252 results about "Log likelihood" patented technology

The invention discloses a distribution line early-stage fault time sequence hidden Markov modeling and identification method and system based on multistage evolution characteristics, and belongs to the field of distribution line early-stage fault identification. Comprising the steps of obtaining a current waveform sample sequence of an early fault in a distribution line; for each current waveform sample in the sequence, extracting a multi-dimensional time-frequency feature, and constructing a feature vector of each sample; performing fault stage identification on each sample by using the first-level hidden Markov model group, and outputting a fault stage tag sequence corresponding to each sample; combining the fault stage label sequences of the current sample and a plurality of previous historical samples to form a stage label sequence window; and respectively inputting the stage label sequence window into a second-level tree line fault hidden Markov model and a second-level non-tree line fault hidden Markov model, calculating a corresponding first average log-likelihood value and a corresponding second average log-likelihood value, comparing the two average log-likelihood values, and judging whether a current sample belongs to a tree line early fault or not.

The invention discloses a dynamic self-adaptive NAND Flash reading method and system based on artificial intelligence, a medium and equipment, and relates to the technical field of solid-state storage, the method comprises the following steps: S1, initiating a reading request, reading page data by using default voltage, decoding the page data through an LDPC algorithm, constructing a feature vector and performing preprocessing, inputting the voltage into a pre-trained neural network model, and carrying out voltage prediction; s2, carrying out fine sampling on the predicted voltage, reconstructing accurate threshold voltage distribution, and calculating a log-likelihood ratio; s3, designing a shared bottom layer backbone network, enabling a neural network model to have independent output branches, predicting the voltage, and evaluating the long-term health state of the block; and S4, designing an online transfer learning mechanism, and enabling the neural network model to quickly adapt to a new environment by using a small amount of new data. According to the method, the delay is greatly reduced, the power consumption is remarkably saved, data loss is prevented through early warning of bad blocks, and the write amplification factor is reduced to 1.26 through dynamic read interference suppression.

The invention relates to the technical field of intelligent traffic, in particular to a traffic abnormal event dynamic detection method and system based on trajectory analysis, and the method comprises the steps: receiving original space-time trajectory data; preprocessing the original spatio-temporal trajectory data, and mapping the original spatio-temporal trajectory data to a road section sequence of a road network through a map matching algorithm; constructing a behavior model based on the track sequence after map matching, learning the track sequence of the normal behavior mode on the road section, and establishing an observation emission model and a state transition matrix; in the online stage, the log-likelihood value of an observation sequence and / or the similarity between the observation sequence and a normal behavior pattern cluster are / is calculated for a real-time track in a set sliding window, and when the log-likelihood value and / or the similarity exceed a set threshold value, the track is marked as abnormal; performing time-space aggregation on abnormal trajectories of the same road section or intersection in unit time according to single vehicle abnormality judgment, and triggering group abnormal event alarm when aggregation data exceed a preset value.

The invention discloses a subcarrier-level diversity combining method and subcarrier-level diversity combining equipment for multi-frequency diversity, and relates to the technical field of wireless communication. The method comprises the following steps: receiving a plurality of frequency points bearing the same service, and respectively carrying out orthogonal frequency division multiplexing demodulation on each frequency point to obtain a symbol estimation value; calculating a soft decision log-likelihood ratio and a reliability vector according to the symbol estimation value; aligning the time deviation, the carrier frequency deviation and the sampling rate deviation generated by frequency crossing of each frequency point based on each symbol estimation value; performing decorrelation processing on the aligned soft decision log-likelihood ratio to obtain a decorrelated soft decision log-likelihood ratio, and performing merging operation on the decorrelated soft decision log-likelihood ratio according to the aligned reliability vector to obtain a merged soft decision log-likelihood ratio; and decoding the merged soft decision log-likelihood ratio to obtain a decoding result, and adjusting the merging operation according to the decoding result. By implementing the technical scheme provided by the invention, the diversity combining accuracy can be improved.

In some example embodiments, there may be provided a method that includes receiving, by a machine learning encoder as part of a training phase, channel state information as data samples; generating, by the machine learning encoder, a latent variable comprising a log likelihood ratio value representation for the channel state information, wherein the latent variable provides a lower dimension binary representation when compared to the received channel state information to enable compression of the received channel state information; generating, by the binary sampler, a binary coding value representation of the latent variable, wherein the binary coding value converts the latent variable to a binary form; and generating, by the machine learning decoder, a reconstructed channel state information, wherein the generating is based in part on the binary coding value representation of the latent variable generated by the binary sampler. Related systems, methods, and articles of manufacture are also disclosed.

The invention discloses an unmanned aerial vehicle identification method and device based on category specific modeling, equipment and a storage medium, and relates to the technical field of mode identification based on digital data processing, and the method comprises the steps: obtaining a radio frequency signal of an unmanned aerial vehicle, and generating a time-frequency graph through short-time Fourier transform; inputting the time-frequency graph into a target residual network, and extracting feature vectors of the unmanned aerial vehicle; calculating a log-likelihood value of the feature vector under each model by using a Gaussian mixture model constructed for the known class of each unmanned aerial vehicle in the training stage, and obtaining a preliminary judgment result according to the maximum log-likelihood value and a preset adaptive threshold value; and for the samples which are preliminarily judged as the unknown class, calculating the minimum feature distance between the samples and all samples in the known class unmanned aerial vehicle sample feature matrix, and if the distance is smaller than a preset recovery threshold value, re-classifying the samples as the known class unmanned aerial vehicle to which the corresponding nearest neighbor sample belongs. According to the invention, known unmanned aerial vehicles can be accurately identified in an open environment, and unknown unmanned aerial vehicles can be effectively distinguished.

The invention discloses a cascade water-wind-light multi-power system modeling method based on time-space correlation, and the method comprises the steps: building an autoregression integral moving average-generalized autoregression condition heterovariance model of wind power photovoltaic power and cascade hydropower station runoff, building an R-vine, C-vine and D-vine Copula function spatial correlation model through employing a generalized autoregression condition heterovariance value, and carrying out the modeling of a cascade water-wind-light multi-power system. And selecting an optimal spatial correlation model according to a Bayesian information criterion, an akaike information criterion and a logarithm likelihood value goodness of fit index. Based on the optimal spatial correlation model, constructing a scene generation method of Latin hypercube-Copula function sampling, performing scene generation on the wind power photovoltaic power and the runoff volume of the cascade hydropower station, and constructing a scene reduction method of contour coefficient optimization Kmeans clustering to reduce the generated scene. The invention provides a time-space correlation cascade water-wind-light multi-power system modeling method, and aims to analyze water-wind-light multi-energy complementary operation characteristics and provide reference for a scheduling mode among water-wind-light multi-energy.

The invention relates to the field of data processing, in particular to a carton production line monitoring method and system, and the method comprises the steps: collecting and preprocessing multi-dimensional time sequence data of a production process; constructing a standard hidden Markov model (HMM) based on historical normal data, and defining the hidden state of the HMM as a microscopic operation mode under a macroscopic process; extracting procedure context features, and constructing a procedure conformity evaluation model (PCAM) to calculate a conformity score; dynamically adjusting the emission probability of the standard HMM in combination with the score to form an improved IHMM; and calculating the log-likelihood probability of the real-time observation sequence by using the improved IHMM, and comparing the log-likelihood probability with a preset threshold to judge the production abnormality. According to the invention, the accuracy and reliability of abnormity monitoring can be effectively improved.

Provided are an iterative detection and decoding (IDD) circuit, which includes a multiple-input and multiple-output (MIMO) detector having low power consumption, an operation method of the IDD circuit, and a modem chip. The IDD circuit, which is configured to receive a signal including a symbol, includes a first detector configured to generate a first log-likelihood ratio based on the symbol by using a linear detection method, a decoding circuit configured to perform a first decoding operation based on the first log-likelihood ratio and to generate a post-log-likelihood ratio when the first decoding operation fails, and a second detector configured to, when the first decoding operation fails, generate a second log-likelihood ratio based on the symbol and the post-log-likelihood ratio by using a nonlinear detection method, wherein the decoding circuit is further configured to perform a second decoding operation based on the second log-likelihood ratio.

The present disclosure relates to a machine learning (ML)-based receiver that is computationally efficient, irrespective of a number NR of receiver antennas used in a Multiple Input Multiple Output (MIMO) scenario. To achieve this, the ML-based receiver is configured to obtain a modified matrix and a modified array of symbols based on a channel state information (CSI) matrix and a received array of symbols. The modified matrix has a dimension NT×NT, where NT is a number of MIMO layers. The modified array of symbols has a dimension NT. After that the ML-based receiver is configured to restore a transmitted array of symbols from the received array of symbols by applying a pre-trained ML model that receives the modified matrix and the modified array of symbols as input data and outputs a set of bit log-likelihood ratio (LLR) estimates for the transmitted array of symbols.

A wireless communication device includes a radio-frequency integrated circuit (RFIC), one or more processors including processing circuitry, and a memory storing instructions. The instructions, when executed by the one or more processors individually or collectively, cause the wireless communication device to receive, via the RFIC, a reception signal including a plurality of subcarriers, and calculate a first log-likelihood ratio (LLR) based on a frequency domain. The plurality of subcarriers include a first subcarrier and a second subcarrier adjacent to the first subcarrier. The calculation of the first LLR includes to measure a channel variation between the first subcarrier and the second subcarrier, determine a second linear detection matrix of the second subcarrier, based on the channel variation, and calculate the first LLR based on at least one of a first linear detection matrix of the first subcarrier and the second linear detection matrix. The first subcarrier is a pivot subcarrier.

The invention belongs to the field of sound anomaly detection, and particularly provides an acoustic water turbine detection method based on unsupervised deep learning, which comprises the following steps: arranging one or more microphones near to-be-detected water turbine part mechanical equipment, and collecting sound signals of equipment operation; the collected sound signals are preprocessed; constructing a multi-branch residual fusion network fusing time domain information, frequency domain information and time-frequency domain information, and performing feature extraction on the sound signals; using the features extracted by the multi-branch residual fusion network as the compression features of the sound signals, fitting the distribution of normal sound signals of each category of the sound signals to be detected by using a Gaussian mixture model, calculating the log-likelihood of the sound signals to be detected for each category, and calculating the Mahalanobis distance from the sound signals to be detected to the center of each category; and taking the nearest mahalanobis distance as an abnormal value score of the sound signal sample to be detected, and if the abnormal value score exceeds a threshold value, judging that the sound signal sample is abnormal. According to the method, on the basis of ensuring rapid and accurate extraction of fault signals, the requirement for real-time online monitoring and detection of the water turbine mechanical equipment is met.

The invention discloses a multi-user joint decoding method for a cellular-free large-scale MIMO system based on model-driven deep learning, and the method comprises the steps: building an optimization problem of a simulation combination matrix with maximization of the reachable rate of the cellular-free large-scale MIMO system as a target in a simulation part of data decoding; on the basis of a heuristic algorithm, an optimization target is simplified, an optimization problem is divided into a plurality of sub-problems, the single AP simulation precoding matrix optimization problem is solved one by one, and an optimal simulation combination matrix is obtained; a multi-user joint LDPC decoding algorithm is expanded to a cellular-free large-scale MIMO scene, multi-user joint LDPC decoding is realized based on an MIMO-LDPC-Net decoding network, an input layer of the MIMO-LDPC-Net decoding network generates an initial value as an input of a hidden layer, the hidden layer is formed by stacking modules composed of a first sub-layer, a second sub-layer and an MIMO detection layer, and the first sub-layer, the second sub-layer and the MIMO detection layer are stacked. The output layer generates a log-likelihood ratio of the variable node. Experiments show that compared with a traditional algorithm, the heuristic algorithm and the AI auxiliary joint decoding algorithm have a signal-to-noise ratio gain of 1dB.

The application provides a low-density parity-check code decoding method and device based on a graph neural network, which comprises the following steps: step 1, constructing a factor graph comprising variable nodes, check nodes and edge relationships, and mapping log-likelihood ratio information of a received signal to initial embedding of the variable nodes; step 2, generating message features according to node embedding, node degree and iteration step length, and constructing attention weights for each edge to measure the importance of the message; step 3, inputting the message features and the attention weights into a gated recurrent unit to update the residual of the edge weight, and feeding back the updated edge weight to the graph structure; step 4, weighting and aggregating the messages from the adjacent nodes according to the edge weight, calculating the updated embedding of the variable nodes and the check nodes, and performing gated modulation combined with the check result; step 5, repeating steps 2 to 4 until a preset iteration number or a decoding convergence condition is reached, and mapping the final variable node embedding to a decoding result to realize LDPC code word recovery. By introducing the attention mechanism and the gated residual update into the message passing process, the application realizes adaptive modeling of the contribution degree of different edge messages, dynamically remembers the historical state, thereby improving the decoding performance and the convergence speed; meanwhile, the application can effectively reduce the bit error rate and is suitable for high-speed reliable data transmission scenarios in a 5G / 6G wireless communication system.

The invention belongs to a price determination method, and provides a method for determining upper and lower limits of contract price fluctuation of a power market and a related device, aiming at the technical problems that a price limit mechanism for setting fixed upper and lower limits of prices is adopted in the current power transaction, a price discovery suppression function exists, and violent price swing within a day cannot be effectively coped with. Calculating a positive excess sample and / or a negative excess sample, then taking the positive excess sample and / or the negative excess sample as input, and obtaining a morphological parameter and a scale parameter when a log-likelihood value of the positive excess sample and / or a log-likelihood value of the negative excess sample converges through generalized Pareto distribution iterative calculation; and finally, positive tail risk measurement index expected loss and / or negative tail risk measurement index expected loss are / is obtained through calculation and serve as an upper limit and a lower limit of contract price fluctuation of the electricity market. And the technical problem of violent price swing within the day can be effectively solved.

The application relates to a Polar code decoding simulation circuit implementation method, system, equipment and medium based on a memory-computing integrated device, and the method comprises the following steps: obtaining a Polar code sparse check matrix and a corresponding probability graph model according to a received Polar code; obtaining an initial log-likelihood ratio of each variable node in the probability graph model; dividing a memory-computing integrated device array into an input part, an output part, an iteration part and a check part according to the Polar code sparse check matrix, and configuring the resistance state of the cross nodes in each part; loading the initial log-likelihood ratio into the input part of the memory-computing integrated device array and keeping it, and the initial input of other parts is 0; checking and iteratively calculating the calculation results output by the iteration part, the check part and the output part in the memory-computing integrated device array by using a peripheral circuit, and outputting a final decoding result after the check condition is met. The application can be widely applied in the field of decoding technology.

Systems, methods, non-transitory computer-readable media to perform operations associated with the storage medium. One system includes a storage medium and an encoding / decoding (ED) system to perform operations associated with the storage medium, the ED system being configured to process a set of log-likelihood ratios (LLRs) and a syndrome vector to obtain a set of confidence values for each bit of a codeword, estimate an error vector based on selecting one or more bit locations with confidence values from the set of confidence values above threshold value and applying hard decision decoding to the selected one or more bit locations, calculate a sum LLR score for the estimated error vector, and output a decoded codeword based on the estimated error vector and the sum LLR score.

The invention provides a large language model staged pre-training method and system. The method comprises the following steps: training a Transform model by using a basic data set, optimizing a negative logarithm likelihood target, and adopting an AdamW optimizer and a cosine attenuation learning rate; continuing training by using the universal knowledge data set based on the first-stage parameters; and weighting the professional data in the training field by adopting an oversampling strategy. By structuring a training target and a data type, the model can efficiently learn language basis, general knowledge and professional skills in stages. Experiments show that according to the method, the training efficiency of the model in the basic stage is improved by 40%, the overall training time is shortened by 30%, and meanwhile the accuracy rate of tasks in the professional field is 15%-20% higher than that of traditional end-to-end training. Finally, model parameters are evaluated through professional ability, and both universal language understanding and domain specialty are achieved.

A method and system for anomaly detection from time-series input data. A Gaussian Mixing Model (GMM) learns distribution parameters in an offline learning stage using sample data. The data used for offline learning and for a subsequent online anomaly detection stage are time-series data collected for multiple parameters of a machine operation, such as a robot performing a repetitive set of operations. The method includes aligning the data with a known good reference data file and taking a difference from it before providing the data to the GMM. In an online anomaly detection stage, the GMM calculates a probability that each time-series data point fits the distribution, and then a log-sum calculation is performed on each data file to determine the likelihood that the file contains anomaly data.The log likelihood of the file is compared with previous values, and an alert is issued if there are statistical deviations from the historical data.

The application provides a cylindrical array sonar system and a target detection method based on high-resolution beam forming. The method comprises the following steps: converting a beam forming weight vector of a received sound wave signal into a Kronecker product, combining a joint probability density model to construct a negative log-likelihood function, and optimizing a vector in an azimuth dimension and an elevation dimension through a preset direction constraint condition to form a high-resolution beam; performing correlation analysis on echo signals reflected by a target at different azimuth angles and elevation angles and on transmitted sound wave signals at different time delays; constructing a three-dimensional matrix containing a distance, an azimuth angle and an elevation angle according to a correlation analysis result, performing normalization processing on the three-dimensional matrix to eliminate signal intensity differences, and outputting a target detection positioning result. Through the combination of Kronecker integral decomposition and three-dimensional matrix modeling, the application realizes high-resolution joint positioning detection of a target in an azimuth angle, an elevation angle and a distance dimension, effectively eliminates signal intensity differences, and improves positioning accuracy.

The application relates to a time delay blind estimation method of asynchronous SCMA. The method comprises the following steps: constructing an asynchronous SCMA system model and setting three hypotheses, designing a time delay blind estimation model based on the three hypotheses, and the model comprises a prior propagation iteration and a rectangular window detection module. In the prior propagation iteration, a probability log likelihood ratio is set and initialized according to a factor graph, an initialization message from a user node to a resource node is calculated, then a message from a resource to a user node is obtained, an update message is generated by aggregating the message of the user node, and the probability node is updated to obtain an optimal likelihood ratio. The rectangular window detection module calculates a detection statistic according to the optimal likelihood ratio, and the maximum value of the detection statistic is a discrete time delay index. The method can realize user node time delay estimation without relying on additional information.

The invention relates to the technical field of communication, in particular to a communication method for inhibiting an overhigh peak-to-average ratio in an orthogonal time-frequency-space system, which comprises the following steps of: 1, converting a signal from a delay-Doppler domain convenient for channel characterization to a time-frequency domain convenient for modulation implementation; step 2, performing preliminary suppression on the peak-to-average ratio by adopting a T-SLM method, generating a plurality of groups of candidate signals, and selecting one group with the peak-to-average ratio lower than a preset threshold or with the minimum peak-to-average ratio; step 3, carrying out Hisenberg transformation on the selected signal to obtain a time domain signal; carrying out iterative amplitude limiting filtering processing, adding a cyclic prefix, and then sending; step 4, removing the cyclic prefix by a receiving end, and recovering to a time delay-Doppler domain through Wigner transform and Sextile Fourier transform; performing preliminary detection by adopting a message passing algorithm to obtain a hard decision value and a log-likelihood ratio; and selecting a high-reliability observation value based on a log-likelihood ratio, and constructing a reliability selection matrix.

Methods and apparatuses for decoding polar codes are provided. A simplified successive cancellation list (SSCL) decoding tree for a polar code is generated. The SSCL decoding tree includes a plurality of nodes. One or more nodes of the plurality of nodes are identified for decoding using a Reed-Muller code. Decoding of received log likelihood ratios (LLRs) is performed at the one or more nodes using the Reed-Muller code. Hard decision values are output from the one or more nodes.

The invention relates to an infant asthma three-branch classification auxiliary judgment method and a medium, and relates to the technical field of data processing. The method comprises the following steps: collecting and preprocessing tidal respiration data and demographic data of infants, and performing heterogeneous feature extraction and grouping log-likelihood ratio acquisition on the preprocessed tidal respiration data and demographic data to obtain a log-likelihood ratio corresponding to each group; and performing hierarchical Bayesian fusion on each log-likelihood ratio to obtain a fusion result, and performing asthma risk auxiliary judgment of three-branch classification based on the fusion result. Compared with the prior art, the method can assist in judging the infant asthma more objectively and accurately based on the tidal respiration data.

The invention discloses an efficient soft demodulation method and system suitable for pulse position modulation, and relates to the technical field of laser communication. Pulse position modulation signals and preset channel parameters are received, and time slot receiving signals are acquired based on the pulse position modulation signals, the preset channel parameters comprise the optical signal intensity and noise power of a Gaussian channel, the pulse optical signal electron number of a Poisson channel and the electron number of an ambient optical signal; inputting the time slot receiving signal and a preset channel parameter into a pre-established soft demodulation model based on a maximum posteriori and a maximum likelihood criterion, and outputting to obtain a log likelihood ratio LLR of a modulation bit; meanwhile, the index logarithm operation of the LLR is simplified by adopting a max-log method, and the calculation complexity is reduced; aI is adopted to optimize the scaling coefficient for the performance loss caused by soft demodulation calculation simplification to balance the performance loss and the implementation complexity, so that the transmission efficiency of PPM modulation is improved.

The application discloses a subspace parameter iterative estimation space-time adaptive detection method for a strong clutter environment. In view of the problem that in a space-time adaptive processing system, a target steering vector falls in a clutter subspace, main and auxiliary data exist power mismatch, and existing methods are difficult to effectively maximize the marginal likelihood, the application projects the main and auxiliary data to a low-dimensional coordinate domain to obtain sufficient statistics, regards the clutter coefficient as a hidden variable, and under two kinds of assumptions, respectively uses EM and ECME algorithms to iteratively maximize the marginal log-likelihood function, and obtains the stationary point estimation of the clutter covariance matrix, the power mismatch factor and the target complex amplitude after convergence, and simultaneously constructs three detection statistics, GLRT, Rao and Wald, according to the stationary point estimation, and compares with a pre-calibrated threshold to complete the judgment. The iterative process of the application has the guarantee of the monotone non-decreasing of the marginal log-likelihood, can effectively compensate for the main and auxiliary power mismatch, and can still maintain good detection performance under the condition of small training samples.

A system and method for learning a predictive function that can automatically learn different operating modes for a multi-modal system and predict the number of operating states for a multi-modal system and additionally the detailed structure for each state. Once learned, the predictive function (model) can be used to determine a mode of a new sample (an asset). Based on the determined components that maximize a log likelihood function, a mode of the new sample is detected into the model via dependency graphs. One aspect includes enforcing a lower bound for the number of sample points to form an operational mode for an asset. While a mode relates to sample points which maximizes like log-likelihood, an ability is provided to remove artifact modes due to noisy data by considering a sufficient sample data condition and maximizing log-likelihood. Domain knowledge can be incorporated into the model via dependency graphs.

An apparatus may include a receiver and one or more processors. The receiver may receive log likelihood ratio (LLR) values corresponding to encoded data that is encoded using a low-density parity-check (LDPC) code. The one or more processors may identify a number of punctured parity bits corresponding to the encoded data, determine that the number of punctured parity bits is less than a threshold, and remove, from a first parity check matrix corresponding to the encoded data, one or more rows corresponding to the number of punctured parity bits to generate a second parity check matrix. The one or more processors may determine, using the LLR values and the second parity check matrix, that there is an error in the LLR values, and an LDPC decoder may decode, based at least on the determination, the LLR values to resolve the error.

The present application relates to a kind of underwater multi-target tracking methods, comprising the following steps: establishing three-dimensional space coordinate system, obtaining the coordinate of sensor, monitoring water volume, water surface in the coordinate Z2 of Z axis, the coordinate Z3 of water bottom in the Z axis, clutter density λ;Sensor periodically obtains measurement information, the measurement information includes azimuth, elevation and the time delay between direct wave, the set of measurement information in the i frame is recorded as Z (i), the number of measurement information that sensor obtains single target is L;Randomly generate K in monitoring water volume Virtual target c K =[c1,c2,...,c K ] In prior formula, the possibility that a measurement information comes from certain propagation path is expressed;According to log likelihood ratio, establish underwater target model;According to underwater target model, the actual target number and the optimal coordinate of each actual target are obtained. Efficiently positioning and tracking underwater multi-target can be carried out.