Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

631 results about "Log likelihood" patented technology

The log-likelihood function is defined to be the natural logarithm of the likelihood function . The log-likelihood function is used throughout various subfields of mathematics, both pure and applied, and has particular importance in fields such as likelihood theory. REFERENCES:

Fire early warning device and fire extinguishing system for electrical power distribution cabinet

ActiveCN119131980A
The invention discloses a fire early warning device and a fire extinguishing system for an electrical power distribution cabinet, and belongs to the technical field of alarm. The fire early warning device comprises a fire identification model which is trained through a Bayesian network, and the training process comprises the following steps: deducing the distribution of a hidden variable Z according to the parameters of the Bayesian network at the current time and the data vector of the ith training set, and calculating the expectation of log-likelihood about Z; acquiring parameters of the Bayesian network at the next time t + 1 according to the expectation; calculating an error of training the Bayesian network by the data vector of the ith training set; further updating Bayesian network parameters according to the error; and the steps are repeatedly executed, training of the Bayesian network is carried out, verification is carried out on the verification set until a stop condition is reached, training of the Bayesian network is completed, and the fire identification model is obtained. According to the method, the Bayesian network is trained into the fire recognition model through the observable variables and the hidden variables, so that the fire recognition capability is improved.
Owner:GUANGDONG ZHONGKE HUIJU TECH CO LTD +1

Low-power wide area network integrated sensing and communication method based on channel sensing and reinforcement learning

The invention provides a low-power wide area network integrated sensing and communication method based on channel sensing and reinforcement learning, and the method comprises the steps: receiving an uplink signal transmitted by a terminal node through a gateway receiver, demodulating the uplink signal through employing a soft demodulation algorithm, and obtaining a bit log-likelihood ratio; the deep reinforcement learning model completes channel reconstruction of the corresponding link during communication according to the bit log-likelihood ratio; and calculating the current optimal network configuration according to the reconstructed channel so as to distribute the current optimal network configuration to each terminal node during next downlink transmission. According to the method, the key environment information is mined by using the deep reinforcement learning model, so that the gateway can allocate optimal network configuration to different terminal nodes according to the current network environment condition, and the decoding capability of the gateway is improved.
Owner:THE HONG KONG UNIV OF SCI & TECH (GUANGZHOU) +2

Receiver-side noise statistics and log likelihood ratio scaling for pre-equalization-based communication schemes

Methods, systems, and devices for wireless communications are described. The described techniques provide for a user equipment (UE) to perform receiver-side noise statistics (Rnn) estimation and for an extended reality (XR) device to perform log likelihood ratio (LLR) scaling estimation. For example, the UE may transmit a first set of reference signals to the XR device. The XR device may sample the reference signals and transmit an indication of quantized samples such that the UE may perform Rnn estimations and generate a pre-equalization matrix. The UE may transmit a set of pre-equalized reference signals to the XR device such that the XR device may estimate LLR scaling coefficients. In some examples, the UE may determine the pre-equalization matrix by performing a filtering of reference signal measurements. The UE may apply whitening to account for momentary interference, and may indicate a whitening matrix to the XR device.
Owner:QUALCOMM INC

Receiver-side noise statistics and log likelihood ratio scaling for pre-equalization-based communication schemes

Methods, systems, and devices for wireless communications are described. The described techniques provide for a user equipment (UE) to perform receiver-side noise statistics (Rnn) estimation and for an extended reality (XR) device to perform log likelihood ratio (LLR) scaling estimation. For example, the UE may transmit a first set of reference signals to the XR device. The XR device may sample the reference signals and transmit an indication of quantized samples such that the UE may perform Rnn estimations and generate a pre-equalization matrix. The UE may transmit a set of pre-equalized reference signals to the XR device such that the XR device may estimate LLR scaling coefficients. In some examples, the UE may determine the pre-equalization matrix by performing a filtering of reference signal measurements. The UE may apply whitening to account for momentary interference, and may indicate a whitening matrix to the XR device.
Owner:QUALCOMM INC

Receiver-side noise statistics and log likelihood ratio scaling for pre-equalization-based communication schemes

Methods, systems, and devices for wireless communications are described. The described techniques provide for a user equipment (UE) to perform receiver-side noise statistics (Rnn) estimation and for an extended reality (XR) device to perform log likelihood ratio (LLR) scaling estimation. For example, the UE may transmit a first set of reference signals to the XR device. The XR device may sample the reference signals and transmit an indication of quantized samples such that the UE may perform Rnn estimations and generate a pre-equalization matrix. The UE may transmit a set of pre-equalized reference signals to the XR device such that the XR device may estimate LLR scaling coefficients. In some examples, the UE may determine the pre-equalization matrix by performing a filtering of reference signal measurements. The UE may apply whitening to account for momentary interference, and may indicate a whitening matrix to the XR device.
Owner:QUALCOMM INC

High-performance low-complexity LDPC (Low Density Parity Check) decoding method for satellite communication

The invention discloses a high-performance low-complexity LDPC (Low Density Parity Check) decoding method for satellite communication, and belongs to the field of digital signal processing. And performing iterative decoding on the received channel information based on a low density parity check (LDPC) code in the 5G standard. The implementation method comprises the following steps of: dynamically setting an erasing range by utilizing the amplitude of current check node information when a variable node is updated; the amplitude of check node information is gradually increased along with the increase of the number of iterations, the whole iteration process is divided into two parts, and the erasing range is set by adopting different parameters. And outputting the decoding result when the decoding result is correct or reaches the maximum number of iterations. According to the method, the reliability of variable node information is improved through the two points, reliable variable information is prevented from being erased, unreliable variable node information is prevented from being spread, and excellent decoding performance is achieved. Non-linear operation in a log-likelihood ratio belief propagation algorithm LLR-BPA does not exist in the decoding algorithm, and the decoding algorithm has the advantages of being high in decoding performance and low in decoding complexity.
Owner:BEIJING INST OF TECH

OFDM-IM-LDPC coding and decoding method based on combination number index

The invention discloses an OFDM-IM-LDPC (Orthogonal Frequency Division Multiplexing-Index-Low Density Parity Check) coding and decoding method based on a combination number index, and belongs to the field of According to the method, the symbol bit information and the index position obtained through calculation are independently coded, original direct coding of information source bits is replaced, the misjudgment rate of the index position is reduced, the accuracy of a subcarrier activation mode is improved, meanwhile, the index position is calculated through the combinatorial number method, the application range of the system is expanded, and the system reliability is improved. Meanwhile, resource waste caused by a preset lookup table is reduced, the implementation difficulty is lowered, and higher flexibility and expansibility are provided. According to the method, the high error correction performance of the LDPC code is utilized, the redundancy of bit information and index positions is increased, noise and attenuation in a channel are effectively resisted, and high precision of data decoding is achieved; according to the method, the log-likelihood ratio soft information is selected for decoding, and the robustness and the accuracy of signal demodulation recovery are improved while the judgment complexity is reduced by arranging the soft information in a descending order and dynamically adjusting an illegal subcarrier mode.
Owner:BEIJING INST OF TECH

LDPC (Low Density Parity Check) encoding and decoding system and method for high-density tape storage

The invention discloses an LDPC (Low Density Parity Check) coding and decoding system and method for high-density tape storage, and the system comprises a main controller which is used for scheduling coding and decoding mode switching and module cooperative control; the storage and calculation integrated array is used for executing zero-carrying storage calculation; the ping-pong buffer framework is composed of an input buffer module Buffer A and an output buffer module Buffer B; the dynamic code length extension module is used for reconstructing a cascade link; the preprocessing module is used for converting the noisy code word into a log likelihood ratio (LLR) value; the hard decision module is used for converting the VN message into user data; the main controller is connected with all the modules through a control bus, and the dynamic code length extension module controls the power state and interconnection topology of the sub-arrays. According to the LDPC coding and decoding system and method based on the storage and calculation integrated array, zero-carrying calculation is achieved, iterative decoding is accelerated through a ping-pong buffer architecture, and the code length is flexibly configured through the dynamic code length extension module, the coding and decoding efficiency of high-density tape storage is improved, delay and energy consumption are reduced, meanwhile, multi-code-length self-adaption is supported, and high performance and flexibility are considered.
Owner:HANGZHOU INTERNATIONAL INNOVATION INSTITUTE OF BEIHANG UNIVERSITY

Unsupervised industrial anomaly detection method based on improved full-convolution cross-scale flow network

The invention provides an unsupervised industrial anomaly detection method based on an improved full-convolution cross-scale flow network. The method comprises the steps of performing data preprocessing on an industrial image data set; industrial image data is used as input, a pre-trained visual backbone network is used for extracting multi-scale features, a resolution-perceived channel attention module is used for enhancing expression of the multi-scale features, and an enhanced multi-scale feature map is obtained; performing multilayer reversible transformation by taking the enhanced multi-scale feature map as input and taking an ICSF-Net model based on hierarchical attention and expansion convolution as a cross-scale normalized flow network, and modeling multi-scale feature distribution; in the training stage, optimizing and updating ICSF-Net model parameters by maximizing the log likelihood of a normal sample under potential Gaussian distribution based on multi-scale feature distribution and combining an LSGR mechanism; in the reasoning stage, probability density estimation is carried out based on multi-scale feature distribution, an abnormal score graph is generated, and defect detection and positioning are achieved. According to the invention, the accuracy, robustness and pixel-level positioning precision of industrial product defect detection are improved.
Owner:HENAN INST OF ENG

Distribution line early fault time sequence hidden Markov modeling and identification method and system based on multi-stage evolution characteristics

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.
Owner:SHANGHAI JIAOTONG UNIV

Machine learning device, prediction control device, machine learning method, prediction control method, and computer-readable recording medium

PCT designated stage expiredWO2024241396A1Machine learningAdaptive controlLogitLog likelihood
A machine learning device 10 comprises: a first gradient value calculation unit 11 that calculates a gradient value for a variance parameter of a mean negative log likelihood, which is calculated using a plurality of training data containing explanatory and objective variables, a variance parameter of a Gaussian distribution variance function model, and a mean parameter of a Gaussian distribution mean function model; a second gradient value calculation unit 12 that calculates a gradient value for a mean parameter of the mean squared error between the output value of the Gaussian distribution mean function model and the objective variable when the explanatory variable and the mean parameter are treated as input, which is calculated using the plurality of training data and the mean parameter of the Gaussian distribution mean function model; and a parameter update unit 13 that updates the variance parameter using the gradient value for the variance parameter of the mean negative log likelihood and updates the mean parameter using the gradient value for the mean parameter of the mean squared error.
Owner:NEC CORP

Resistive random access memory ReRAM data judgment method based on sneak path correlation

The invention provides a resistive random access memory ReRAM data judgment method based on sneak path correlation. The method comprises the implementation steps that parameters are initialized; calculating a reading resistance value of the ReRAM storage unit; classifying the ReRAM storage units based on the sneak path correlation; calculating a log-likelihood ratio and a judgment threshold of each type of storage unit; and obtaining a ReRAM data judgment result. The storage units are classified according to the sneak path correlation values of the row and the column where each storage unit is located, and the likelihood ratio function of each category of storage units is calculated according to the conditional probability distribution of the resistance value read by the storage data of each storage unit under different value conditions. According to the method for calculating the likelihood ratio function by using the sneak path correlation, the influence of the difference of interference suffered by different storage units on the bit error rate is fully considered, and the judgment precision is effectively improved.
Owner:XIDIAN UNIV

Receiver for adjusting log likelihood ratio and method of operating the same

A receiver configured to receive a signal may include a symbol, the receiver may include: at least one memory storing instructions; and at least one processor configured to execute the instructions to: generate a first posteriori log likelihood ratio corresponding to a bit in the symbol based on a channel log likelihood ratio corresponding to the bit; apply a first scaled value to a first extrinsic log likelihood ratio in the first posteriori log likelihood ratio; generate a first priori log likelihood ratio by selectively using a first reference value based on a comparison result; generate a second posteriori log likelihood ratio corresponding to the bit based on the first priori log likelihood ratio and the channel log likelihood ratio; and decode the signal based on the second posteriori log likelihood ratio.
Owner:SAMSUNG ELECTRONICS CO LTD

FEC decoding for channels with memory

An optical receiver includes a receiving component, a log-likelihood ratio component, and a forward error correction decoder. The receiving component may be configured to receive a data transmission of transmitted data and convert a portion of the data transmission into one or more digital samples. The log-likelihood ratio component may be configured to generate a log-likelihood ratio value using a set of the one or more digital samples. The forward error correction decoder may be configured to correct errors in in the log-likelihood ratio value and recover the transmitted data using the log-likelihood ratio value.
Owner:MAXLINEAR INC

Quick list decoding method for polarization-adjusted convolutional code special nodes

Provided are a quick list decoding method and apparatus for polarization-adjusted convolutional code special nodes, a computer electronic device, a computer-readable storage medium, a computer program product, and a computer program. The method comprises: determining special nodes in a decoding tree for list decoding of a polarization-adjusted convolutional code, wherein the special nodes comprise a source node and a code rate-0 or repetition code sub-node sequence, and the special nodes are identified by means of a sequence set; calculating a log-likelihood ratio of the source node corresponding to the special node, and determining the number of paths; on the basis of the log-likelihood ratio and the sequence set corresponding to the special nodes, calculating the metrics of the paths, and reserving the paths having the minimum metric, the number of reserved paths being the set number of lists; decoding the source node by taking the reserved paths as inputs to obtain an estimated code word of each path of the source node; and calculating an estimated output bit and an estimated information bit of the special node by means of the estimated code word.
Owner:ZHEJIANG UNIV

Communication data transmission system and method based on rateless code

The invention discloses a communication data transmission system and method based on rateless codes. The system comprises a sending end device and a receiving end device, the sending end device obtains source data, codes the source data by using a RaptorQ encoder to obtain a coded symbol set, and modulates the coded symbol set into a modulation signal; determining link distribution information of the modulation signal according to the link communication quality, and transmitting the link distribution information to a receiving end; the receiving end determines channel state information according to the signal quality, a block code statistical method is used for analysis, decoding parameters are obtained and fed back, and the decoding parameters reflect the minimum number of symbols needed by decoding of the receiving end; performing soft demodulation on the received signal to obtain a log-likelihood ratio, performing cascade hierarchical statistical decoding to obtain a decoding result, verifying the decoding result, and feeding back a verification result; and the sending end dynamically adjusts the link distribution information of the modulation signal according to the decoding parameter and the verification result. The technical problem that a traditional coordinated multi-point transmission technology based on a fixed rate code is difficult to adapt to an ultra-reliable low-delay communication scene is solved.
Owner:CHINA TELECOM CORP LTD

Fine-grained Chinese calligraphy style recognition model construction method, system and device and storage medium

The invention discloses a fine-grained Chinese calligraphy style recognition model construction method, system and device and a storage medium. The method comprises the steps of obtaining and constructing a data set containing multiple Chinese famous calligrapher works; preprocessing the calligraphy pictures in the data set; selecting a deep learning model, combining the preprocessed data set with a general image data set, and retraining the deep learning model; inputting the preprocessed data set into the model for CLS feature extraction, and then combining all the CLS features of all the images into a new CLS feature vector; performing principal component analysis dimensionality reduction on the CLS feature vector; a Gaussian mixture model is used to fit the CLS feature vectors after dimension reduction; calculating a log-likelihood value of the preprocessed data set; the model is not only limited to the recognition of common font styles, but also can more accurately recognize which calligrapher's style is.
Owner:WUHAN KEJIAN CULTURE TECH CO LTD

Method and system for calculating size of beam spot of deformable electron beam lithography machine

The invention discloses a deformable electron beam lithography machine beam spot size calculation method and system, and the method comprises the steps: S1, employing a hyperbolic cosine function as a primary function according to a relation curve between the current intensity and the beam spot position, and constructing a composite function model through translation and superposition; s2, assuming that the observation value is a noise version of the composite function model, and establishing a likelihood function; a logarithm likelihood function is maximized and converted into a minimum residual sum of squares, partial derivatives of the minimum residual sum of squares about all parameters are solved, and a gradient equation set is obtained; s3, calculating the measurement signal, and extracting key feature points in the curve; calculating an initial parameter value according to the position of the feature point: taking the initial parameter as input, and iteratively solving a gradient equation set to obtain an exact solution of each parameter; and S4, calculating the actual size, the central position and the sharpness of the left and right edges of the electron beam spot according to the exact solutions of the parameters. The method has the advantages of high calculation precision and the like.
Owner:48TH RES INST OF CHINA ELECTRONICS TECH GROUP CORP

Low-complexity polarization code construction method

The invention relates to the technical field of communication channel coding, and discloses a low-complexity polar code construction method. According to the method, the reliability of each polarization channel is represented by utilizing a log-likelihood ratio output by noise-free SC decoding, and firstly, a reliability sequence of all N polarization channels is obtained by utilizing a binary expansion reliability sequence generation algorithm provided by the invention; and then, according to the code rate R, in the reliability sequence of the N polarization channels, selecting the indexes of the most reliable NR polarization channels as information bits, and selecting the indexes of the remaining N-NR polarization channels as freezing bits, thereby completing the construction of the polarization code with the code length N and the code rate R. The method is independent of an actual channel, the complexity of polar code construction can be effectively reduced, and the performance of a medium-high signal-to-noise ratio region is improved.
Owner:SOUTHWEST PETROLEUM UNIV

Multi-channel-oriented adaptive signal recovery method, device and equipment

The embodiment of the invention provides a multi-channel-oriented adaptive signal recovery method, device and equipment, which are applied to the technical field of communication, and the method comprises the following steps: carrying out frame synchronization on a data frame of each channel according to a unique identification code and a channel identifier in frame header information; performing quality evaluation and threshold comparison on effective information of each channel according to key information for guiding soft information combination in the frame header information, and adaptively matching a soft information combination mode according to a quality evaluation and threshold comparison result; merging according to a data frame synchronization result and the matched soft information merging mode to obtain an optimal channel receiving soft value; demodulating the channel receiving soft value, and calculating a log-likelihood ratio; and performing iterative decoding based on the log-likelihood ratio to recover an original signal. In this way, the compatible processing capacity of a communication system on signals with different qualities in a multi-channel transmission scene can be improved, the quality of soft information required by iterative decoding is improved, and the reliability of the communication system is improved.
Owner:SHANDONG PROMOTE MECHANICAL & ELECTRICAL TECHNOLOGY CO LTD

Dynamic self-adaptive NAND Flash reading method and system based on artificial intelligence, medium and equipment

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.
Owner:JIANGSU XINSHENG INTELLIGENT TECH CO LTD

Continuous training methods for systems identifying anomalies in an image of an object

A system identifying anomalies in an image of an object is first trained using first sets of images corresponding to first anomaly types for the object. A model of the object is formed in a latent space. A label for each anomalous image is used to calculate vectors containing means and standard deviations for each first anomaly types. The means and standard deviations are used to calculate a log-likelihood loss for each first anomaly type. The system is retrained using second sets of images corresponding to second anomaly types for the object. The vectors are supplemented using labels for each second anomaly types. A statistically sufficient sample of information in the means and standard deviations vectors is supplied to the latent space. A log-likelihood loss for each of the first and second anomaly types is calculated based on their respective mean and standard deviation.
Owner:SERVICENOW INC

Intelligent recognition system for suspension state defects of overhead line system

The invention belongs to the technical field of intelligent detection and identification, and particularly relates to a contact network suspension state defect intelligent identification system, which comprises an abnormal index calculation unit, a defect category probability analysis unit, a data enhancement unit and a spatial proximity probability field inference unit, the abnormal index calculation unit is used for obtaining a normalized value and obtaining a composite abnormal index based on the normalized value and in combination with the position parameter; the defect category probability analysis unit is used for forming defect category probability distribution through logarithm likelihood function mapping; the data enhancement unit is used for obtaining a defect triple consisting of a defect category, a defect barycentric coordinate and a defect severity quantized value; and the spatial proximity probability field inference unit is used for automatically giving an alarm when the probability of any grid node exceeds an alarm threshold value. According to the invention, the defect detection rate, the positioning precision and the intelligent maintenance efficiency are obviously improved.
Owner:CHENGDU NUOBIKAN TECH CO LTD

Large language model generation content detection method based on retrieval enhancement

The invention provides a large language model generation content detection method based on retrieval enhancement. Comprising three parts: the first part is construction and caching of a corpus, and a human writing corpus and a large model rewriting corpus are constructed in an offline stage; setting the label of the human writing text as y = 0, and setting the label of the large language model generation text as y = 1; the second part is used for retrieving related texts and rewritten texts and respectively taking the related texts and the rewritten texts as prefixes of to-be-detected texts; and the third part is that the generation probability difference of the to-be-detected text under different context conditions is calculated through a detection method, and is combined with the log-likelihood detection features of the to-be-detected text to finally obtain a judgment conclusion that whether the text is generated by the large language model or not.
Owner:RENMIN UNIVERSITY OF CHINA

Diversity combining method and system for soft demodulation output

The invention discloses a diversity combining method and system for soft demodulation output, and relates to the technical field of wireless communication. The method comprises the following steps: receiving a plurality of antenna signals, and acquiring a signal-to-noise ratio of each signal receiving path as an input parameter; establishing a Taylor expansion model based on the relationship between the signal-to-noise ratio and the log-likelihood ratio, and searching an optimal polynomial fitting formula coefficient; calculating a weighting coefficient of each signal receiving path according to an optimization formula, wherein the weighting coefficient is generated by a polynomial function of a signal-to-noise ratio; and carrying out weighted summation on the log-likelihood ratios of all the signal receiving paths according to the calculated weighting coefficient. According to the method, the weighting coefficient of bit-level signal combination is optimized, and the optimal weighting coefficient is determined in a manner of Taylor expansion at a zero point, so that the defect that the diversity gain is reduced when the signal-to-noise ratio difference is relatively large in a traditional method is overcome, the bit error rate performance under a complex channel condition is remarkably improved, and the reliability and the anti-interference capability of a system are enhanced.
Owner:上海飞机试飞工程有限责任公司 +1

Soft decoding method, equipment, medium and product

The invention discloses a soft decoding method and device, a medium and a product, and relates to the technical field of error correction coding, and the method is applied to a low-density parity check code, and comprises the steps: obtaining a target low-density parity check code with an information node and a check node containing an erasure bit; and allocating a preset non-zero log-likelihood ratio to the erasure bit on the check node, and executing first soft decoding. And if the first soft decoding fails, judging whether to carry out second soft decoding or not based on the weight of the syndrome generated by the first soft decoding and the number of the erasure bits on the information node. If it is judged that the second soft decoding needs to be carried out, a preset non-zero log likelihood ratio is distributed to the erasure bit located on the information node, and the second soft decoding is carried out. According to the method and the device, the erasing bits on different nodes are conditionally processed in stages, so that the success rate and the efficiency of decoding are effectively improved.
Owner:INSPUR SUZHOU INTELLIGENT TECH CO LTD

Trajectory analysis-based traffic abnormal event dynamic detection method and system

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.
Owner:AI SUPER EYE TECH CO LTD

Large model fine tuning method and system based on suffix prompt learning and ensemble learning

The invention discloses a large model fine tuning method and system based on a suffix prompt learning method and ensemble learning, and the method comprises the steps: carrying out the preprocessing of an existing data set, obtaining K subsets through the replacement sampling, enabling each subset to account for a specified percentage of the total, and constructing an artificial prompt; according to the teacher model and the student model, a predicted target word classification score is obtained through suffix prompt learning, through probability normalization, teacher knowledge is distilled to the student model by using a KL divergence loss function, and a prompt vector is initialized; an input task vector is obtained through context learning, and knowledge of the input task vector is distilled to a student model prediction vector by using a mean square error loss function; constructing a mixed loss function, and optimizing knowledge distillation and task feature learning; performing end-to-end optimization on the initialized prompt vector by using suffix prompt in combination with a negative log-likelihood loss function; during reasoning, after the optimized prompt vectors are spliced to the input vectors, suffix prompts share the input vector key value cache to generate output. According to the method, the adaptability and reasoning performance of the model are remarkably improved.
Owner:NANJING UNIV

Deviation correction Bayesian model averaging-based flood flow prediction method and system

The invention discloses a flood flow prediction method and system based on deviation correction Bayesian model averaging, and the method comprises the steps: constructing a hydrodynamic model through employing the obtained river section data, hydrodynamic model parameter data, historical flow data or hydrological model output flow data, and constructing a model set through employing the hydrodynamic model, outputting a flow prediction result of the model set; for K candidate models in the model set in the step 1, averaging and post-processing Bayesian models based on correction coefficients according to a total probability formula, and constructing a log-likelihood function; maximizing a log-likelihood function by using an SCE-UA algorithm, and calculating an average optimal parameter of a Bayesian model; and calculating a final predicted value and a confidence interval according to the calculated average optimal parameter estimation result of the Bayesian model. According to the method, the accuracy of flood flow prediction can be improved.
Owner:HOHAI UNIV

Data error correction method and device, electronic equipment and computer readable storage medium

The embodiment of the invention provides a data error correction method and device, electronic equipment and a computer readable storage medium, and relates to the technical field of storage, the bias reading voltage value of a storage unit of a memory is obtained, hard decoding is carried out according to the bias reading voltage value and a reference reading voltage value, and the data error correction efficiency is improved. The method comprises the following steps: acquiring a read voltage characteristic diagram of a memory, acquiring a reference log-likelihood ratio according to the read voltage characteristic diagram, a bias read voltage value and a reference read voltage value of the memory, if hard decoding fails, offsetting the reference read voltage value by the bias read voltage value in the read voltage characteristic diagram, and determining the CELL change quantity between the offset reference read voltage value and the reference read voltage value according to the read voltage characteristic diagram, and adjusting the reference log-likelihood ratio according to the CELL change number, and performing soft decoding according to the adjusted reference log-likelihood ratio. The log-likelihood ratio is adjusted according to the actual CELL change number, the more accurate log-likelihood ratio is obtained, the error correction performance of the LDPC is improved, the reliability of a memory is improved, and the service life of the memory is prolonged.
Owner:HANGZHOU CORE POWER SEMICON CO LTD