Abstract: Disclosed herein is a signal processing apparatus for carrying out signal processing to convert input data into output data with a quality higher than the quality of the input data, the data processing apparatus including: a first data extraction section; a nonlinear feature quantity computation section; a processing-coefficient generation section; a second data extraction section; and a data prediction section.

Abstract: A method of training a neural network to perform decoding of a time-varying signal comprising a sequence of input symbols, which is coded by a coder such that each coded output symbol depends on more than one input symbol, characterized by repetitively: providing a plurality of successive input symbols to the neural network and to the coder, comparing the network outputs with the input signals; and adapting the network parameters to reduce the differences therebetween.

Abstract: A signal processing device for learning operations made by a user, and for generating signal optimal to the user based on the learning results. A learning unit monitors supplied operating signals generated based on user operations, and judges whether to learn the operating signals. When the operating signals are to be learned, the learning unit learns a correction norm for correcting input signals, based on the learning operating signals and outputs the correction norm to a correcting unit. The correcting unit corrects input signals based on the correction norm and outputs the corrected signals as output signals. The present invention can be applied to an NR (Noise Reduction) circuit which removes noise.

Abstract: A system for inducing an effect in a raw audio signal comprises a computing device for receiving a first audio signal and a second audio signal from a signal source, and the second audio signal comprises the first audio signal induced with an effect. The system further comprises logic that parameterizes the effect in the second audio signal into an artificial neural network (ANN).

Abstract: A neural network component includes a plurality of inputs, at least one processing element, at least one output, and a digital memory storing values at addresses respectively corresponding to the at least one processing element, wherein the at least one processing element is arranged to receive a value from the digital memory in response to an input signal, and is instructed to execute one of a plurality of operations by the value that is received from the digital memory.

Abstract: A system for inducing an effect in a raw audio signal comprises a computing device for receiving a first audio signal and a second audio signal from a signal source, and the second audio signal comprises the first audio signal induced with an effect. The system further comprises logic that parameterizes the effect in the second audio signal into an artificial neural network (ANN).

Abstract: A method for operating a communications device for separating an independent source signal from a mixture of source signals provided by M signal sources includes receiving at an antenna array at least M different summations of the M source signals. The at least M different summations define the mixture of source signals, where N<M. A signal separation processor processes the mixture of source signals by taking samples of the mixture of source signals over time and storing each sample as a data vector to create a data set, and assigning each data vector within the data set to one class within a plurality of classes based on its similarity to other data vectors in the one class. The data vectors assigned to the one class are analyzed to separate the independent source signal from other signals in the mixture of source signals.

Abstract: A boost table stores adjusted target levels for pairs of original and target pixel levels. The adjusted target levels can be used to as a substitute for the target pixel level to improve pixel response in reaching the desired target pixel level. A reduced boost table can be used, storing a subset of the adjusted target levels. Fuzzy logic control rules can be used to calculate adjusted target levels not actually stored in the reduced boost table.

Abstract: A neural prosthesis for providing a signal indicative of a predicted event from a cycle of events includes a signal-acquisition system for receiving a neural signal, and a fuzzy-logic inference system for receiving, from the signal acquisition system, a signal indicative of a current location within the cycle of events. The fuzzy-logic inference system is configured to predict a successive event in the cycle of events.

Abstract: A system and method responsive to input stimuli is provided by incorporating a computer software program, hardware processing engine, or a specialized ASIC chip processor apparatus to capture concurrent inputs that are responsive to training stimulation, store a model representing a synthesis of the captured inputs, and use the stored model to generate outputs in response to real-world stimulation. Human user forced-choice approval/disapproval generated descriptions and decisions may be dynamically mapped with conventionally presented information and sensor and control data. The model mapping is stored into and out of a conventional mass storage device, such as is used in a relational database for use in generating a response to the stimuli.

Abstract: A matrix includes samples associated with a first signal and samples associated with a second signal. The second signal includes a first portion associated with the first signal and a second portion associated with at least one disturbance, such as white noise or colored noise. A projection of the matrix is produced using canonical QR-decomposition. Canonical QR-decomposition of the matrix produces an orthogonal matrix and an upper triangular matrix, where each value in the diagonal of the upper triangular matrix is greater than or equal to zero. The projection at least substantially separates the first portion of the second signal from the second portion of the second signal.

Abstract: A method and an apparatus for predicting intake manifold pressure are presented, to compensate for a large lag or a large time delay without producing an overshot or discontinuous behaviors of a predicted value. The method comprises the step of obtaining a difference of values of a variable to be predicted and a difference of values of another variable ahead of the variable to be predicted. The method further comprises the step of filtering the differences with adaptive filters. The method further comprises the step of obtaining a predicted difference of values of the variable to be predicted, through algorithm of estimation with fuzzy reasoning. The method further comprises the step of adding the predicted difference of values of the variable to be predicted, to a current value of the variable to be predicted, to obtain a predicted value of the variable to be predicted.

Abstract: The present invention provides techniques for transmitting at least one signal through an element of a classification system. One or more input signals are received at the element. One or more functional components are extracted from the one or more input signals, and one or more membership components are extracted from the one or more input signals. An output signal is generated from the element comprising a functional component and a membership component that correspond to one or more functional components and membership components from one or more input signals.

Abstract: A monitoring system including a baseline model that automatically captures and models normal system behavior, a correlation model that employs multivariate autoregression analysis to detect abnormal system behavior, and an alarm service that weights and scores a variety of alerts to determine an alarm status and implement appropriate response actions. The baseline model decomposes the input variables into a number of components representing relatively predictable behaviors so that the erratic component e(t) may be isolated for further processing. These components include a global trend component, a cyclical component, and a seasonal component. Modeling and continually updating these components separately permits a more accurate identification of the erratic component of the input variable, which typically reflects abnormal patterns when they occur.

Abstract: In a method for detecting the modes of a dynamic system with a large number of modes that each have a set ? (t) of characteristic system parameters, a time series of at least one system variable x(t) is subjected to modeling, for example switch segmentation, so that in each time segment of a predetermined minimum length a predetermined prediction model, for example a neural network, for a system mode is detected for each system variable x(t), whereby modeling of the time series is followed by drift segmentation in which, in each time segment in which there is transition of the system from a first system mode to a second system mode, a series of mixed prediction models is detected produced by linear, paired superimposition of the prediction models of the two system modes.

Abstract: A method and an apparatus of designing a set of wavelet basis trained to fit a particular problem. The method and apparatus include constructing a neural network of arbitrary complexity using a discrete and finite Radon transform, feeding an input wavelet prototype through the neural network and its backpropagation to produce an output, and modifying the input wavelet prototype using the output.

Abstract: The present invention relates to a system and methodology providing layered probabilistic representations for sensing, learning, and inference from multiple sensory streams at multiple levels of temporal granularity and abstraction. The methods facilitate robustness to subtle changes in environment and enable model adaptation with minimal retraining. An architecture of Layered Hidden Markov Models (LHMMs) can be employed having parameters learned from stream data and at different periods of time, wherein inferences can be determined relating to context and activity from perceptual signals.

Abstract: By applying time-frequency analysis to a given standard iterative learning control or ILC an adaptive filter for the learned feed-forward loop is designed. This time varying filter varies according to the momentary frequency content of the error signal and allows to discriminate between areas of deterministic and stochastic error. Its application results in selective application of ILC to those intervals where error signals of high level are concentrated and allows application of a single ILC acquisition to different setpoint trajectories. The adaptive filter finds particular use in lithographic scanning systems where it is used for varying scan length.

Abstract: The invention provides a method of calculating uncertainty in the system represented as two or more modules comprising the steps of passing uncertainty information from each module to at least one further module; and calculating uncertainty in the measurement result from the information exchanged between the modules. The invention also provides a method of propagating uncertainty in a measurement system and also provides related systems for calculating uncertainty and propagating uncertainty.

Abstract: A Q-Filter is a reconfigurable technique that performs a continuum of linear and nonlinear filtering operations. It is modeled by unique mathematical structure, utilizing a function called the Q-Measure, defined using a set of adjustable kernel parameters to enable efficient hardware and software implementations of a variety of useful, new and conventional, filtering operations. The Q-Measure is is based on an extension of the well-known Sugeno ?-Measure.

Abstract: In the field of electronic or computerised control apparatuses, known methods of compensating for disturbance signals rely on less than ideal analytical or empirical models. There is provided a controller (200) or method of controlling which observes and learns the correlation between various measured signals and automatically learns how to control the apparatus.

Abstract: An online Gaussian mixture learning model for dynamic data utilizes an adaptive learning rate schedule to achieve fast convergence while maintaining adaptability of the model after convergence. Experimental results show an unexpectedly dramatic improvement in modeling accuracy using an adaptive learning schedule.

Abstract: A local synchronization type parallel pulse signal processing circuit has a plurality of neurons connected to each other based on a predetermined rule and disposed in parallel, executing a predetermined arithmetic process with respect to input signals and outputting, a phase synchronization signal generation circuit outputting phase synchronization signals to the predetermined vicinal neurons, and a synchronization detection portion detecting synchronization within an allowable phase difference between the outputs of the predetermined vicinal neurons. The phase synchronization signal generation circuit functions also as a neuron executing the predetermined arithmetic process and outputting in accordance with a result of the synchronization detection by the synchronization detection portion. With this architecture, the synchronization circuit operating stably without any contradiction in a way that brings neither an increase in circuit scale nor an increase in consumption of electric power, is actualized.

Abstract: The present invention provides methods and apparatus to stably separate and extract an original signal from multiple signals by a few calculation steps when multiple signals have been observed in a mixed state. In an example embodiment, signals are separated by introducing a function having a monotonously increasing characteristic like an exponential type function as a cost function, and applying an adaptive algorithm that minimizes that cost function in terms of a signal separation matrix. Then, an error signal e(t) is calculated based on y(t) formed by this nonlinear function, the estimated separation matrix W(t?1) estimated at the previous cycle, and the observed signal x(t) at that time. Then, based on the calculated error signal e(t), the update of the separation matrix W(t) at that time is performed such that consideration weight is increased when estimation errors are large using the cost function having a monotonously increasing characteristic.

Abstract: A neural network system is provided that models the system in a system model (12) with the output thereof providing a predicted output. This predicted output is modified or controlled by an output control (14). Input data is processed in a data preprocess step (10) to reconcile the data for input to the system model (12). Additionally, the error resulted from the reconciliation is input to an uncertainty model to predict the uncertainty in the predicted output. This is input to a decision processor (20) which is utilized to control the output control (14). The output control (14) is controlled to either vary the predicted output or to inhibit the predicted output whenever the output of the uncertainty model (18) exceeds a predetermined decision threshold, input by a decision threshold block (22).

Abstract: A signal processing system is provided which includes one or more receivers for receiving signals generated by a plurality of signal sources. The system has a memory for storing a predetermined function which gives, for a set of input signal values, a probability density for parameters of a respective signal model which is assumed to have generated the signals in the received signal values. The system applies a set of received signal values to the stored function to generate the probability density function and then draws samples from it. The system then analyses the drawn samples to determine parameter values representative of the signal from at least one of the sources.

Abstract: A method and apparatus are disclosed for recommending items of interest by fusing a plurality of recommendation scores from individual recommendation tools using one or more Radial Basis Function neural networks. The Radial Basis Function neural networks include N inputs and at least one output, interconnected by a plurality of hidden units in a hidden layer. A unique neural network can be used for each user, or a neural network can be shared by a plurality of users, such as a set of users having similar characteristics. A neural network training process initially trains each Radial Basis Function neural network using data from a training data set. A neural network cross-validation process selects the Radial Basis Function neural network that performs best on the cross-validation data set. A neural network program recommendation process uses the selected neural network(s) to recommend items of interest to a user.

Abstract: A method and an apparatus of designing a set of wavelet basis trained to fit a particular problem. The method and apparatus include constructing a neural network of arbitrary complexity using a discrete and finite Radon transform, feeding an input wavelet prototype through the neural network and its backpropagation to produce an output, and modifying the input wavelet prototype using the output.

Abstract: Method of suppressing audible noise in speech transmission by means of a multi-layer self-organizing fed-back neural network comprising a minima detection layer, a reaction layer, a diffusion layer and an integration layer, said layers defining a filter function F(f,T) for noise filtering.

Abstract: A computer-implemented method and apparatus that adapts class parameters, classifies data and separates sources configured in one of multiple classes whose parameters (i.e. characteristics) are initially unknown. A mixture model is used in which the observed data is categorized into two or more mutually exclusive classes. The class parameters for each of the classes are adapted to a data set in an adaptation algorithm in which class parameters including mixing matrices and bias vectors are adapted. Each data vector is assigned to one of the learned mutually exclusive classes. The adaptation and classification algorithms can be utilized in a wide variety of applications such as speech processing, image processing, medical data processing, satellite data processing, antenna array reception, and information retrieval systems.

Abstract: The method and system described herein use a biologically-based signal processing system for noise removal for signal extraction. A wavelet transform may be used in conjunction with a neural network to imitate a biological system. The neural network may be trained using ideal data derived from physical principles or noiseless signals to determine to remove noise from the signal.

Abstract: A method determines the probabilities of states of a system represented by a model including of nodes connected by links. Each node represents possible states of a corresponding part of the system, and each link represents statistical dependencies between possible states of related nodes. The nodes are grouped into arbitrary sized clusters such that every node is included in at least one cluster. A minimal number of marginalization constraints to be satisfied between the clusters are determined. A super-node network is constructed so that each cluster of nodes is represented by exactly one super-node. Super-nodes that share one of the marginalization constraints are connected by super-links. The super-node network is searched to locate closed loops of super-nodes containing at least one common node. A normalization operator for each closed loop is determined, and messages between the super-nodes are defined.

Abstract: An integrated sensor processing cell device capable of transforming, reshaping, and modulating an original sensed image includes a sensing medium. At least one memory device stores weight bits. Multiplexers are associated with at least one of the memory devices. At least one transconductance amplifier is associated with at least one of the multiplexers. A multiple input logic gate, associated with at least one of the memory devices, is configured to store a signed pixel output derived from the sensing medium output.

Abstract: A physics based neural network (PBNN) for detecting trends in a series of data inputs comprising a neural filter comprising a plurality of nodes for receiving the series of data inputs and outputting a plurality of averaged outputs, at least one standard deviation node for receiving one of the plurality of averaged outputs and the series of data inputs to produce at least one standard deviation output, wherein at least one of the average outputs is a delayed average output and at least one of the standard deviation outputs is a delayed standard deviation output, and a neural-detector comprising a plurality of neural detector nodes receiving the plurality of averaged outputs and the delayed average output and outputting a neural detector output, a neural level change node receiving the plurality of averaged outputs and outputting a neural level change estimate output, a neural confidence node receiving a counter input, the delayed standard deviation output, and the neural level change estimate output and outpu

Abstract: A system and method for a neural network is disclosed that is trained to recognize noise characteristics or other types of interference and to determine when an input waveform deviates from learned noise characteristics. A plurality of neural networks is preferably provided, which each receives a plurality of samples of intervals or windows of the input waveform. Each of the neural networks produces an output based on whether an anomaly is detected with respect to the noise, which the neural network is trained to detect. The plurality of outputs of the neural networks is preferably applied to a decision aid for deciding whether the input waveform contains a non-noise component. The decision aid may include a database, a computational section and a decision module. The system and method may provide a preliminary processing of the input waveform and is used to recognize the particular noise rather than a non-noise signal.

Abstract: An artificial neural network is utilized as a filter in an optical heterodyne balanced receiver system. Such an artificial neural network can be adapted or trained to correct for non-ideal behavior and imperfections in the optical heterodyne system.

Abstract: Two neural networks are used to control adaptively a vibration and noise-producing plant. The first neural network, the emulator, models the complex, nonlinear output of the plant with respect to certain controls and stimuli applied to the plant. The second neural network, the controller, calculates a control signal which affects the vibration and noise producing characteristics of the plant. By using the emulator model to calculate the nonlinear plant gradient, the controller matrix coefficients can be adapted by backpropagation of the plant gradient to produce a control signal which results in the minimum vibration and noise possible, given the current operating characteristics of the plant.

Abstract: An application development environment for developing end user applications for supporting management of operational environment. The environment consists of a number of subsystems. An organization modeler for defining the operational processes and the organizational units that are involved. A connection builder for defining connections to collect data from other systems. A view builder to define the views on the data. A distributor to create the end user application. A steering process modeler for defining a steering process by the steps of identifying a steering process, identifying at least one steering step for each steering process and linking the steering steps in order to build the steering process model. A model builder for defining steering models by the steps of identifying a steering model, defining input variables, defining output variables, defining the relation between the input and output variables.

Abstract: A fuzzy logic estimator for minimizing signal measuring inaccuracy effects in a signal processing system preferably includes a microprocessor-based system operable to receive a number of measured signal values and estimate a solution to an overdetermined system of equations that minimizes differences between the measured signal values and corresponding model values. The fuzzy logic solution estimate process includes assigning a probability distribution to delta values representing differences between the measured signal values and corresponding model values to form a corresponding number of probability distribution functions, associating at least some of the probability distribution functions with each equation of a system of equations defining a number of unknown parameter values, solving the system of equations for a domain of possible solutions, and determining a unique solution for the unknown parameter values from the domain of possible solutions.

Abstract: When neuron operations are computed in parallel using a large number of arithmetic units, arithmetic units for neuron operations and arithmetic units for error signal operations need not be provided separately, and a neural network arithmetic apparatus that consumes the bus band less is provided for updating of synapse connection weights. Operation results of arithmetic units and setting information of a master node are exchanged between them through a local bus. During neuron operations, partial sums of neuron output values from the arithmetic units are accumulated by the master node to generate and output a neuron output value, and an arithmetic unit to which neuron operations of the specific neuron are assigned receives and stores the neuron output value outputted from the master node.

Abstract: Electronic device (10) to calculate linear functions and to calculate and generate non-linear functions, intended to process signals. The electronic device (10) is provided with a calculator unit (11) to calculate linear functions, a device (12) to generate arbitrary functions, including nonlinear functions, and a selector device (13) to selectively put the electronic device (10) in a first mode to calculate linear functions and a second mode to generate non-linear functions.

Abstract: Artificial neural network systems where each signal processing junction connected between signal processing elements is operable to, in response to a received impulse action potential, operate in at least one of three permitted manners: (1) producing one single corresponding impulse, (2) producing no corresponding impulse, and (3) producing two or more corresponding impulse. A preprocessing module may be used to filter the input signal to such networks. Various control mechanism may be implemented.

Abstract: An information processing system having signal processors that are interconnected by processing junctions that simulate and extend biological neural networks. Each processing junction receives signals from one signal processor and generates a new signal to another signal processor. The response of each processing junction is determined by internal junction processes and is continuously changed with temporal variation in the received signal. Different processing junctions connected to receive a common signal from a signal processor respond differently to produce different signals to downstream signal processors. This transforms a temporal pattern of a signal train of spikes into a spatio-temporal pattern of junction events and provides an exponential computational power to signal processors. Each signal processing junction can receive a feedback signal from a downstream signal processor so that an internal junction process can be adjusted to learn certain characteristics embedded in received signals.

Abstract: A robust Artificial Neural Network controller is proposed for the motion control of a magnetic disk drive voice coil motor (voice coil motor). The neural controller is used to approximate the nonlinear functions (actuator electromechanical dynamics) of the voice coil motor while having on line training. One main advantage of this approach, when compared with standard adaptive control, is that complex dynamical analysis is not needed. Using this design, not only strong robustness with respect to uncertain dynamics and non-linearities can be obtained, but also the output tracking error between the plant output and the desired reference can asymptotically converge to zero. Additionally, standard offline training, utilizing training vectors to stimulate the voice coil motor, is not required.

Abstract: Signal separation processing includes a signal separation architecture defining a relationship between at least one input signal and at least one output signal. The signal separation architecture includes a state space representation that establishes a relationship between the input and output signals. At least one output signal is based on the signal separation architecture.

Abstract: The invention herein provides a supervisory circuit which is adapted to monitor an input signal and produce as an output signal, a parametric signal corresponding to the input signal. The circuit includes an input for receiving the input signal, and a stochastic processor coupled to the input for receiving the input signal and processing it to derive a signal that represents a parametric measure of the input signal. An output connected to said stochastic processor provides the parametric output signal as an output for supervisory purposes.

Abstract: A system for bearinqs-only contact state estimation in response to target bearing and ownship speed and course information provided for a plurality of observation legs at successive points in time, includes a plurality of neural networks and a data fusion circuit. Each of the neural networks generates range-normalized parameter estimate information for one of the observation legs in response to target bearing and ownship speed and course information for an associated one of the observation legs, provided thereto at each point in time and information generated for the previous point in time. The data fusion system receives the range-normalized parameter estimate information from the neural networks and generates the contact state estimate in response thereto.

Abstract: A system and method for preprocessing input electronic commerce data to a non-linear model for use in an electronic commerce (e-commerce) system. The non-linear model includes parameters that define the representation of the e-commerce system, and operates in two modes: run-time and training. A data preprocessor preprocesses received data in accordance with predetermined preprocessing parameters, and outputs preprocessed data. The data preprocessor includes an input buffer for receiving and storing the input data. The input data may include one or more outlier values. A data filter detects and removes, and may optionally replace, any outlier values in the input data, generating corrected input data. An output device outputs the corrected data from the data filter as preprocessed data, which may be input to the non-linear model in training mode to train the non-linear model, and/or in run-time mode to generate control parameters and/or predictive output information for the e-commerce system.

Abstract: The present invention discloses an implementation of the selective attention mechanism occurring in the human brain using a conventional neural network, multi-layer perceptron and the error back-propagation method as a conventional learning method, and an application of the selective attention mechanism to perception of patterns such as voices or characters. In contrast to the conventional multi-layer perceptron and error back-propagation method in which the weighted value of the network is changed based on a given input signal, the selective attention algorithm of the present invention involves learning a present input pattern to minimize the error of the output layer with the weighted value set to a fixed value, so that the network can receive only a desired input signal to simulate the selective attention mechanism in the aspect of the biology.

Abstract: A neural system is disclosed for processing an exogenous input process to produce a good outward output process with respect to a performance criterion, even if the range of one or both of these processes is necessarily large and/or keeps necessarily expanding during the operation of the neural system. The disclosed neural system comprises a recurrent neural network (RNN) and at least one range extender or reducer, each of which is a dynamic transformer. A range reducer transforms dynamically at least one component of the exogenous input process into inputs to at least one input neuron of said RNN. A range extender transforms dynamically outputs of at least one output neuron of said RNN into at least one component of the outward output process. There are many types of range extender and reducer, which have different degrees of effectiveness and computational costs.