36 results about "Vector mode" patented technology

The invention provides a mutual inductor test data cloud edge cooperative processing method and device, and relates to the technical field of data processing, and the method comprises the steps: carrying out the time domain and frequency domain combined feature extraction of the original measurement data of a mutual inductor test through a multi-mode decoupling preprocessing model, and forming a data feature vector set, further generating a confidence label flow through state recognition and Bayesian inference, evaluating a prediction error, and realizing data classification screening and priority queue construction; the data are uploaded to a cloud end in a semantic compression and multi-resolution representation vector mode, so that the transmission efficiency is improved; carrying out deep modeling and model performance monitoring at the cloud, and if degradation is detected, returning edge original data to update the model; and finally, generating a scheduling weight factor according to the prediction error distribution graph, and dynamically optimizing an edge cloud task allocation proportion. According to the invention, the problem of low resource allocation efficiency caused by lack of a dynamic scheduling mechanism based on prediction errors and confidence driving in existing mutual inductor test data edge cloud cooperative processing can be solved.

Discloses an arbitrary polarization mode generator based on a photonic integrated circuit, which includes a plurality of polarization structures arranged on the photonic integrated circuit and forming an array, each polarization structure in the array is configured to disassemble a polarization state in a target mode into the plurality of units correspondingly, and a polarization state of each unit after disassembly is independently completed by one of the polarization structure correspondingly, so that generating any required polarization state is realized. According to the disclosure, the polarization state of the non-uniformity mode in a polarization space is disassembled, and an independent polarization is generated by using the polarization structure, so as to generate any polarization mode including a vector mode can be realized by using on-chip design, having a plurality of advantages including low cost, good stability and high integration degree, and convenient for productization.

Embodiments of the present disclosure provide a solution for video processing. In a method for video processing, for a conversion between a current video block of a video and a bitstream of the video, a block vector (BV) candidate of the current video block associated with a reference block of the current video block is determined. A validation of the BV candidate is determined based on at least one reconstructed sample and at least one unreconstructed sample of the reference block. The conversion is performed based on the validation. The BV candidate is in at least one of the following coding modes or processes: a combined intra block copy and intra prediction (IBC-CIIP) mode, an IBC with geometry partitioning mode (IBC-GPM) mode, an IBC with local illumination compensation (IBC-LIC) mode, a direct block vector (DBV) mode, or a BV difference (BVD) or BVD sign prediction process.

The invention discloses a lane attribute making system, a lane attribute making method and a computer program product. The lane attribute making system comprises a map element encoder and a large language model. An output layer of the map element encoder is connected to an input layer of the large language model, the map element encoder is used for processing a vectorized map to output a vector encoding result, the vectorized map uses vector features to represent map elements, and the map elements comprise lanes; the input layer of the large language model further receives at least one of the image coding result and the text coding result, and the large language model is configured to generate lane attribute data according to at least one of the image coding result and the text coding result and the vector coding result. According to the scheme provided by the embodiment of the invention, the data of various modes including the vector mode can be processed, so that the driving rules are matched to the corresponding lanes, and accurate and detailed lane attribute data are provided for constructing the traffic rule layer.

Embodiments of the disclosure provide a solution for video processing. A method for video processing is proposed. The method includes: deriving, for a conversion between a video unit of a video and a bitstream of the video, a block vector of a direct block vector (DBV) mode for a chroma block of the video unit based on one of: block vectors of luma blocks which are at predefined positions, block vectors of luma subblocks in an N×N granularity, where N is an integer, or a list of block vector candidates; and performing the conversion based on the block vector of the DBV mode for the chroma block.

Photonic units for vector-vector multiplication, matrix-vector multiplication, matrix-matrix multiplication, batch matrix-matrix multiplication, and tensor-tensor multiplication are described. Multiplications are through coherent mixing and square-law detection. There are many dimensions—wavelength, vector mode, quadrature, and three dimensions of space—that can be used to construct photonic accelerators. The encoded input vector or input matrix is fanned out into a desired number of copies and mixed with the corresponding encoded local oscillators containing the weight vectors comprising the weight matrix. Any subset of two (three) dimensions can be used to construct photonic accelerators for matrix-vector (matrix-matrix) multiplications. Multiple dimensions can be combined into a hyperdimension to increase the scalability. Each dimension, each non-overlapping subset of a dimension, or each non-overlapping subset of a hyperdimension, can be used independently to construct a photonic tensor accelerator (PTA) for batch matrix multiplication, or tensor multiplication operations.

The invention discloses a high fundamental frequency permanent magnet motor low carrier wave ratio control method and device based on an optimal switching vector mode. The method comprises the steps of sampling a motor current and a rotor position, and obtaining a dq-axis reference current; calculating a reference stator flux linkage, a reference modulation ratio and a reference voltage phase angle; looking up the table according to the carrier wave ratio and the reference modulation ratio to obtain an optimal switching vector mode and an optimal flux linkage apocentric distance of offline optimization; determining an optimal switching vector mode of the current control time domain; calculating an optimal reference stator flux linkage; calculating a reference voltage vector according to sampling and calculation delay; calculating a switching vector sequence and action time; a switching signal is generated. Through off-line optimization, on-line table look-up and real-time synthesis calculation of the optimal switching vector mode, low-current harmonic and high-dynamic response control of the permanent magnet motor under a low carrier wave ratio is realized, the complexity of a control algorithm is simplified, and the efficiency and reliability of the system are improved.

The application discloses a permanent magnet synchronous motor model prediction torque control method based on a hybrid mode optimization. The method first constructs a reference voltage vector through a torque error target function and a discretization prediction model of the permanent magnet synchronous motor, and then designs a hybrid mode optimization algorithm. The algorithm first designs a double-vector mode optimization on the sector boundary of a voltage space vector diagram, and then designs a three-vector optimization mode based on a discrete space vector modulation inside the sector. The optimal synthesized voltage vector obtained by the hybrid mode optimization algorithm is transmitted to a voltage error target function to realize the switching of the two modes, so that the control execution mode and the optimal synthesized voltage vector are determined. The control method disclosed by the application has high freedom to adjust the amplitude and position of the synthesized voltage vector, and thus has high steady-state performance of the control system and fast torque dynamic response in the full speed range.

The embodiment of the invention provides a vector data processor, an instruction processing method and a system on chip, and the method comprises the steps: determining an instruction calculation mode of a vector instruction according to vector mode configuration information after obtaining the vector instruction needing to be subjected to vector operation; the vector configuration information comprises configuration information for adjusting a data path executed by the current vector through the size of the data block. According to an instruction calculation mode, determining one or more execution assembly lines in the N execution assembly lines for the vector instruction; and executing vector operations corresponding to the vector instructions by execution assembly lines in the vector floating point execution unit respectively. The scheme provided by the invention supports a mixed width calculation mode, so that the vector data processor has the capability of adjusting the data path mode executed by the current vector through the size of the data block, and the throughput of processing small block data is improved; the performance of a large bit width vector register architecture processor when processing software loads with mixed size data block characteristics is improved.

The application provides a method, device and system for light field vector mode conversion, the method comprising: acquiring a cylindrical vector beam field, a fundamental mode Gaussian light array field and a first geometric phase multi-plane light transformation field; determining a first light field and a second light field based on the cylindrical vector beam field, the fundamental mode Gaussian light array field and the first geometric phase multi-plane light transformation field; acquiring a second geometric phase corresponding to a preset condition based on a difference between orthogonal basis vectors of the first geometric phase, the first light field and the second light field and the preset condition; determining a second geometric phase multi-plane light transformation field by using a vector wavefront matching mode based on the second geometric phase; and converting the fundamental mode Gaussian light array field and the cylindrical vector beam field based on the second geometric phase multi-plane light transformation field, thereby realizing multiplexing and demultiplexing of the cylindrical vector beam field, greatly improving the channel capacity of the optical fiber and improving the transmission efficiency of the optical fiber communication.

Embodiments of the present disclosure provide a solution for video processing. In a method for video processing, for a conversion between a current video block of a video and a bitstream of the video, a block vector (BV) candidate of the current video block associated with a reference block of the current video block is determined. A validity of the BV candidate is determined based on at least one reconstructed sample point and at least one non-reconstructed sample point of the reference block. The conversion is performed based on the validity. The BV candidate is in at least one of the following codec modes or processes: a combined Intra Block Copy and Intra Prediction (IBC-CIIP) mode, an IBC (IBC-GPM) mode with a geometric partition mode, an IBC (IBC-LIC) mode with local illumination compensation, a Direct Block Vector (DBV) mode, or a BV Difference (BVD) prediction process or a BVD symbol prediction process.

The embodiment of the present application provides a vector data processor, an instruction processing method and a system on chip, after a vector instruction needing vector operation is acquired, a vector mode configuration information is used to determine an instruction calculation mode of the vector instruction, the vector configuration information includes configuration information of a data path of current vector execution adjusted by a data block size, according to the instruction calculation mode, one execution pipeline or multiple execution pipelines in the N execution pipelines are determined for the vector instruction, and the execution pipelines in the vector floating point execution unit respectively execute vector operations corresponding to the vector instruction. The scheme provided by the present application supports a mixed width calculation mode, so that the vector data processor has the ability to adjust the data path mode of current vector execution by the data block size, improves the throughput of processing small block data, and improves the performance of a large bit width vector register architecture processor when processing a software load with a mixed size data block characteristic.

Embodiments of the present disclosure provide a solution for video processing. A method for video processing is presented. The method includes, for a conversion between a video unit of a video and a bitstream of the video unit, deriving a block vector for a direct block vector (DBV) mode of a chroma block of the video unit based on one of: a block vector of a luma block at a predefined location, a block vector of a luma sub-block of NxN granularity, where N is an integer, or a list of block vector candidates; and performing a conversion based on the block vector of the DBV mode for the chroma block.

Procedure, comprehensive: decoding the first block of a current frame from an encoded video stream, wherein decoding the first block includes the following: decoding a first motion vector from the encoded video stream; decoding (820) an identifier of a first interpolation filter from the encoded video stream; and Reconstructing the first block using the first motion vector and the first interpolation filter; decoding, with a processor, in response to instructions stored on a non-transitory computer-readable medium, a second block of the current frame from the encoded video stream, wherein decoding the second block includes: the identification (840) of the first motion vector from the first block as the selected motion vector for the prediction of the second block in response to the decoding (830) of an identifier of an inter-prediction mode from various possible prediction modes for decoding the second block; Identifying the first interpolation filter as a selected interpolation filter for predicting the second block in response to identifying the first motion vector from the first block as the selected motion vector for predicting the second block; and the reconstruction (850) of the second block using the first motion vector and the first interpolation filter; and output or storage of the first block and the second block; including the various possible prediction modes: a prediction mode with a new motion vector mode in which a new motion vector for the current block is explicitly signaled in the encoded video stream, a prediction mode with a zero-motion vector mode where no motion is present, a prediction mode with a next motion vector mode in which a motion vector that was used to encode an adjacent, previously encoded block is identified as the next motion vector and used to predict the current block, and a prediction mode with a near motion vector mode, in which a motion vector that was used for encoding an adjacent, previously encrypted block and identified as the near motion vector is used for the prediction of the current block.

Embodiments of the present disclosure provide a solution for video processing. In a method for video processing, for a conversion between a current video block of a video and a bitstream of the video, a block vector (BV) candidate of the current video block associated with a reference block of the current video block is determined. A validation of the BV candidate is determined based on at least one reconstructed sample and at least one unreconstructed sample of the reference block. The conversion is performed based on the validation. The BV candidate is in at least one of the following coding modes or processes: a combined intra block copy and intra prediction (IBC-CIIP) mode, an IBC with geometry partitioning mode (IBC-GPM) mode, an IBC with local illumination compensation (IBC-LIC) mode, a direct block vector (DBV) mode, or a BV difference (BVD) or BVD sign prediction process.

The invention belongs to the technical field of image coding, and relates to a phase image coding method and system based on random vector optical complex coherence degree measurement. Phase information is loaded into a spatial distribution phase difference of an incoherent light source orthogonal complex electric field; constructing a cross-spectral density matrix function of the partially coherent vector light field by using a coding system; obtaining a plurality of orthogonal vector electric field mode bases based on a mode decomposition theory and a cross-spectral density matrix function of the partially coherent vector light field so as to obtain a plurality of modes of partially coherent vector light beams on a light source surface; measuring each vector mode of the partially coherent vector light beam from the interference light intensity by using an off-axis holographic technology, and constructing an orthogonal polarization field of the partially coherent vector light beam of each mode; full-dimensional complex coherence degree information is obtained based on an orthogonal polarization field of a partially coherent vector light beam, so that phase information is recovered by utilizing a decoding system, and image coding and complex coherence degree full-dimensional measurement based on vector partially coherent light are realized.

This invention belongs to the field of bearing fault monitoring technology, specifically relating to a precision bearing full-lifecycle fault early warning method based on multi-source vibration signal decoupling. The method includes: acquiring vibration signals from three channels of the bearing; calculating the impact activity level for the current time window based on the local energy of all micro-segments of the spatial vector mode sequence and the number of times the spatial vector mode in the spatial vector mode sequence is greater than the mean of the spatial vector mode sequence; calculating the autocorrelation coefficient and skewness coefficient of the spatial vector mode sequence and obtaining the background noise coupling strength; obtaining the envelope spectral entropy of the spatial vector mode sequence and weighting it using the impact activity level and the background noise coupling strength to obtain a fault characteristic index; and outputting a graded early warning based on the fitting slope of the fault characteristic index for multiple consecutive time windows and an early warning threshold. This invention overcomes the masking of weak features by strong background noise, significantly reduces early missed and false alarms, and achieves high-precision monitoring throughout the entire lifecycle.

The application relates to the technical field of digital signal processing, and discloses a high-precision and low-delay Nth root solving method and a hardware implementation circuit, wherein the method comprises the following steps: calculating the natural logarithm lnR of a radicand R through a generalized hyperbolic high-base CORDIC algorithm in a vector mode; calculating the ratio lnR / N of the natural logarithm lnR and an Nth root of the radicand through a linear high-base CORDIC algorithm in the vector mode; calculating the hyperbolic cosine value cosh(lnR / N) and the hyperbolic sine value sinh(lnR / N) of the ratio lnR / N through a generalized hyperbolic high-base CORDIC algorithm in a rotation mode; and calculating the sum of the hyperbolic cosine value cosh(lnR / N) and the hyperbolic sine value sinh(lnR / N) to obtain the Nth root solution R^(1 / N) of the radicand R. The method and the device can reduce the iteration times in the calculation process, thereby reducing the delay of the Nth root solution, making the scheme adapt to scenes with high timeliness requirements, and reducing the chip area required for implementation, and the implementation power consumption and cost are reduced under the premise of ensuring the precision.

The invention discloses a vector mode generation method and system based on a polarization-maintaining fiber grating. The method comprises the following steps: S1, determining dispersion characteristics of a few-mode fiber and a grating period; s2, on the basis of the dispersion inflection point, a carbon dioxide laser welding machine is adopted, a long-period fiber grating is inscribed on the polarization maintaining fiber, and an LP01 mode is converted into an LP11ax broadband high-order mode, an LP11ay broadband high-order mode, an LP11bx broadband high-order mode and an LP11by broadband high-order mode; and S3, preparing a gradient spiral structure on the polarization-maintaining fiber, packaging the polarization-maintaining broadband long-period fiber grating, and integrating the polarization-maintaining broadband long-period fiber grating with the gradient spiral structure. By adopting the technical scheme of the invention, the technical problems of mode distortion and polarization crosstalk of a traditional device in a broadband scene are solved.

Embodiments of the present disclosure provide a solution for video processing. A method for video processing is proposed. The method comprises: deriving, for a conversion between a video unit of a video and a bitstream of the video unit, a block vector of a direct block vector (DBV) mode for a chroma block of the video unit based on one of: block vectors of luma blocks which are at predefined positions, block vectors of luma subblocks in an NxN granularity, where N is an integer, or a list of block vector candidates; and performing the conversion based on the block vector of the DBV mode for the chroma block.

The invention relates to a simulation method for a fuel-saving orbit change task for "one-to-many" approach to space targets. The objective function is designed with the minimum velocity increment vector modulus required in the orbit change process as the goal; the optimization levels are set, the first level: setting approach sequence optimization variables, and determining the approach sequence of a service aircraft arriving at a target satellite through the variable size; the second and third levels: determining the value range of the approach orbit change end time optimization variable for each target satellite that satisfies both a measurement and control window and an illumination window; determining the value range of the approach orbit change start time optimization variable for each target satellite; the fourth level: under a given approach sequence, optimizing the given approach orbit change end time and approach orbit change start time according to the objective function to obtain an optimization result; according to the set optimization level order, starting the optimization by level, selecting the best result from the optimization results to obtain a fuel-saving orbit change strategy.

The invention provides a dynamic flight information pushing method and system based on machine learning, and belongs to the technical field of artificial intelligence. The method comprises the following steps: collecting flight demand information input by a user, and forming three types of structured arrays in combination with predefined supply and demand characteristics, time sequence characteristics and temporary characteristics; inputting the supply-demand feature array and the time sequence feature array into a prediction model of machine learning training to obtain an initial price fluctuation curve; retrieving price fluctuation vector mode values of the historical flight sample sets meeting the temporary features, and setting the price fluctuation vector mode values as correction vectors; and dynamically adjusting the initial curve according to the correction vector to obtain an updated price fluctuation curve, and pushing the updated price fluctuation curve to the user according to a predicted price sequence. Through accurate prediction of the flight price, more accurate flight information pushing is provided, and the user experience is improved.

The invention discloses a magneto-dependent sensor immunity test device and method for simulating a complex multi-dimensional magnetic field. The device comprises a signal simulation and generation unit, a magnetic field generation unit, a magnetic field measurement and monitoring unit and a closed-loop control unit. The signal simulation and generation unit generates a customizable driving electric signal according to the target waveform data; the magnetic field generation unit generates composite time-varying magnetic fields which are different in direction and can be superposed in a vector mode in a magnetic field generation chamber through at least three independent excitation coils which are arranged in a nested mode. The magnetic field measuring and monitoring unit monitors and feeds back an actual magnetic field signal in the field in real time; the closed-loop control unit dynamically adjusts signal output by comparing an actual signal with a target signal to form high-precision closed-loop control. By means of multi-axis coil nesting and a closed-loop vector control mechanism, flexible generation and accurate control of an ultra-wideband and high-strength space vector magnetic field are achieved, and the efficiency and reliability of product research and development and quality verification are improved.

The invention discloses a cubic root solving method and solving device based on an extended CORDIC algorithm, which considers the characteristics of digital signal processing in a super-large-scale integrated circuit, and divides the structure of cubic root calculation into three steps: solving a natural logarithm ln (R), where R is an input digital signal and represents the amplitude of a signal to be processed, and R is an output digital signal and represents the amplitude of the signal to be processed; in the step, the signal is converted into a logarithmic form, and a natural logarithmic signal is obtained. The step is realized by expanding a vector mode of the CORDIC algorithm in a hyperbolic coordinate system. And further processing the natural logarithm signal, obtaining ln (R) / 3 by a solving method, and obtaining a division result signal. The step is realized by expanding a vector mode of the CORDIC algorithm in a linear coordinate system. And an e index is solved, and index processing is carried out on the division result signal. The device comprises a natural logarithm calculation module, a division calculation module and an e index calculation module. According to the method, a plurality of calculation stages are processed in parallel, the delay is reduced, and the requirements of real-time processing and high-performance calculation are met.

The application discloses an image pre-analysis method, applied to a pre-analysis module of an encoder. The method includes: performing downsampling on a to-be-processed image, and dividing the to-be-processed image into square blocks of a same size; performing inter-frame prediction on a current block based on a single-motion-vector mode, to determine a best cost and a corresponding prediction direction; performing inter-frame prediction on the current block based on a pseudo-affine transformation mode by using the prediction direction as a search direction of the pseudo-affine transformation mode, to obtain an affine transformation cost; and comparing a value of the best cost with a value of the affine transformation cost, and determining, based on a comparison result, a best mode for performing pre-analysis inter-frame prediction on the current block. The application further discloses an image pre-analysis system, an electronic apparatus, and a computer-readable storage medium.

The invention discloses a production and transportation collaborative optimization task set allocation method based on cloud manufacturing, and relates to the technical field of cloud manufacturing and production, and the method comprises the steps: coding the allocation conditions of manufacturing service providers of all task sets through employing a one-dimensional positive integer vector mode, and randomly generating an initial solution; according to the method, efficient modeling of the distribution relation of the task set group on the distributed manufacturing service provider is realized by adopting a one-dimensional positive integer vector coding mode; in combination with an optimal sorting method for minimizing completion time of a single MSP, determining an optimal workpiece group processing sequence and a processing sequence of workpieces in the group at the same time; and a closed expression of the completion time of each machine is obtained based on mathematical derivation, so that the performance of the whole scheduling scheme is effectively evaluated. By introducing a contrast disturbance mechanism, a multi-neighborhood search structure and an iterative greedy strategy, a multi-level and multi-angle search cooperation mechanism is formed, and the diversity retention ability and the deep search ability of the algorithm in a solution space are significantly enhanced.

The invention relates to the technical field of data anomaly detection, in particular to a hardware acceleration method and device for data anomaly detection, and the method comprises the steps: inputting two groups of 16Bit-wide data a (xa, ya) and b (xb, yb) into a pipeline adder for complement addition calculation, and obtaining the calculation results of xa-xb and ya-yb; and solving a data module value by utilizing a vector mode of a Cordic algorithm through iterative calculation of original data. The construction method of the pipeline adder comprises the following steps of: cascading two 4Bit carry lookahead adders to form an 8Bit carry lookahead adder, and forming a 16Bit pipeline adder by two 8Bit carry lookahead adders and three registers; wherein one 8Bit carry lookahead adder is used for low 8Bit addition calculation, and the other 8Bit carry lookahead adder is used for high 8Bit addition calculation; and the three registers are used for adjusting a synchronous time sequence. According to the method and the device, the problems that the calculation speed for solving the Euclidean distance is relatively low and the detection time consumption is relatively long under the condition that the LOF anomaly detection algorithm is used for processing a large data volume can be solved.

The invention provides a super lens, an optical element and a light beam modulation method, which can be applied to the field of super lenses. The super lens comprises a plurality of micro-units which are arranged at intervals in a first direction; wherein the plurality of micro-units comprise at least one first micro-unit, and the first micro-unit has a first rotating angle relative to the first direction, so that the extension direction of the first micro-unit is vertical to the polarization direction of the first light beam with the first high-order vector mode; the first micro unit is used for selectively allowing the first light beam to transmit under the irradiation of at least two light beams including the first light beam and reflecting other light beams in the at least two light beams; in the process of transmitting the first light beam, wavefront modulation is carried out on the first light beam based on the structure size of the first micro-unit, so that a first target light field is formed based on the modulated first light beam; wherein the at least two light beams have different high-order vector modes with respect to each other.

Photonics units for vector-vector multiplication, matrix-vector multiplication, matrix-matrix multiplication, batch matrix-matrix multiplication, and tensor-tensor multiplication are described. Multiplication is performed through coherent mixing and square-law detection. There are many dimensions-wavelength, vector mode, quadrature, and spatial-three that can be used to build photonic accelerators. An encoded input vector or input matrix is fan-out into the required number of copies and mixed with the corresponding encoded local oscillators containing weight vectors, where the weight vectors comprise a weight matrix. Any subset of two (three) dimensions can be used to construct a photonic accelerator for matrix-vector (matrix-matrix) multiplication. Multiple dimensions can be combined into hyper-dimensions to increase scalability. Each dimension, each non-overlapping subset of dimensions, or each non-overlapping subset of hyper-dimensions can be independently used to construct a photonic tensor accelerator (PTA) for batch matrix multiplication or tensor multiplication operations.

The present invention discloses a chiral multi-dimensional, multi-band controllable OAM beam generator, comprising a laser light source and a multi-core twisted optical fiber. The core of the multi-core twisted optical fiber has a certain twist rate. Light from the laser light source enters the multi-core twisted optical fiber, where the central core and adjacent cores couple to each other, enabling the central core to selectively couple with a specific "satellite core" over a wide wavelength range, thereby exciting the higher-order modes of the "satellite core." Because the twisted higher-order modes transform from scalar modes to vector modes, carrying orbital angular momentum, the selective coupling of the central core of the twisted optical fiber with the "satellite core" excites multiple orbital angular momentum modes of the "satellite cores," thereby generating a multi-dimensional, multi-band OAM beam. By varying the twist rate of the twisted optical fiber, the wavelength of the light, and the distance between the cores, the central core can selectively couple with a specific "satellite core," achieving flexible selection of the excitation OAM beam.