15 results about "Fourier transform" patented technology

A system and method Fast Fourier Transform (FFT) method in a multi-mode wireless processing system. The method can include loading an input vector into an input buffer, initializing a second counter and a variable N, where N=log₂ (input vector size), and s is the value of the second counter, performing an FFT stage, and comparing s to N and performing additional FFT stages until s=N. Performing the FFT stage can include performing vector operations on data in the input buffer and sending results to an output buffer, the data in the input buffer comprising a plurality of segments, advancing the value of the second counter; and switching roles of the input and output buffers. The vector operations can include performing Radix-4 FFT vector operations on the four input data at a time and multiplying the resulting output vectors with a Twiddle factor.

The invention belongs to numerical simulation of discharge of a Tokamak device in the field of magnetic confinement controlled nuclear fusion, and relates to a simulation method for controlling a newclassic tearing mode through resonance magnetic disturbance in Tokamak. According to the current data in the Tokamak upper coil assembly, the Biot-Savart law is used for solving an induced magnetic field; integrating is performed to obtain an expression of a polar magnetic flux corresponding to the induced magnetic field according to approximation of a large ring diameter ratio; a Fourier transform pseudo-spectrum method is utilized to transform magnetic flux into components of different moduli to obtain three-dimensional distribution of the components in a tokamak median form, the three-dimensional distribution is coupled into a magnetofluid equation, and the Crank-Nicosolson method is utilized to solve the evolution of the magnetic field with the resonance magnetic disturbance along with time, so that the influence of the magnetic field on the new classical tearing mode is studied. The method can simulate the physical processes of new classic mode locking, inhibition, field penetration and the like of the tearing mode, is high in calculation efficiency and relatively good in numerical stability, and is an efficient numerical method capable of accurately simulating an experiment.

A method for removing truncation artifacts in magnetic resonance images based on missing data reconstruction belongs to the technical field of image processing and comprises the following steps: detecting the magnetic resonance images with the truncation artifacts to obtain low frequency K data; carrying out Fourier inversion on the low frequency K data in the frequency encoding direction to obtain a truncated spectrum G (kx, y) of one-dimensional Fourier transform in the phase encoding direction; extracting singularities and singular value in the images of each row of data of G (kx, y) and reconstructing each row of data to obtain the high frequency partial K data; and carrying out Fourier inversion on the high frequency partial K data Fy(k) to obtain the row information fy(x) of the magnetic resonance images with the artifacts removed and storing and combining fy(x) based on rows to form the magnetic resonance images g(x, y) with the artifacts removed. The method overcomes the artifact problem existing during imaging by zero padding, ensures high signal to noise ratio of the images, effectively reduces the image errors, precisely displays the original magnetic resonance images, provides high quality reliable image information for medical nuclear magnetic resonance detection and is conductive to development and popularization of medical imaging detection technology.

The invention discloses a high-precision detection method for SBS (styrene butadiene styrene) contents in SBS modified asphalt. The high-precision detection method includes the steps: preparing various SBS modified asphalt with the known SBS contents; respectively testing an infrared spectrum based on Fourier transformation for the SBS modified asphalt, calculating a specific value A of a characteristic absorption peak area A1 of an SBS modifying agent in the SBS modified asphalt and the sum of the characteristic absorption peak area A1 of the SBS modifying agent and a characteristic absorption peak area A2 of base asphalt, and building a standard curve equation of A values of the SBS modified asphalt corresponding to the SBS contents; testing the SBS modified asphalt to be detected by the aid of the infrared spectrum based on Fourier transformation, calculating corresponding A values, and substituting the corresponding A values into the standard curve equation to obtain the SBS contents in the SBS modified asphalt to be tested. The high-precision detection method has the advantages that the detection method is simple, the influence of the thickness of a modifying asphalt test sample on a test result is eliminated by the aid of the specific value of the characteristic absorption peak areas, section between characteristic absorption peak areas of an original creation infrared spectrum is selected, and the SBS contents in the SBS modified asphalt are rapidly and precisely detected.

The present invention relates to a single horizontal array passive speed measurement and distance measurement method based on sound field interference fringe, belonging to the field of underwater acoustic engineering and sonar signal processing. The method comprises a step of carrying out condition processing and collection on the signal received by each array element of a sonar, a step of carrying out beam formation on the collected signals and obtaining an azimuth time journey map after accumulation, a step of tracking an interested object, taking a tracking beam as object azimuth estimation, carrying out short-time Fourier transform spectrum analysis on the output of the tracking beam, obtaining the interference structural map of the interested object, selecting two equal time intervals of a single horizontal array in a motion process, forming interference structural maps, carrying out preprocessing and Radon transform on the two interference structural maps, and estimating an object heading angle with the combination of the waveguide invariant theory, and a step of realizing the estimation of the distance and speed of the interested object. According to the method, the physical nature of an underwater sound field is fully dug, the prior knowledge of a marine environment is not needed, the algorithm is simple and easy and is convenient and flexible to use, and the application prospect is wide.

The invention discloses a multi-terminal flexible DC network longitudinal protection method and system based on current limiting inductor voltage polarity, which comprises the following steps: collecting voltage values of current limiting inductor at both ends of a DC line in real time; detecting voltage values of current limiting inductor at both ends of the DC line in real time; Voltage gradientalgorithm is used to deal with the voltage, and then to detect whether there is a fault; After the protection is started, the current-limiting inductor voltage in the sampling data window is integrated, and the current-limiting inductor voltage polarity is judged by comparing the integrated value S with the set threshold value. The current limiting inductor voltage polarity R at both ends of theline is compared, and if both polarities are positive, the fault is judged to be in-area fault, otherwise, the fault is judged to be out-of-area fault. The protection principle and algorithm of the invention are simple, and do not need mathematical tools such as fast Fourier transform, wavelet transform and the like for frequency division or traveling wave capture, and also do not need data synchronization at both ends of the line, and are easy to realize. The data window of the protection criterion is short, and no additional fault pole identification step is needed, so the protection criterion is fast.

The invention discloses a microphone array-oriented channel attention weighted speech enhancement method. The method comprises the following steps: respectively acquiring a path of noisy speech time domain signal from each microphone in a microphone array; constructing a multi-channel speech enhancement model; taking a short-time Fourier transform spectrum of each path of noisy speech of the microphone array as an input feature of a multi-channel speech enhancement model, and pre-training the multi-channel speech enhancement model; and inputting multiple paths of noisy speech acquired by the microphone array into the multi-channel speech enhancement model, and realizing microphone array speech enhancement by using the multi-channel speech enhancement model. According to the method, a UNETnetwork is adopted to process the voice frequency spectrum features, the parameter quantity is small, the calculation complexity is low, attention mechanism weighting operation is adopted for the channel dimension on the basis, attention weighting is carried out on each layer of encoder and decoder in the channel dimension, the features are weighted in deep structure simulation beam forming, and the performance of speech enhancement is improved.

A method of calculating the fast Fourier transform or the inverse fast Fourier transform of a series of N real samples x(n), with N power of two, operating according to a time interleaving algorithm and providing the sample series X(n) in ascending order to index n and using limited calculating storage means. A method of calculating the fast Fourier transform or the inverse fast Fourier transform of a series of N conjugated complex samples X(n), with N power of two, operating according to a frequency interleaving algorithm.

Methods and apparatus for forming solar cells with selective emitters are provided. A method includes positioning a substrate on a substrate receiving surface. The substrate has a surface comprising a first patterned heavily doped region having a first dopant concentration that defines the selective emitters, and a second doped emitter region having a second dopant concentration that is less than the first dopant concentration, wherein the second doped emitter region surrounds the first patterned heavily doped region. The method further comprises determining a position of the first patterned heavily doped region by using a Fourier transform to process a filtered optical image, aligning one or more distinctive elements in a screen printing mask with the first patterned heavily doped region by using information received from the determined position of the first patterned heavily doped region, and depositing a layer of material on a portion of the first patterned heavily doped region.

A method and an apparatus are disclosed to detect faults in a rotary machine having a rotor and at least two stator windings. The method comprises generating an input signal and applying the input signal to stator windings while the rotor is locked at a rotor position, measuring output signals from the stator windings and processing all the output signals by Fourier transform to obtain frequency responses of the stator windings, forming derived quantities from all the frequency responses so that the derived quantities are independent of the rotor position, comparing the derived quantities with corresponding reference quantities, determining that there is a fault in the rotary machine if a magnitude or phase difference between the derived quantities and the corresponding reference quantities exceeds a threshold number. The apparatus is designed to test a rotary machine using the inventive method.

The invention discloses a speech synthesis method capable of being used for a chip end. The speech synthesis method comprises the following steps: step 1, calculating a pulse excitation seed signal and a noise excitation seed signal; step 2, solving an excitation signal according to the fundamental frequency F0 of the given voice and the band-shaped aperiodic ratio ap; and step 3, for each frame of frequency spectrum envelope of the given audio, calculating corresponding audio data, and then performing superposition according to frame shift to obtain a final voice waveform. According to the method, the required pulse excitation seed signal is calculated off line in advance, only multiplication and addition operation is performed when periodic excitation and non-periodic excitation are calculated, Fourier transform and inverse Fourier transform are not involved, and the operation speed of the vocoder at a chip end is increased.