公开(公告)号：US20070247629A1

公开(公告)日：2007-10-25

申请号：US11407216

申请日：2006-04-20

申请人： Byung Kim ,  Mark Rood ,  Ke Du

发明人： Byung Kim ,  Mark Rood ,  Ke Du

IPC分类号： G01N21/00

CPC分类号： G01N21/59 ,  G01N21/278 ,  G01N2021/1772 ,  G01N2021/1793 ,  G01N2021/5957

摘要： Photography is employed to objectively quantify opacity of fluids such as smoke plumes and dust via a method termed the Digital Optical Method (DOM™). The DOM™ quantifies the ratio of radiance values by means of a camera response curve obtained using objective measures. The radiance ratios are then used to calculate opacity of target fluids such as smoke plumes. The DOM™ quantifies opacity during both daytime and nighttime conditions with a much broader range of subject types, e.g., white, gray and black smoke plumes, and environmental conditions, e.g., non-blue-sky, building, and mountain backgrounds, than existing systems while not requiring human interpretation for any application. In one embodiment, the DOM™ quantifies opacity from digital photos using a pre-designed algorithm and an inexpensive digital camera. Very little training is needed to implement the DOM™ and it yields consistent objective quantitative results, while providing a permanent photographic record easily digitally archived.

摘要翻译： 摄影用于通过称为数字光学方法（DOM）的方法客观地量化诸如烟雾和灰尘的流体的不透明度。 DOM TM通过使用客观测量获得的相机响应曲线来量化辐射值的比率。 然后使用辐射比来计算目标流体（如烟羽）的不透明度。 DOM（TM）在白天和夜间条件下量化不透明度，具有更广泛的主题类型，例如白色，灰色和黑色烟雾羽毛以及环境条件，例如非蓝天，建筑和山脉背景， 比现有系统，而不需要任何应用程序的人为解释。 在一个实施例中，DOM TM使用预先设计的算法和便宜的数码相机来量化来自数字照片的不透明度。 实施DOM（TM）需要很少的培训，并且可以产生一致的客观量化结果，同时提供易于数字归档的永久性照片记录。

公开(公告)号：US6055121A

公开(公告)日：2000-04-25

申请号：US920592

申请日：1997-08-29

申请人： Ke Du ,  William C. Little ,  Matthew C. Burton

发明人： Ke Du ,  William C. Little ,  Matthew C. Burton

IPC分类号： G11B5/596

CPC分类号： G11B5/59644

摘要： An apparatus and method for optimizing detection thresholds used to identify addresses of tracks in a disc drive having a head and a disc upon which a plurality of tracks are defined, the tracks including track address fields in which track address information is stored. A comparator is provided which generates track address sequence signals in response to a corresponding plurality of readback signals obtained as the head is maintained in a fixed relationship to a selected track. A demodulator identifies erroneous track address sequence signals therefrom and increments a count in an accumulator for each detected, erroneous signal. A detection threshold circuit sequentially provides the comparator with a set of detection thresholds from a population of detection thresholds, each of which are used in turn by the comparator. An optimal set of thresholds is thereby selected for use by the disc drive in accordance with the relative number of erroneous track address sequence signals accumulated for each of the detection thresholds from the population.

摘要翻译： 一种用于优化检测阈值的装置和方法，用于识别具有头和盘的盘驱动器中的轨道的地址，轨道地址信息被存储在轨道地址字段中。 提供比较器，其响应于当头被保持与所选轨道的固定关系时获得的对应的多个回读信号而产生轨道地址序列信号。 解调器从其识别错误的轨道地址序列信号，并为每个检测到的错误信号增加累加器中的计数。 检测阈值电路依次向比较器提供一组检测阈值，每个检测阈值由比较器依次使用。 因此，根据从群体中的每个检测阈值积累的错误轨道地址序列信号的相对数目，选择最佳阈值集合供磁盘驱动器使用。

公开(公告)号：US08189582B2

公开(公告)日：2012-05-29

申请号：US11996641

申请日：2006-08-08

申请人： Ke Du ,  Chen Wu ,  Chengfa Fan ,  Yanfeng Qu

发明人： Ke Du ,  Chen Wu ,  Chengfa Fan ,  Yanfeng Qu

IPC分类号： H04L12/28 ,  H04L12/56

CPC分类号： H04L12/4641 ,  H04L12/185 ,  H04L12/1886

摘要： A multicast-supported virtual local area network switching system and a method thereof comprise a switching engine, a CPU unit, further comprising a mapping storage unit for storing the mapping relationship of the virtual local area network and the multicast stream pushed to the switching system by it, and a switching table storage unit for storing the corresponding relation between ingress virtual local area network and egress virtual local area network; the switching engine receives the internet group management protocol message from the user virtual local area network and transmits the message to the CPU unit for processing, if it is Internet group management protocol join/leave message, the CPU unit looks up the mapping storage unit and forwards it in the looked-up multicast virtual local area network; the switching engine receives the Internet group management protocol messages or multicast data frames from the multicast virtual local area network and forwards them in the looked-up user virtual local area network by looking up the switching table storage unit.

摘要翻译： 一种组播支持的虚拟局域网交换系统及其方法包括：交换引擎，CPU单元，还包括：映射存储单元，用于存储虚拟局域网的映射关系，以及通过 以及用于存储入口虚拟局域网和出口虚拟局域网之间的对应关系的交换表存储单元; 交换引擎从用户虚拟局域网接收因特网组管理协议消息，并将消息发送到CPU单元进行处理，如果是因特网组管理协议加入/离开消息，则CPU单元查找映射存储单元和 在查找到的组播虚拟局域网中转发; 交换引擎从组播虚拟局域网接收因特网组管理协议消息或组播数据帧，并通过查看交换表存储单元将其转发到查找用户虚拟局域网中。

公开(公告)号：US20110026530A1

公开(公告)日：2011-02-03

申请号：US12854584

申请日：2010-08-11

申请人： Ke Du ,  Mingyi Wang ,  Chunhong Huang ,  Jun Feng ,  Weidong Zhong

发明人： Ke Du ,  Mingyi Wang ,  Chunhong Huang ,  Jun Feng ,  Weidong Zhong

IPC分类号： H04L12/28

CPC分类号： H04L12/4641 ,  H04L45/507

摘要： A pseudo wire label reflector, based on a existing label distribution router, further includes: a mode management unit, for determining the mode for switching and transferring a pseudo wire label message; a first session management unit, for saving the support information that LDP signaling session connection supports the pseudo wire label reflecting capacity; a first pseudo wire label receiving unit, for receiving a pseudo wire label message; a first pseudo wire management unit, for saving the switched pseudo wire state and deciding whether to modify the pseudo wire label message according to the switch and transfer mode determined by the mode management unit; a first pseudo wire label sending unit, for sending the pseudo wire label message. The invention also provides an edge apparatus and a L2 VPN. The present invention makes PE equipments participating pseudo wire service to set up PSN tunnel and LDP signaling session connection with the pseudo wire label reflector by introducing the pseudo wire label reflector and extending the capacity that the current edge apparatus supports the pseudo wire label reflector.

摘要翻译： 基于现有标签分发路由器的伪线标签反射器还包括：模式管理单元，用于确定用于切换和传送伪线标签消息的模式; 第一会话管理单元，用于保存LDP信令会话连接支持伪线标签反映能力的支持信息; 第一伪线标签接收单元，用于接收伪线标签消息; 第一伪线管理单元，用于保存切换伪线状态，并根据由模式管理单元确定的切换和传送模式来决定是否修改伪线标签消息; 第一伪线标签发送单元，用于发送伪线标签消息。 本发明还提供边缘设备和L2 VPN。 本发明使PE设备参与伪线业务，通过引入伪线标签反射器并扩展当前边缘设备支持伪线标签反射器的容量，建立伪线标签反射器的PSN隧道和LDP信令会话连接。

公开(公告)号：US07495767B2

公开(公告)日：2009-02-24

申请号：US11407216

申请日：2006-04-20

申请人： Byung J. Kim ,  Mark J. Rood ,  Ke Du

发明人： Byung J. Kim ,  Mark J. Rood ,  Ke Du

IPC分类号： G01N21/00

CPC分类号： G01N21/59 ,  G01N21/278 ,  G01N2021/1772 ,  G01N2021/1793 ,  G01N2021/5957

摘要： Photography is employed to objectively quantify opacity of fluids such as smoke plumes and dust via a method termed the Digital Optical Method (DOM™). The DOM™ quantifies the ratio of radiance values by means of a camera response curve obtained using objective measures. The radiance ratios are then used to calculate opacity of target fluids such as smoke plumes. The DOM™ quantifies opacity during both daytime and nighttime conditions with a much broader range of subject types, e.g., white, gray and black smoke plumes, and environmental conditions, e.g., non-blue-sky, building, and mountain backgrounds, than existing systems while not requiring human interpretation for any application. In one embodiment, the DOM™ quantifies opacity from digital photos using a pre-designed algorithm and an inexpensive digital camera. Very little training is needed to implement the DOM™ and it yields consistent objective quantitative results, while providing a permanent photographic record easily digitally archived.

摘要翻译： 摄影用于通过称为数字光学方法（DOM）的方法客观地量化诸如烟雾和灰尘的流体的不透明度。 DOM TM通过使用客观测量获得的相机响应曲线来量化辐射值的比率。 然后使用辐射比来计算目标流体（如烟羽）的不透明度。 DOM（TM）在白天和夜间条件下量化不透明度，具有更广泛的主题类型，例如白色，灰色和黑色烟雾羽毛以及环境条件，例如非蓝天，建筑和山脉背景， 比现有系统，而不需要任何应用程序的人为解释。 在一个实施例中，DOM TM使用预先设计的算法和便宜的数码相机来量化来自数字照片的不透明度。 实施DOM（TM）需要很少的培训，并且可以产生一致的客观量化结果，同时提供易于数字归档的永久性照片记录。

公开(公告)号：US20080165681A1

公开(公告)日：2008-07-10

申请号：US11813187

申请日：2005-07-29

申请人： Jinhong Huang ,  Zhijian Feng ,  Ke Du

发明人： Jinhong Huang ,  Zhijian Feng ,  Ke Du

IPC分类号： H04L1/00

CPC分类号： H04L47/10 ,  H04L12/185 ,  H04L45/16 ,  H04L47/12 ,  H04L47/15 ,  H04L47/20 ,  H04L47/24

摘要： The present invention relates to a method for preventing IP multicast data stream from overloading the communication system by distinguishing multiple services. When a multicast data packet ‘packet (key)’ with its key value being ‘key’ reaches the forwarding layer, if it doesn't find the matching multicast route entry or matching multicast forwarding entry, and it satisfies the requirement of multicast protocol Assert state machine, then it searches for current-limiting policy list L (type, key) according to the key, and applies the searched current-limiting policy to multicast forwarding and packet up-sending. If it finds the matching multicast route entry and the protocol requests sending the multicast data packet to the protocol processor, the new multicast forwarding entry P_M(key) is formed with the current-limiting policy of distinguishing the service type(key), and P_M(key) is used to forward and up-send. The above-mentioned three current-limiting types may be used separately or combined freely. The method of the invention can use the corresponding policy based on distinguishing different multicast service streams so as to avoid a big flow of multiservice IP multicast data stream from overloading the IP communication devices.

摘要翻译： 本发明涉及一种通过区分多个业务来防止IP组播数据流超载通信系统的方法。 当其密钥值为“密钥”的组播数据包“密钥”到达转发层时，如果没有找到匹配的组播路由条目或匹配的组播转发表项，则满足组播协议的要求Assert 根据密钥搜索限流策略列表L（类型，密钥），并将搜索到的限流策略应用于组播转发和报文上发。 如果发现匹配的组播路由表项和协议请求向协议处理器发送组播数据报文，则新的组播转发表项P_M（key）由形成服务类型（密钥）的限流策略和P_M （键）用于转发和向上发送。 上述三种限流类型可以单独使用或组合使用。 本发明的方法可以基于区分不同的多播服务流来使用相应的策略，以避免多业务IP组播数据流的大流量过载IP通信设备。

公开(公告)号：US06307884B1

公开(公告)日：2001-10-23

申请号：US09087168

申请日：1998-05-29

申请人： Ke Du ,  Yujie Xu

发明人： Ke Du ,  Yujie Xu

IPC分类号： H03H730

CPC分类号： H04L25/061 ,  H04L25/03057 ,  H04L2025/0349 ,  H04L2025/03566

摘要： An apparatus and method for improving error rate performance in a communication channel utilizes a dual decision feedback equalizer (DDFE) through the application of selective attenuation. An improved DDFE comprises first and second decision feedback equalizers (DFEs) operably connected in parallel to receive a series of input signals zk. The DFEs independently generate two series of decisions â ;1k and â ;2k in relation to peak magnitudes of the series of transmitted signals zk by filtering the series of decisions â ;1k and â ;2k to generate filtered outputs (a*p)1k and (a*p)2k based on previous decisions, and summing the filtered outputs (a*p)1k and (a*p)2k with the transmitted input series zk to generate two series of decision variables ã1k and ã2k, to remove effects of intersymbol interference. The improved DDFE further comprises first and second attenuation blocks, so that when a mismatch is detected between the series of decisions â ;1k and â ;2k, a selective attenuation gain less than one is introduced into each feedback filter loop. This results in a reallocation of some of the decision energy associated with the correct series of decisions to the erroneous series and helps the erroneous DFE to make correct decisions sooner, thereby improving the error rate performance of the channel.

摘要翻译： 用于提高通信信道中的误码率性能的装置和方法通过应用选择性衰减来利用双判决反馈均衡器（DDFE）。 改进的DDFE包括并行可操作地连接以接收一系列输入信号zk的第一和第二判决反馈均衡器（DFE）。 DFE通过过滤一系列决策＆acirc; 1k和＆acirc; 2k来独立生成两系列决策与acirc; 1k和＆acirc; 2k相关于一系列传输信号zk的峰值幅度，以生成滤波输出（a * p）1k 和（a * p）2k，并且将所滤波的输出（a * p）1k和（a * p）2k与发送的输入序列zk相加以生成两个系列的决定变量ã1k和ã2k，以去除效果 的符号间干扰。 改进的DDFE还包括第一和第二衰减块，使得当在一系列决策与acirc 1k和＆acirc; 2k之间检测到失配时，在每个反馈滤波器环路中引入小于1的选择性衰减增益。 这导致将与正确的一系列决策相关联的一些决策能量重新分配给错误的系列，并且有助于错误的DFE更早地作出正确的决定，从而提高了频道的错误率性能。

公开(公告)号：US6108152A

公开(公告)日：2000-08-22

申请号：US898871

申请日：1997-07-22

申请人： Ke Du ,  Robert D. Cronch ,  Kenneth R. Burns ,  Bernardo Rub

发明人： Ke Du ,  Robert D. Cronch ,  Kenneth R. Burns ,  Bernardo Rub

IPC分类号： G11B20/14 ,  G11B5/00 ,  G11B5/09 ,  G11B19/04 ,  G11B20/10 ,  G11B20/12 ,  G11B20/18 ,  G11B27/30

CPC分类号： G11B20/1217 ,  G11B19/04 ,  G11B20/1833 ,  G11B27/3027 ,  G11B20/10009 ,  G11B20/1816 ,  G11B2005/0016 ,  G11B2020/1476 ,  G11B2220/20

摘要： An apparatus and method are disclosed for reducing the effects of thermal asperities and other anomalous conditions on disc drive read operations. An improved data block format is provided having first and second sync fields in which sync patterns are stored, the first and second sync fields being separated by a selected distance within the data block. The sync pattern of the second sync field enables synchronization of a disc drive read channel to the user data in the data block when the read channel fails to be synchronized by the sync pattern of the first sync field. In one embodiment, control information used by the read channel to prepare for receipt of the user data is stored in a field between the first and second sync fields; in an alternative embodiment, a portion of the user data is stored between the first and second sync fields.

摘要翻译： 公开了用于减少热凹凸和其他异常条件对磁盘驱动器读取操作的影响的装置和方法。 提供了改进的数据块格式，其具有其中存储同步模式的第一和第二同步字段，第一和第二同步字段在数据块内被选择的距离分隔开。 当读通道不能通过第一同步字段的同步模式同步时，第二同步字段的同步模式使得盘驱动器读通道能够与数据块中的用户数据同步。 在一个实施例中，由读取信道用于准备接收用户数据的控制信息存储在第一和第二同步字段之间的字段中; 在替代实施例中，用户数据的一部分被存储在第一和第二同步字段之间。

公开(公告)号：US6055118A

公开(公告)日：2000-04-25

申请号：US866689

申请日：1997-05-30

申请人： Ke Du

发明人： Ke Du

IPC分类号： G11B19/04 ,  G11B20/10 ,  G11B20/18 ,  G11B5/09

CPC分类号： G11B20/18 ,  G11B19/04 ,  G11B20/10009

摘要： An apparatus and associated method are disclosed for recovering user data from a data block in a disc drive in which user data are stored by way of a plurality of blocks on tracks of a disc, the blocks including data blocks in which the user data are stored in predefined increments. A head generates readback signals from the blocks and a read channel reconstructs the user data from the readback signals. An automatic gain control stage establishes a gain used to amplify the readback signal to a level suitable for use by the read channel. When an uncorrected read error is detected, the disc drive proceeds to read a previous block sequentially preceding the affected data block and sets the gain to a fixed value based upon the readback signals determined from the previous block. The fixed value of gain is then used to recover the user data from the affected data block.

摘要翻译： 公开了一种装置和相关方法，用于从其中通过多个块在盘的轨道上存储用户数据的盘驱动器中的数据块来恢复用户数据，所述块包括存储用户数据的数据块 以预定义的增量。 头从块生成回读信号，读通道从回读信号重构用户数据。 自动增益控制级建立用于将回读信号放大到适于读取通道使用的电平的增益。 当检测到未校正的读取错误时，盘驱动器继续在受影响的数据块之前顺序地读取先前的块，并且基于从先前块确定的回读信号将增益设置为固定值。 然后使用固定的增益值从受影响的数据块中恢复用户数据。