57 results about "Non linear estimation" patented technology

A method of estimating the navigational state of a system entails acquiring observation data produced by noisy measurement sensors and providing a probabilistic inference system to combine the observation data with prediction values of the system state space model to estimate the navigational state of the system. The probabilistic inference system is implemented to include a realization of a Gaussian approximate random variable propagation technique performing deterministic sampling without analytic derivative calculations. This technique achieves for the navigational state of the system an estimation accuracy that is greater than that achievable with an extended Kalman filter-based probabilistic inference system.

A method and an apparatus for determining a location of a mobile receiver including the steps of measuring a plurality of signal strengths received by a mobile receiver, wherein the plurality of signal strengths are associated with a plurality of cellular stations, wherein the plurality of signal strengths is associated with a specific point in time, combining the plurality of signal strengths with a plurality of signal path modeling parameters to create a propagation path loss model of the path between the plurality of cellular stations and the mobile receiver, applying a non-linear estimation algorithm to the propagation path loss model, generating a plurality of distances, wherein each distance is associated with the mobile receiver and each of the plurality of cellular stations and computing the location of the mobile receiver by iterating the non-linear estimation algorithm and resulting mobile receiver position until converged.

A method and system for providing performance estimations for a specified power budget provides an indication of the impact on processing performance when closed-loop power/performance control is employed to meet the specified power budget. A workload, which may be the actual workload, or a test workload is run to determine actual power consumption at intervals during the execution of the workload. The power values are examined and if they exceed the specified budget, which may be one of multiple possible budget values, an estimate of the amount by and duration for which the closed-loop power/performance control would have to reduce the performance of the system for each interval in order to provide an estimate of actual performance for the budgetary level(s). The estimate is informed by tests of the workload at each power/performance level to provide a non-linear estimate of the relationship between performance and power for the particular workload.

The present invention includes methods for the reduction of speckle noise in an image and methods for segmenting an image. Each of the methods disclosed herein includes steps for analyzing the uniformity of a pixel within a plurality of pixels forming a portion of the image and, based on the uniformity of the intensity of the plurality of pixels, adjusting and/or replacing the pixel in order to produce a speckle-noise reduced image, a segmented image, or a segmented and speckle-noise reduced image. The methods of the present invention can employ for example conditional probability density functions, nonlinear estimator functions, convex energy functions and simulated annealing algorithms in the performance of their respective steps.

The invention provides a system information sending method, which comprises the following steps of: calculating the autocorrelation matrix of a feedback signal X; and calculating the cross correlation matrix of the conjugate matrix Y* of the feedback signal X and a sending signal Y and calculating a pre-distortion parameter W, wherein RW is equal to B. The invention also provides a system information sending device. The problem of huge calculation amount during matrix decomposition is solved according to the technical scheme provided by the invention. In the technical scheme provided by the invention, based on a geometrical orthogonality principle the feedback signal and the sending signal of digital pre-distortion (DPD) form a nonlinear correlation matrix and an optimal nonlinear evaluation parameter is obtained, so that the acquisition time of the DPD coefficient after the optimization is greatly shortened, and a highly efficient and feasible algorithm is provided for the application of the DPD to products.

The present invention includes methods for the reduction of speckle noise in an image and methods for segmenting an image. Each of the methods disclosed herein includes steps for analyzing the uniformity of a pixel within a plurality of pixels forming a portion of the image and, based on the uniformity of the intensity of the plurality of pixels, adjusting and/or replacing the pixel in order to produce a speckle-noise reduced image, a segmented image, or a segmented and speckle-noise reduced image. The methods of the present invention can employ for example conditional probability density functions, nonlinear estimator functions, convex energy functions and simulated annealing algorithms in the performance of their respective steps.

The invention discloses an ocean remote sensing image water color and water temperature monitoring method based on compression sampling. The monitoring method comprises the following steps of: obtaining original remote sensing data of ocean water color or water temperature to be detected, and carrying out compression sampling, sparse transformation and image processing on the original remote sensing data to obtain a water color or water temperature data set; sequentially conducting denoising and reconstruction processing on the water color or water temperature data set through an SOM (self-organized mapping) algorithm to obtain a processed water color or water temperature data set; conducting denoising processing on the processed water color or water temperature data set and reconstructing missing image data by utilizing the continuous interpolation property; and outputting and displaying the reconstructed missing image data. The monitoring method utilizes the compression sampling principle for reducing the amount of processed data, and utilizes the advantages of nonlinear estimation of the SOM algorithm as well as linear estimation and continuous interpolation of an improved empirical orthogonal decomposition (EOF) algorithm and the like so as to improve the efficiency of reconstructing missing images, expand the scope of the reconstructed images, and enable the precision and efficiency of ocean remote sensing water color and water temperature monitoring to be high.

The invention provides a multi-target tracking method and system based on an automobile radar. The method comprises the following steps of: using density clustering to generate each effective target for each detection target clustering; calculating the degree of correlation between each effective target and each previous cycle track and generating a first covariance matrix; generating an evaluation matrix by dynamic [alpha] filtering according to the first covariance matrix and a second covariance matrix of the previous cycle; performing Hungarian assignment based on each priority and evaluation matrix to generate matching track of each effective target; performing Kalman filtering according to an effective target state, previous cycle tracks, and a second covariance matrix of the previouscycle to generate a second covariance matrix in a current cycle; and according to each effective target state, matching track state corresponding to each effective target, the first covariance matrixand the second covariance matrix in the current cycle, generating each track and a resampled object set in the current cycle by Monte Carlo multivariate probability sampling. The multi-target tracking method and system based on the automobile radar has the beneficial effects of taking into account nonlinear estimation accuracy and real-time performance.

A spring design method is disclosed. The method begins with inputting a first set of design parameters for a spring. The design parameters include a parameter that provides an estimate of non-linearity in the spring. The spring design method determines a spring design based on the first set of design parameters. A spring design and analysis method is disclosed. The method begins with creating a spring design. The spring design includes a parameter that provides an estimate of non-linearity in the spring design. The spring design and analysis method creates a spring animation file that enables stress levels in a spring design to be identified at the coil level. The spring design method next identifies the coil in the spring design having the lowest dynamic fatigue factor and determines whether the lowest dynamic fatigue factor is acceptable.

The invention relates to a variance minimizing method of a satellite image local area network based on nonlinear estimation of a restriction function. The method comprises the following steps: S01, building a function model of image positioning of an image square space based on RFM (Rational Function Model); S02, introducing a restriction condition; S03, solving parameters to be estimated. The method has the advantages that the restriction condition is introduced on a conventional local area network least variance mathematical model, so that the problem of non-unique solution caused by loss of necessary initial data under a condition of no control points is solved, and meanwhile, unknown parameters are optimally estimated by adopting a restriction function method, an original non-linear form of the functional model is preserved, the non-linearity problem with the restriction condition is converted into the problem of the solution to unconstrained function extreme value, and a model error problem caused by a linear optimal estimation algorithm is avoided, and therefore, high precision of least variance solution of the satellite image local area network under the condition of no control points is ensured.

The invention provides a non-uniform nonlinear system cooperative control method and control system capable of achieving control parameter adaptive compensation in order to solve the problem that thecontrol effect of an existing nonlinear system cooperative control method with fixed control coefficients is poor. The control method comprises the steps that a controlled nonlinear system model withtime-varying control parameters is built; synchronous tracking errors of various controlled nonlinear systems are acquired; a neural network nonlinear estimation model is built, and a neural network adaption law is acquired and used for estimating kinetic parameters in the controlled nonlinear system model; according to the control quantities of the controlled nonlinear systems and the acquired synchronous tracking errors, a time-varying control parameter adaption law is acquired and used for estimating control parameters in the controlled nonlinear system model; and a distributed control lawis acquired by combining the synchronous tracking errors, the neural network adaption law and the time-varying control parameter adaption law according to the controlled nonlinear system model, and the control quantities of the various controlled nonlinear systems are obtained according to the acquired distributed control law.

The invention provides an LS-SVM-based sensorless control system of a bearingless induction motor. According to the non-linear estimation model of radial displacement variable, formulas (4) and (5) are given. The generalization ability of SVM is used to approximate the nonlinear relationship between radial displacement and related physical variables, so that the real-time estimation of the radialdisplacement of the rotor can be realized, the estimation accuracy of the radial displacement does not depend on the accurate mathematical model of the motor, the radial displacement estimation errorcaused by the inaccurate motor parameters can be effectively overcome, and the complicated signal extraction algorithm is not needed, and the LS-SVM rotor radial displacement estimator is constructed,thereby getting rid of that mechanical radial displacement sensor, getting rid of the mechanical radial displacement sensor, overcoming the influence of modeling error of nonlinear radial displacement estimation model, and avoiding the radial displacement estimation error caused by inaccurate motor parameters. A new control system is constructed based on LS-SVM rotor radial displacement estimator, which can effectively reduce the cost of bearingless induction motor control system.

Method for the digital compensation of nonlinearities in a communication system that includes a transmitter, a transmission channel and a receiver, comprising:

at least one stage of estimating the nonlinearities induced by the transmitter and/or the receiver from at least one learning sequence received at the receiver end and distorted by said nonlinearities,

at least one stage of compensating for said nonlinearities distorting a signal received at the receiver end from the previous estimation of nonlinearities.

The invention discloses a target tracking method under DoS attack. The method comprises the following steps: establishing a state model and a measurement model of a system, and initializing a system state, sampling time and control parameters; modeling a signal received by the monitoring center under the DoS attack; defining a local state estimate (LSE) and a distributed fusion estimate (DFE); designing two optimization problems for the networked nonlinear estimation system, and respectively calculating to obtain a local estimation gain and an optimal weighting matrix; iterative updating is carried out to obtain the optimal estimation of the target position; the distributed fusion estimation algorithm is provided for a networked nonlinear system under DoS attack, and compared with an existing target tracking method, the distributed fusion estimation algorithm effectively improves the target tracking precision on the premise of ensuring the calculation complexity.

The invention discloses a vector miss distance parameter estimation method based on GA. The vector miss distance parameter estimation method based on GA comprises the steps of building a solving target function; according to signal quality, selecting a channel with the best signal quality as a reference channel, calculating the phase difference of the channels relative to reference channel, and onthe basis of the phase difference, building a target function; building a constraint function; using a genetic algorithm for optimizing the set target function, and the optimal solution is obtained;serving the obtained optimal solution as iteration initialization value of a nonlinear estimation function, and obtaining a global optimum. Advantages and shortcomings of the GA algorithm and a traditional nonlinear estimation algorithm are combined, a rapid and accurate parameter estimation method is developed, the defect that GA local optimizing capacity is weak, and the global optimization capacity of the traditional nonlinear estimation algorithm is weak is overcome, the GA global optimizing capacity and the traditional nonlinear estimation local optimizing capacity are sufficiently used,the characteristics of GA randomness are considered, repeated GA synchronous optimization thought is adopted, and the problem about parameter estimation of complicated problems can be effectively solved.

Various embodiments include methods and systems having a frequency-modulated continuous wave radar operable to compensate a return signal for nonlinearity in the associated radar signal that is transmitted. The radar signal can be mixed with a delayed version of the radar signal such that the mixed signal can be used to generate an estimate of the nonlinearity. The estimate can be used to compensate the return signal from an object that reflects the associated transmitted radar signal. Additional apparatus, systems, and methods can be implemented in a variety of applications.

This invention relates to frequency-offset compensation method and a device immobile communication system. The device includes a pre-frequency set compensator, a frequency-offset and phase shift discriminator, a phase shift compensator and a ring circuit filter. The said pre-offset compensator utilize the evaluation to the offset from the filer to carry out pre-compensation to the current two modulation symbols and output it to the discriminator and phase shift compensator, the discriminator utilizes two-dimension non-linear estimation for offset estimation and correction, the compensator compensates the current modulation symbols.

The invention discloses an active distribution network probabilistic stability analysis method in consideration of distributed photovoltaic randomness. In view of the distributed photovoltaic random fluctuation characteristics, a new strategy for analyzing the active distribution network small interference stability is provided based on a probability distribution method. The method can establish the polynomial estimation model of the critical mode damping ratio of the active distribution network with respect to each photovoltaic random solar irradiance, and reveals the influence of photovoltaic output power uncertainty caused by the random variation of solar irradiance on the small interference stability of the active distribution network. On this basis, by obtaining the photovoltaic solarirradiance data, the probability density function of the system critical mode damping ratio is estimated and the probability of system instability is further calculated. The invention uses a nonlinear estimation model and a probability analysis method to improve the accuracy of the stability analysis result.

The invention discloses a construction method of a large-scale MIMO system detection model, and the model is formed by iterating K layers of same networks, each layer of network is divided into a linear module and a nonlinear module, the linear modules are combined with a linear combination of known information, training parameters are added, and linear estimation is carried out on a sending signal x. And the nonlinear module performs multi-segment mapping on the linear estimation value to obtain a nonlinear estimation value, and in addition, a residual structure with a variable residual coefficient is added into each layer of network. The constructed detection model is simple in network structure, and has the characteristics of strong network adaptive capacity, high convergence speed andlow complexity.