57 results about "Windowed fourier transform" patented technology

Provided is a configuration capable of executing a detection test for a comfort level including the quality of sleep, which is measurable at home without requiring the measurement of brain waves or electrocardiogram. The respiratory waveform of a subject during sleep is continuously measured and recorded from the respiratory gas flow, etc., and is window-Fourier transformed at each measurement time to generate a frequency spectrum, and a bandwidth including a respiratory frequency is extracted. The index indicating the regularity of the respiratory period of the subject is also calculated at each time point during the sleep, and the time-dependency of this index during the sleep is represented as a graph. A medical device includes a sleep evaluation system equipped with a control means for performing control so that a sleep cycle repeated at a cycle of about 90 minutes is clearly observed if the comfort level including the quality of sleep of the subject is favorable.

The invention provides a harmonic analysis method for a power system. The method comprises the following steps: calculating fundamental frequency of the power system and windowing and performing Fourier transform on a sampled signal sequence of the power system; taking the calculated fundamental frequency of the power system as an actual frequency of the power system so as to obtain an approximate formula for windowed Fourier transform spectrum; calculating the frequency of the harmonic wave of the power system according to the calculated fundamental frequency of the power system and calculating the amplitudes of the fundamental frequency and the harmonic wave of the power system according to the approximate formula; and calculating the phases of the fundamental frequency and the harmonic wave of the power system according to the frequency difference between the frequency of the sampled signal of the power system and the frequency of the power system. The method can calculate the harmonic wave of the power system with an extremely high accuracy; and meanwhile, the method has simple calculation and meets the requirements for real-time detection of the power system.

An analyzing method for deriving phase information by analyzing a periodic pattern of moiré comprises steps of: subjecting at least a part of the periodic pattern of moiré to a windowed Fourier transform by a window function; calculating analytically, based on the moiré subjected to the windowed Fourier transform, information of a first spectrum carrying the phase information, and information of a second spectrum superimposed on the information of the first spectrum; and separating the information of the first spectrum from the information of the second spectrum, to derive the phase information.

The invention discloses a broadband signal arriving direction estimation method based on iteration spectral reconfiguration. The realizing process comprises the steps of transforming array antenna receiving data into a frequency domain through windowing Fourier transform, conducting multi-sub-band frequency domain segmentation to the data in the frequency domain, initializing each covariance matrix of each sub-band, obtaining an initialized power matrix of a frequency point according to a covariance matrix of one frequency point in the sub-bands, updating the covariance matrix of a signal at the position of the frequency point according to the initialized power matrix, obtaining corresponding optimum weight of a search angle space by the adoption of the least-squares method, updating the power of each corresponding angle according to the optimum weight on each search angle to obtain the initialized power matrix of a next frequency point and conducting iteration process, and estimating the signal arriving direction through a power spectral peak of the search angle space. The broadband signal arriving direction estimation method based on the iteration spectral reconfiguration can estimate the signal arriving direction of an incidence signal source under the conditions of weak signals and short data, is high in precision, and has the advantage of decorrelation.

The present invention discloses a multiple human-targets location and tracking method based on multichannel phase comparison location. The method comprises: based on a linear frequency modulation continuous wave radar system of single input multiple output, employing the moving target indication and de-mean method to perform preprocessing of experiment data and perform diploid interpolation refinement distance dimension windowing Fourier transform of each period of the echo data so as to obtain object distance information according to the distance spectrum of an object; obtaining the object distance information according to the object distance; extracting the phase information corresponding to the target from the peak value where a frequency spectrum object is located; fusing the phase information of different receiving channels, calculating the phase differences among different receiving channels, calculating the azimuth or the pitch angle of the object, and combining the object distance information to calculate the location result of the object at a two-dimensional plane or a three-dimensional plane; and finally, employing the Kalman filtering method to realize the tracking of the target. The multiple human-targets location and tracking method based on the multichannel phase comparison location can perform detection, location and tracking of a plurality of human body objects indoors and outdoors so as to effectively reject multiple paths while keeping the target and perform accurate location of the object.

An analyzing method for deriving phase information by analyzing a periodic pattern of moire comprises steps of: subjecting at least a part of the periodic pattern of moire to a windowed Fourier transform by a window function; calculating analytically, based on the moire subjected to the windowed Fourier transform, information of a first spectrum carrying the phase information, and information of a second spectrum superimposed on the information of the first spectrum; and separating the information of the first spectrum from the information of the second spectrum, to derive the phase information.

The present invention discloses an MIMO (Multiple-Input Multiple-Output) radar waveform modulation and demodulation method. The method comprises the following steps of: according different transmitting channels at a transmitting end, modulating periodic target detection baseband signals to form different phase codes at the Doppler dimension; converting echo signals received by each receiving signals of a receiving end to digital echo signals after the echo signals are preprocessed, and performing windowed Fourier transform of the digital echo signals according to distance dimensions; converting the digital echo signals after windowed Fourier transform to a matrix in a distance-Doppler two-dimensional form, and performing decoding of the matrix in the Doppler dimension; and performing Doppler dimension coherent processing of the decoded signals to separate each transmitting channel to obtain two-dimensional output of the distance-Doppler frequency. The waveform modulation and demodulation method generates different frequency offsets at the Doppler dimension through phase encoding so as to achieve the purpose of distinguishing different transmitting channels and solve the mutual interference problem between the channels due to increasing of the transmit-receive antennas; and moreover, the signal real-time bandwidth and the computing resource consumption are not increased.

The invention provides a method and a system for recognizing an object type. The method comprises the following steps: intercepting the area of a target to be recognized from an obtained image; obtaining the windowed Fourier transform characteristic pattern of the image in the area of the target to be recognized; utilizing a pre-training object model to detect the windowed Fourier transform characteristic pattern so as to obtain the object type. The windowed Fourier transform characteristic pattern is obtained by extracting windowed Fourier transform characteristics on the image, and then the windowed Fourier transform characteristic pattern is detected by utilizing the pre-training object model so as to obtain the object type, so that the discriminative degree of the object model for different objects can be improved, and the recognition accuracy rate for the object type is improved by 5%.

The invention discloses a power signal synchronized phasor measurement method based on a frequency domain dynamic model. The power signal synchronized phasor measurement method comprises the following steps of collecting the voltage or current signals of a power grid measured by a voltage transformer or a current transformer to obtain a power signal discrete series x(n) of the power grid, then carrying out windowed Fourier transform on the power signal discrete series x(n) to obtain 2L+1 phasor measurement pre-estimated values X(omega) of a deviation 50Hz fundamental wave omega radian; then using the dynamic model based on a frequency domain to carry out synchronized phasor measurement to obtain the corrected values of phasor pre-estimated values, namely obtaining the phasor value of the power signals of the power grid at the current moment trep through phase shift calculation. The method can effectively reduce the measurement errors, the influence of time-varying characteristics of signal fundamental components on phasor measurement can be eliminated or weakened under the dynamic conditions of low frequency oscillation, frequency deviation and the like, the phasor measurement precision of the signals is greatly improved, and the dynamic performance is improved.

The invention discloses a wide-nosed dolphin click signal detection method based on a Gaussian mixture model. Windowing Fourier transform is used to carry out time-frequency analysis on collected ocean signals to acquire an initial signal time-frequency map of the ocean sound signals, so that the implementation of subsequent steps is simple and fast. The soft clustering method of the Gaussian mixture model is used to carry out second clustering on the ocean sound signals, which improves the detection flexibility. The method has the advantages of high detection result accuracy and good integrity, and can be analogously used to detect sound signals of various dolphins.

The invention relates to detection methods of harmonics / inter-harmonics. The detection methods of harmonics / inter-harmonics comprise the following ones: A. distributing signals to different channels 50% of which overlap, and extracting in M times convolution of a filter corresponding to the harmonics / inter-harmonics from the channels; B. performing windowed Fourier transform on the output frequency spectrum after the M times extraction of the harmonics / inter-harmonics; C. calculating power spectrum; D. searching the power spectrum; if the maximum of the power spectrum is more than noise power, then this signal is a harmonic / inter-harmonic, or otherwise the signal is noise; E. adjacent channels with high amplitude is the real ones where lie harmonics / inter-harmonics; F. performing approximate calculations on the points adjacent to the maximum of the power spectrum, and outputting the amplitude, the initial phase and the frequency of corrected channels. The detection methods of harmonics / inter-harmonics can carry out parallel processing on harmonics / inter-harmonics in different channels, and obviously increase real-time performance of the detection and estimation accuracy of parameters. In addition, the methods have the advantages of being easy in algorithm, small in computing load and easy to realize.

A method for identifying and determining the frequency of scattered radiation fringe patterns for applications such as particle or droplet sizing and laser Doppler velocimetery. The method utilizes a series of windowed Fourier transforms performed on an intensity profile of the scattered radiation to locate a segment where a fringe pattern is located on the intensity profile. A standard Fourier transform is then performed on the segment to determine a dominant frequency of the fringe pattern, from which a physical quantity such as diameter or velocity of the particle or droplet may be derived. The method may be utilized with a line or an array sensor to measure the fringe patterns in a spatial domain, or with a point detector to measure the fringe patterns in the time domain.

The invention discloses a frequency-domain blind deconvolution method for a voice signal, comprising the following steps of: converting a time-domain convolution mixed voice signal to a frequency domain and then performing blind separation; converting and transforming the time-domain convolution mixed voice signal to a frequency-domain linear instantaneous mixture model via windowed Fourier transform according to the short-time stability of the voice signal; after performing pre-processing such as filtering and whitening in the frequency domain, realizing segmented blind separation for the voice signal by adopting a method of the approximate joint diagonalization of correlation matrices under different time delays; and after solving the problem of the fuzziness of the blind separation for the signal, performing segmented recombination for the separated signals in the time domain via inverse Fourier transform. Via the frequency-domain blind deconvolution method disclosed by the invention, a good separation effect is realized for 2*2 real-time recoded mixed voice signal, and the recognition accuracy of the voice signal of a human-computer interaction system in an environment with the speech interference of other people can be efficiently improved.

The invention provides an active power distribution network multi-terminal fault identification method and system based on transient signals. The method comprises the following steps: firstly, transient voltage being adopted to construct a protection starting criterion; comparing the integral average value of the zero-mode voltage in the protection time window with a zero-mode setting threshold value, judging whether a grounding fault occurs or not, comparing the integral average value of the transient voltage of each phase after the zero-mode component is filtered out in the protection time window with a set threshold value, and identifying a fault phase; then, according to the fault phase and the fault type, selecting a corresponding transient signal to perform windowed Fourier transform, extracting the voltage and current of a corresponding characteristic frequency band, calculating the transient impedance under the required characteristic frequency band, and judging the fault direction by comparing the root-mean-square value of the transient impedance with the protection setting impedance value; and finally, the fault position being judged through the multi-terminal fault direction information. The method disclosed by the invention is basically not influenced by factors such as transition resistance, DG permeability, network operation state and the like, has relatively lowrequirements on communication conditions, and can realize fault identification and comprehensive protection of the active power distribution network.

The invention discloses a large grounded screen power-frequency grounding impedance measurement method based on a multiple-synchronous different-frequency method. The Large grounded screen power-frequency grounding impedance measurement method specifically includes, by a multiple synchronous measurement method, synchronously measuring multiple groups of different-frequency voltages and different-frequency currents in the process of measuring large grounded screen power-frequency grounding impedance through the different-frequency method; calculating to obtain multiple groups of different-frequency voltage phasors and multiple different-frequency current phasors by means of a windowed Fourier transform filter algorithm; calculating the grounding impedance of the large grounded screen under different frequency through a designated computing method; and calculating the grounding impedance, including resistive component and inductive component, of the large grounded screen under power frequency by an interpolation method. In the large grounded screen power-frequency grounding impedance measurement method, influences of electromagnetic coupling effects between current electrode lead and voltage electrode lead and between parallel overhead lines of a neutral-point grounding system surrounding a measurement circuit and the voltage electrode lead to measurement results are taken into consideration when an outage line is utilized as the measurement circuit for measurement of the large grounded screen grounding impedance, thus precision of the measurement results of the large grounded screen power-frequency grounding impedance is improved greatly.

The invention relates to a method for setting an optical superlattice structure aiming at a coupling nonlinear optical process. Fourier transformation of a window is adopted to replace common whole fourier transformation; a local effective nonlinear coefficient and a spatial weighting function of a reciprocal lattice vector of an optical superlattice are introduced; the competitive relation between the fourier coefficients of each reciprocal lattice vector is considered; the spatial weighting function of the reciprocal lattice vector is subjected to optimized matching; for a triple-frequency coupling process, the spatial weighting function of the reciprocal lattice vector leans to the reciprocal lattice vector of the frequency doubling process in the initial stage and leans to the reciprocal lattice vector of the sum frequency process in the late stage; and through the optimization of the spatial distribution of the local fourier coefficients of two reciprocal lattice vectors, the method realizes the optimization of nonlinear conversion efficiency in the coupling process and carries out the setting of the optical superlattice structure on the basis.

A method for compressing voice frequency of digital deaf-aid includes carrying out time frequency domain conversion by utilizing a windowing Fourier transform, using slope of signal to digital spectrum to judge energy distribution of voice signal and dynamically regulating voice spectrum, realizing low-pass filter by removing off high spectrum coefficient and compressing voice band by regulating spectrum envelope.

The invention provides a single-channel speech enhancement method. The method is realized through a speech enhancement network model. The speech enhancement network model comprises an analysis layer,an encoder, a time convolution module, a decoder and a synthesis layer. According to the single-channel speech enhancement method provided by the invention, an analysis layer of quasi-short-time windowed Fourier transform based on convolution layer design and a synthesis layer of quasi-inverse short-time windowed Fourier transform are added, so that the characteristics of speech are better mined in a transform domain. Besides, the encoder and the decoder are constructed by adopting the gating convolution layer so as to expand the receptive field, transmission of information in the hierarchicalstructure can be better controlled, and the time convolution module is added between the encoder and the decoder so as to better learn the long-term memory characteristic of the speech, so that the speech enhancement effect can be enhanced. Meanwhile, the invention provides a single-channel speech enhancement system and a speech enhancement network model.

The invention provides a distribution cable defect positioning algorithm adopting windowing Fourier transform. The distribution cable defect positioning algorithm comprises the following steps: step 1, testing cable data; and step 2, positioning a local defect / intermediate joint of the cable by inputting an impedance spectrum at the head end of the cable. The distribution cable defect positioning algorithm has the beneficial effects that windowing Fourier transform is applied, the performance of different window functions is analyzed and compared, and the optimal window function is selected to improve the recognition sensitivity and positioning precision of the cable local defect / intermediate joint.

The invention provides a broadband multi-signal detection method, a broadband multi-signal detection device and broadband multi-signal detection equipment. The broadband multi-signal detection method comprises the following steps: collecting broadband signals of a preset frequency band; after windowing Fourier transform is carried out on the broadband signal, obtaining a signal power spectrum of the broadband signal through a logarithm method; preprocessing the signal power spectrum; according to the change trend of the preprocessed signal power spectrum, carrying out edge detection identification on each signal, and marking a left edge point, a right edge point and a signal overlapping point of each signal in the preprocessed signal power spectrum; correcting the left edge point and the right edge point of the signal with signal overlapping according to the signal overlapping point; and outputting a signal detection result according to a matching result of the corrected left edge point and the corrected right edge point and the preprocessed signal power spectrum. According to the scheme, the influence caused by noise, signal intensity difference and signal overlapping can be reduced, and then the signal detection performance and the signal detection accuracy are improved.

Provided is an optical fiber white-light interferometry endoscope three-dimensional measurement system, which comprises an optical fiber interference device and a measuring probe. The optical fiber interference device is formed by a bandwidth coupling optical fiber LED, an optical fiber coupler, piezoelectric ceramics and an optical detector. The measuring probe is formed by optical fiber interference arms and an electronic endoscope CCD. The two optical fiber interference arms are connected with the two output ends of the optical fiber coupler respectively and used for projecting interference fringes to a measured object; the electronic endoscope CCD collects fringe images modulated by the measured object; and three-dimensional measurement of a static object and a dynamic object is realized through a seven-step phase shift algorithm and a windowing Fourier transform method. The system adopts the optical fiber devices and has the advantages of small size, compact structure and good flexibility and the like; the optical fiber white-light interferometry device is utilized to project the fringes, so that fringe density can be improved, and measurement precision is improved, and the system is suitable for high-precision discontinuous surface three-dimensional shape measurement; and with the combination of the endoscope technology and the optical fiber white-light interferometry, the system is allowed to have the advantages of high flexibility and measurement precision.

The invention discloses a DC transient state step response time delay test method. The method comprises the following steps: sampling a standard source signal and a second step oscillation attenuationcurrent signal; solving steady-state direct-current components of the standard source signal and the second-step oscillation attenuation current signal; acquiring step moments of the standard sourcesignal and the second-step oscillation attenuation current signal according to the steady-state direct-current component; solving the oscillation frequency of the second-order step oscillation attenuation current signal after the step and before the stability according to the steady-state direct-current component; adopting windowed Fourier transform to solve fundamental wave phases of the standardsource signal and the second-order step oscillation attenuation current signal; and according to the fundamental wave phase and the oscillation frequency, solving the DC transient state step responsetime delay. The invention further discloses a corresponding system and device. According to the method, direct-current transient step response delay calculation is realized through alternating-current phase compensation of windowed Fourier transform, and errors caused by interpolation calculation in a traditional method are avoided.

The invention belongs to the field of sonar signal processing, and discloses a windowing two-dimensional unrolling multi-beam power spectrum estimation algorithm, which comprises the following steps of: pre-calculating and storing a guide vector universal set, namely a point spread function, of each scanning direction (theta, F) according to an array structure and a processing frequency band; performing windowing Fourier transform by adopting the k-lattice element time domain data to obtain the nth batch of array element frequency domain data; adopting a pre-stored guide vector set to carry out windowing beam former processing to obtain a multi-beam power spectrum estimation result; carrying out two-dimensional unrolling processing on the multi-beam power spectrum windowed in the space domain-time domain by using an R-L iterative algorithm in a discrete two-dimensional integral form, and estimating the real energy distribution estimation PBout of the target; and after the unrolling processing is completed, carrying out two-dimensional spectrum peak search on the PBout, carrying out background estimation by taking the minimum value in the extreme points as background estimation, and carrying out background smoothing on all non-extreme points to obtain new multi-beam power spectrum estimation. According to the method, the performance of passive sonar narrowband warning is effectively improved, and the method has high engineering practical significance.

The invention relates to a frequency-response function measurement method employing pulse-like excitation, and belongs to the technical field of electric system and mechanical system frequency-response function measurement. The method comprises the steps: determining a frequency range; constructing a pulse-like excitation signal as an input signal of a measured system; measuring the input signal of the measured system and an output signal of the measured system; carrying out the windowing Fourier transform of the input and output signals of the measured system through employing a central zero window analysis method, and obtaining a windowed Nyquist frequency spectrum; employing the windowed Nyquist frequency spectrum for calculation, and obtaining the frequency-response function of the measured system. The method can achieve the quick measurement of the frequency-response function. Moreover, the method can obtain high precision which is difficult to achieve through other methods the frequency-response function of a strong resonance system is measured.

A method for identifying and determining the frequency of scattered radiation fringe patterns for applications such as particle or droplet sizing and laser Doppler velocimetery. The method utilizes a series of windowed Fourier transforms performed on an intensity profile of the scattered radiation to locate a segment where a fringe pattern is located on the intensity profile. A standard Fourier transform is then performed on the segment to determine a dominant frequency of the fringe pattern, from which a physical quantity such as diameter or velocity of the particle or droplet may be derived. The method may be utilized with a line or an array sensor to measure the fringe patterns in a spatial domain, or with a point detector to measure the fringe patterns in the time domain.

The invention discloses an optical stripe phase spatial domain demodulation method. The demodulation method comprises the following steps: (1) demodulation is carried out on phases of single-row image signals, and phase values of all pixel points of the row are obtained; (2) phases of signals of images of all rows in a two-dimensional optical stripe are demodulated according to the method in the first step, and the phase values of all pixel points of the entire optical stripe graph are obtained. The method is characterized in that the specific demodulation method in the first step is that a formula as described in the specification is adopted for carrying out demodulation on the phase phi(Xm) of the Xm point; and in the formula, Im1, Im2, Im3, and Im4 are total values of light intensity of four adjacent sub regions. Compared with the prior phase demodulation technology, the method of the invention only operates in the spatial domain, locality of the spatial domain is smaller than two stripe periods, and the highest can reach four pixels. The locality is poorer than the time domain phase shift technology but higher than the windowed Fourier transform technology and the wavelet transformation technology. only one stripe graph is needed for the phase, and the method is suitable for dynamic measurement.

The present invention provides a device and a method for surface shape measurement based on structured light illumination, belonging to the technical field of optical microscopy imaging and measurement. The device is characterized in that: the device comprises a structured light illumination module, an axial scanning module and a detection module. A conventional structural light illumination microscopic system is added with an axial scanning device formed by polarization spectroscopes, low-aperture object lenses, tube lenses and a planar mirror so as to achieve high-speed axial movement of structural light illumination stripes in a space for samples in observation, windowed Fourier transform is employed to perform processing of images shot under different z-directional position stripe projections, the correlated coefficient of each sub area image at a projection strip frequency is calculated, each transverse position sharpness axial response curve is obtained, the peak of the curve isa relative height of the sample at the transverse position, and a surface shape of the sample is finally obtained. The device and the method for surface shape measurement based on the structured lightillumination are simple in mounting and regulation and fast in axial scanning speed, the influence of sample surface reflectivity difference on measurement result is small, a signal to noise ratio ishigh.

The invention discloses a harmonic resonance suppression method and system of a flexible direct current power transmission system, and relates to the technical field of power systems. The method comprises the following steps of: acquiring a three-phase current signal and sending the signal to an analog-to-digital converter to obtain a current digital signal; performing Hanning windowing Fourier transform processing on the current digital signal to obtain resonance information; according to the resonance information, constructing an ideal same-frequency positive-sequence three-phase voltage; performing phase locking processing on the same-frequency positive-sequence three-phase voltage to obtain a phase locking result; performing proportional integral control on a specific harmonic component of the current digital signal based on a phase locking result to obtain a preliminary voltage reference value; and superposing the initial voltage reference value with a voltage reference value generated by the current inner loop to obtain a final voltage reference value. According to the harmonic resonance suppression method, resonance mode information is determined through broadband resonance detection, harmonic current control is corrected according to the resonance mode information, and a set of complete closed-loop control strategy is formed; and the problem of suppression failure caused by the change of the system operation mode in the existing suppression method is avoided.