公开(公告)号：US08054969B2

公开(公告)日：2011-11-08

申请号：US11675345

申请日：2007-02-15

Applicant: Akshay Adhikari ,  Sachin Garg ,  Anjur Sundaresan Krishnakumar ,  Navjot Singh

Inventor： Akshay Adhikari ,  Sachin Garg ,  Anjur Sundaresan Krishnakumar ,  Navjot Singh

IPC: H04N7/167

CPC classification number: H04L63/08 ,  G10L19/018 ,  H04L63/168

Abstract: A method is disclosed that enables the transmission of a digital message along with a corresponding media information signal, such as audio or video. A telecommunications device that is processing the information signal from its user, such as a speech signal, encodes the information signal by using a model-based compression coder. One such device is a telecommunications endpoint. Then, based on an evaluation of the perceptual significance of each encoded bit, or on some other meaningful characteristic of the signal, the endpoint's processor: (i) determines which encoded bits can be overwritten; and (ii) intersperses the digital message bits throughout the encoded signal in place of the overwritten bits. The endpoint then transmits those digital message bits as part of the encoded information signal. In this way, no additional bits are appended to the packet to be transmitted, thereby addressing the issue of compatibility with existing protocols and firewalls.

Abstract translation: 公开了一种能够传送数字消息以及对应的媒体信息信号（诸如音频或视频）的方法。 正在处理其用户的信息信号（例如语音信号）的电信设备通过使用基于模型的压缩编码器对信息信号进行编码。 一个这样的设备是电信端点。 然后，基于对每个编码比特的感知意义的评估，或在信号的一些其它有意义的特征上，终点的处理器：（i）确定可以覆盖哪些编码比特; 并且（ii）将整数编码信号中的数字信息位散置在代替覆盖位。 端点然后将这些数字消息比特作为编码信息信号的一部分进行发送。 以这种方式，没有额外的比特附加到要发送的分组，从而解决与现有协议和防火墙的兼容性的问题。

公开(公告)号：US20110196803A1

公开(公告)日：2011-08-11

申请号：US12701480

申请日：2010-02-05

Applicant: John Reynolds ,  Kurt R. Schedler ,  Sachin Garg ,  Balaji Gandhi ,  Christopher P. Gray

Inventor： John Reynolds ,  Kurt R. Schedler ,  Sachin Garg ,  Balaji Gandhi ,  Christopher P. Gray

IPC: G06Q50/00 ,  G06Q10/00

CPC classification number: G06Q10/10 ,  G06Q50/2053

Abstract: An evaluation of an external course may be based on course-to-requirement mappings that have been previously established for the target institution. As a result of the evaluation, the external course may be mapped to a particular program requirement for a particular student, which allows the external course to fulfill the particular program requirement at the institution for the particular student. The course evaluation may be automatic or may be performed by a human evaluator. Course-to-requirement mappings may be globally-established, and applicable to all students and applicants of the target institution, or may be applicant-specific, and only applicable to the student for which the mapping was established. Information about both globally-established and applicant-specific course-to-requirement mappings that have been previously established for the institution may be used to evaluate an external course for a particular student.

Abstract translation: 对外部课程的评估可以基于以前为目标机构确定的课程到要求的映射。 作为评估的结果，外部课程可能被映射到特定学生的特定课程要求，这允许外部课程在特定学生的机构满足特定的课程要求。 课程评估可以是自动的，也可以由人类评价者执行。 课程到要求的映射可能是全球建立的，适用于目标机构的所有学生和申请人，或者可能是申请人特定的，并且仅适用于映射建立的学生。 关于以前为该机构确定的全球确定和申请人特定课程到要求映射的信息可用于评估特定学生的外部课程。

公开(公告)号：US20090274143A1

公开(公告)日：2009-11-05

申请号：US12115196

申请日：2008-05-05

Applicant: Sachin Garg ,  Navjot Singh ,  Akshay Adhikari ,  Yu-Sung Wu

Inventor： Sachin Garg ,  Navjot Singh ,  Akshay Adhikari ,  Yu-Sung Wu

IPC: H04L12/66

CPC classification number: H04L65/1079 ,  H04L51/12 ,  H04M3/436 ,  H04M2201/18

Abstract: An apparatus and method for detecting potentially-improper call behavior (e.g., SPIT, etc.) are disclosed. The illustrative embodiment of the present invention is based on finite-state machines (FSMs) that represent the legal states and state transitions of a communications protocol at a node during a Voice over Internet Protocol (VoIP) call. In accordance with the illustrative embodiment, a library of FSM execution profiles associated with improper call behavior is maintained. When there is a match between the behavior of a finite-state machine during a call and an execution profile in the library, an alert is generated.

Abstract translation: 公开了一种用于检测潜在不适当的呼叫行为（例如，SPIT等）的装置和方法。 本发明的说明性实施例基于表示在因特网协议语音（VoIP）呼叫期间在节点处的通信协议的合法状态和状态转换的有限状态机（FSM）。 根据说明性实施例，保持与不正当呼叫行为相关联的FSM执行简档库。 当在调用期间有限状态机的行为与库中的执行配置文件之间存在匹配时，生成警报。

公开(公告)号：US20090103701A1

公开(公告)日：2009-04-23

申请号：US11874162

申请日：2007-10-17

Applicant: Sachin Garg ,  Navjot Singh

Inventor： Sachin Garg ,  Navjot Singh

IPC: H04M3/436

CPC classification number: H04M3/436 ,  H04M3/42059 ,  H04M2201/10 ,  H04M2201/36

Abstract: A method is disclosed that enables the screening of unwanted telephone calls, such as voice or video calls, for one or more called parties. In accordance with the illustrative embodiment of the present invention, an anti-SPAM system receives signaling information for one or more telephone calls made to one or more called parties by a calling party. Although the calling party can be a human caller, in a SPAM-over-Internet-Telephony context the calling party can alternatively be a server or other network element that originates SPAM voice calls for advertising purposes; both possibilities are accounted for in the illustrative embodiment. The anti-SPAM system then observes the behavior of the called party or parties that is exhibited in response to receiving the telephone calls. Based on the observed behavior, the anti-SPAM system then updates one or more rules for handling future telephone calls made to the protected called parties.

Abstract translation: 公开了一种能够对一个或多个被叫方筛选不需要的电话呼叫（例如语音或视频呼叫）的方法。 根据本发明的说明性实施例，反SPAM系统接收由主叫方对一个或多个被叫方进行的一个或多个电话呼叫的信令信息。 虽然呼叫方可以是人类呼叫者，但是在因特网上的电话上下文中，呼叫方可以替代地是发起用于广告目的的SPAM语音呼叫的服务器或其他网络元件; 在说明性实施例中考虑了两种可能性。 反垃圾邮件系统然后观察响应于接收电话而展示的被叫方或行为者的行为。 基于观察到的行为，反垃圾邮件系统然后更新一个或多个规则，以处理对受保护的被叫方的未来电话呼叫。

公开(公告)号：US20080319940A1

公开(公告)日：2008-12-25

申请号：US11874161

申请日：2007-10-17

Applicant: Sachin Garg ,  Navjot Singh ,  Shalini Yajnik ,  Ranjith Vasireddy ,  Sridhar Vasireddy

Inventor： Sachin Garg ,  Navjot Singh ,  Shalini Yajnik ,  Ranjith Vasireddy ,  Sridhar Vasireddy

IPC: G06F17/30

CPC classification number: H04L12/66

Abstract: A technique is disclosed that enables the run-time behavior of a data-processing system to be analyzed and, in many cases, to be predicted. In particular, the illustrative embodiment of the present invention comprises i) transforming the messages that constitute an unstructured log into a numerical series and ii) applying a time-series analysis on the resultant series for the purpose of pattern detection. Indeed, it is recognized in the illustrative embodiment that the problem really is to detect patterns that depict aspects of system behavior, regardless of the textual content of the individual log messages. In other words, by analyzing the totality of the messages in the log or logs—as opposed to looking for pre-defined patterns of the individual messages—system behavior can be mapped and understood. The mapping helps in characterizing the system for the purposes of predicting failure, determining the time required to reach stability during failure recovery, and so forth.

Abstract translation: 公开了一种能够分析数据处理系统的运行时行为并且在许多情况下被预测的技术。 特别地，本发明的说明性实施例包括：i）将构成非结构化日志的消息变换为数字序列，以及ii）为了模式检测的目的，对所得到的序列应用时间序列分析。 实际上，在说明性实施例中认识到，问题确实是检测描绘系统行为方面的模式，而不管各个日志消息的文本内容如何。 换句话说，通过分析日志或日志中的消息的总体，而不是寻找单个消息的预定义模式，可以映射和理解系统行为。 映射有助于表征系统的目的是为了预测故障，确定在故障恢复期间达到稳定所需的时间等等。

公开(公告)号：US20080144499A1

公开(公告)日：2008-06-19

申请号：US11610487

申请日：2006-12-13

Applicant: Akshay Adhikari ,  Sachin Garg ,  Anjur Sundaresan Krishnakumar ,  Navjot Singh

Inventor： Akshay Adhikari ,  Sachin Garg ,  Anjur Sundaresan Krishnakumar ,  Navjot Singh

IPC: H04L12/56

CPC classification number: H04L47/10 ,  H04L47/22 ,  H04L49/3054

Abstract: A method is disclosed that enables the avoidance of a processor overload of a telecommunications endpoint device that is susceptible to traffic floods. An enhanced network switch sets the speed on one of its data ports as a specific function of the speeds of the devices that are connected to one or more of its other data ports. This behavior is different from that of network switches in the prior art, in which the data rate of a port in the prior art is auto-negotiated to the highest speed that can be supported by the network elements at either end of the port's connection, regardless of the other devices present. By considering the specific devices that are connected, the enhanced network switch is able to limit the amount of traffic that is directed by an upstream device, such as a router, towards a device with limited processor capability, such as a packet-based phone.

Abstract translation: 公开了一种能够避免容易遭受交通洪泛的电信端点设备的处理器过载的方法。 增强型网络交换机将其数据端口之一的速度设置为连接到其一个或多个其他数据端口的设备的速度的特定功能。 这种现象不同于现有技术中的网络交换机，其中现有技术中的端口的数据速率被自动协商为端口连接的任一端可由网络元件支持的最高速度， 不管其他设备存在。 通过考虑连接的特定设备，增强型网络交换机能够将诸如路由器之类的上游设备指向的业务量限制为具有有限处理器能力的设备，例如基于分组的电话。

公开(公告)号：US06871233B1

公开(公告)日：2005-03-22

申请号：US09610630

申请日：2000-07-05

Applicant: Mark Joseph Bearden ,  Sachin Garg ,  Woei-Jyh Lee ,  Aad Petrus-Antonius van Moorsel

Inventor： Mark Joseph Bearden ,  Sachin Garg ,  Woei-Jyh Lee ,  Aad Petrus-Antonius van Moorsel

IPC: H04L12/24 ,  H04L12/26 ,  H04L12/56 ,  G06F15/173

CPC classification number: H04L47/10 ,  H04L41/0893 ,  H04L41/5019 ,  H04L41/5032 ,  H04L43/00 ,  H04L43/045 ,  H04L43/08 ,  H04L43/0817 ,  H04L43/0823 ,  H04L43/0852 ,  H04L47/20 ,  H04L47/2458

Abstract: The “what” and “how” of policy-based management is integrated in a single framework that enables a system administrator to specify service-level quality of service (QoS) goals for automatic enforcement. Automatic enforcement of the specified “high-level” QoS goals is realized through the execution of “low-level” rule-based or procedural logic, without the client having to specify the low level logic. Specifically, one embodiment of the invention employs a management server including a graphical interface that allows a user, e.g. a system administrator, to easily specify parameters for service-level QoS goals. A QoS goal is defined by the administrator selecting a client, a service and a QoS expression that specifies the desired service-level QoS. The state of the network is monitored and one or more defined QoS goals are selected for evaluation in a continuous process. The QoS delivered for the selected goal is determined and compared to the desired QoS for the selected QoS goal. Then, prescribed actions are taken or not depending whether the delivered QoS is equal to the selected QoS goal. If not, and the delivered QoS exceeds the selected QoS goal, a set of actions is determined and executed to reduce network resources assigned to the client and service of the selected goal. Similarly, if the delivered QoS is worse than the selected QoS goal, a set of actions is determined and executed to increase network resources assigned to the client and service of the selected goal.

Abstract translation: 基于策略的管理的“什么”和“如何”被集成到一个单一的框架中，使系统管理员可以指定自动执行服务级别的服务质量（QoS）目标。 通过执行“低级”规则或程序逻辑来实现指定的“高级”QoS目标的自动实现，而客户端不必指定低级逻辑。 具体地，本发明的一个实施例采用管理服务器，其包括图形界面，其允许用户，例如， 系统管理员，轻松指定服务级QoS目标的参数。 QoS目标由管理员选择客户端，服务和指定所需服务级QoS的QoS表达式来定义。 监视网络的状态，并选择一个或多个定义的QoS目标以在连续过程中进行评估。 确定为选定目标传递的QoS，并将其与所选QoS目标的期望QoS进行比较。 然后，根据传送的QoS是否等于所选择的QoS目标来执行规定的动作。 如果不是，并且所传递的QoS超过所选择的QoS目标，则确定并执行一组动作以减少分配给所选目标的客户端和服务的网络资源。 类似地，如果所递送的QoS比选择的QoS目标更差，则确定并执行一组动作以增加分配给所选择的目标的客户端和服务的网络资源。

公开(公告)号：US20050060424A1

公开(公告)日：2005-03-17

申请号：US10662728

申请日：2003-09-15

Applicant: Sachin Garg ,  Martin Kappes

Inventor： Sachin Garg ,  Martin Kappes

IPC: H04L12/28 ,  H04L12/56 ,  G06F15/16

CPC classification number: H04L47/10

Abstract: A technique for lessening the likelihood of congestion in a congestible node is disclosed. In the illustrative embodiment, the proxy node resides in the path of the protocol data units en route to a congestible node and the proxy node decides whether to drop protocol data units en route to the congestible node. In some embodiments of the present invention, the proxy node comprises a larger queue for the protocol data units than does the congestible node. The illustrative embodiment of the present invention is useful because it enables the manufacture of “lightweight” nodes without large queues and without the horsepower needed to run an algorithm, such as the Random Early Detection algorithm, for deciding which protocol data units to drop. Furthermore, the illustrative embodiment is useful because it can lessen the likelihood of congestion in legacy nodes.

Abstract translation: 公开了一种减少拥堵节点拥挤可能性的技术。 在说明性实施例中，代理节点驻留在路由到可拥塞节点的协议数据单元的路径中，并且代理节点决定是否将路由上的协议数据单元丢弃到可拥塞节点。 在本发明的一些实施例中，代理节点包括比可拥塞节点更大的协议数据单元队列。 本发明的说明性实施例是有用的，因为它能够制造没有大队列的“轻量级”节点，并且不需要运行诸如随机早期检测算法的算法所需的功率来决定哪些协议数据单元丢弃。 此外，说明性实施例是有用的，因为它可以减少传统节点中拥塞的可能性。

公开(公告)号：US20050060423A1

公开(公告)日：2005-03-17

申请号：US10662724

申请日：2003-09-15

Applicant: Sachin Garg ,  Martin Kappes

Inventor： Sachin Garg ,  Martin Kappes

IPC: H04L12/24 ,  H04L12/56 ,  G06F15/16

CPC classification number: H04L47/10

Abstract: A technique for lessening the likelihood of congestion in a congestible node is disclosed. In accordance with the illustrative embodiments of the present invention, one node—a proxy node—drops protocol data units to lessen the likelihood of congestion in the congestible node. In some embodiments of the present invention, the proxy node receives a metric of a queue at a congestible node and, based on the metric, decides whether to drop protocol data units en route to the congestible node. In some other embodiments of the present invention, the proxy node estimates a metric of a queue at a congestible node and, based on the metric, decides whether to drop protocol data units en route to the congestible node.

Abstract translation: 公开了一种减少拥堵节点拥挤可能性的技术。 根据本发明的说明性实施例，一个节点 - 代理节点丢弃协议数据单元，以减少拥塞节点中的拥塞的可能性。 在本发明的一些实施例中，代理节点在可拥塞节点处接收队列的度量，并且基于该度量决定是否将路由上的协议数据单元丢弃到可拥塞节点。 在本发明的一些其他实施例中，代理节点估计在可拥塞节点处的队列的度量，并且基于该度量决定是否将路由上的协议数据单元丢弃到可拥塞节点。

公开(公告)号：US06161193A

公开(公告)日：2000-12-12

申请号：US044054

申请日：1998-03-18

Applicant: Sachin Garg ,  Yennun Huang ,  Sampath Rangarajan

Inventor： Sachin Garg ,  Yennun Huang ,  Sampath Rangarajan

IPC: G06F11/14 ,  G06F11/20 ,  G06F11/00

CPC classification number: G06F11/203 ,  G06F11/1492 ,  G06F11/2035

Abstract: A distributed computing system includes a number of computers, workstations or other computing machines interconnected by a network. A non-interactive process arriving in a host machine of the system is migrated for execution to at least two remote machines. For example, first and second executions of the process may be started on respective first and second remote machines. One of the first and second executions of the process is then used to provide an on-demand checkpoint for the other execution of the process in the event the other execution is terminated, such that an additional execution of the process can be started from the on-demand checkpoint. This on-demand checkpointing is augmented with periodic checkpointing performed on at least one of the multiple executions of the process. The period of the periodic checkpointing for a given execution of the process may be fixed without regard to the status of the on-demand checkpointing for that execution, or alternatively may be reset each time an on-demand checkpoint is taken for that execution.

Abstract translation: 分布式计算系统包括由网络互连的多个计算机，工作站或其他计算机。 到达系统主机的非交互过程被迁移以执行至少两台远程机器。 例如，可以在相应的第一和第二远程机器上开始进程的第一和第二执行。 然后，该过程的第一次和第二次执行之一用于在其他执行终止的情况下为进程的另一个执行提供按需检查点，使得可以从开始执行进程的附加执行 -demand检查点。 在对进程的多个执行中的至少一个执行的周期性检查点处理上增加了该按需检查点。 对于给定执行过程的周期性检查点的周期可以是固定的，而不考虑该执行的按需检查点的状态，或者可以在每次对该执行采取按需检查点时被重置。