98 results about "Discrete frequency domain" patented technology

The invention relates to a seismic wave absorption and attenuation compensation method, and belongs to the geophysical exploration data processing technology. The method comprises the following steps: selecting one of VSP (vertical seismic profiling) initial data as reference data, carrying out generalized S-transform on the reference data, selecting sampling points as reference points, and recording the time and time-frequency spectrums of the sampling points; carrying out generalized S-transform on data of each sampling point, and dividing the time-frequency spectrum of corresponding frequency in the time-frequency spectrums of the reference points by the time-frequency spectrum of each sampling point so as to obtain an absorption and attenuation curve; determining a natural logarithm to figure out a phase compensation operator according to a formula of frequency difference of absorption and attenuation signals in a time frequency domain; and carrying out absorption and attenuation compensation and phase compensation in the time frequency domain to obtain a fine VSP wave field profile. The method has the advantages that noise can be eliminated during the process of fitting the absorption and attenuation curve; generalized S-inverse transform without energy loss is realized; signal-to-noise ratio and resolution of the data are improved after compensation; log resolution control is realized; the reference points are selected flexibly; and algorithm is simple and efficient.

The invention discloses a method for identifying a sound fault based on a mel energy spectrum and a convolution neural network. The method comprises the following steps: first, performing pre-emphasison initially input audio data; then, performing framing and windowing processing on the data; after that, performing fast Fourier transform on framed and windowed data; extracting energy features ona frequency domain, processing an energy spectrum by using a set of mel-scale triangular filter banks; and after that, converting the data into a Mel energy spectrum by using energy in different frequency domains corresponding to each frame as a Y axis, and different frames in a time domain as an X axis. After that, the energy spectrum needs to be further framed to adapt to an input of a CNN (convolution neural network). Each frame is a sample, and one-hot coding of a label corresponding to each sample is used as an output of the CNN to train a CNN model until a network training error reachesthe minimum. During prediction, a probability value of each type of label is output, and a label with the largest probability value is taken as a final discrimination result.

The invention provides an OFDM carrier frequency offset estimation method for packet transmission, which comprises the following steps: 1) performing IFFT conversion on frequency domain symbols to obtain time domain training symbols of two same repetition modules; 2) adding cyclic prefixes with the length of L on the front of the time domain training symbols to form integral OFDM symbols, and obtaining received OFDM signals carrying frequency offset epsilon through a multipath channel; 3) rotating phases with pi epsilon by N / 2 spaced received sample values in the received OFDM signals to obtain fine frequency offset estimation epsilon [fine]; 4) performing fine frequency offset compensation on the received OFDM signals to obtain OFDM signals after the compensation; 5) performing FFT conversion on training symbol parts of the compensated OFDM symbols after the cyclic prefixes are removed so as to obtain receiving end modulation frequency domain symbols; 6) detecting peaks of cyclic shift of the receiving end frequency domain symbols and sending end frequency domain symbols to obtain coarse frequency offset estimation; and 7) performing coarse frequency offset compensation on the receiving end modulation frequency domain symbols in a receiving end frequency domain, and summing up the coarse frequency offset estimation and the fine frequency offset estimation to obtain the carrier frequency offset estimation.

A method for communicating across first and second frequency domains of an integrated microchip is provided. The method initiates with determining a clock ratio between the first frequency domain and the second frequency domain. The first frequency domain is associated with a faster clock cycle. Then, a synchronizing signal based upon the clock ratio is generated. The synchronizing signal coordinates communication of data between the first and second frequency domains. Next, the data is transferred between respective frequency domains according to the synchronizing signal. A microchip and a system enabling synchronous data transfer across different frequency domains are also provided.

An impact load monitoring system for a wind turbine for wind power generation is provided with: an acceleration sensor 28 attached to a step-up gear 18 connected to a main shaft 20; a frequency filter 30 for extracting a monitoring-object component contained in a target frequency domain from vibration data representing a temporal change of amplitude of the acceleration obtained by the acceleration sensor 28; and a determination unit 32 for determining whether or not amplitude of acceleration of the monitoring-object component exceeds a reference value which is set in advance, by comparing the amplitude of the acceleration with the reference value.

This invention discloses an adaptive frequency domain receive method used in a super-wide-band radio communication system using a band-pass filter set, among which, difference of central frequencies of adjacent filters is a fixed frequency interval and the central frequency of each band-pass filter is adjusted adaptively by feedback including: every filter gets the spectrum information of related frequency point signal in received signals, after filtering the filers with the symbol frequency of the received signal, the signal from sampling is converted in module to get a set of frequency domain digital signals, maximum likelihood estimation is carried out to the digital signal and the current local formwork signals to get new local formwork signals to be correlated with the digital signal to get the sent information.

The invention discloses a high-precision method for acquiring an FMCW radar distance measurement value and aims to discloses a distance measurement value method with high ranging precision and capableof reducing frequency spectrum errors. The method is implemented by adopting the following technical scheme; sending target beat signals subjected to analog-digital conversion into four parallel timedomain extraction module; sending digital beat signals and sampling rate data acquired respectively into a data zero fill module for time domain zero filling; acquiring discrete frequency domain signals having a length of Ns through discrete Fourier transform (DFT) on time domain signals having a length of Ns acquired by the four time domain extraction module through corresponding DFT modules; sending the discrete frequency domain signals are sent into a data interception and splicing module by the four DFT modules in common-end manner for frequency domain data splicing operation, and an FMCWradar distance measurement value is acquired through a distance measurement value acquiring module based on a mathematic interpolation operation result.

The invention discloses an adjacent space slow platform synthetic aperture radar big scene imaging method. The method particularly comprises the following steps: according to the principle that time domain convolution of two signals is equivalent to frequency domain multiplication, convoluting the boost echo and the dechirp reference signal in the orientation time domain, and converting the result to obtain the Doppler domain, multiply a compensating factor in the Doppler domain to accomplish the Doppler aliasing step, overcome Doppler aliasing problem, and obtain the echo signal with non-aliased frequency domain; conducting reference phase function multiply in the two-dimensional frequency domain to accomplish consistent compression, and accurately accomplish residual distance ACTS nmo correction, secondary range compression and residual azimuth compression through Stolt interpolation, so as to accomplish focusing; obtaining the distance and orientation two-dimensional time domain focused image through orientation time domain aliasing step; obtaining the boost image, splicing to form a continuous wide surveying and mapping band image output. The method provided by the invention has the advantages of being high in surveying and mapping band width and imaging algorithm accuracy.

The invention discloses a communication radiation source individual identification method and system. According to the technical scheme, the method comprises the steps of: receiving communication radiation source signal, carrying out radio frequency preselection amplification, carrying out frequency mixing, carrying out intermediate frequency filtering, carrying out A / D conversion, carrying out digital orthogonal demodulation, carrying out rectangular integration double-spectrum transformation, dividing rectangular integration double-spectrum characteristic vectors into a training sample set and a testing sample set, constructing a rectangular integration double-spectrum characteristic dictionary, carrying out non-linear transformation, carrying out mapping to a cooperation projection matrix, constructing a classifier, obtaining classification residual errors, and using the type corresponding to the smallest classification residual error as type of a communication radiation source individual. According to the invention, a small sample problem in a communication radiation source individual identification process is solved, the time complexity of the algorithm is lowered, and the phase and amplitude information distortion problem in the process when an existing method based on the time domain, frequency domain or time frequency domain is used to process non-stable or non-Gaussian signals is adopted.

The invention discloses a squint SAR ground moving target extracting method. The squint SAR ground moving target extracting method comprises the following steps that 1, echo signals of a stationary target and a moving target are computed; 2, range direction Fourier transformation is carried out; 3, consistent range walk correction is carried out on a range frequency domain and an orientation time domain; 4, range direction residual phase compensation is carried out; 5, variable parameter extension azimuth nonlinear CS is carried out; 6 the echo signals subjected to the extension azimuth nonlinear CS are multiplied by corresponding azimuth minus factors respectively and are transformed to the azimuth frequency domain; 7, the amplitudes of the two echoes are subtracted, the echo of the stationary target is removed, and stationary target clutter suppression is achieved; 8, amplitude detection and moving target extraction are carried out. The method solves the problems that the range and azimuth are coupled seriously in a ground moving target under the single-channel squint mode and moving target movement parameters are unknown, estimation of movement parameters is not needed, efficiency is higher, and the method can be widely used in the fields of clutter suppression, moving target extraction and the like.

The invention discloses a full-resolution range model-based one-stationary bistatic SAR (synthetic aperture radar) imaging method. The method of the invention include the following steps that: 1, a one-stationary bistatic SAR system geometric model is accurately constructed, and echo signals are obtained; 2, range-direction pre-processing is performed on the signals in a range frequency domain; 3,a new full-resolution range model is constructed according to the result of the range-direction processing, and azimuth space-variation characteristics between point targets which are located on thesame range unit and distributed along an azimuth direction and the center distance of a receiver can be accurately described; and 4, accurate modeling is performed for the azimuth space-variation characteristics of complex Doppler phases according to the new full-resolution range model, and azimuth equalization is realized through an ENLCS algorithm, a filter is designed to perform azimuth compression, and a final focused image is obtained. The method of the invention enables a better focusing effect in processing the high-resolution echo data of a large-squint one-stationary bistatic SAR.

The invention discloses a low-coherence interferometric demodulation method based on spectrum nonlinear effect dispersion compensation. The method comprises the following steps: step 1, achieving discrete Fourier transform of interference fringe signals to obtain the magnitude spectrum and the phase spectrum under the discrete frequency domain; step 2, setting a threshold H, taking a region of which the amplitude is larger than H, except for the low-frequency part, as the selected discrete frequency band, calculating wave numbers corresponding to all points in the frequency band, and transforming the phase spectrum from the discrete frequency domain to the wave number domain; step 3, selecting the wave number corresponding to the maximum amplitude in the magnitude spectrum, within the wave number band in which the optical spectrum of a light source is distributed, as the specific wave number, carrying out phase unwrapping in the wave number band, calculating the interference level of the specific wave number, reconstructing the absolute phase of the specific wave number, which directly reflects the range information, according to the relative phase value and the interference level of the specific wave number. The low-coherence interferometric demodulation method realizes high-precision distance demodulation and is suitable for light sources with any wideband.

The invention discloses a speech denoising method based on a time frequency domain. An input signal is divided into data blocks according to a sampling sequence, and each data block is converted into a frequency domain signal through discrete Fourier transform. A time frequency domain is formed by the data which is subjectd to discrete Fourier transform. A time frequency domain signal is subjected to statistical analysis and identification, a speech signal and a noise signal can be roughly separated, and the modeling of the two signals is carried out and characteristic values are extracted. According to the extracted characteristic values, a filter is designed and adjusted in real time, and the time frequency domain signal is filtered. Finally, the filtered time frequency domain signal is subjected to discrete Fourier transform to be a time domain audio signal. The method has the advantages that the system mechanical structure design and production process are simplified, the system cost is reduced, the frequency domain filter can be designed and adjusted in real time, and the input voice signal is filtered to achieve purpose of denoising, the cumbersome filter operation is simplified as single frequency domain gain multiplication operation, the processing calculation amount is reduced, and the processing efficiency is improved.

The invention belongs to the technical field of radar imaging, and discloses a curvilinear track SAR squint imaging method based on an improved slant distance model. The curvilinear track SAR squint imaging method based on the improved slant distance model comprises the specific steps that an equivalent slant distance model is established; a two-dimensional spectrum is derived; the distance walking momentum is calculated and is transformed to the distance frequency domain to complete the distance walking correction in the distance direction; an echo signal in the distance frequency domain is subjected to the fourier transformation in the azimuth direction, and a two-dimensional spectrum of the echo signal is obtained; distance compression, secondary distance compression and distance bending correction are carried out, and the echo signal is subjected to inverse fourier transformation in the distance direction; and an azimuth frequency domain signal is subjected to azimuth compression and inverse fourier transformation in the azimuth direction, and imaging results are obtained. According to the curvilinear track SAR squint imaging method based on the improved slant distance model, on the basis of motion characteristics of a curvilinear track platform, the equation of motion of the curvilinear track platform is subjected to high-order approximation, a four-order equivalent slantdistance model expression is established, based on the slant distance model, the analytical solution of the two-dimensional spectrum of the curvilinear track platform is derived, the conciseness and the high precision of a frequency spectrum is ensured, and full-aperture high-resolution imaging can be achieved.

The invention provides a control method and a control device for controlling soft start of an induction motor, which are applied to three-phase antiparallel thyristor AC (Alternating Current) voltage regulation control circuits. The control method comprises the following steps: firstly, the induction motor is started under the low frequency in a discrete frequency control manner; in the starting process, whenever a starting current of the induction motor is less than a rated current, the discrete frequency is gradually increased, wherein the relation between the discrete frequency f(HZ) and the working frequency is that f(HZ)=50HZ / K, the working frequency is 50HZ, K is a positive integer and when the K value is decreased, the discrete frequency is increased; after the discrete frequency reaches half of the working frequency, the induction motor is controlled to be switched to the working frequency to operate. According to the invention, a variable frequency speed regulation principle is utilized; the rotating speed of the induction motor is gradually increased by gradually increasing the starting frequency, so that when the induction motor is started, the air gap magnetic flux of the induction motor can be controlled to be close to the rated magnetic flux, and thus, not only is the starting current reduced, but also the starting torque is increased.

The invention discloses a low SNR (signal noise ratio) signal parameter extraction method based on pulse cutting, and a problem that parameters of a discrete frequency coded signal are difficult to extract under the condition of a low SNR in the prior art mainly is solved. The implementation scheme comprises a step of intercepting one section of digital pulse signals received by a radar to be a reference signal and performing short-time Fourier transform on the reference signal to obtain a time-frequency diagram, a step of obtaining a pulse repetition time estimation value of the radar pulse signal and the number of frequency codes by using the time-frequency diagram, and a step of grouping the digital pulse signals by the pulse repetition time, and sequentially performing pulsed alignmentcorrection, phase compensation and coherent accumulation on the grouped signals to obtain accumulated signals with a high SNR, and obtaining estimated results of pulse width, bandwidth, carrier frequency and frequency code of a radar pulse signal. According to the method, the parameters of a discrete frequency coded signal can be effectively extracted under the condition of the low SNR, the parameter precision is high, and the method can be used for processing and analyzing a discrete frequency code pulse signal.

The invention discloses a rotary sound source recognition method based on a mixed time frequency domain. The method comprises the following steps that 1, a sound source recognized by using a DAS; and2, a sound source recognition result is cleared based on a frequency domain deconvolution method, in the unconstrained optimization formula (please see the equation, described in the specification), aregularization parameter is selected (please see the equation, described in the specification, about lambda), wherein II . II infinite represents an infinite norm; and the sound source recognition anti-convolution method adopts an iterative method for solving the sound source intensity q. The method has the technical effects that the sound source recognition performance of the existing hybrid time domain method under the high time domain sampling frequency can be achieved by adopting a lower time domain sampling frequency.

The present invention discloses a full frequency prediction method for vibration noise on a deepwater semi-submersible drilling platform, wherein the method includes the following steps: S1, determining a vibration source and noise source of the deepwater semi-submersible drilling platform under the condition of loading, and obtaining data relating to the vibration source and noise source; S2, performing a vibration prediction for the deepwater semi-submersible drilling platform, and providing, by the vibration prediction, services for structural noise caused by vibration; S3, predicting noise on the deepwater semi-submersible drilling platform; S4, determining the transfer path of the noise, and evaluating noise levels. The full frequency prediction method of the present invention is an innovation without needing to separately build an acoustic model under different frequency-domain conditions. The vibration of the platform structure and cabin noise ratio under different frequency domains can be obtained only by modifying element conditions on the input boundary and dividing subsystem, and the weight ratio can be directly obtained, which provide an effective measurement for reducing vibration and noise.

The invention discloses a PDN capacitor optimizing method based on lossless resonant cavity power supply horizon plane modeling. The method comprises the steps of 1) designing related parameters input in a power supply distribution network; 2) calculating an initial power supply distribution network impedance value according to the result of the step 1); 3) calculating number of decoupling capacitors according to the parameter designed in the step 1); 4) calculating an actual power supply distribution network impedance value Z according to the results of the steps 2) and 3); 5) determining a relation between a discrete frequency point f<n> and the maximum target frequency f<max>, if the f<n> is less than the f<max>, executing a step 6); and if the f<n>is equal to the f<max>, outputting the result of the step 3); and 6) comparing the Z with the target impendence Z<t> designed in the step 1), if Z is greater than Z<t>, returning to the step 3), and or otherwise, adding 1 to the subscript n of the discrete frequency point, and returning to step 4). By adoption of the PDN capacitor optimizing method, the target frequency band is expanded, and the number of the capacitors is reduced while the power supply distribution network design requirement is satisfied, so that the PDN capacitor optimizing method can be used for design and analysis of a high-speed system.

Provided is an inversion range migration method for a synthetic aperture radar (SAR) large-scale scene deception jamming. The method comprises steps of: 1, constructing a jammer deception template according to a false electromagnetic characteristic which is planned to be generated by a jamming party in a radar image; 2, computing equidistant sampling values of a complementary focusing filter HHdiff (k, kx) on a two-dimension frequency domain (k, kx) plane; 3, computing equidistant sampling values of a consistent focusing filter HHbulk (k, kx) on a two-dimension frequency domain (k, kx) plane, and multiplying the equidistant sampling values of the consistent focusing filter HHbulk (k, kx) on the two-dimension frequency domain (k, kx) plane by the equidistant sampling values of the complementary focusing filter HHdiff (k, kx) on the two-dimension frequency domain (k, kx) plane acquired in the step 2; and 4, performing azimuth inverse Fourier transformation on a result acquired in the step 3 and multiplying a transformed result by the equidistant sampling values of a jammer blanking filter Hhelim (k, x) on a distance frequency domain (k, x) plane. The invention provides the SAR large-scale scene deception jamming method for overcoming defects in the prior art. The method computes jammer system frequency response accurately and high efficiently in order that a radar, when having high resolution, a large squint angle, and a long synthetic aperture, may satisfy computational complexity and focusing effects.

The invention relates to a distance frequency domain polynomial-phase transform based space mobile target detection method. The method comprises the steps of carrying out fast Fourier transform along a quick time direction of received mobile target echo data so as to acquire distance frequency domain-slow time data, then correcting a Doppler migration phenomenon occurred in the mobile target echo signals by using a distance frequency domain polynomial-phase transform method on the basis of the acquired distance frequency domain-slow time data, and carrying out phase-coherent accumulation on the echo data on the basis of correction so as to be applied to target detection. A simulation result is compared with processing results of a moving target detection method, a Radon-Fourier transform method, a fractional Fourier transform method and a Radon-FRFT method; and the simulation result indicates that the method can detect a target effectively under a condition that the coherent accumulation pulse number is limited and the signal-to-noise ratio is low, thereby verifying the effectiveness of the method provided by the invention.

The present invention discloses a blind separation hybrid matrix estimation method based on synchronous compression short-time Fourier transform. The purpose of the invention is to improve the sparsity of the time frequency domain of the observation signal, improve the energy of the useful signal, further reduce the influence of noise, and correctly estimate the hybrid matrix. Subjecting to STFT synchronous compression conversion, the observation signal is converted from the time domain to the time frequency domain, and the energy of each time frequency point in frequency is rearranged to improve the energy aggregation of the signal to be analyzed in the time frequency plane, thereby increasing the sparsity in the time frequency domain and reducing the effects of noise. In the process of extracting a single source point, a sparse coding method is used to search for a time frequency point corresponding to the only one non-zero element of coding coefficients, so as to search for time frequency points located in the same 1-D subspace. By minimizing the l1 norm of the coding coefficients and constructing the minimal objective function to extract the single source point, the noise suppression function is further enhanced. Finally, the hybrid matrix estimation is implemented by the hierarchical clustering method.

The invention discloses a visibility measurement method based on machine learning and belongs to image processing technology field, The method comrpiseses extracting ROI images, reprocessing images, establishing a binary tree-based multi-classification support vector machine model; processing the sample image. The technical problem that the recognition accuracy of four grades of good air quality,light piollution, medium pollution and heavy pollution is low since traditional method for extracting the ROI of the image is adopted is solved, a salient image obtained based on an image frequency domain is utilized, an interest pane extracted according to a salient region in the extraction process has representativeness in a picture, the characteristics of the image can be fully reflected, and the characteristic value extracted from the interest pane has high distinction; for images with low visibility, edge information characteristics need to be enhanced to improve the matching accuracy, the method adopts contrast linear broadening to enhance the edge information characteristics, and the method is simple and convenient to install, low in price, high in sensitivity and simple and convenient to operate.

The invention provides a positioning and navigation method and device based on ground texture. The positioning and navigation method comprises: acquiring a ground texture image of a current frame; performing first registration of registration processing on the ground texture image of the current frame and the ground texture image of the reference frame, wherein the registration processing system is used for solving discrete frequency domains corresponding to feature points on frequency spectrums of two ground texture images at least through a discrete phase correlation method so as to obtain relative displacement and a relative angle between the two ground texture images; and in response to the successful first registration, determining the position information of the current frame based on the position information of the reference frame and the relative displacement and the relative angle between the reference frame and the current frame. The invention also provides a positioning andnavigation device for realizing the positioning and navigation method. The positioning navigation method and device provided by the invention have the advantages that the ground texture matching calculation speed is high, the local and global positioning is accurate, and the stability and the adaptability are higher.

The invention discloses a frequency search matching tracking method and system applied to a broadband ultrasonic power supply. The method mainly comprises the following steps: judging whether an absolute value of a difference between a current effective value of a current moment and a maximum value of a current effective value detected before the current moment is greater than a first threshold, if so, calculating a search frequency value at the next moment by using a fuzzy reasoning variable step size algorithm, or otherwise, calculating the search frequency value by using a linear variable step size algorithm; judging whether the search frequency value at the next moment reaches the highest search frequency value, if so, determining the search frequency value corresponding to the currentmaximum value as a resonant frequency value, or otherwise, continuing to iterate until the search frequency value at the next moment reaches the highest search frequency value; and adjusting the frequency value at the current moment according to the resonant frequency value, until the same-phase properties of a voltage phase and a current phase are satisfied when an ultrasonic transducer is at aresonant frequency point. The search matching tracking of the resonant frequency of the ultrasonic transducer within a wider frequency domain is achieved, the timeliness and accuracy of the search tracking are improved, and it is ensured that a resonant system is always in a stable resonant state.