32 results about "Coefficient quantization" patented technology

The invention provides a TLF-GV signal correlation earthquake three-element joint probability prediction method and system, belongs to the technical field of earthquake prediction, and aims to solve the problems that traditional three-element prediction lacks joint fusion, evidence conflicts are difficult to process, and weights are fixed. Comprising the following steps: extracting physical features of TLF-GV signals, carrying out preorder three-element prediction results and model credibility parameters, splicing the physical features into feature vectors, inputting the feature vectors into a dynamic weight prediction model constructed by a shallow MLP, and outputting time, magnitude and position weights; dividing the three elements into 3 * 3 * 3 = 27'time-magnitude-position 'propositions, generating basic probability assignment (BPA) of each proposition based on weights, and quantifying evidence contradictions through a conflict coefficient K; and adopting a Dempster combination rule to obtain a joint BPA, calculating a mean value of a reliability function and a likelihood function of each proposition as a comprehensive probability, and extracting a maximum probability proposition as a coupling result. According to the method, the scientificity and practicability of strong earthquake short-term and temporary prediction are remarkably improved.

The invention relates to a low latency low frequency effect codec. In some embodiments, a method of encoding a low frequency effect (LFE) channel includes receiving a time domain LFE channel signal; filtering the time domain LFE sound channel signal by using a low-pass filter; converting the filtered time domain LFE channel signal into a frequency domain representation of the LFE channel signal comprising a number of coefficients representing a spectrum of the LFE channel signal; arranging coefficients into a number of subband groups corresponding to different frequency bands of the LFE channel signal; quantizing coefficients in each secondary frequency band group according to a frequency response curve of the low-pass filter; encoding the quantized coefficients in each subband group using an entropy decoder that is tuned for the subband groups; and generating a bitstream comprising the encoded quantized coefficients; and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

Methods and apparatus for image compression / decompression using tensor product B-spline (TPB) representations. According to an example embodiment, an image compression method includes generating a plurality of TPB models representing an image, the plurality of TPB models including a TPB model for approximating a spatial remodeling function and a plurality of block-by-block TPB models for estimating an image signal. In some examples, the plurality of tile-by-tile TPB models includes a plurality of luma channel TPB models and a plurality of chroma channel TPB models. The spatial remodeling function is configured to transform a non-uniform local content complexity distribution in the image towards a more uniform transition. The method also includes generating a metadata stream carrying metadata representing a set of coefficients for the various TPB models. At least a portion of the metadata is generated by quantizing corresponding coefficients to a smaller number of bits and applying arithmetic coding to the quantized coefficients.

The invention discloses a non-uniform quantization method and device based on density sensing clustering, equipment and a storage medium, and relates to the technical field of network communication, and the method comprises the steps: obtaining a tap coefficient of an equalizer; performing clustering division on the tap coefficient according to a preset quantization bit width to obtain a plurality of target clusters; determining a quantization parameter corresponding to the target cluster according to the dynamic range of the intra-cluster coefficient in the target cluster, wherein the quantization parameter comprises a scaling factor and a zero offset parameter; and independently and uniformly quantizing the tap coefficient in the target cluster according to the quantization parameter to obtain a recovery coefficient. According to the invention, the coefficient is adaptively divided into a plurality of clusters based on density sensing clustering, and uniform quantization is independently performed on each cluster, so that high-precision quantization of the equalizer coefficient under extremely low bits is completed, and the storage efficiency of coefficient quantization and the performance of the equalizer are improved.

The invention relates to a low latency low frequency effect codec. In some embodiments, a method of encoding a low frequency effect (LFE) channel includes receiving a time domain LFE channel signal; filtering the time domain LFE sound channel signal by using a low-pass filter; converting the filtered time domain LFE channel signal into a frequency domain representation of the LFE channel signal comprising a number of coefficients representing a spectrum of the LFE channel signal; arranging coefficients into a number of subband groups corresponding to different frequency bands of the LFE channel signal; quantizing coefficients in each secondary frequency band group according to a frequency response curve of the low-pass filter; encoding the quantized coefficients in each subband group using an entropy decoder that is tuned for the subband groups; and generating a bitstream comprising the encoded quantized coefficients; and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

The application discloses a kind of stability detection method, device, equipment and medium of multi-conversion station system, it is related to power system technical field, method includes: obtaining the impedance matrix of multi-conversion station system;Obtain the port admittance matrix of converter in multi-conversion station system;According to impedance matrix and port admittance matrix, construct the return difference matrix of multi-conversion station system;Generalized radius coefficient of return difference matrix is defined;If generalized radius coefficient is less than preset threshold in research frequency band, then stability criterion is simplified based on generalized diagonal dominance, whether multi-conversion station system is stable according to the stability criterion after simplification is detected.The application constructs the return difference matrix of multi-conversion station system, and the coupling relationship between converter is quantified according to generalized radius coefficient of return difference matrix, if generalized radius coefficient is less than preset threshold, then it can be determined as weak coupling relationship, and then stability criterion is simplified based on generalized diagonal dominance, can reduce the complexity of detecting the stability of multi-conversion station system and improve detection efficiency.

The invention discloses a time-frequency fusion Hokes dynamic wireless sensor network topology hierarchical inference method, which comprises the following steps of: converting radio frequency signals acquired by a multi-sensor array into a time-frequency domain characteristic space, and establishing a dynamic mapping relation between node space positions and electromagnetic characteristics; the method comprises the following steps: constructing an event triggering relationship between nodes based on a Horkes process, quantifying space-time correlation through a baseline intensity parameter and a dynamic excitation coefficient, and setting a historical event influence attenuation rate; subnets are initialized through boundary sensing nodes, the mass center of the subnets is calculated according to the node degree inverse proportion weight, uncovered nodes which are closest to the mass center and have the feature similarity reaching the standard are preferentially expanded, and the node overlapping rate between the subnets is controlled to be not lower than a preset value; and confidence weighted fusion is carried out on the reasoning results of the subnets, time-frequency feature backtracking verification is carried out on conflicting links, and a dynamic adjacency matrix with a timestamp is generated. According to the method, high-precision link detection can be realized in a dynamic network, and the method has remarkable dynamic adaptability and calculation efficiency.

The invention relates to a low latency low frequency effect codec. In some embodiments, a method of encoding a low frequency effect (LFE) channel includes receiving a time domain LFE channel signal; filtering the time domain LFE sound channel signal by using a low-pass filter; converting the filtered time domain LFE channel signal into a frequency domain representation of the LFE channel signal comprising a number of coefficients representing a spectrum of the LFE channel signal; arranging coefficients into a number of subband groups corresponding to different frequency bands of the LFE channel signal; quantizing coefficients in each secondary frequency band group according to a frequency response curve of the low-pass filter; encoding the quantized coefficients in each subband group using an entropy decoder that is tuned for the subband groups; and generating a bitstream comprising the encoded quantized coefficients; and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

Systems, methods, and devices for wireless communication are described. A method of wireless communication includes receiving, by a wireless device, configuration signaling. The method also includes determining, by the wireless device, a precoding matrix indicator (PMI), where the PMI is based on a codebook indicated by the configuration signaling and the PMI corresponds to a precoding matrix applicable to more than 32 antenna ports of a network device. The method also includes transmitting, by the wireless device, the PMI. In some embodiments, the method includes configuration of oversampling factors, configuration of codebook subset restriction, selection of a spatial domain (SD) / frequency domain (FD) basis, non-zero coefficient indication, and non-zero coefficient quantization.

The invention provides a ship sensing data optimization method and device, and the method comprises the steps: determining a correlation calculation coefficient of each parameter sensor based on the distribution characteristics of to-be-optimized ship data of each parameter sensor, and then decomposing the to-be-optimized ship data based on a variational mode decomposition algorithm, the correlation between the ship data to be optimized and a plurality of intrinsic mode functions is quantitatively evaluated based on correlation calculation coefficients corresponding to the parameter sensors, and finally partition optimization reconstruction is performed based on the correlation, so that multi-source ship data optimization which automatically adapts to data features and dynamically processes real-time data is realized. The method effectively overcomes the dependence of a traditional method on artificial experience, can carry out efficient and accurate denoising processing on sensor data of a marine nuclear power system under a complex background, remarkably improves the data quality, provides more reliable data input for subsequent ship state analysis, evaluation and control decision, and improves the reliability of data processing. Therefore, safe and stable operation of the nuclear power system is guaranteed.

The invention relates to a low latency low frequency effect codec. In some embodiments, a method of encoding a low frequency effect (LFE) channel includes receiving a time domain LFE channel signal; filtering the time domain LFE sound channel signal by using a low-pass filter; converting the filtered time domain LFE channel signal into a frequency domain representation of the LFE channel signal comprising a number of coefficients representing a spectrum of the LFE channel signal; arranging coefficients into a number of subband groups corresponding to different frequency bands of the LFE channel signal; quantizing coefficients in each secondary frequency band group according to a frequency response curve of the low-pass filter; encoding the quantized coefficients in each subband group using an entropy decoder that is tuned for the subband groups; and generating a bitstream comprising the encoded quantized coefficients; and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

Methods and apparatus for image compression / decompression using tensor product B-spline (TPB) representations. According to an example embodiment, an image-compression method includes generating a plurality of TPB models representing an image, including a TPB model for approximating a spatial reshaping function and a plurality of patch-wise TPB models for estimating the image signal. In some examples, the plurality of patch-wise TPB models includes a plurality of luma-channel TPB models and a plurality of chroma-channel TPB models. The spatial reshaping function is configured to shift a non-uniform distribution of local content complexity in the image toward being more uniform. The method also includes generating a metadata stream carrying metadata representing sets of coefficients of various TPB models. At least a portion of the metadata is generated by quantizing the corresponding coefficients to a smaller number of bits and applying arithmetic coding to the quantized coefficients.

A channel feedback information transmission method and apparatus, relating to the technical field of wireless communications. The method comprises: a second device receives a first signal from a first device; and the second device sends channel feedback information to the first device, the channel feedback information indicating non-zero coefficient quantization information and non-zero coefficient selection information in a coefficient matrix, and the coefficient matrix being obtained on the basis of a spatial domain basis and a frequency domain basis. The spatial domain basis and the frequency domain basis satisfy a first option, a second option, or a third option; the first option is that the spatial domain basis consists of all the vectors in a spatial domain basis matrix, and the frequency domain basis consists of all the vectors in a frequency domain basis matrix; the second option is that the spatial domain basis consists of a subset of vectors in the spatial domain basis matrix, and the frequency domain basis consists of all the vectors in the frequency domain basis matrix; and the third option is that the spatial domain basis consists of all the vectors in the spatial domain basis matrix, and the frequency domain basis consists of a subset of vectors in the frequency domain basis matrix.

This invention provides a method and system for joint probabilistic prediction of three earthquake elements associated with TLF-GV signals, belonging to the field of earthquake prediction technology. It aims to solve the problems of traditional three-element prediction lacking joint fusion, difficulty in handling evidence conflicts, and fixed weights. The method includes: extracting the physical characteristics of the TLF-GV signal, taking the prediction results of the previous three elements and model confidence parameters, concatenating them into a feature vector, inputting it into a dynamic weighted prediction model constructed by a shallow MLP, and outputting time, magnitude, and location weights; dividing the three elements into 3×3×3=27 "time-magnitude-location" propositions, generating a basic probability assignment (BPA) for each proposition based on the weights, and quantifying evidence contradictions through a conflict coefficient K; obtaining the joint BPA using the Dempster combination rule, calculating the mean of the confidence function and likelihood function of each proposition as the comprehensive probability, and extracting the proposition with the highest probability as the coupling result. This method significantly improves the scientific rigor and practicality of short-term strong earthquake prediction.

The invention relates to a low latency low frequency effect codec. In some embodiments, a method of encoding a low frequency effect (LFE) channel includes receiving a time domain LFE channel signal; filtering the time domain LFE sound channel signal by using a low-pass filter; converting the filtered time domain LFE channel signal into a frequency domain representation of the LFE channel signal comprising a number of coefficients representing a spectrum of the LFE channel signal; arranging coefficients into a number of subband groups corresponding to different frequency bands of the LFE channel signal; quantizing coefficients in each secondary frequency band group according to a frequency response curve of the low-pass filter; encoding the quantized coefficients in each subband group using an entropy decoder that is tuned for the subband groups; and generating a bitstream comprising the encoded quantized coefficients; and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

The invention relates to a low latency low frequency effect codec. In some embodiments, a method of encoding a low frequency effect (LFE) channel includes receiving a time domain LFE channel signal; filtering the time domain LFE sound channel signal by using a low-pass filter; converting the filtered time domain LFE channel signal into a frequency domain representation of the LFE channel signal comprising a number of coefficients representing a spectrum of the LFE channel signal; arranging coefficients into a number of subband groups corresponding to different frequency bands of the LFE channel signal; quantizing coefficients in each secondary frequency band group according to a frequency response curve of the low-pass filter; encoding the quantized coefficients in each subband group using an entropy decoder that is tuned for the subband groups; and generating a bitstream comprising the encoded quantized coefficients; and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

The present disclosure relates to low-latency low-frequency effect codecs. In some embodiments, a method of encoding a low-frequency effect (LFE) channel includes receiving a time-domain LFE channel signal, filtering the time-domain LFE channel signal using a low-pass filter, converting the filtered time-domain LFE channel signal to a frequency-domain representation of the LFE channel signal including a number of coefficients representing a spectrum of the LFE channel signal, arranging coefficients into a number of sub-band groups corresponding to different frequency bands of the LFE channel signal, quantizing coefficients in each sub-band group according to a frequency response curve of the low-pass filter, encoding the quantized coefficients in each sub-band group using an entropy coder tuned for the sub-band group, and generating a bitstream including the encoded quantized coefficients, and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

Methods and apparatus for image compression / decompression using tensor product B-spline (TPB) representations. According to an example embodiment, an image-compression method includes generating a plurality of TPB models representing an image, including a TPB model for approximating a spatial reshaping function and a plurality of patch-wise TPB models for estimating the image signal. In some examples, the plurality of patch-wise TPB models includes a plurality of luma-channel TPB models and a plurality of chroma-channel TPB models. The spatial reshaping function is configured to shift a non-uniform distribution of local content complexity in the image toward being more uniform. The method also includes generating a metadata stream carrying metadata representing sets of coefficients of various TPB models. At least a portion of the metadata is generated by quantizing the corresponding coefficients to a smaller number of bits and applying arithmetic coding to the quantized coefficients.

The application relates to a voltage-current oversampling and dynamic interpolation method, device and storage medium, wherein during output, a target voltage is used to calculate a DAC digital quantity, the DAC digital quantity is decomposed into an integer reference value and a decimal part, a decimal error is converted into a time domain adjustment parameter through coefficient quantization, and finally in a continuous output period, two adjacent levels are alternately output by the DAC according to the parameter, the output time proportion is accurately controlled, and the average voltage is forced to approach a set value. During back reading, voltage and current signals are synchronously collected in a fixed high-frequency period, high-signal-to-noise ratio intermediate data units are formed through real-time accumulation, a dynamic circular buffer is used to adaptively average a plurality of units, a high-precision conversion value is output, and the dynamic circular buffer is quickly reset when data suddenly changes. The application cooperates a common-precision ADC and a DAC through a pure software algorithm, effectively breaks through the limitation of inherent resolution of hardware, and improves the system measurement precision and output accuracy without increasing the hardware cost.

The invention provides an interlayer fluid channeling coupling interpretation method and device for layered injection well test data and electronic equipment. Comprising the following steps: identifying whether fluid channeling exists between layers based on pressure time sequence data collected by each layer in a discharge and mining test process, and quantifying a channeling channel; the quantized fluid channeling result is used for correcting the actually-measured flow data of all the layers so as to eliminate fluid channeling interference, and the real stratum parameters of all the layers are inversed based on the corrected data; and establishing a coupling relation model of the layered real flow and the interlayer cross flow, performing comparison verification on the total flow calculated by the model and the actually measured total flow, and performing iterative optimization on model parameters according to a verification result. Interlayer fluid channeling can be quantified through pressure curve synchronous fluctuation characteristics and fluid channeling coefficients, the fluid channeling which is regarded as interference in a traditional model is converted into interpretable parameters, and the interpretation precision is remarkably improved.

Various aspects of the present disclosure generally relate to wireless communications. For example, time-varying channel state information (CSI) content may be associated with group 2 of CSI portion 2, such that the time-varying CSI content may be first omitted from CSI portion 2. Some aspects are more particularly directed to associating content related to Doppler domain base selection, coefficient quantization associated with Doppler domain bases having an index greater than zero, and / or a non-zero coefficient selection bitmap, and / or content related to channel quality indicators for slots in a CSI window other than a first slot, with CSI Part 2, Group 2. Thus, the time-varying CSI content associated with group 2 of CSI part 2 may be finally packed and / or first omitted in the event that an uplink resource allocation for CSI part 2 is insufficient for all of the CSI content extrapolated by the UE for the CSI window.

The application discloses a kind of Σ-ΔADC downsampling digital filter floating point coefficient fixed-point unit design method, with all-digital implementation, fusion is based on the optimal coding of graph theory, hybrid coding combination and multiplication-addition operation optimization technique.By the hybrid coding method based on graph theory, multiplication-addition law operation hardware resource is reduced, operation speed is improved, and while guaranteeing the slight drop of delay, the area and power consumption are significantly optimized.In addition, the optimization carry chain rearrangement algorithm is used to maximize the fanout number of adder and the efficiency of connection.Compared with the traditional floating point coefficient fixed-point encoding and adder method, the present application can effectively realize the multiplication-addition operation optimization of downsampling filter in Σ-ΔADC digital module, significantly reduce the power delay product on the basis of keeping similar area, significantly enhance the comprehensive performance of downsampling digital filter floating point coefficient fixed-point unit, and be suitable for multiplication-addition operation optimization of downsampling filter in Σ-ΔADC digital module.

A method and apparatus for image compression / decompression using a tensor product B-spline (TPB) representation. According to one example embodiment, the image compression method includes generating a plurality of TPB models representing an image, including a TPB model for approximating a spatial reshaping function and a plurality of patch-wise TPB models for estimating an image signal. In some examples, the plurality of patch-wise TPB models include a plurality of luma channel TPB models and a plurality of chroma channel TPB models. The spatial reshaping function is configured to shift a non-uniform distribution of local content complexity within the image to a more uniform distribution. The method also includes generating a metadata stream carrying metadata representing sets of coefficients of the various TPB models. At least a portion of the metadata is generated by quantizing corresponding coefficients to a fewer number of bits and applying arithmetic coding to the quantized coefficients.

The invention relates to a voltage and current oversampling and dynamic interpolation method and device and a storage medium, and the method comprises the steps: calculating a DAC digital quantity according to a target voltage during output, decomposing the DAC digital quantity into an integer reference value and a decimal part, and converting a decimal error into a time domain adjustment parameter through coefficient quantization; and finally, in a continuous output period, dynamically controlling the DAC to alternately output two adjacent levels according to the parameter, and accurately regulating and controlling the output time proportion to enable the average voltage to approach a set value. Voltage and current signals are synchronously acquired at a fixed high-frequency period during backward reading, and an intermediate data unit with a high signal-to-noise ratio is formed through real-time accumulation; a dynamic circular buffer area is adopted to carry out adaptive average processing on a plurality of units, a high-precision conversion value is output, and rapid resetting is carried out when data suddenly changes; according to the invention, a pure software algorithm cooperates with the ADC and the DAC with common precision, the limitation of the inherent resolution of hardware is effectively broken through, and the measurement precision and the output accuracy of the system are improved on the premise of not increasing the hardware cost.

The invention relates to a low latency low frequency effect codec. In some embodiments, a method of encoding a low frequency effect (LFE) channel includes receiving a time domain LFE channel signal; filtering the time domain LFE sound channel signal by using a low-pass filter; converting the filtered time domain LFE channel signal into a frequency domain representation of the LFE channel signal comprising a number of coefficients representing a spectrum of the LFE channel signal; arranging coefficients into a number of subband groups corresponding to different frequency bands of the LFE channel signal; quantizing coefficients in each secondary frequency band group according to a frequency response curve of the low-pass filter; encoding the quantized coefficients in each subband group using an entropy decoder that is tuned for the subband groups; and generating a bitstream comprising the encoded quantized coefficients; and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

An encoding device encodes each encoding-target block. The encoding device includes: a predictor configured to generate, for each of the components, a prediction block corresponding to the encoding-target block; a residual generator configured to generate, for each of the components, a prediction residual that represents a difference between the encoding-target block and the prediction block; a color space transformer configured to perform a color space transform process on the prediction residual of each of the components; a transformer configured to generate transform coefficients by performing a transform process on the prediction residual; a quantization controller configured to determine a scaling list to be used in a quantization process on the transform coefficients; and a quantizer configured to perform the quantization process on the transform coefficients by using the determined scaling list, wherein the quantization controller is configured to determine the scaling list, based on the color space transform process.

The invention discloses an image classification method based on sparse and quantization of an addition neural network, and the method comprises the steps: carrying out the sparse and quantization of the addition neural network, constructing an image classification prediction network model, carrying out the lossless truncation of a weight in the network to a distribution range which is the same as an input image, carrying out the calculation of a truncation value in advance, and enabling the truncation value to serve as a bias term, therefore, the sparseness of the weight is increased; secondly, mixing precision sharing coefficient quantification is carried out on the sparse addition neural network, grouping is carried out along a weight input channel, scaling factors of the weight are selected, and input and the weight are quantified at the same time; and calculating input image data through the image classification prediction network model, extracting feature information of a computer image, and outputting an image classification result. The method can reduce the reasoning energy and hardware overhead of the image classification prediction network model, and improves the efficiency and precision of image classification and recognition.

The invention relates to the technical field of power system control, in particular to a physical guidance primary frequency modulation adaptive control method and system, and the method comprises the steps: constructing a linearized offshore wind power system transfer function model; determining a dominant pole and carrying out characteristic root analysis on the dominant pole; introducing a modal response regulation and control coefficient, and quantifying the regulation and control capability of the control parameters on the dominant modal of the system so as to determine the differentiated adjustment step length proportion of the control parameters; defining an energy function describing the system frequency deviation, and analyzing the derivative of the energy function to represent the evolution direction of the system state; constructing a reinforcement learning agent based on a double-delay depth deterministic strategy gradient TD3 algorithm; applying the modal response regulation and control equilibrium principle to the action space design of the intelligent agent, and fusing the derivative information of the energy function into the reward function design; and the trained reinforcement learning agent is utilized to adaptively adjust the control parameters of the network construction type energy storage according to the real-time state of the system so as to complete the adaptive optimization of the primary frequency modulation.

In some implementations, a method of encoding a low frequency effect (LFE) channel includes receiving a time domain LFE channel signal, filtering the time domain LFE channel signal using a low pass filter, converting the filtered time domain LFE channel signal to a frequency domain representation of the LFE channel signal including a number of coefficients representing a spectrum of the LFE channel signal, arranging coefficients into a number of sub-band groups corresponding to different frequency bands of the LFE channel signal, quantizing coefficients in each sub-band group according to a frequency response curve of the low pass filter, encoding the quantized coefficients in each sub-band group using an entropy coder tuned for the sub-band group, and generating a bitstream including the encoded quantized coefficients, and storing the bitstream on a storage device or streaming the bitstream to a downstream device.

The invention discloses an oscillation risk early warning method and system based on multi-electrical element correlation evaluation, and the method comprises the steps: achieving the time sequence synchronous collection and rolling analysis of key electrical quantities, such as power, voltage, current and power factor angle, through a sliding window mechanism, and extracting the extreme value fluctuation characteristics and frequency characteristics of each quantity; a dimension threshold system, a frequency correlation verification rule and a weight dynamic adaptation mechanism are designed, a coupling correlation model among multiple elements is established, a frequency consistency coefficient is introduced to quantify the feature matching degree among quantities, technical upgrading from single-quantity independent analysis to multi-element correlation judgment is achieved, and multi-electric-quantity frequency modal collaboration is achieved. Potential risks are captured in advance through multi-quantity feature fusion analysis, risk signals are accurately recognized at the initial stage of formation of an oscillation fault chain, intervention time is bought for power grid dispatching, oscillation upgrading risks caused by multi-quantity collaborative fluctuation in a novel power system are effectively reduced, and stable operation of sensitive loads and power grid safety are guaranteed.