18 results about "Non linear estimation" patented technology

The invention provides a model training method, device and equipment, a medium and a computer program product, and relates to the technical field of artificial intelligence. The method comprises the steps that a pulse model trained according to a dynamic visual data set collected by a dynamic visual sensor and a predefined truncation function are acquired, and an initial quantization pulse model is acquired; according to the first moment of the input variable of the initial quantization pulse model, the credibility estimation value of the initial quantization pulse model is obtained, the initial quantization pulse model is updated, and a quantization pulse model for converting continuous data collected by the dynamic vision sensor into pulse sequence data is obtained. A truncation function is adopted to truncate the pulse model to obtain an initial quantization pulse model, and the prediction accuracy is improved while model memory compression is preliminarily achieved; the credibility estimation value is obtained through the first moment of the input variable, the nonlinear estimation has a closed-form solution, accurate estimation can be realized, and the time sequence data prediction consistency is further improved.

The invention relates to a GNSS array antenna-based phase multi-path real-time correction method and system. The method comprises the following steps of: constructing a spatial correlation array antenna system; based on an array antenna system, calculating geometric position relations of other antenna units relative to a reference antenna unit, forming single-difference carrier phase residual observation values of different antenna units and the reference antenna unit, and constructing a carrier phase multipath error parameter estimation model; based on a carrier phase multi-path error parameter estimation model, jointly processing a multi-frequency GNSS observation value by using the consistency and frequency correlation of propagation paths of frequency points of a multi-frequency GNSS signal, and establishing a multi-frequency multi-path error redundancy estimation model; based on the multi-frequency multi-path error redundancy estimation model and particle filtering, performing nonlinear estimation on parameters of the carrier phase multi-path error parameter estimation model to obtain multi-path error parameters; and obtaining a carrier phase multipath error value of each frequency point by using the multipath error parameter, and correcting the error value in real time in GNSS positioning calculation.

A first communication device (e.g., a transmitter) configured to communicate with a second communication device (e.g., a receiver), the first communication device comprising circuitry configured to: transmit a first training field, the first training field comprising a channel estimation sequence for use by the second communication device to estimate a channel between the first communication device and the second communication device; transmitting a second training field, the second training field comprising a non-linear estimation sequence for a second communication device to estimate non-linearity between the first communication device and the second communication device, the second training field having a higher peak-to-average power ratio and / or a different average power than the first training field; receiving non-linear feedback regarding the estimated non-linearity from the second communication device; and configuring subsequent transmissions using the received non-linear feedback.

The present application is applicable to the field of communications, and provides a communication method and a communication apparatus. In the technical solution provided by the present application, decoupling a signal for nonlinear estimation from a signal for channel estimation can prevent a nonlinear estimation result from being affected by channel estimation, such that the accuracy of the nonlinear estimation result can be improved, thereby improving the accuracy of nonlinear compensation of data, ultimately improving the transmission performance of the data.

The present application relates to a kind of well gas invasion and loss parameter inversion methods based on unscented Kalman filter, belong to petroleum engineering technical field, comprising: step 1: establish overflow wellbore multiphase flow model;Step 2: establish inversion model;Step 3: solve inversion model;Calculate the inversion result of downhole gas invasion and loss parameter.The application method combines wellbore multiphase flow equation with state space model, uses the nonlinear estimation characteristics of UKF algorithm, carries out dynamic filtering and state updating to wellhead measurable data, realizes the real-time inversion of downhole gas volume fraction, gas invasion flow and loss amount.UKF method can accurately capture the nonlinear characteristics of system without linearization, thereby improving the inversion accuracy and calculation stability under complex conditions.Through example verification, this method has good applicability under single reservoir, multiple reservoir and overflow and leakage coexistence conditions, provides reliable boundary conditions and intelligent control basis for well control system.

The application discloses an uplink multi-user random modulation and OAMP detection method, which comprises the following steps: independently modulating and transmitting signals of each user, and receiving the sum of signals of each user; performing linear estimation according to prior distribution of the transmitted signals of each user, state information of a channel matrix, estimation error of the channel matrix and channel noise power; performing nonlinear estimation according to the prior distribution of the transmitted signals of each user and the estimation and variance obtained by linear estimation, so as to obtain the nonlinear estimation and variance in the present iteration; continuously iterating the linear estimation and the nonlinear estimation, and performing hard decision according to the linear estimation and variance obtained by the last iteration to obtain a final decoding result. The application independently modulates all users, considers the channel estimation error, does not require that the entire equivalent channel matrix has right unitary invariant characteristics, reduces performance loss, and obtains detection performance which is obviously superior to that of the prior OAMP.

The application discloses a CE-OFDM communication and sensing integrated anti-nonlinear DOA estimation method, belongs to the field of communication sensing, and comprises the following steps: generating a 0dB peak-to-average power ratio CE-OFDM signal at a transmitting end, so that the CE-OFDM signal can drive a power amplifier to work in a saturation region, thereby avoiding nonlinear distortion while ensuring high power amplifier efficiency. A uniform linear array is used to receive signals at a receiving end, the signals are decomposed into a frequency domain through discrete Fourier transform, narrowband MVDR spatial spectra are independently calculated for each subcarrier frequency point, wideband spatial spectra are obtained through non-coherent fusion, and finally high-precision DOA estimation is realized through peak value searching. The application fundamentally suppresses power amplifier nonlinear distortion by using the constant envelope characteristic of CE-OFDM, avoids eigenvalue decomposition and source number estimation by combining the IMVDR algorithm, supports frequency point level parallel calculation, and realizes high-precision and low-complexity DOA estimation while ensuring communication reliability.

A starboard laser terminal turntable pointing zero offset calibration method using double star sensors.A kind of starboard laser terminal turntable pointing zero offset calibration method using double star sensors is disclosed.The invention establishes a complete attitude transfer chain model, fuses the inertial attitude data output by the platform star sensor and the terminal star sensor, constructs the measurement relative attitude of the terminal star sensor relative to the satellite platform, aligns the measurement attitude with the pre-established theoretical attitude chain model with high precision, constructs a nonlinear estimation model about the azimuth encoding angle zero offset and the pitch encoding angle zero offset of the turntable, solves the problem, autonomously estimates the azimuth and pitch zero offset values of the turntable with high precision in orbit, and finally realizes the on-orbit autonomous calibration of the turntable pointing zero offset.The whole process does not need ground intervention, effectively improves the autonomy and long-term stability of satellite pointing control.

This application provides a nonlinear compensation method and an apparatus. A terminal device receives first indication information from a network device, where the first indication information indicates that a demodulation reference signal (DMRS) sequence includes a sequence whose peak to average power ratio (PAPR) is higher than that of a Zadoff-Chu (ZC) sequence. The terminal device sends the sequence and data to the network device, where the data is carried on a physical uplink shared channel (PUSCH). A nonlinear estimation result of the sequence is used by the network device to perform nonlinear compensation on the PUSCH. The sequence and the data are transmitted by using a discrete Fourier transform spread orthogonal frequency division multiplexing (DFT-s-OFDM) waveform.

The invention provides a communication method and a communication device, which can be applied to the field of communication. In the technical scheme provided by the invention, the signal for nonlinear estimation and the signal for channel estimation are decoupled, so that the nonlinear estimation result can be prevented from being influenced by channel estimation, the accuracy of the nonlinear estimation result can be improved, the nonlinear compensation accuracy of data can be improved, and the accuracy of the nonlinear compensation of the data can be improved. And finally, the data transmission performance can be improved.

The invention discloses a satellite-borne laser terminal rotary table pointing zero offset calibration method using double star sensors. Relates to the technical field of spacecraft attitude measurement and pointing control crossing, in particular to a satellite-borne laser terminal rotary table pointing zero offset calibration method using double star sensors. According to the method, a complete attitude transfer chain model is established, inertial attitude data output by a platform star sensor and inertial attitude data output by a terminal star sensor are fused, so that the measurement relative attitude of the terminal star sensor relative to a satellite platform is constructed, and the measurement attitude and a pre-established theoretical attitude chain model are subjected to high-precision alignment; a nonlinear estimation model about the azimuth coding angle zero offset and the pitching coding angle zero offset of the rotary table is constructed, the azimuth and pitching zero offset values of the rotary table can be autonomously estimated in an on-orbit mode with high precision by solving the problem, and finally on-orbit autonomous calibration of the pointing zero offset of the rotary table is achieved. The whole process does not need ground intervention, and the autonomy and long-term stability of satellite pointing control are effectively improved.

The invention discloses a neural stimulation artifact real-time dynamic tracking and removing method based on deep learning and recursive Bayesian estimation. The method comprises the following steps that S1, electroencephalogram signals with electrical stimulation artifacts are collected; s2, training a neural network model, inputting the electroencephalogram signals with the electrical stimulation artifacts into the trained neural network model, and outputting amplitude estimation of the electrical stimulation artifacts; s3, performing state estimation on the electroencephalogram signals with the electrical stimulation artifacts by adopting a recursive Bayesian estimator to obtain phase estimation of the electrical stimulation artifacts; s4, performing calculation through the amplitude estimation of the electrical stimulation artifacts and the phase estimation of the electrical stimulation artifacts to obtain artifact estimation; s5, artifact estimation is removed from the electroencephalogram signals with the electrical stimulation artifacts, and clean electroencephalogram signals are obtained. According to the method, the accurate prediction capability of the CNN-BiLSTM-AM deep neural network on the amplitude and the nonlinear estimation capability of the unscented Kalman filtering on the phase are combined, so that the high-precision reconstruction and removal of the non-stationary tACS artifacts are realized.

Systems, methods, and computer media for battery sys- tem management and non-linear estimation of battery state of charge are provided herein. Battery data is received for a time period over which a battery system has operated. The battery data represents the actual performance of the battery system over the time period. Sub- periods of charging or discharging can be identified in the time period. For the sub-periods of time, a curve can be fit to the battery data. Using the curves for the battery data for the sub-periods of time, an expected performance of the battery system, over a range of states-of-charge, can be determined. Operating instructions for the battery system can be provided based on the expected performance.

This invention provides a method for estimating the driving range of electric vehicles based on battery aging levels. The method includes: collecting operational data of the electric vehicle, including average driving speed and battery usage intervals; obtaining the battery aging level corresponding to each battery usage interval using incremental capacity analysis; grouping the battery usage intervals based on temperature conditions; constructing a nonlinear estimation model of the electric vehicle's speed versus unit energy consumption driving range based on the operational data; and constructing a driving range calculation model within the battery usage interval based on the nonlinear estimation model of speed versus unit energy consumption driving range, thereby calculating the driving range of the electric vehicle under different battery aging levels and temperature conditions. This invention achieves accurate estimation of the electric vehicle's driving range while effectively incorporating the characteristics of the actual operating environment, making it suitable for driving range detection and prediction systems for electric vehicles in complex urban traffic environments.

This application provides a method, apparatus, device, medium, and aircraft for estimating aircraft disturbance parameters, relating to the field of aircraft control technology. The method includes: acquiring flight state data of the aircraft in the current flight cycle; the flight state data includes flight speed, flight acceleration, and thrust; constructing a residual measurement model based on the flight speed, flight acceleration, and thrust; the residual measurement model is constructed from residual measurements, wind speed, and load changes; and estimating the disturbance in the current cycle using the residual measurement model to obtain estimated wind speed and load change values. This application, by constructing a residual measurement model, transforms the complex coupling relationship between aerodynamic drag and mass change into a linear measurement model concerning wind speed and load changes, avoiding the cumbersome and unstable nature of nonlinear estimation. The estimation of wind speed and load changes is achieved by solving the residual measurement model.

In one aspect, a method performed by a transmitter in a first device for signaling a nonlinearity profile of distortion of transmissions in a communications network is provided. The method includes generating an amplitude tracking reference signal (ATRS), wherein the ATRS comprises information indicating a nonlinearity profile of distortions of transmissions by the transmitter, and transmitting the generated ATRS towards a second device. In another aspect, a method performed by a receiver in a second device for identifying a nonlinearity profile of distortion of transmissions in a communications network by a transmitter in a first device is provided. The method includes receiving an ATRS from the first device, wherein the ATRS comprises information indicating a nonlinearity profile of distortions of transmissions by the transmitter and identifying the nonlinearity profile of distortion of transmissions by the transmitter based on the ATRS.

The invention relates to an underground gas cut and leakage parameter inversion method based on unscented Kalman filtering, and belongs to the technical field of petroleum engineering, and the method comprises the steps: 1, building an overflow well killing shaft multiphase flow model; 2, establishing an inversion model; 3, solving an inversion model; and calculating an underground gas cut and leakage parameter inversion result. According to the method, a shaft multiphase flow equation and a state space model are combined, dynamic filtering and state updating are conducted on measurable data of a wellhead by means of the nonlinear estimation characteristic of the UKF algorithm, and real-time inversion of the underground gas volume fraction, the gas cut flow and the leakage amount is achieved. The UKF method can accurately capture the nonlinear characteristics of the system without linearization, thereby improving the inversion precision and calculation stability under a complex working condition. Through example verification, the method has good applicability under the conditions of single reservoir, multiple reservoirs and simultaneous overflow and leakage, and provides reliable boundary conditions and intelligent regulation and control basis for a well control system.

The invention discloses an adjacent well pipe column space position while-drilling anti-collision identification system based on multi-source information fusion, and particularly relates to the technical field of petroleum and natural gas drilling engineering, and the method comprises the following steps: obtaining measurement while drilling data, adjacent well historical data and electromagnetic distance measurement data of an adjacent well pipe column in real time; performing space-time synchronization and registration on the data, and constructing a space geometric model and a representation measurement error model; calculating relative distance and direction information of a current drill bit and an adjacent well pipe column and an anti-collision risk index through a nonlinear estimation algorithm; performing graded early warning based on an anti-collision risk index, and generating a three-dimensional visual picture of a current well track, a space position of an adjacent well pipe column and a real-time relative position relationship; and generating a collision avoidance control instruction. Through the data acquisition module, the data processing and fusion core module, the anti-collision early warning and visualization module and the closed-loop control interface module, the multi-source information fusion adjacent well pipe column space position while-drilling anti-collision identification system is constructed.