Using Matrix Operation Patents (Class 702/196)
-
Patent number: 6853961Abstract: Process for estimating the correlation matrix of signals of unknown characteristics in an array of N sensor. The process determines the correlation matrix {circumflex over (R)}s of the signals of known characteristics, estimates the correlation matrix {circumflex over (R)}x of the sensor signals, forms a matrix A equal to A={circumflex over (R)}s?1/2{circumflex over (R)}x {circumflex over (R)}s?1/2 and decomposes the matrix A into eigenelements, constructs a matrix B based on the eigenelements of the matrix A, and determines the estimated correlation matrix {circumflex over (R)}b of the interfering signals based on the estimated matrix {circumflex over (R)}s and of the matrix B, such that {circumflex over (R)}b={circumflex over (R)}s1/2 B {circumflex over (R)}s1/2. Such a process may find application to antenna processing.Type: GrantFiled: March 16, 2001Date of Patent: February 8, 2005Assignee: ThalesInventors: Pascal Chevalier, François Pipon
-
Patent number: 6847917Abstract: Disclosed are techniques used in connection with determining a health indicator (HI) of a component, such as that of an aircraft component. The HI is determined using condition indicators (CIs) which parameterize characteristics about a component minimizing possibility of a false alarm. Different algorithms are disclosed which may be used in determining one or more CIs. The HI may be determined using a normalized CI value. Techniques are also described in connection with selecting particular CIs that provide for maximizing separation between HI classifications. Given a particular HI at a point in time for a component, techniques are described for predicting a future state or health of the component using the Kalman filter. Techniques are described for estimating data values as an alternative to performing data acquisitions, as may be used when there is no pre-existing data.Type: GrantFiled: December 4, 2001Date of Patent: January 25, 2005Assignee: Simmonds Precision Products, Inc.Inventor: Eric Robert Bechhoefer
-
Patent number: 6829568Abstract: A method and apparatus is described for fusing a plurality of signals corresponding to estimates of the state of an object, system, or process. The method and apparatus is specialized or programmed for (1) receiving estimates, each of which can be expressed in the form of a state vector and an error covariance matrix, at least one estimate of which can be decomposed into a sum of an independent error covariance matrix and a potentially correlated error covariance matrix, and (2) transmitting a resulting signal corresponding to an estimate, which can be expressed in the form of a state vector and an error matrix, in order to evoke a physical response from a system receiving the signal. The method and apparatus provides advantages over the prior art by improving the accuracy of fused estimates while guaranteeing consistency.Type: GrantFiled: April 26, 2002Date of Patent: December 7, 2004Inventors: Simon Justin Julier, Jeffrey Kent Uhlmann
-
Patent number: 6826575Abstract: A system and method for selecting forecasting performance metrics from among a plurality of performance metrics includes a data mining kernel that receives performance metrics from a collection agent. The data mining kernel performs a factor analysis on the performance metrics to identify one or more factors and associates each performance metric with a corresponding one of the one or more factors. For each of the one or more factors, the data mining kernel selects the performance metrics having a weight greater than a predetermined value thus pairing the selected performance metrics down to the forecasting performance metrics. The forecasting performance metrics are then stored in a database for access by a capacity planning system.Type: GrantFiled: September 13, 2001Date of Patent: November 30, 2004Assignee: MCI, Inc.Inventor: Anthony C. Waclawski
-
Publication number: 20040225478Abstract: The aim of the present invention is to estimate EDR directions in a single index model composed of a large number of variables with simple calculations without using the inverse matrix of the variance-covariance matrix and principle component analysis. Data conversion means 21 receives, from an input device 3, data to be analyzed, the data composed of sets of response variables and explanatory variables, standardizes the explanatory variables, and sends them to slice average calculating means 22. The slice average calculating means 22 divides the data into two slices with reference to the median of the response variables to calculate the mean vector of the explanatory variables on a slice basis. The calculated mean vectors are sent to EDR direction calculating means 23. The EDR direction calculating means 23 calculates the difference between the mean vectors for respective slices to estimate an EDR direction.Type: ApplicationFiled: October 30, 2003Publication date: November 11, 2004Applicants: NEC CORPORATION, Megu OhtakiInventors: Masataka Andoh, Akira Saitoh, Megu Ohtaki, Kenichi Satoh
-
Patent number: 6807510Abstract: A process for coordinated control of machine direction MD and cross direction CD actuators in a sheetmaking machine for manufacturing a sheet of material is disclosed. The process involves measuring a plurality of sheet properties at regular intervals to collect sheet measurement data. The sheet measurement data is manipulated to establish a plurality of sheet property measurement arrays, which are then mapped to a common resolution. The common resolution sheet property measurement arrays are concatenated into one larger one-dimensional common resolution measurement array. The common resolution measurement array and an array of past changes in actuator set point are used as inputs to a paper machine process model state observer to generate the estimated current internal state of the sheet manufacturing process. A plurality of future-sheet property target arrays are concatenated into one target array.Type: GrantFiled: May 5, 2003Date of Patent: October 19, 2004Assignee: Honeywell ACSA Inc.Inventors: Johan U. Backstrom, Pengling He
-
Patent number: 6807517Abstract: A method for estimating one or more parameters of a propagation channel, with a priori knowledge of at least one signal, in a system having one or more sensors. The method includes correlating one or more signals received by the sensors with the known signal, sampling the received signals at a sampling period and selecting a number of samples per concatenation, and determining at least one parameter of the propagation channel which enables the most efficient reconstruction of the signals via a maximum likelihood method.Type: GrantFiled: February 1, 2002Date of Patent: October 19, 2004Assignee: ThalesInventors: Pascal Chevalier, Marc Chenu-Tournier, Claude Adnet, Pascal Larzabal
-
Publication number: 20040181351Abstract: Described are techniques for processing data sets produced by analyzing a sample. The input data set is represented as rows of intensities over time for a particular mass to charge (m/z) range. A correlation matrix is produced in which each row of the input data set is correlated with every other row in the input data set. The correlation matrix is clustered or grouped such that those highly correlated m/z ranges are included in the same group. A set of one or more scans is selected for each group representing periods of interest within each group. Using the m/z values included in each cluster, a resultant sample spectra is created for each of the selected scans. The processing techniques may be used to identify parent and related fragment ions in the input data set and as a preprocessor producing resultant sampled spectra used as input to subsequent processing.Type: ApplicationFiled: June 24, 2003Publication date: September 16, 2004Inventors: Dean R. Thompson, William M. Old, David Lee Gines
-
Publication number: 20040073408Abstract: The invention relates to a method for identifying the state (point of rupture 2) of electrically conductive oblong tensioning elements (1) involving the following steps: launching an electromagnetic measurment signal into a tensioning element (1a); changing the frequency; measuring the reflection spectrum, and; indentifying the state of the tensioning element (1a) according to the resonance frequencies. Said signal is lauched on the fore-part or on the periphery. In the event of coupled tensioning element (1), a scattering matrix system of equations is iteratively devised. The invention is used in the construction industry for prestressed concrete structures and rear anchored systems.Type: ApplicationFiled: October 29, 2003Publication date: April 15, 2004Inventors: Bert Jannsen, Arne Jacob, Harald Budelmann, Guido Schneider, Hans-Joachim Wichmann, Karim Hariri
-
Patent number: 6721692Abstract: An LED-based spectrophotometer uses a reconstruction algorithm, based on spectral information of an illumination source and a reference spectrophotometer, to convert integrated multiple illuminant measurements from a non-fully illuminant populated color sensor into a fully populated spectral curve using a reference database. A dynamic, Least Squares-based spectral reconstruction algorithm, used to reconstruct spectra, gives greater importance to the data from the reference database in the neighborhood of the color sample under measurement. This is done using linear operators.Type: GrantFiled: August 30, 2001Date of Patent: April 13, 2004Assignee: Xerox CorporationInventors: Lingappa K. Mestha, Sohail A. Dianat
-
Patent number: 6721693Abstract: The correction device comprises a block for decomposition into total proper elements, according to the Kato spectral decomposition method, of the covariance matrix to be corrected. The block supplies to a correction block the total proper values and the total projectors of the covariance matrix, as well as first coefficient vectors for the combination of the source signals of the covariance matrix. The correlation block estimates the reconstructed variances of combinations of the source signals using the first coefficient vectors and second used-defined coefficient vectors. It looks for the corrected matrix closest to the matrix to be corrected, in relation to the variances and total projectors while is applying a quasi-Newton type method to the dual of a semi-defined lesser error square programme.Type: GrantFiled: September 18, 2002Date of Patent: April 13, 2004Assignee: Raise PartnerInventor: François Oustry
-
Publication number: 20040054507Abstract: A reference matrix contains valid measurements characterizing operation of a multivariate process (220). Modeling parameters of the reference matrix are derived (222-232). The final model parameters, balanced with respect to measuring and modeling uncertainties (234, 202), are applied to model (204) a new set of measurements (200). If the new set has no faults (206) then all modeled values and modeling uncertainties (208) can be used to control the process (218). If the new set has only one fault (210) then the modeled value and modeling uncertainty of the faulted measurement plus the measured values and measuring uncertainties of the unfaulted measurements (212) can be used to control the process (218) while repair procedures are implemented for the identified fault (216). If the new set has more than one fault (214) then the process (218) should be shut down, and repair procedures should be implemented (216) for all identified faults.Type: ApplicationFiled: September 24, 2003Publication date: March 18, 2004Inventor: Jack Edward Mott
-
Patent number: 6694280Abstract: A method for overflow testing of a blind equalizer. The method uses the following steps; providing a primary located signal in a primary signaling point of a period by a signal generating loop; multiplying the primary located signal by a set of continuously decreasing signals to get a primary signal; providing a interfering located signal in a plurality of interfering signaling points of the period by a ISI generating loop, wherein the interfering signaling points are different from the primary signaling point; multiplying the interfering located signal by a set of continuously increasing/decreasing signals to get an interference signal; adding the primary signal and the interference signal to get the input signal; inputting the input signal to the blind equalizer to adapt the input signal, wherein the tap weights corresponding to the interference signal in a blind equalizer overflow.Type: GrantFiled: April 4, 2002Date of Patent: February 17, 2004Assignee: Faraday Technology CorporationInventor: Jyh-Ting Lai
-
Patent number: 6675106Abstract: A method of determining the properties of a sample from measured spectral data collected from the sample by performing a multivariate spectral analysis. The method can include: generating a two-dimensional matrix A containing measured spectral data; providing a weighted spectral data matrix D by performing a weighting operation on matrix A; factoring D into the product of two matrices, C and ST, by performing a constrained alternating least-squares analysis of D=CST, where C is a concentration intensity matrix and S is a spectral shapes matrix; unweighting C and S by applying the inverse of the weighting used previously; and determining the properties of the sample by inspecting C and S. This method can be used to analyze X-ray spectral data generated by operating a Scanning Electron Microscope (SEM) with an attached Energy Dispersive Spectrometer (EDS).Type: GrantFiled: June 1, 2001Date of Patent: January 6, 2004Assignee: Sandia CorporationInventors: Michael R. Keenan, Paul G. Kotula
-
Publication number: 20030216949Abstract: A monitoring and reporting system employing field sensor packages with telemetry to a processing system having input software creating cardinal matrices for the sensor data and conversion elements for compatibility with a geographical information system (GIS) to produce real-time generate geostatistically rendered contour diagrams that display the spatial and temporal distribution of environmental parameters of interest.Type: ApplicationFiled: April 4, 2003Publication date: November 20, 2003Inventors: Mark Kram, Sanya Sirivithayapakorn, Ralph Edward Beighley
-
Patent number: 6647790Abstract: A method and system for dynamically balancing a rotating system or rotating device utilizing data collection and correlation techniques to extract balance measures from sensor data is disclosed. Sensor data from the rotating system, containing information indicative of the dynamics of the rotating system, are filtered and sampled. A select time span of sampled data, adjusted upward to a nearest number of full revolutions of the rotating system, which is of sufficient size to extract balance measures, is collected. Data may be collected as a fixed number of samples per revolution, independent of a rotational speed, so as to simplify later computation. Several unique select time spans may be used for different ranges of rotational speeds so as to maintain the amount of data to be processed while still serving its original purpose. Window function may be applied to modify sampled data in the select time span so as to improve extraction performance.Type: GrantFiled: November 15, 2001Date of Patent: November 18, 2003Assignee: Honeywell International Inc.Inventor: Kevin J. Stalsberg
-
Patent number: 6651018Abstract: In a method for correcting a calibration table T(n,k) of a CT apparatus that contains calibration values, the CT apparatus having a detector system with N≧2 rows of detector elements following one another in the z-direction that include a first active row of detector elements in the z-direction and a last active row of detector elements in the z-direction, for correcting the calibration values of the aforementioned first and last active rows of detector elements, a reference vector R(k) is produced, the error is determined with respect to the first and last active rows of detector elements, and the error F(n,k) of the first and last active rows of detector elements is subtracted from the corresponding calibration values of the calibration table T(n,k) for determining corrected calibration values Tcor(n,k) with respect to the first and last active rows of detector elements.Type: GrantFiled: March 5, 2002Date of Patent: November 18, 2003Assignee: Siemens AktiengesellschaftInventors: Helmut Kropfeld, Karl Schwarz
-
Patent number: 6647343Abstract: A sample may be analyzed by sampling a quantity of the sample so as to provide a respective quantity of vapor phase molecules of the respective complex sample to a, mass sensor; performing a succession of scans of the sample, so as to generate a respective series of mass spectra; compiling a k×n vector matrix representative of a plurality of temporal profiles derived from the plurality of mass spectra; performing one or more multivariate data analysis routines with respect to the vector matrix; and, in response to the multivariate data analysis routine, reporting the results of the analysis.Type: GrantFiled: April 29, 1999Date of Patent: November 11, 2003Assignee: Agilent Technologies, Inc.Inventors: Edward J. Guthrie, Charles R. Knipe, Jr.
-
Publication number: 20030204382Abstract: A method and apparatus is described for fusing a plurality of signals corresponding to estimates of the state of an object, system, or process. The method and apparatus is specialized or programmed for (1) receiving estimates, each of which can be expressed in the form of a state vector and an error covariance matrix, at least one estimate of which can be decomposed into a sum of an independent error covariance matrix and a potentially correlated error covariance matrix, and (2) transmitting a resulting signal corresponding to an estimate, which can be expressed in the form of a state vector and an error matrix, in order to evoke a physical response from a system receiving the signal. The method and apparatus provides advantages over the prior art by improving the accuracy of fused estimates while guaranteeing consistency.Type: ApplicationFiled: April 26, 2002Publication date: October 30, 2003Inventors: Simon Justin Julier, Jeffrey Kent Uhlmann
-
Patent number: 6634207Abstract: The invention is a method and apparatus for improving the accuracy of an inertial navigation system. The method comprises (1) obtaining a measure of the angular velocity of a body frame of reference having a first axis, a second axis, and a third axis, (2) obtaining a measure of the acceleration of a first reference point in the direction of the first axis, a second reference point in the direction of the second axis, and a third reference point in the direction of the third axis, the first, second, and third reference points being fixed in the body frame, and (3) determining compensated acceleration values. A compensated acceleration value is the difference of the measure of acceleration of a reference point and a compensation quantity. A compensation quantity is an estimate of the portion of the acceleration of the reference point resulting from the rotation of the body frame. The method further comprises establishing the optimum navigation center based on a criterion of goodness.Type: GrantFiled: October 10, 2000Date of Patent: October 21, 2003Assignee: Litton Systems, Inc.Inventors: Brian T. Lottman, Daniel A. Tazartes
-
Patent number: 6625569Abstract: A general method of anomaly detection from time-correlated sensor data is disclosed. Multiple time-correlated signals are received. Their cross-signal behavior is compared against a fixed library of invariants. The library is constructed during a training process, which is itself data-driven using the same time-correlated signals. The method is applicable to a broad class of problems and is designed to respond to any departure from normal operation, including faults or events that lie outside the training envelope.Type: GrantFiled: March 6, 2002Date of Patent: September 23, 2003Assignee: California Institute of TechnologyInventors: Mark L. James, Ryan M. E. Mackey, Han G. Park, Michail Zak
-
Patent number: 6618696Abstract: The present invention is a method of simulating the effects of a plurality of channels on a signal that includes the steps of acquiring a state transition matrix for each of said plurality of channels; acquiring an error matrix for each of said plurality of channels; selecting the first channel to be simulated; assuming that the signal is in a particular state; receiving the signal; generating a first number; determining the state to which the signal transitions; transitioning the signal to the state determined in the last step; generating a second number; determining what errors, if any, to inject into the signal by comparing the second number to entries in the column of the error matrix of the corresponding channel that matches the state of the signal; if one of the errors determined in the last step is lost signal then discarding the signal, not injecting any other error into the signal, selecting another channel if the user desires, assuming that the next signal selected is in the same state to which theType: GrantFiled: June 14, 1999Date of Patent: September 9, 2003Assignee: The United States of America as represented by the National Security AgencyInventors: Richard A. Dean, Steven William Roberts, Mark George Jacobs, Ronald E. Krebs
-
Patent number: 6615163Abstract: A method for developing testing configurations and a computer system operable to implement the same. The method includes the steps of identifying factors to be tested and levels for each factor; selecting a Hadamard matrix of an appropriate size such that for any integer 1, 2 . . . n, the Hadamard matrix has a size 2n and at least (2n−1) columns, and maintains orthogonality of at least strength n when columns are selected sequentially; and creating a configuration matrix from the selected Hadamard matrix. The configuration matrix is created by first creating an integer matrix from the Hadamard matrix by combining columns of the Hadamard matrix where necessary, and replacing unassigned variables in the integer matrix in a manner that improves coverages. In one embodiment, a frequency matrix and a proportionality matrix are used to improve coverage when replacing unassigned variables.Type: GrantFiled: December 13, 1999Date of Patent: September 2, 2003Assignee: Dell USA, L.P.Inventors: Hamid Rasoulian, Stephen A. Osella
-
Patent number: 6594606Abstract: A method and apparatus to calibrate an LED matrix display such that a driver will provide a proper precharge voltage to LED elements within the display during a scan period. A current is driven through a calibration element, and a voltage reflecting the steady-state element voltage is measured and stored as a calibration value. A processor controls whether to precharge during the calibration cycle, and determines when the calibration cycle is completed. During subsequent normal scans, a driver applies a voltage based on the stored calibration value to rapidly precharge parasitic capacitance associated with a display element to a proper value, and also drives a selected current through the device.Type: GrantFiled: May 9, 2001Date of Patent: July 15, 2003Assignee: Clare Micronix Integrated Systems, Inc.Inventor: James Everitt
-
Patent number: 6584435Abstract: An LED-based spectrophotometer uses a reconstruction algorithm, based on the spectral characteristics of the illumination source and the color sensing system, to convert integrated multiple illuminant measurements from a non-fully illuminant populated color sensor into a fully populated spectral curve using a reference database. The reference database contains training samples that indicate reflectance spectra and their corresponding LED sensor output. A dynamic, Karhunen-Loeve-based (DKL) spectral reconstruction algorithm, used to reconstruct spectra, gives greater importance to the data from the training samples in the neighborhood of the color sample under measurement. This is done using linear operators and basis vectors.Type: GrantFiled: August 30, 2001Date of Patent: June 24, 2003Assignee: Xerox CorporationInventors: Lingappa K. Mestha, Sohail A. Dianat
-
Patent number: 6584413Abstract: An apparatus and system for determining the properties of a sample from measured spectral data collected from the sample by performing a method of multivariate spectral analysis. The method can include: generating a two-dimensional matrix A containing measured spectral data; providing a weighted spectral data matrix D by performing a weighting operation on matrix A; factoring D into the product of two matrices, C and ST, by performing a constrained alternating least-squares analysis of D=CST, where C is a concentration intensity matrix and S is a spectral shapes matrix; unweighting C and S by applying the inverse of the weighting used previously; and determining the properties of the sample by inspecting C and S. This method can be used by a spectrum analyzer to process X-ray spectral data generated by a spectral analysis system that can include a Scanning Electron Microscope (SEM) with an Energy Dispersive Detector and Pulse Height Analyzer.Type: GrantFiled: June 1, 2001Date of Patent: June 24, 2003Assignee: Sandia CorporationInventors: Michael R. Keenan, Paul G. Kotula
-
Patent number: 6564164Abstract: A method for optimally determining sensor positions for an acoustic pyrometry, capable of obtaining a minimum number of sensors satisfying a given error limit and determining positions of the obtained number of sensors for error minimization, in which a predetermined number of sensors are arbitrarily arranged on sides of a rectangular sectional area to be measured and a transfer matrix is calculated with respect to positions of all the sensors, with effective independence values obtained for respective paths of each of the sensors by performing singular value decomposition with respect to the calculated transfer matrix and a total effective independence value of each of the sensors is obtained by summing up the effective independence values obtained for the respective paths, followed by having a smallest one of the obtained total effective independence valves removed from among the sensors and the above steps are repeated until the number of the remaining sensors becomes equal to a target valve.Type: GrantFiled: October 18, 2000Date of Patent: May 13, 2003Inventors: Jeong Guon Ih, Chul Min Park
-
Publication number: 20030088387Abstract: A method of measuring similarity of a first object represented by first set of feature values to a second object represented by a second set of feature values, comprising determining respective feature distance values between substantially all corresponding feature values of the first and second sets of feature values, selecting a subset of the determined feature distance values in which substantially all feature distance values that are selected to be within the subset are smaller in value than feature distance values that are not selected to be within the subset, and summing the feature distance values in the subset to produce a partial feature distance measure between the first and second objects.Type: ApplicationFiled: September 24, 2002Publication date: May 8, 2003Inventors: Edward Y. Chang, Beitao Li
-
Publication number: 20030088386Abstract: The correction device comprises a block for decomposition into total proper elements, according to the Kato spectral decomposition method, of the covariance matrix to be corrected. The block supplies to a correction block the total proper values and the total projectors of the covariance matrix, as well as first coefficient vectors for the combination of the source signals of the covariance matrix. The correlation block estimates the reconstructed variances of combinations of the source signals using the first coefficient vectors and second used-defined coefficient vectors. It looks for the corrected matrix closest to the matrix to be corrected, in relation to the variances and total projectors while is applying a quasi-Newton type method to the dual of a semi-defined lesser error square programme.Type: ApplicationFiled: September 18, 2002Publication date: May 8, 2003Applicant: RAISE PARTNERInventor: Francois Oustry
-
Publication number: 20030055611Abstract: An LED-based spectrophotometer uses a reconstruction algorithm, based on spectral information of an illumination source and a reference spectrophotometer, to convert integrated multiple illuminant measurements from a non-fully illuminant populated color sensor into a fully populated spectral curve using a reference database. A dynamic, Least Squares-based spectral reconstruction algorithm, used to reconstruct spectra, gives greater importance to the data from the reference database in the neighborhood of the color sample under measurement. This is done using linear operators.Type: ApplicationFiled: August 30, 2001Publication date: March 20, 2003Applicant: Xerox CorporationInventors: Lingappa K. Mestha, Sohail A. Dianat
-
Publication number: 20030050768Abstract: An LED-based spectrophotometer uses a reconstruction algorithm, based on the spectral characteristics of the illumination source and the color sensing system, to convert integrated multiple illuminant measurements from a non-fully illuminant populated color sensor into a fully populated spectral curve using a reference database. The reference database contains training samples that indicate reflectance spectra and their corresponding LED sensor output. A dynamic, Karhunen-Loeve-based (DKL) spectral reconstruction algorithm, used to reconstruct spectra, gives greater importance to the data from the training samples in the neighborhood of the color sample under measurement. This is done using linear operators and basis vectors.Type: ApplicationFiled: August 30, 2001Publication date: March 13, 2003Applicant: XEROX CORPORATIONInventors: Lingappa K. Mestha, Sohail A. Dianat
-
Patent number: 6529859Abstract: The phase noise of an oscillator described by a known set of differential algebraic equations (DAEs) can be predicted by a) finding the steady state waveform of the oscillator, e.g., by using harmonic balance techniques or so-called “shooting” techniques, either of which involves developing a mathematical quantity known as the augmented Jacobian matrix; b) solving a prescribed linear system of equations that uses the augmented Jacobian matrix, the solution being called a perturbation projection vector (PPV), c) plotting a graph of the phase noise of the oscillator as a Lorentzian function of the solution of the prescribed linear system of equations. The prescribed linear system of equations which is used is the system of equations formed by setting a matrix product equal to a unit vector.Type: GrantFiled: October 1, 1999Date of Patent: March 4, 2003Assignee: Agere Systems Inc.Inventors: Alper Demir, Jaijeet Shankar Roychowdhury
-
Patent number: 6522996Abstract: Principal features are inventively derived from a non-stationary time series signal, and can be used to classify the attributes of the signal. Intrinsic to the invention is its recognition of the value of information contained within singular vectors. The invention therefore seeks to render singular vectors as proper density functions for purposes of formulating at least one feature of a signal, which is or has been sensed. To this end, the signal is typically evaluated thusly according to the invention: a time-frequency distribution (e.g., positive time-frequency distribution) matrix is generated; the time-frequency distribution matrix is decomposed; an element-by-element square of singular vectors is performed; the non-principal singular values are sorted and truncated; density functions are obtained; the density functions are normalized; and, at least one feature from the normalized density functions is formulated.Type: GrantFiled: January 18, 2000Date of Patent: February 18, 2003Assignee: The United States of America as represented by the Secretary of the NavyInventor: F. Dale Groutage
-
Patent number: 6487524Abstract: Systems and methods for designing or determining parameters for a system are disclosed. These systems and methods exploit the Tensor Convolution Block Toeplitz (TCBT) structure involved in certain systems wherein a symmetric positive definite linear system of equations needs to be solved in the design or determination of the parameters for the system. Further, these methods and systems employ a TCBT-PCG process that uses a fast matrix-vector multiply in order to improve the computational efficiency in the design or determination of these system parameters.Type: GrantFiled: June 8, 2000Date of Patent: November 26, 2002Assignee: BBNT Solutions LLCInventor: Robert D. Preuss
-
Patent number: 6476605Abstract: Inductive sensor for non-contact detection of discontinuities of a conductive or ferromagnetic target for determination of target position, movement and speed, or generally of a magnetic image, including a plurality of receiving secondary windings associated with each exciting primary winding, configured to obtain electric signals optimized as a function of the discontinuities of the target. The sensor, combined with a suitable electric measuring circuit, allows global analysis of the signals provided by the secondary windings and configuration of a network in accordance with a predefined selection. An adaptive circuit allows auto-calibration of the sensor.Type: GrantFiled: March 13, 2000Date of Patent: November 5, 2002Assignee: CSEM Centre Suisse d'Electronique et de Microtechnique SAInventor: Yves de Coulon
-
Patent number: 6477485Abstract: Sensors (12) attached to various parameters of a system (11), such as an elevator system or a chiller, provide values (14) of corresponding parameters which are utilized to build (15) an empirical distribution of the process, such as by means of bootstrapping methodology using a five-dimensional Markov chain model. In normal operation thereafter, the sensors are read periodically and in response to events, and an abnormality is determined by comparison of current information against the empirical distribution of the process. Deviations from normal behavior provide quantitative measure of system malfunction or abnormality; eliminating data from one or more sensors in each iteration of processing identifies one or more sensors associated with the abnormality. By utilizing cumulative distribution norm of deviation from normal behavior, the relative health of one system can be compared with the relative health of other, similar or dissimilar systems.Type: GrantFiled: October 27, 2000Date of Patent: November 5, 2002Assignee: Otis Elevator CompanyInventors: Dragan Radulovic, Robert E. LaBarre
-
Publication number: 20020161560Abstract: An impluse response hmn(q) of each transmission path is estimated from N received signals rm (m=1, . . . , M) and a known signal (for a number of users equal to N, n=1, . . . , N). M×N matrix H (q) having hmn(q) as an element and a Q×Q matrix H having H(q) as an element are determined (where Q represents a number of multipaths of each transmitted wave and q=0, . . . , Q−1). A soft decision value b′n(k) is determined from decoded &lgr;2 [bn(k)], and this is used to generate an interference component matrix B′(k) to generate an interference replica H·B′(k). The interference replica H·B′(k) is subtracted from a received matrix y(k) to determine y′(k). y(k) and H are used to determine an adaptive filter coefficient wn(k) to be applied to an n-th user in order to eliminate residual interference components in y′(k) according to the minimum mean square error criteria.Type: ApplicationFiled: February 19, 2002Publication date: October 31, 2002Applicant: NTT DoCoMo, IncInventors: Tetsushi Abe, Hiromasa Fujii, Shigeru Tomisato, Tasashi Matsumoto
-
Patent number: 6415247Abstract: A system identification device is provided capable of performing a sequential updating process and preventing the quantity of operation from becoming too complex. The system identification device includes a first delay for delay of y2 (n), a matrix generator, a plurality of matrix calculators, a second delay, a third delay and a matrix separator for identifying an unknown system.Type: GrantFiled: September 7, 1999Date of Patent: July 2, 2002Assignee: Matsushita Electric Industrial Co., Ltd.Inventors: Takashi Kimura, Hideaki Sasaki, Hiroshi Ochi
-
Publication number: 20020077789Abstract: In a simultaneous-linear-equations solving method of calculating the numerical solutions of simultaneous linear equations having a coefficient matrix, all the elements of coefficient matrix elements including zero elements and all the elements of right-side vector elements are stored into an array. Next, a non-zero-structure-specifying index table is created which indicates the row number of a terminal-end non-zero element in each column and the column number of a terminal-end non-zero element in each row within the array. Moreover, a decomposition processing is executed toward the elements existing within a range indicated by the created index table. Finally, a forward/backward substitution processing is executed toward the coefficient matrix elements subjected to the decomposition processing and the right-side vector elements stored into the array, thereby determining the numerical solutions.Type: ApplicationFiled: October 19, 2001Publication date: June 20, 2002Applicant: Hitachi, Ltd.Inventors: Saori Mitsunaga, Schinichi Tanaka, Hiroki Kawamura
-
Patent number: 6377907Abstract: A system and method for selecting forecasting performance metrics from among a plurality of performance metrics includes a data mining kernel that receives performance metrics from a collection agent. The data mining kernel performs a factor analysis on the performance metrics to identify one or more factors and associates each performance metric with a corresponding one of the one or more factors. For each of the one or more factors, the data mining kernel selects the performance metrics having a weight greater than a predetermined value thus pairing the selected performance metrics down to the forecasting performance metrics. The forecasting performance metrics are then stored in a database for access by a capacity planning system.Type: GrantFiled: November 17, 1999Date of Patent: April 23, 2002Assignee: MCI WorldCom, Inc.Inventor: Anthony C. Waclawski
-
Patent number: 6370490Abstract: In determining an intrinsic spectrum from a measured spectrum using the Maximum Entropy Algorithm, it is hardly or even not at all practical to determine the eigenvalues of an N×N matrix of large dimensions (N of the order of from 104 to 105). According to the invention such a large matrix is subdivided into a large number of much smaller partial matrices that are located on the diagonal or trace of the large matrix. The set of eigenvalues to be determined then consists of all eigenvalues of the partial matrices which can be determined much faster. Because of the Toeplitz-like character of the partial matrices, their eigenvalues can be determined very fast by Fourier transformation of a single row of such a matrix. Using the set of eigenvalues thus obtained, the intrinsic spectrum is determined by means of a minimizing algorithm.Type: GrantFiled: June 12, 2000Date of Patent: April 9, 2002Assignee: U.S. Philips CorporationInventors: Derk Reefman, William Jacques Jean Rey, Augustus Josephus Elizabeth Maria Janssen
-
Patent number: 6317662Abstract: The stability of a recursive estimator process (e.g., a Kalman filter is assured for long time periods by periodically resetting an error covariance P(tn) of the system to a predetermined reset value Pr. The recursive process is thus repetitively forced to start from a selected covariance and continue for a time period that is short compared to the system's total operational time period. The time period in which the process must maintain its numerical stability is significantly reduced as is the demand on the system's numerical stability. The process stability for an extended operational time period To is verified by performing the resetting step at the end of at least one reset time period Tr whose duration is less than the operational time period To and then confirming stability of the process over the reset time period Tr.Type: GrantFiled: July 24, 2000Date of Patent: November 13, 2001Assignee: Hughes Electronics CorporationInventors: Rongsheng Li, Yeong-Wei Andy Wu
-
Patent number: 6292760Abstract: A method of measuring non-coherent electrical signals using either windowed or non-windowed digital signal processing. The method includes the steps of providing a digitized version of the non-coherent electrical signal; generating a matrix A of correlations between sine and cosine components of known correlation frequencies and sine and cosine components of signal frequencies; generating an inverse matrix A−1 of the correlation matrix; generating a second matrix B of correlations between sine and cosine components of the correlation frequencies and the digitized non-coherent electrical signal; and generation of a third matrix C which represents the measured amplitudes of the sine and cosine components of the non-coherent electrical signal.Type: GrantFiled: August 11, 1998Date of Patent: September 18, 2001Assignee: Texas Instruments IncorporatedInventor: Mark Burns
-
Patent number: 6249755Abstract: A computer implemented method is provided for determining the source of a problem in a complex system of managed components based upon symptoms. The problem source identification process is split into different activities. Explicit configuration non-specific representations of types of managed components, their problems, symptoms and the relations along which the problems or symptoms propagate are created that can be manipulated by executable computer code. A data structure is produced for determining the source of a problem by combining one or more of the representations based on information of specific instances of managed components in the system. Computer code is then executed which uses the data structure to determine the source of the problem from one or more symptoms.Type: GrantFiled: July 15, 1997Date of Patent: June 19, 2001Assignee: System Management Arts, Inc.Inventors: Yechiam Yemini, Shaula Yemini, Shmuel Kliger
-
Patent number: 6202033Abstract: The invention is based on the use of the principles of Lange's Fast Kalman-Filtering (FKF™) for large process control, prediction or warning systems where other computing methods are either too slow or fail because of truncation errors. The invented method makes it possible to exploit the FKF method for adaptive Kalman Filtering of dynamic multiparameter systems with large moving data-windows.Type: GrantFiled: May 15, 1998Date of Patent: March 13, 2001Inventor: Antti A. Lange
-
Patent number: 6094626Abstract: A method and system for the identification of genetic information from a polynucleotide sequence is described. Genetic information from a raw polynucleotide sequence is identified by assigning values to nucleotide base changes along the raw polynucleotide sequence and processing these values. Particularly, a distribution of the values of the base changes is generated and the variance from a random distribution of this distribution is calculated by determining the root mean square variance of the distribution. The types of genetic information that can be identified using this system and method include the presence of a protein encoding region, or gene; the number of genes in a given overall polynucleotide sequence; the reading frame for the gene sequence; and tRNA and rRNA sites.Type: GrantFiled: February 25, 1997Date of Patent: July 25, 2000Assignee: Vanderbilt UniversityInventors: Thomas W. Kephart, Robert W. Cutler
-
Patent number: 6092033Abstract: A method and apparatus are described for fusing a plurality of signals, corresponding to mean and covariance estimates, into a single signal when the given signal estimates are correlated to an unknown degree. The core components of the method and apparatus are (1) capability to express the means and covariances of the set of signals in inverse form in a common coordinate frame and (2) capability to form a signal determined by a convex combination of the means and covariances, so expressed. The set of estimates to be fused may comprise previously fused estimates as well as measurements of a physical system. The signal derived from the fused result can be transmitted and used to physically respond to the measured state of the system of interest.Type: GrantFiled: April 16, 1997Date of Patent: July 18, 2000Inventor: Jeffrey K. Uhlmann
-
Patent number: 6014614Abstract: A method and mechanism for performing a spatial join between two sets of objects employs a two-pass primary filter. The objects are decomposed into variable-sized cells no larger than a predetermined cell size and stored in respective spatial indexes. The spatial indexes include a code for the variable-size cells of the object and a code for the fixed-size supercell of the variable-size cells. The first pass can be implemented as an equijoin filtering operation using the fixed-size cell codes corresponding to the predetermined cell size, and the second pass as a join operation using the variable-size cell codes at smaller cell sizes.Type: GrantFiled: May 29, 1998Date of Patent: January 11, 2000Assignee: Oracle CorporationInventors: John Herring, Nipun Agarwal, Panagiotis Vretanos
-
Patent number: 5943642Abstract: A method of designing a subband digital filter bank having first and second analysis filters with a 2-band linear phase satisfies a perfect reconstruction (PR) condition and has transfer functions H0(z) and H1(z) separated into first and second polyphase components Hi0(z) and Hi1(z). The method includes designing H0(z) which is odd-symmetric or antisymmetric and has a degree of 2m+1, converting the PR condition into a matrix form, obtaining H11(z) according to the matrix obtaining H10(z) using H11(z), obtaining H1(z) from H10(z) and H11(z), and obtaining a general second transfer function H1'(z) using H0(z) and H1(z). The filter characteristic of the second analysis filter is adjusted by adjusting x1, and the symmetry or antisymmetry thereof is determined by K.Type: GrantFiled: November 3, 1997Date of Patent: August 24, 1999Assignee: SamSung Electronics Co., Ltd.Inventor: Han-Mook Choi
-
Patent number: 5937369Abstract: A thermal energy apparatus and methodology 10 including a computer 12, operating under stored program control, to selectively allow elements 26 and 28 to be created and partitioned and to further allow the partitioning to be dispensed with upon calculating the radiation aspect of the thermal model. The thermal calculation and measurement apparatus and methodology can further include a means for allowing elements 102-108 to be created and thereafter allow an amorphous potting solution or other model wide elements 120 to be integrated into the model without detrimentally affecting the previously created elements 102-108.Type: GrantFiled: January 28, 1997Date of Patent: August 10, 1999Assignee: Thermal Engineering Services, Inc.Inventor: Bahram Zandi