105 results about "Comb filter" patented technology

The invention provides a directional pickup method, system and device and a storage medium, and relates to the field of pickup. The method comprises the following steps: performing MVDR beam forming processing on an initial voice signal input by a microphone array, and then extracting a BFCC feature and a fundamental tone frequency feature of the voice signal; and processing the BFCC features by using a pre-trained voice / noise separation DNN to obtain a time-frequency mask. And according to the fundamental tone frequency characteristics, adaptively designing comb filter parameters, and carrying out comb filtering on the initial voice signal. And taking the time-frequency mask as a weighting coefficient, respectively multiplying the weighting coefficient with the voice signal after comb filtering and the initial voice signal, and estimating target voice energy and noise energy. And calculating a Bark gain coefficient related to the frequency on the Bark frequency scale, wherein the Bark gain coefficient is used for enhancing the voice signal in the time-frequency domain. And finally, converting the enhanced voice signal into a time domain signal through inverse BFCC conversion to obtain a target enhanced voice pointing to the target direction. Through the scheme, the directional pickup effect is improved.

The invention discloses a power transmission system internal source excitation load and vibration transmission analysis method and system. The method comprises the steps that firstly, gear dynamic meshing rigidity, spline equivalent bearing rigidity and a flexible shaft model are obtained based on parameters of a gear, a spline and a rotating shaft, and then a power transmission system vibration transmission analysis model is established by combining bearing parameters; the method comprises the following steps: respectively obtaining a vibration acceleration signal of each bearing seat position through experiments and simulation, extracting a signal corresponding to each meshing excitation from each bearing seat measuring point signal through a defined comb filter, and calculating the vibration energy contribution of each meshing excitation; finally, through meshing excitation vibration contribution of multiple measuring points and a load transmission rule among parts obtained through simulation, a transmission rule of endogenous excitation load and vibration in the power transmission system is disclosed, vibration characterization of each measuring point in the power transmission system can be accurately obtained based on the method, a theoretical basis is provided for selection of sensitive measuring points, and the method has a wide application prospect. Therefore, support is provided for digital twin modeling and health management of the power transmission system.

The invention provides a three-phase grid-connected phase-locked loop two-stage filter and a filtering method thereof, and belongs to the related technical field of effective detection and extraction of power grid voltage phase signals. The invention provides a three-phase grid-connected phase-locked loop two-stage filter and a filtering method thereof in order to solve the problem that a traditional phase-locked loop filter cannot completely filter harmonic waves and direct-current offset voltage due to insufficient filtering capacity under a non-ideal power grid condition. In order to enhance the anti-interference capability of a phase-locked loop filter, the invention designs a mixed-order generalized integrator and constructs a double-mixed-order generalized integrator. Meanwhile, the double-mixed-order generalized integrator serves as a first-stage filtering unit, and a Parker transformation unit and a comb filter are introduced to serve as a second-stage filtering unit, so that a double-stage filter is formed through combination. By applying the filtering method in the phase-locked loop, it is proved that the method can effectively filter interference in a power grid. The method is mainly used for the grid-connected phase-locked loop filtering process under the non-ideal power grid condition.

A method performed by a programmed processor of an audio system, the method includes retrieving a head-related transfer function (HRTF) dataset from memory, processing the HRTF dataset to remote comb filtering effects during audio playback; applying, to each HRTF in the processed HRTF dataset, a gain to increase a magnitude of the HRTF at a frequency range; and storing the processed HRTF dataset in memory.

The invention discloses a power grid voltage sag detection method based on improved discrete Fourier transform, and belongs to the field of power quality detection, and the method comprises the steps: S1, converting a three-phase power grid voltage signal under a three-phase static coordinate system (a, b, c) into a two-phase static coordinate system (alpha, beta) through 3s / 2s coordinate transformation; s2, enabling the voltage signal under the coordinate system (alpha, beta) to sequentially pass through a comb filter Gf (z), a resonator # imgabs0 # and an adjustment factor Ga (z) so as to obtain a phasor signal of a positive sequence fundamental frequency component; and S3, calculating the mode length of the positive sequence fundamental frequency component phasor signal to obtain the amplitude of the positive sequence fundamental frequency voltage, and comparing the amplitude with a power grid voltage sag detection threshold to judge whether the power grid voltage sags or not. According to the algorithm provided by the invention, the accurate detection of the voltage sag event of the power grid can be realized under the non-ideal power grid condition of harmonic distortion and three-phase voltage imbalance.

The invention discloses a high-order digital decimation filter with a variable decimation multiple, which comprises a cascade comb filter used for downsampling and anti-aliasing filtering, filtering high-frequency components and obtaining low-frequency density signals, and a compensation filter connected to the output end of the cascade comb filter and used for compensating passband attenuation and performing double decimation on the signals, and the output end of the compensation filter is connected with a half-band filter for realizing third-stage signal extraction. The invention further discloses a filtering method of the high-order digital decimation filter with the variable decimation multiple. Aiming at the defects that a traditional single-stage FIR filter is high in hardware requirement, complex in design, high in filter order and not easy to implement and the like, and aiming at the defect that a CIC filter is insufficient in passband flatness, a CFIR compensation filter is used for compensation, and finally a half-band filter is used for frequency reduction.

The invention provides a digital filter circuit and method of a sigma-delta analog-to-digital converter. Relates to the technical field of digital decimation filters, and solves the problems that in the prior art, the line decimation rate is single, and output of a first data is delayed. The input data of the sigma-delta analog-to-digital converter is sent to one of the integrator comb filter and the low-delay filter, and the filter receiving the input data is used as a decimation filter; and the decimation filter is used for opening the enabling of the decimation filter according to the configuration information, adjusting the decimation rate of the decimation filter according to the configuration information, and filtering and sampling the input data according to the decimation rate to obtain filter output data. The digital decimation filtering circuit is suitable for the requirements of low time delay and multiple decimation rates, the time delay for outputting the first data can be reduced by selecting the filter path, and the digital decimation filtering with multiple decimation rates and low time delay can be realized.

The invention relates to a permanent magnet synchronous motor rotor position estimation method based on high-frequency signal injection, and the method comprises the following steps: carrying out the amplitude modulation of a preset high-frequency voltage carrier signal through an orthogonal rotation modulator, and obtaining an injection signal vector; performing voltage excitation on a stator winding of the permanent magnet synchronous motor based on the injection signal vector to obtain a high-frequency current response signal; performing harmonic separation on the high-frequency current response signal through a comb filter bank to obtain a rotor position modulation current component; orthogonal demodulation is carried out on the rotor position modulation current component through a complex demodulator, and a position-dependent phase difference value is obtained; non-linear mapping processing is carried out on the position-dependent phase difference value through an adaptive state observer, a real-time rotor position angle value is obtained, and the key problem of how to effectively extract information related to the rotor position from a high-frequency current response signal is solved.

The invention discloses a cone motion test data processing method based on parallel calculation and harmonic wavelet filtering, and the method comprises the steps: collecting an original voltage signal matrix measured by a wind tunnel force measurement six-component balance through a data collection system when a test model carries out the cone motion in a cone motion wind tunnel test; a principal coefficient matrix and a secondary coefficient matrix are constructed by using a six-component balance formula matrix, and a parallel resolving matrix is constructed by using the principal coefficient matrix and an original voltage signal matrix; iteratively solving a time-domain balance load matrix at each moment in the conical motion wind tunnel test based on the original voltage signal matrix, the principal coefficient matrix, the secondary coefficient matrix and the parallel solution matrix; the method comprises the following steps: determining the center frequency of a harmonic wavelet according to the prior information of a conical motion wind tunnel test, constructing a generalized harmonic wavelet, constructing a comb filter in a frequency domain based on the generalized harmonic wavelet, filtering a frequency domain balance load matrix by using the constructed comb filter, and then converting the frequency domain balance load matrix into a time domain, thereby completing the extraction and reconstruction of load data.

The present application provides a directional sound pickup method, system, device, and storage medium, relating to the field of sound pickup. The method comprises: performing MVDR beamforming processing on the initial speech signal input by the microphone array, and then extracting the BFCC features and fundamental frequency features of the speech signal. The BFCC features are processed using a pre-trained speech / noise separation DNN to obtain a time-frequency mask. Comb filter parameters are adaptively designed based on the fundamental frequency features, and the initial speech signal is comb filtered. The time-frequency mask is used as a weighting coefficient and multiplied with the comb-filtered speech signal and the initial speech signal, respectively, to estimate the target speech energy and noise energy. The frequency-dependent Bark gain coefficient is calculated on the Bark frequency scale and used to enhance the speech signal in the time-frequency domain. Finally, the enhanced speech signal is converted into a time-domain signal through an inverse BFCC transform to obtain the target enhanced speech pointing in the target direction. The above scheme improves the directional sound pickup effect.

The utility model discloses a sound guide rear seat of a condenser microphone and the condenser microphone, and particularly relates to the sound guide rear seat of the condenser microphone and the condenser microphone comprising the sound guide rear seat, the sound guide rear seat comprises a conical head, a base and two identical connecting parts, the conical head is a rotating body; the base is a hollow cylinder; the conical head and the base have the same rotating central axis; the first end of the base is connected with the conical head, and the second end of the base is connected with the two connecting parts. The condenser microphone is provided with the sound guide rear seat, so that the focusing and diffusion degree and direction of rear incoming sound can be changed, the frequency response characteristic of the condenser microphone can be improved, the directional characteristic can be optimized, the comb filtering effect and the sidelobe phenomenon can be inhibited, howling is reduced, and the condenser microphone has the advantages of simple structure, easiness in assembly, good acoustic performance and the like.

The invention relates to the technical field of data acquisition, in particular to a data acquisition method and system for a high-low voltage transformer substation. The method comprises the following steps: acquiring an original electrical signal sequence, calculating an initial integer sampling point number based on a zero crossing point interval, determining a fundamental wave period by carrying out autocorrelation operation on the original electrical signal sequence in a preset window taking the fundamental wave period as a center, and obtaining a signal after primary filtering by adopting first comb filtering taking the fundamental wave period as a delay parameter; and performing short-time Fourier transform on the signal after primary filtering, forming a non-integer harmonic frequency point set in a preset non-harmonic frequency band, constructing a second filtering link according to the non-integer harmonic frequency point set, and filtering non-integer harmonics in the signal after primary filtering to obtain collected data. According to the scheme, the fundamental wave period can be accurately determined to deeply suppress the integer harmonics, the non-integer harmonics can be recognized and filtered out in the environment with the low signal-to-noise ratio, and meanwhile original signal damage is avoided.

The invention belongs to the technical field of optical fiber lasers, and particularly relates to a random laser based on a biconical optical fiber and a multi-wavelength laser output method. The random laser based on the single-mode biconical optical fiber comprises an optical fiber reflector, a biconical optical fiber, an optical fiber coupler, a wavelength division multiplexer, an erbium-doped optical fiber, an optical fiber random Rayleigh scattering feedback unit and an optical fiber 8-degree chamfer which are connected in sequence. A signal light port of the wavelength division multiplexer is connected with an input port of the optical fiber coupler, a public port of the wavelength division multiplexer is connected with an input port of the erbium-doped optical fiber, and a pump light port of the wavelength division multiplexer is connected with an output port of the 980nm laser. According to the invention, multi-wavelength selection in the random laser is realized by using the comb filtering function of the biconical fiber; random Rayleigh scattering in the optical fiber random Rayleigh scattering feedback unit is used as optical feedback, and the random Rayleigh scattering feedback unit and the reflector form a laser cavity. And multi-wavelength laser output is formed.

The invention provides a digital filter, a parameter configuration method and a chip. The digital filter comprises a multi-order sinc filter and a multi-order comb filter which are connected, the order N and the down-sampling rate of the multi-order sinc filter, the order b of the multi-order comb filter and a plurality of corresponding weight coefficients are pre-determined and configured according to the frequency to be filtered; the multi-order sinc filter receives the coded signal, samples the coded signal for N times according to a preset down-sampling rate, and calculates an average value of each sampling result to obtain an average value signal; the multi-order comb filter performs weighting operation on the mean value signal through the first signal processing unit, the second signal processing unit and the corresponding weight coefficients, and sums the mean value signal through the accumulator to obtain an output signal; the output signal is a coded signal after the signal with the frequency to be filtered is filtered. The parameters of the digital filter are flexibly configured based on different application scenes.

The invention discloses an oxyhemoglobin saturation estimation method and system based on a PPG signal, and relates to the technical field of oxyhemoglobin saturation estimation, and the method comprises the following steps: S1, designing a band-pass filter based on a Butterworth filter, designing a comb filter according to the characteristics of motion artifacts in the PGG signal, motion artifacts in the PGG signals are removed through a band-pass filter and a comb filter; s2, performing preliminary screening on PGG signal features through a threshold filtering method based on correlation coefficients; s3, performing wrapping method feature screening from three dimensions of time domain, frequency domain and statistics through a recursive feature elimination method, and extracting a PGG signal feature set; s4, estimating the PGG signal feature set through an integrated regression device based on a random forest algorithm; the blood oxygen saturation information can be extracted from human body PPG signals, and real-time blood oxygen saturation monitoring is achieved.

The invention discloses a mechanical transmission fault prediction method and system based on a probability box theory. The method comprises the steps of data acquisition and preprocessing, probability box model construction, fault feature extraction, fault occurrence probability prediction and equipment maintenance. The invention belongs to the technical field of intelligent production, and particularly relates to a mechanical transmission fault prediction method and system based on the probability box theory, and the method comprises the steps: carrying out the fault feature extraction, segmenting a signal, carrying out the processing of a comb filter, calculating a discrete sequence and frequency response, and defining the change of a noise signal component to evaluate a filtering effect. A sliding window is introduced to capture time dimension information, so that fault features are extracted more accurately and comprehensively; the fault occurrence probability is predicted, specifically, trend analysis and influence factor analysis are carried out, a mechanical transmission effect prediction value is calculated, and prediction of the model on the mechanical transmission effect is closer to the real situation by minimizing an objective function.

The invention provides a tumor in-situ pathology detection system based on an endoscopic multi-mode probe. The tumor in-situ pathology detection system comprises a main controller, an SC laser, a DUV laser, a comb filter, a polarizer, a microarray lens, a multiplexing beam combiner, a conjugate coupling module, an endoscopic optical fiber bundle, a FAST optical fiber bundle, an SC spectrograph and a DUV spectrograph. Aiming at G14-level bronchial navigation surgical robot in-situ pathology detection, the invention provides a detection system based on an endoscopic multi-mode optical probe, FAST hyperspectral image spectrum integrated detection of SC, DUVRS, DUVLIF and mIHC can be realized, a multi-mode hyperspectral fusion image reflecting spatial distribution of histochemical components of a diagnosis point and corresponding spectral feature data are output, and the detection system is used for detecting the in-situ pathology of the G14-level bronchial navigation surgical robot. And real-time, accurate and multi-mode data or information is provided for diagnosis of doctors.

The application discloses a kind of electric energy metering system and method with magnetic field anti-interference ability, belong to electric energy metering technical field.The system includes current connection terminal, multi-winding current transformer, first analog-digital conversion unit, voltage connection terminal, voltage division conversion unit, second analog-digital conversion unit and electric energy metering unit, further include external magnetic field detection unit, active compensation control unit and micro power amplifier.The multi-winding current transformer of the application adopts the structure of double magnetic core+multi-winding, which can effectively detect magnetic field interference current, and also can realize accurate compensation, and has the advantages of simple structure, small size, etc.The active compensation control unit of the application accurately extracts interference signals through multi-stage feedback filter and four-stage comb filter, and combines active compensation control, which can effectively suppress the interference components in the difference current and ensure the accuracy of measurement.

The invention provides a communication antenna steering control system and method, and relates to the technical field of antenna control, and the system comprises a parameter configuration module which is used for obtaining the initial state of a communication antenna, and collecting hardware state information in real time; the signal capturing and tracking module is used for configuring the initial frequency of the numerical control oscillator, starting the parallel correlator group to scan the carrier frequency and calculating frequency domain power spectrum data; and the angle calculation module is used for mapping the resampling phase difference and the baseline parameter into an azimuth angle based on a quaternion solution, and superposing IMU angular velocity data for dynamic compensation. According to the invention, a multi-stage phase-locked loop architecture, a dynamic threshold adaptive algorithm, cascade frequency division coefficient compensation and an FIR comb filter are adopted, so that high-sensitivity acquisition of signals in a broadband is realized; an extended Kalman filtering algorithm and a mechanical lag compensation algorithm are fused, so that the phase lag of the steering engine system is greatly reduced; and in combination with real-time fusion of the IMU angular velocity, the system can still keep relatively high pointing precision in a dynamic scene.

OFDM receiver that determines a time-domain channel impulse response, the receiver comprising: a filter (620) receiving OFDM symbols (610) containing data information and actual pilot information (520, 540, 560), the filter (620) being configured to comb filter and puncture the received OFDM symbols (610) to provide punctured OFDM symbols comprising the actual pilot information (520, 540, 560) and in which data information has been punctured; a memory (630) for receiving and storing the punctured OFDM symbols; a generator (650) coupled to the memory for generating additional pilot information (510) for the OFDM symbols based on the actual pilot information (520, 540, 560) of one or more of the punctured OFDM symbols; a time-domain channel estimator (670) that processes a first punctured OFDM symbol comprising the additional pilot information (510) to generate a channel impulse response for the first punctured OFDM symbol; and a frequency equalizer (690) that equalizes the first punctured OFDM symbol based on the channel impulse response for the first punctured OFDM symbol.

The invention relates to a method for reducing the bias in a periodic signal having a theoretically zero mean value and provided by a sensor associated with a rotary electric machine, in particular by a sensor for determining the angular position of a rotor of a rotary electric machine, the method comprising: - receiving the signal provided by the sensor; - multiplying this signal by a signal proportional to the rotational speed of the rotor of the rotary electric machine in order to obtain a first signal; - at least one integration over half a period of the first signal in order to obtain a second signal; - applying, to the second signal, a finite impulse response filter, in particular a comb filter, so as to add, to the second signal, a delayed version of this second signal in order to obtain a third signal representative of the bias, in particular equal to the bias of the signal provided by the sensor; and - subtracting the third signal from the signal provided by the sensor, so as to reduce the bias in this signal provided by the sensor.

A method operates a hearing aid, in particular a classic hearing aid. The hearing aid includes an input transducer, a signal processing device, and an output transducer. Wherein an occurrence of a comb filter effect is determined by the signal processing device, and wherein a countermeasure is controlled by the signal processing device as a function of the determined occurrence of the comb filter effect.

The invention relates to the technical field of railway electric traction transmission, in particular to an irregular wheel wear detection method based on stator current analysis, which comprises the following steps of: constructing an electromechanical coupling model of a high-speed train traction transmission system based on a traction variable-current driving model and a train-track coupling power model; secondly, based on an electromechanical coupling model, the influence of irregular wear of the wheel on electrical parameters in an electric driving system is analyzed, and the frequency relation between irregular wear excitation and the electrical parameters (stator harmonic current) is established; and by constructing a comb filter based on variable mode decomposition, accurate capture of wheel wear related harmonic waves in stator current is realized, so that irregular wear of the wheel is detected, additional installation of a sensor is not needed, the cost is low, the method is not liable to be interfered by the external environment, and the method has relatively good engineering applicability.

The present invention relates to a Mach-Zehnder interferometer cascade comb filter and a wavelength division multiplexer / demultiplexer to solve the technical problems of wavelength sensitivity introduced by the wavelength division multiplexer of the existing cascade Mach-Zehnder interferometer and its unsuitability for transverse magnetic waves. The filter comprises n-stage comb filter units connected in series, and the m-th stage comb filter unit comprises 2 m A comb filter is provided, each comb filter includes a Mach-Zehnder interferometer, and the Mach-Zehnder interferometer includes a first optical coupling component, a transmission waveguide, and a second optical coupling component arranged in sequence along the transmission direction of the light wave; the first transmission waveguide in the transmission waveguide is provided with a phase delay waveguide, and after TE polarized light and TM polarized light pass through the phase delay waveguide, the phase difference between the TE polarization and the TM polarization is 2πA, where A is a positive integer, and the group refractive index difference between the TE polarization and the TM polarization is within 15%.

When an observation signal including noise the pitch of which changes with the passage of time has been received, a comb filter selection unit (21) selects, from a plurality of comb filters having different pitches to be removed, a comb filter having a pitch to be removed, which matches the pitch of noise, as a selected comb filter in response to the change of the pitch of the noise with the passage of time. A comb filter application unit (23) applies the selected comb filter to the observation signal to remove noise from the observation signal.

The application relates to a digital wave trap, a digital low-pass filter and a design method, a medium and an electronic device. The digital wave trap comprises a wave limit boundary anchoring unit and a wave limit zero point modulation unit which are cascaded in sequence, the wave limit boundary anchoring unit adopts a cascaded integral comb filter structure and is used for generating a first attenuation zero point at a boundary frequency of a target wave trap frequency band; the wave limit zero point modulation unit adopts a finite-length unit impulse response filter structure and is used for generating at least one second attenuation zero point in the target wave trap frequency band; and a continuous suppression band for the target wave trap frequency band is formed in the amplitude-frequency response of an output signal of the wave limit zero point modulation unit by the first attenuation zero point and the at least one second attenuation zero point. By using the digital wave trap provided by the application, a digital low-pass filter which can deeply and widely suppress power frequency interference and maintain strict linear phase and excellent low-pass characteristics can be designed without significantly increasing hardware resources and power consumption.

The invention relates to a multi-wavelength narrow-linewidth laser, comprising: a reflective optical amplifier for generating seed light; the input end of the comb filter is connected with the output end of the reflective optical amplifier; the wavelength division multiplexer is provided with an input end and at least two output ends, and the input end of the wavelength division multiplexer is connected with the output end of the comb filter; the multi-wavelength narrow linewidth laser meets the following condition: [delta] fch (WDM) = m * [delta] f (filter), delta f (WDM) lt; [delta] f (filter); and fFSR (WDM) gt; [delta] f (gain); wherein delta f (filter) is the frequency interval of the comb filter, delta fch (WDM) is the frequency channel interval of the wavelength division multiplexer, delta f (WDM) is the 3dB bandwidth of a single channel of the wavelength division multiplexer, m is a positive integer not less than 2, delta f (gain) is the 3dB gain bandwidth of the reflective optical amplifier, and fFSR (WDM) is the free spectral range of the wavelength division multiplexer.

The invention relates to underwater acoustic signal processing, in particular to a low signal-to-noise ratio underwater acoustic line spectrum enhancement system based on deep learning, which comprises a learnable comb filter module, and is characterized in that a series of learnable multi-channel comb filters with fixed frequency are constructed through a convolutional layer to scan input signals, enhance target frequency and attenuate non-target frequency; the channel fusion module is used for carrying out feature extraction and feature fusion on the output signal of the multi-channel comb filter; the frequency estimation module is used for identifying the line spectrum component of the target frequency from the feature map output by the channel fusion module, screening the frequency bin by introducing a learnable threshold parameter, and generating a frequency index of each frame; the mask enhancement module is used for carrying out adaptive filtering on the input signal by utilizing the frequency index sequence output by the frequency estimation module; according to the technical scheme provided by the invention, the defect of insufficient line spectrum enhancement capability in a complex ocean noise environment in the prior art can be effectively overcome.

A method for reducing bias in a periodic signal with a theoretically zero average value provided by a sensor associated with a rotating electrical machine, in particular by a sensor for determining the angular position of a rotating electrical machine rotor, the method comprising: - receiving the signal provided by the sensor - multiplying this signal by a signal proportional to the rotational speed of the rotating electrical machine rotor to obtain a first signal - at least one integration over half a period of the first signal to obtain a second signal - applying a finite impulse response filter to the second signal, in particular a comb filter, to add a delayed version of this second signal to obtain a third signal representative of the bias, in particular equal to the bias of the signal provided by the sensor, and - subtracting the third signal from the signal provided by the sensor.in order to reduce the bias of this signal provided by the sensor. Abbreviated figure: Fig 3,

The invention discloses a high-resolution spectrometer which comprises a comb filter, a band-pass filter, a detector and data analysis equipment which are connected in sequence, adjusting the transmission peak position of the comb filter and the center frequency of the band-pass filter, performing combined filtering on the incident light through the comb filter and the band-pass filter, measuring the incident light subjected to combined filtering through the detector, generating an analysis signal, and analyzing the analysis signal through the data analysis equipment to obtain a spectrum analysis result; the ultra-narrow bandwidth of the comb-shaped adjustable filter is cascaded with the common adjustable band-pass filter to serve as a scanning engine of the scanning type spectrograph, so that the ultra-high resolution spectrograph is formed.