公开(公告)号：US07577548B1

公开(公告)日：2009-08-18

申请号：US11713561

申请日：2007-03-01

Applicant: Krzysztof W Przytula ,  Shubha Kadambe ,  Narayan Srinivasa

Inventor： Krzysztof W Przytula ,  Shubha Kadambe ,  Narayan Srinivasa

IPC: G06F19/00 ,  G06F17/40

CPC classification number: G06N7/005

Abstract: Described is a system for diagnosis and prognosis of a component. The system is configured to receive a signal from a component. The signal is representative of a current health observation of the component. The system also computes a present likelihood of the component failure based on the signal. Additionally, the system computes a future likelihood of failure of the component for a given future mission. Through diagnosis, a user can determine the present health of the component, and based on the present health and future mission, determine whether or not the component will fail in the future mission.

Abstract translation: 描述了用于组件的诊断和预后的系统。 系统被配置为从组件接收信号。 该信号代表组件的当前健康观察。 该系统还基于该信号计算组件故障的当前可能性。 此外，该系统计算组件在给定未来任务中的未来可能性。 通过诊断，用户可以确定组件的当前健康状况，并根据当前健康状况和未来任务，确定组件是否在将来的任务中失败。

公开(公告)号：US07379870B1

公开(公告)日：2008-05-27

申请号：US11051747

申请日：2005-02-03

Applicant: Robert Belvin ,  Shubha Kadambe

Inventor： Robert Belvin ,  Shubha Kadambe

IPC: G10L15/18

CPC classification number: G10L15/18

Abstract: The present invention relates to techniques for contextual filtering for improving an output of a speech recognizer. The techniques comprise receiving a representation of a speech utterance into a shallow parser component in the form of a first word lattice including a set of potential language matches for the utterance, where the language matches include at least one word; the parser component operative for receiving and analyzing the first word lattice to produce a second word lattice, and including a filter that assigns a probability match for a potential word match in relation to another potential word match in the utterance based on a particular filter-specific criteria; and outputting the probability matches from the shallow parser component as a portion of a second word lattice for further processing to determine most likely combinations of words present in the speech utterance.

Abstract translation: 本发明涉及用于改善语音识别器的输出的语境滤波技术。 这些技术包括以包括一组用于话语的潜在语言匹配的第一单词格式的形式，将语音发音的表示接收到浅解析器组件中，其中语言匹配包括至少一个单词; 所述解析器组件可操作用于接收和分析所述第一字阵列以产生第二字网格，并且包括滤波器，所述滤波器基于特定的特定滤波器特定来分配关于所述话语中的另一潜在词匹配的潜在词匹配的概率匹配 标准 并且将来自浅解析器组件的概率匹配作为第二字网格的一部分输出，用于进一步处理以确定语音话语中存在的词的最可能组合。

公开(公告)号：US07478071B2

公开(公告)日：2009-01-13

申请号：US10171280

申请日：2002-06-11

Applicant: Shubha Kadambe

Inventor： Shubha Kadambe

IPC: G06Q99/00

CPC classification number: G06K9/6217 ,  G06Q40/00 ,  G06Q40/06 ,  G06Q99/00

Abstract: A method, apparatus, and computer program product for determining and assessing information for collection from information sources for a desired level of decision accuracy are presented. Operations include: receiving a partial set of information; performing a minimax entropy-based test to determine a source with useful information; performing a mutual information or a conditional entropy-based test check minimax test validity. With an invalid result, the information source is excluded from further consideration and the minimax test is repeated; with a valid result, a cost/benefit analysis is determines whether to gather the information. If the cost/benefit analysis succeeds, the information is gathered. Otherwise, the information source is excluded and the minimax test is performed again. A consistency check ensures validity of the information prior to restarting the process. Thus, the set of information is iteratively augmented until there is no information to add or until adding information would be cost-prohibitive.

Abstract translation: 提出了一种用于确定和评估用于从信息源收集以获得期望水平的决策精度的信息的方法，装置和计算机程序产品。 操作包括：接收部分信息; 执行基于最小熵的测试以确定具有有用信息的源; 执行相互信息或基于条件熵的测试检查最小化测试有效性。 无效结果，信息源被排除在进一步考虑之外，并且重复进行极小值测试; 具有有效结果，成本/效益分析决定是否收集信息。 如果成本/效益分析成功，则收集信息。 否则，排除信息源，并再次执行最小化测试。 一致性检查确保重新启动过程之前信息的有效性。 因此，该信息集被迭代地增加，直到没有信息添加或直到添加信息将是成本高昂的。

公开(公告)号：US07899761B2

公开(公告)日：2011-03-01

申请号：US11113962

申请日：2005-04-25

Applicant: Shubha Kadambe ,  Leandro G. Barajas ,  Youngkwan Cho ,  Pulak Bandyopadhyay

Inventor： Shubha Kadambe ,  Leandro G. Barajas ,  Youngkwan Cho ,  Pulak Bandyopadhyay

IPC: G06F15/18

CPC classification number: G06K9/6297 ,  G05B23/0232 ,  G06K9/6219

Abstract: Disclosed herein are a system and method for trend prediction of signals in a time series using a Markov model. The method includes receiving a plurality of data series and input parameters, where the input parameters include a time step parameter, preprocessing the plurality of data series according to the input parameters, to form binned and classified data series, and processing the binned and classified data series. The processing includes initializing a Markov model for trend prediction, and training the Markov model for trend prediction of the binned and classified data series to form a trained Markov model. The method further includes deploying the trained Markov model for trend prediction, including outputting trend predictions. The method develops an architecture for the Markov model from the data series and the input parameters, and disposes the Markov model, having the architecture, for trend prediction.

Abstract translation: 这里公开了使用马尔可夫模型的时间序列中的信号趋势预测的系统和方法。 该方法包括接收多个数据序列和输入参数，其中输入参数包括时间步长参数，根据输入参数对多个数据序列进行预处理，以形成分类和分类数据序列，并处理分类和分类数据 系列。 该处理包括初始化用于趋势预测的马尔科夫模型，并训练马尔可夫模型用于仓位和分类数据序列的趋势预测，形成训练马尔可夫模型。 该方法还包括部署用于趋势预测的经过训练的马尔可夫模型，包括输出趋势预测。 该方法从数据系列和输入参数开发了Markov模型的架构，并将具有架构的马尔科夫模型用于趋势预测。

公开(公告)号：US07813654B1

公开(公告)日：2010-10-12

申请号：US11117656

申请日：2005-04-27

Applicant: Willie W. Ng ,  Robert R. Hayes ,  Shubha Kadambe

Inventor： Willie W. Ng ,  Robert R. Hayes ,  Shubha Kadambe

IPC: H04B10/04 ,  H04B10/12

CPC classification number: H04B10/505 ,  H04B10/58

Abstract: Transversal equalization is used to obtain broadband linearization of photonic modulation. A photonic link comprises a signal path and a feed-forward path. The feed-forward path includes an optical linearizer and a transversal equalizer connected with the optical linearizer. In this way, amplitude and phase matching of the error in the signal path is obtained over a wide bandwidth. This, in turn, enables a broadband enhancement of the link's spur free dynamic range (SFDR).

Abstract translation: 使用横向均衡来获得光子调制的宽带线性化。 光子链路包括信号路径和前馈路径。 前馈路径包括光学线性化器和与光学线性化器连接的横向均衡器。 以这种方式，在宽带宽上获得信号路径中的误差的幅度和相位匹配。 这反过来又能够实现链路无刺激动态范围（SFDR）的宽带增强。

公开(公告)号：US07085711B2

公开(公告)日：2006-08-01

申请号：US10007322

申请日：2001-11-09

Applicant: Shubha Kadambe

Inventor： Shubha Kadambe

IPC: G10L19/00 ,  G10L21/00 ,  H04Q7/20

CPC classification number: G06K9/624

Abstract: A data processing system blind source separation of an overcomplete set of signals generally includes means for storing input from sensors in a mixed signal matrix X 200, noise in a noise matrix V 202, an estimate of the individual signals from the mixture of signals from the signal sources in a source signal estimate matrix Ŝ 204, and an estimate of environmental effects in a mixing matrix  206, the matrices related by X=ÂŜ+V; generating an initial estimate of  208; determining the number of, and associated lines of correlation of, each source from Â, and representing the sources in the source signal estimate matrix Ŝ 210; jointly optimizing Ŝ and  in an iterative manner to generate an optimized source signal estimate matrix Ŝ 212 and a final estimated mixing matrix Â; and restoring the separated source signals from the optimized source signal estimate matrix Ŝ 214.

Abstract translation: 数据处理系统盲信号分离过度信号集合通常包括用于存储来自混合信号矩阵X 200中的传感器的输入的装置，噪声矩阵V 202中的噪声，来自所述信号的混合信号的各个信号的估计 源信号估计矩阵S 204中的信号源，以及混合矩阵206中的环境影响的估计，X = S + V相关的矩阵; 产生了208的初始估计值; 确定每个源的数量和相关的相关行，并表示源信号估计矩阵S 210中的源; 以迭代方式共同优化S和Â以产生优化的源信号估计矩阵S 212和最终的估计混合矩阵; 以及从优化的源信号估计矩阵S 214恢复分离的源信号。

公开(公告)号：US07620529B2

公开(公告)日：2009-11-17

申请号：US10685352

申请日：2003-10-14

Applicant: George C. Valley ,  Peter Petre ,  Shubha Kadambe ,  Todd S. Kaplan

Inventor： George C. Valley ,  Peter Petre ,  Shubha Kadambe ,  Todd S. Kaplan

IPC: G06F7/60 ,  G06F17/10

CPC classification number: G06F17/5036

Abstract: The present invention comprises a method, an apparatus, and a computer program product for simulating a mixed-signal system. The invention comprises a first operation of generating a matrix-based wavelet operator representation of equations characterizing a system, with the matrix-based wavelet operator representation including wavelet connection coefficients. A second operation is performed by selecting a number of wavelets and a set of wavelet basis functions with which to represent a performance of the system, whereby the wavelet operator, the number of wavelets and the set of wavelet basis functions represent a wavelet model of the system. A third operation is performed by iteratively applying the wavelet model over a series of clock cycles to develop a behavioral model of the system. The invention has particular use in the area of computer-aided design and may be applied to any suitable system, whether electrical, mechanical, or other.

Abstract translation: 本发明包括用于模拟混合信号系统的方法，装置和计算机程序产品。 本发明包括基于矩阵的小波算子表示包括小波连接系数的基于矩阵的小波运算符表示系统特征的方法的第一操作。 通过选择多个小波和一组用于表示系统的性能的小波基函数来执行第二操作，由此小波运算符，小波的数量和小波基函数集合表示小波基函数的小波模型 系统。 通过在一系列时钟周期上迭代地应用小波模型来开发系统的行为模型来执行第三操作。 本发明在计算机辅助设计领域中具有特别的用途，并且可以应用于任何合适的系统，无论是电气的，机械的还是其他的。

公开(公告)号：US07457745B2

公开(公告)日：2008-11-25

申请号：US10728106

申请日：2003-12-03

Applicant: Shubha Kadambe ,  Ron Burns ,  Markus Iseli

Inventor： Shubha Kadambe ,  Ron Burns ,  Markus Iseli

IPC: G10L19/00

CPC classification number: G10L15/07

Abstract: A fast on-line automatic speaker/environment adaptation suitable for speech/speaker recognition system, method and computer program product are presented. The system comprises a computer system including a processor, a memory coupled with the processor, an input coupled with the processor for receiving acoustic signals, and an output coupled with the processor for outputting recognized words or sounds. The system includes a model-adaptation system and a recognition system, configured to accurately and efficiently recognize on-line distorted sounds or words spoken with different accents, in the presence of randomly changing environmental conditions. The model-adaptation system quickly adapts standard acoustic training models, available on audio recognition systems, by incorporating distortion parameters representative of the changing environmental conditions or the speaker's accent. By adapting models already available to the new environment, the system does not need separate adaptation training data.

Abstract translation: 提出适用于语音/扬声器识别系统，方法和计算机程序产品的快速在线自动扬声器/环境适配器。 该系统包括计算机系统，包括处理器，与处理器耦合的存储器，与处理器耦合的用于接收声信号的输入，以及与处理器耦合以输出识别的字或声音的输出。 该系统包括模型适配系统和识别系统，配置为在存在随机变化的环境条件的情况下，准确有效地识别在线失真的声音或用不同口音说出的单词。 模型适应系统通过结合代表不断变化的环境条件或扬声器口音的失真参数，快速适应音频识别系统上可用的标准声学训练模型。 通过调整已经可用于新环境的模型，系统不需要单独的适应训练数据。

公开(公告)号：US20070255668A1

公开(公告)日：2007-11-01

申请号：US11413924

申请日：2006-04-27

Applicant: Shubha Kadambe ,  Qin Jiang

Inventor： Shubha Kadambe ,  Qin Jiang

IPC: G06F15/18 ,  G06E1/00

CPC classification number: G06K9/623

Abstract: The present invention describes a method, system, and computer program product for determining the minimum-dimension of a feature set that is needed for optimal signal representation. The present invention is configured to consider a set of N features to determine a minimum number of features for optimal signal representation. Once the minimum number of features for optimal signal representation is determined, the present invention determines the smallest subset of features that provides for optimal signal classification. Upon determining the smallest subset of features that provide for optimal signal classification, a user may provide those features to a signal classifier for signal classification.

Abstract translation: 本发明描述了用于确定最佳信号表示所需的特征集的最小维度的方法，系统和计算机程序产品。 本发明被配置为考虑一组N个特征以确定用于最佳信号表示的特征的最小数量。 一旦确定了用于最佳信号表示的特征的最小数量，本发明确定提供最佳信号分类的特征的最小子集。 在确定提供最佳信号分类的特征的最小子集时，用户可以将这些特征提供给用于信号分类的信号分类器。

公开(公告)号：US20060241927A1

公开(公告)日：2006-10-26

申请号：US11113962

申请日：2005-04-25

Applicant: Shubha Kadambe ,  Leandro Barajas ,  Youngkwan Cho ,  Pulak Bandyopadhyay

Inventor： Shubha Kadambe ,  Leandro Barajas ,  Youngkwan Cho ,  Pulak Bandyopadhyay

IPC: G06F13/10

CPC classification number: G06K9/6297 ,  G05B23/0232 ,  G06K9/6219

Abstract: Disclosed herein are a system and method for trend prediction of signals in a time series using a Markov model. The method includes receiving a plurality of data series and input parameters, where the input parameters include a time step parameter, preprocessing the plurality of data series according to the input parameters, to form binned and classified data series, and processing the binned and classified data series. The processing includes initializing a Markov model for trend prediction, and training the Markov model for trend prediction of the binned and classified data series to form a trained Markov model. The method further includes deploying the trained Markov model for trend prediction, including outputting trend predictions. The method develops an architecture for the Markov model from the data series and the input parameters, and disposes the Markov model, having the architecture, for trend prediction.

Abstract translation: 这里公开了使用马尔可夫模型的时间序列中的信号趋势预测的系统和方法。 该方法包括接收多个数据序列和输入参数，其中输入参数包括时间步长参数，根据输入参数对多个数据序列进行预处理，以形成分类和分类数据序列，并处理分类和分类数据 系列。 该处理包括初始化用于趋势预测的马尔科夫模型，并训练马尔可夫模型用于仓位和分类数据序列的趋势预测，形成训练马尔可夫模型。 该方法还包括部署用于趋势预测的经过训练的马尔可夫模型，包括输出趋势预测。 该方法从数据系列和输入参数开发了Markov模型的架构，并将具有架构的马尔科夫模型用于趋势预测。