicided to form a silicided strip. In one embodiment the entire single crystal silicon strip is silicided to create a silicide strip. Second insulator(s) is/are formed on the silicide strip, so as to isolate the silicided strip from surrounding structures. Before or after forming the second insulators, the method forms electrical contacts through the second insulators to ends of the silicided strip. By utilizing a single crystal silicon strip, any form of semiconductor, such as a diode, conductor, insulator, transistor, etc. can form the underlying portion of the fuse structure. The overlying silicide material allows the fuse to act as a conductor in its unprogrammed state. However, contrary to metal or polysilicon based eFuses which only comprise an insulator in the programmed state, when the inventive eFuse is programmed (and the silicide is moved or broken) the underlying semiconductor structure operates as an active semiconductor device.
摘要翻译： eFuse从在第一绝缘体层上具有单晶硅层的单晶硅绝缘体（SOI）结构开始。 将单晶硅层图案化成条带。 在构图之前或之后，单晶硅层掺杂有一种或多种杂质。 至少单晶硅层的上部然后被硅化以形成硅化带。 在一个实施例中，整个单晶硅带被硅化以产生硅化物条。 在硅化物条上形成第二绝缘体，从而将硅化物带与周围结构隔离。 在形成第二绝缘体之前或之后，该方法通过第二绝缘体形成与硅化带的端部的电接触。 通过使用单晶硅条，任何形式的半导体，例如二极管，导体，绝缘体，晶体管等都可以形成熔丝结构的下面部分。 上覆的硅化物材料允许熔丝作为未编程状态的导体。 然而，与仅编程状态的仅包含绝缘体的金属或多晶硅基eFuse相反，当本发明的eFuse被编程（并且硅化物被移动或断开）时，下面的半导体结构作为有源半导体器件工作。
-
公开(公告)号：US07381655B2
公开(公告)日：2008-06-03
申请号：US11226726
申请日：2005-09-14
IPC分类号： H01L21/302 , H01L21/461
CPC分类号： H01L21/28132 , H01L21/0337 , H01L21/0338 , H01L29/66795
摘要： Disclosed herein is an imaging method for patterning component shapes (e.g., fins, gate electrodes, etc.) into a substrate. By conducting a trim step prior to performing either an additive or subtractive sidewall image transfer process, the method avoids the formation of a loop pattern in a hard mask and, thus, avoids a post-SIT process trim step requiring alignment of a trim mask to sub-lithographic features to form a hard mask pattern with the discrete segments. In one embodiment a hard mask is trimmed prior to conducting an additive SIT process so that a loop pattern is not formed. In another embodiment an oxide layer and memory layer that are used to form a mandrel are trimmed prior to the conducting a subtractive SIT process. A mask is then used to protect portions of the mandrel during etch back of the oxide layer so that a loop pattern is not formed.
摘要翻译： 本文公开了一种用于将部件形状（例如散热片，栅电极等）图案化成基板的成像方法。 通过在执行加法或减损侧壁图像转移处理之前进行修整步骤，该方法避免了在硬掩模中形成环形图案，并且因此避免了后SIT工艺修整步骤，其需要将修剪掩模对准 亚光刻特征以形成具有离散片段的硬掩模图案。 在一个实施例中，在进行添加SIT处理之前修剪硬掩模，使得不形成环形图案。 在另一个实施例中，用于形成心轴的氧化物层和记忆层在进行减法SIT处理之前被修整。 然后在氧化层的回蚀刻期间使用掩模来保护心轴的部分，使得不形成环形图案。
-
-
-
-
-
-
-
-
-
搜索结果
244 条记录 (耗时 0.035 秒)
