71 results about "Time-varying network" patented technology

The invention provides a dynamic space-time air quality prediction method based on a physical constraint graph attention network, and the method comprises the steps: carrying out the collection and preprocessing of multi-source data, and obtaining the pollutant concentration and meteorological element historical sequence of a monitoring station; constructing a dynamic space-time diagram network fusing distance attenuation and a wind direction driving transmission path based on the longitude and latitude of the station and the real-time meteorological field; multi-cycle time sequence features are extracted in a self-adaptive mode through an enhanced time sequence network ETNet, and attention weight distribution is guided through physical priori; time sequence features are embedded into the dynamic graph, a physically constrained graph diffusion attention network PC-GDAN is input, and space-time diffusion modeling is achieved through a differential operator embedded into an atmospheric diffusion equation; time sequence and space features are fused, mass conservation and diffusion smoothness constraints are introduced in decoding prediction, and a concentration prediction result conforming to a physical rule is generated. The method effectively fuses physical mechanisms and data, has strong space-time modeling capability and physical interpretability, and significantly improves the precision and robustness of air quality prediction.

The invention discloses a multi-layer heterogeneous sequential network feature alignment and clustering method and device based on tensor self-representation. The method comprises the steps that sequential feature matrix sequences {},..., {} of data views in a multi-view heterogeneous network in time windows are obtained; based on the matrix sequence of the first time window, constructing a Laplacian regular term by introducing a self-representation learning mechanism, modeling multiple views by adopting a tensor structure and introducing a Schatten p-norm regularization mode to construct an optimization objective function for optimization, and representing shared feature representation of each data view in the first time window; and solving the function and carrying out clustering processing on the obtained function, wherein the clustering result of each time window is used for carrying out characteristic analysis on nodes in the network. According to the method, on the premise that the original structure and semantics of the network are guaranteed, unified modeling can be carried out on the incomplete multi-view heterogeneous network, and feature expression is aligned.

The invention discloses a sports market demand prediction method and device based on deep learning and a medium, and relates to the technical field of market demand prediction, and the method comprises the steps: collecting sports demand data, and carrying out the preprocessing; performing relation mining and graph structure learning on the preprocessed sports demand data through a graph attention space-time network to generate a macroscopic demand potential energy graph; performing potential area identification on the macroscopic demand potential energy diagram by adopting a pre-trained sports market simulation model, and outputting local demand prediction data; performing weighted fusion and error correction on the macroscopic demand potential energy map and the local demand prediction data, and outputting a sports demand prediction score; and making a sports market demand strategy according to the sports demand prediction score and the multi-granularity demand prediction report, and transmitting the sports market demand strategy to an enterprise manager through an enterprise decision support interface. According to the method, multi-level accurate prediction and decision support of sports market demands are realized through dual-mechanism cooperation of the graph attention space-time network and the space-time convolution.

The invention discloses a satellite secure transmission optimization method and system based on joint anomaly detection, mainly solving the problem of low end-to-end transmission reliability under the condition of network attack in the prior art, and the implementation scheme comprises the following steps: constructing a time-varying network topology according to constellation configuration; periodically updating a network topology and a satellite neighbor node set according to the law of satellite orbital motion; the network is monitored in real time based on an anomaly detection mechanism, whether the current network is subjected to interference attack and DoS attack is judged in time, and whether the current satellite can safely transmit data is judged; and according to indexes such as node states, link hops, detection precision, node load capacity and link transmission delay, and according to a network topology and a neighbor node set at the current moment, determining a routing control optimization problem, and guiding each satellite to select an optimal transfer satellite along a transmission direction according to an optimal solution to obtain an end-to-end optimal transmission path. The method can actively trigger a detection mechanism to quickly identify abnormal nodes, obtains a safe and reliable end-to-end transmission path, and can be used for a low earth orbit satellite network.

The invention provides an interest point recommendation method based on a dynamic hierarchical attention fusion space-time network, and the method comprises the following steps: 1, obtaining user historical sign-in and interest point feature data, and carrying out the preprocessing of the user data; 2, learning robust spatio-temporal information representation of the interest points through an adaptive spatio-temporal double-graph convolutional network module; 3, constructing a'interest point-functional unit-regional unit 'multi-level semantic hierarchical structure, and learning hierarchical semantic information representation of interest points by using a bidirectional hierarchical graph attention network module; 4, after fusing the two interest point representations, coding a user track and inputting the user track into an attention-frequency fusion Transform module, so as to obtain a multi-level semantic hierarchical structure; the module separates long and short term dynamic signals in a user behavior sequence through frequency analysis and dynamically fuses the long and short term dynamic signals with a self-attention mechanism to predict a next point of interest; and 5, training by adopting a combined loss function. According to the method, the complex characteristics of the user and the POI are effectively modeled, and the accuracy and the capability of capturing the dynamic preference of the user are improved.

The invention discloses an electrocardiosignal noise reduction method based on deep convolution and a sequential network, which comprises the following steps of: firstly, obtaining a pure electrocardiosignal through a synthetic function or a public data set, and constructing a pure-noisy electrocardiosignal pair by adding myoelectricity noise, power frequency noise, baseline drift or a combination of the myoelectricity noise, the power frequency noise and the baseline drift; designing a time sequence network structure comprising an input layer, a multi-level residual layer, a channel pruning layer, a full connection layer and an output layer, extracting time sequence features in the residual layer by using causal convolution, improving the performance of the model in combination with normalization, an activation function and a residual connection and discarding mechanism, training the model in a supervised learning mode, and obtaining a time sequence network structure; a mean square error or a mean absolute error is adopted as a loss function, network parameters are optimized through back propagation, and finally, the trained model is deployed on electrocardiogram monitoring equipment, so that real-time noise reduction processing on actual electrocardiogram signals is realized. According to the method, the time sequence characteristics of the electrocardiosignals can be effectively reserved, the noise reduction precision and robustness are improved, and the method is suitable for various medical and health monitoring scenes.

The invention relates to a wind power prediction method, system and device based on a multi-site dual-space-time network and a medium. The method comprises the following steps: acquiring historical data of a multi-site wind power plant, and preprocessing the historical data of the multi-site wind power plant to obtain preprocessed data; performing feature selection processing on the preprocessed data to obtain an optimized feature set; inputting the optimized feature set into a preset multi-site dual-time-space network to obtain a first future predicted value of each site and a second future predicted value of each site; the multi-site dual-time-space network comprises a long short-term memory network branch and a time convolution network branch; and carrying out weighted fusion on the first future predicted value and the second future predicted value to obtain a multi-station wind power predicted value. By adopting the method, the precision and robustness of multi-station wind power prediction can be improved, the clustering development requirement of the wind power plant can be better adapted, and powerful support is provided for power grid dispatching strategy optimization and safe and stable operation.

The invention relates to a computer-implemented method for transporting data (40) between an end-user device (20) and a server device (24) for providing a network service via a time-varying network (12), the time-varying network (12) including a plurality of nodes (14) that are interconnected intermittently in time.

This invention relates to the field of fertility data analysis technology and discloses an Internet-based regional fertility intelligent census system. By modeling data completion tasks through Markov decision processes, predicting completion value using deep Q-networks, achieving cross-period data calibration based on Bayesian temporal networks, and applying temporal smoothing constraints to ensure historical continuity, this system solves the technical problems of uneven distribution of missing historical fertility data, inconsistent detection technology standards, and limited supplementation costs. It achieves high-quality enhancement and spatiotemporal distribution analysis of regional fertility historical data.

The invention relates to the technical field of action recognition, and provides a hand action recognition method and system based on a lightweight cross-modal space-time network, and the method comprises the following steps: carrying out the preprocessing of a hand-washing video stream, and obtaining an RGB image sequence and a hand key point coordinate sequence; performing separation convolution operation of multi-layer space convolution and time convolution on the RGB image sequence, extracting motion-related features in combination with a channel attention mechanism, and generating a visual feature mark sequence; performing semantic extraction and time sequence compression on the key point coordinate sequence to obtain a structural feature mark sequence aligned with the X; and dynamically calculating a gating coefficient based on context information, fusing X and H to obtain fusion features, and outputting classification probability distribution of the hand washing steps. According to the method, a lightweight visual feature extraction structure is adopted, and a dynamic gating mechanism is introduced to realize adaptive collaborative fusion of visual and structural modes, so that high-precision and low-delay hand action recognition is realized on edge equipment.

The invention discloses a sea condition classification method and system based on a double-branch graph enhanced space-time network, and relates to the technical field of computers, and the method comprises the steps: receiving a multi-channel ship motion time sequence, carrying out the down-sampling of the multi-channel ship motion time sequence into an interlaced subsequence, and carrying out the down-sampling of the interlaced subsequence; performing feature extraction on each staggered subsequence based on a multi-layer perceptron, outputting compressed feature representation, and splicing all the compressed feature representations of the staggered subsequences to form time compression features; dynamically constructing an adjacent matrix between channels based on the time compression feature, and executing a graph convolution operation for information propagation to obtain a graph enhancement feature; and performing two-way fusion and classification based on the graph enhancement feature and the time compression feature, and outputting a sea condition classification result. According to the invention, through an efficient and lightweight network model, rapid classification of sea condition levels is completed on a resource-limited embedded device.

This invention discloses a low-latency speech enhancement method based on a microphone array. The method includes: setting a set of initial pole parameters; optimizing the initial pole parameters using an artificial neural network to obtain real poles; constructing orthogonal basis function models for each channel of the microphone array using the real poles and calculating the responses of filters of each order; performing frame segmentation and temporal feature extraction on the received signal from the microphone array, and estimating the adaptive beamformer weights constructed from the orthogonal basis function models using an improved temporal network; calculating the system response of each channel of the beamforming network based on the filter responses and beamformer weights to obtain the enhanced complete speech signal. This invention, by using an orthogonal basis structure beamforming network, can flexibly adjust the poles, increase the network's degrees of freedom, shorten the filter length, and reduce network latency; achieving better speech enhancement results with shorter filter lengths.

The invention discloses a low earth orbit satellite network deterministic multicast routing method based on a time expansion graph, which relates to satellite and communication technologies, and comprises the following steps: constructing a time expansion graph to represent time slot characteristics, dynamic network topology and time-varying network resources of a low earth orbit satellite network; pruning the original time expansion graph according to the working characteristics of the CSQF mechanism and multicast service transmission requirements, and providing a solution space for deterministic multicast routing calculation; and by taking the pruned and enhanced time extension graph as input, calculating a deterministic multicast route from a source node, and selecting a transmission edge with sufficient capacity and a storage edge conforming to a CSQF mechanism hop by hop until all virtual nodes are reached, so that the obtained route has the maximum bottleneck capacity on the premise of meeting the multicast service time delay upper limit requirement. According to the method, a low-complexity algorithm is designed, the link with the optimal resource and the transmission time slot with the time delay guarantee are selected hop by hop in the graph, and the multicast tree with the time attribute is constructed, so that end-to-multi-end deterministic guarantee is provided.

The application discloses an activity recognition method based on a variational autoencoder and a time sequence network, and belongs to the technical field of behavior feature recognition, and comprises the following steps: S1, collecting an original data set of activity signals and performing pretreatment; S2, extracting latent features based on S1 by using a VAE, and outputting a reconstruction error; S3, obtaining time dynamic features based on S1 by using a CNN-GRU; S4, performing feature fusion based on the results of S2 and S3 by using a self-attention mechanism, performing soft assignment by using a deep embedding clustering layer, and performing classification by using a classifier, and completing model training; S5, testing, judging whether input is OOD data, and outputting a category. The application effectively recognizes and processes OOD data by judging the reconstruction error and the output of the deep embedding clustering layer, and enhances the robustness and generalization ability of the model; by combining the VAE, the CNN, the GRU and the self-attention mechanism, space-time features in IMU signals are comprehensively extracted, the accuracy of activity recognition is improved, time sequence information is effectively reserved, and the model training efficiency is improved.

The invention discloses an online car-hailing OD demand prediction method based on a self-adaptive normalized space-time network, and the method comprises the steps: dividing a research region, constructing a graph adjacency matrix, carrying out the statistics of OD demands on the basis, diagnosing the distribution offset of data, and finally inputting a space-time OD matrix into a self-adaptive normalized space-time network model for training prediction. The adaptive normalized space-time network comprises the following steps: a RevIN space-time decoupling normalization layer eliminates cross-sample distribution offset; fully extracting space-time features through GCN-LSTM space-time diagram convolution; and finally, the predicted value is recovered to the size the same as that of the true value through a RevIN anti-normalization layer. According to real drop-out data prediction, the improved RevIN network is introduced into the space-time prediction model to provide the ST-RevIN model, and the model can relieve the distribution offset problem of online car-hailing OD data, improve the prediction accuracy and reduce the prediction error.

The invention discloses an unbalanced data fault diagnosis method for an aero-engine gas path system based on a space-time network. The method comprises the steps of obtaining limited unbalanced data of gas path faults of an aero-engine in different operation states; constructing an STFD model for the unbalanced data fault diagnosis of the aero-engine gas path system based on the space-time network; the system comprises a spatio-temporal feature extraction network, a spatio-temporal feature distribution alignment module and a class balance label distributed sensing marginal module. And performing model training on the STFD model according to the source domain data set and the data training set, determining an optimal STFD model according to the data verification set based on the constructed model total loss function, and inputting the data test set into the optimal STFD model, thereby realizing fault prediction of the unbalanced data of the aero-engine gas path system. The problems that according to an existing unbalanced data fault diagnosis method, the features of the time dimension and the space dimension cannot be effectively extracted, feature information in limited data cannot be fully mined, and the problem of data imbalance cannot be solved in combination with an advanced algorithm are solved.

The invention relates to a distributed heterogeneous unmanned aerial vehicle time-sensitive task allocation and flight path collaborative planning method, which comprises a fast path cost dynamic estimation module, a distributed task allocation and path planning coupling module and a local reset module for new tasks in a dynamic environment. The method has the advantages that a real-time evaluation mechanism of the path cost is embedded into a task allocation process, deep fusion of task allocation and path reachability is achieved, and decision deviation caused by decoupling processing is effectively avoided. On the basis, a local reset strategy based on a task time window and path cost is introduced, the response capability of the system to dynamic task changes is remarkably improved, and the calculation overhead is reduced. Meanwhile, an improved consensus-based task coordination framework is adopted, and an adaptive communication mechanism is fused, so that the robustness and convergence speed of the algorithm under a time-varying network condition are enhanced.

The invention discloses an emotion recognition method and system based on a hierarchical bidirectional depth sequential network. The method comprises the steps that firstly, electroencephalogram signals of a subject are collected through electroencephalogram collection equipment, noise reduction and feature extraction are conducted on the collected signals, and feature vectors are formed; secondly, screening channels of electroencephalogram signals, and selecting key channels to reduce redundant features; then, constructing a hierarchical bidirectional depth time sequence network, capturing global time sequence dependence features through a bidirectional time sequence modeling layer, and cascading multiple unidirectional time sequence modeling layers to hierarchically extract emotional features; and finally, inputting the extracted emotion features into a classifier for emotion classification, and outputting an emotion recognition result. According to the method, cross-subject consistency channels are screened by optimizing a feature selection strategy, and redundant features are reduced; meanwhile, transient fluctuation and long-range emotion migration of the EEG signals are captured; on the premise of ensuring high accuracy, model complexity and computing resource consumption are reduced.

The application relates to the technical field of computer communication, and provides a satellite time-varying network performance prediction method based on a graph neural network. Specifically, network topology and state data of a LEO satellite network are acquired from a data plane through SDN technology; a graph network with heterogeneous nodes is constructed according to the network topology and network flow; and a network performance prediction model based on a message passing neural network (MPNN) and adding a node feature LSTM mechanism and a graph attention mechanism is used to realize efficient prediction of the network performance of the LEO satellite network. The method can accurately predict key performance indicators (KPI) of end-to-end network flow in the satellite network, such as delay, jitter and packet loss rate, so as to optimize the quality of network service (QoS).

The invention relates to the field of computer vision, in particular to a weak supervision video anomaly method and system based on prompt correction and a hierarchical sequential network. Performing feature extraction on the input video to obtain an initial visual feature sequence; performing hierarchical time sequence processing on the initial visual feature sequence to generate an enhanced visual feature sequence, and calculating an initial visual anomaly score of each video clip according to the enhanced visual feature sequence; inputting the enhanced visual feature sequence into a video classifier which shares a weight with the fragment classifier, and generating a video-level anomaly score; on the basis of similarity calculation of the text prompt features and the enhanced visual feature sequence, a prompt correction score is generated through dynamic allocation, and the initial visual anomaly score is corrected to obtain a final anomaly score; and realizing video anomaly detection according to the final anomaly score. According to the method, the layered sequential network and the prompt correction mechanism are introduced, so that the accuracy and robustness of weak supervision video anomaly detection are remarkably improved.

The invention discloses a modeling and decomposition method for collaborative distribution of trucks and unmanned aerial vehicles based on a high-dimensional network. The method comprises the following steps: respectively constructing space-time networks belonging to trucks and unmanned aerial vehicles; the method comprises the following steps: acquiring a truck-unmanned aerial vehicle operation state, and constructing a space-time-state high-dimensional network and an unmanned aerial vehicle sortie set based on a ground network and an air network; taking the minimum completion time as a target function, and establishing an arc-based truck unmanned aerial vehicle distribution path integer linear programming model; and solving the model by using generalized Benders decomposition, and accelerating the solving process in combination with an improved Benders feasible cut and branch and bound framework to obtain an optimal distribution path of the truck unmanned aerial vehicle. According to the method, the blank of insufficient space-time state dimension modeling of vehicle-machine collaborative path optimization in existing research is filled up, and a solution with expandability and high practical value is provided for landing application of an intelligent distribution system.

The invention discloses a structure evolution measurement and anomaly tracing method and system of a temporal network, and belongs to the technical field of dynamic graph data mining. In order to solve the problem that in the prior art, it is difficult to quantify the network evolution strength and locate structural abnormity, an evolution measurement model based on the tense graph editing distance is constructed. According to the method, six types of editing operations covering topology mutation and time sequence deviation are defined, and a non-uniform cost function is constructed based on a time interval overlapping relation. A lightweight evaluation and accurate solution coupling strategy is adopted in the solution process. Firstly, a global lower bound is utilized to quickly evaluate the volatility so as to eliminate normal disturbance; and solving the minimum evolution cost through heuristic search and inert evaluation, and constructing an abnormal traceability path. According to the invention, real-time quantification of large-scale temporal network evolution behaviors and accurate positioning of key abnormal events are realized.

The application discloses an in-bound tourism risk prediction and dynamic service optimization method based on big data, aiming at solving the problems of heterogeneous risk data perception difficulty, inaccurate spatio-temporal evolution prediction and disconnection between risk and service response in the prior art. The method predicts the risk evolution trend by constructing a unified risk tensor and using a deep spatio-temporal network combining graph convolution and a gated recurrent unit, and realizes dynamic mapping from risk state to service action through a comprehensive adaptation function weighted by reinforcement learning, and then combines the threshold drift and cross-cultural content generation of the tourist cultural portrait to form a whole-process intelligent solution from global perception, trend prediction, strategy mapping, differentiated intervention to closed-loop adaptive optimization, effectively improving the precision of in-bound tourism safety management and the level of personalized service.

The invention discloses a millimeter wave radar point cloud moving target segmentation method based on a layered space-time network. The method comprises the following steps: step 1, obtaining a radar point cloud sequence data set; 2, constructing a layered space-time consistency moving target segmentation network HSTC-MoSeg, wherein the layered space-time consistency moving target segmentation network HSTC-MoSeg comprises a layered context attention block, a space adaptive point convolution module and a time consistency enhancement module; 3, training the improved model and obtaining a test result; and 4, performing an ablation experiment to verify the effectiveness of each module. According to the method, local and global semantic contexts are balanced through the hierarchical context attention block, and the problem of unbalanced feature expression caused by uneven point cloud density is solved; local aggregation of density robustness is realized through spatial adaptive point convolution, and feature degradation of a sparse region is reduced; through a time consistency enhancement module, the segmentation consistency is enhanced by using cross-frame time information, and the jitter of a motion boundary is inhibited. Experiments on a RadarMOSEVE data set show that the method provided by the invention is obviously superior to the existing method in indexes of IoU-Avg, F1-Avg and Movin-F1, and the accuracy and robustness of radar point cloud moving target segmentation are effectively improved.

This invention provides a satellite network topology management method and system based on feeder link switching. The method includes: constructing a continuously time-varying network topology model, and storing the availability sequence, available bandwidth sequence, and propagation delay of each link; responding to new service requests and obtaining the source satellite node and the target ground station; selecting the target feeder link and its effective connection period based on link status information; screening inter-satellite links that are continuously connected within the time period to form a candidate set, and calculating the average propagation delay and minimum available bandwidth of each link to construct a temporary topology; calculating the transmission path from the source satellite node through inter-satellite links and the target feeder link to the target ground station on the temporary topology; and locally updating the available bandwidth sequence of each link on the path according to the path and bandwidth requirements. This invention achieves efficient management of highly dynamic satellite-ground networks through feeder link event-driven and continuously time-varying modeling, improving service continuity and reducing management overhead.

The application relates to a dynamic time sequence network clustering method and device based on contrast learning, electronic equipment and a storage medium, and belongs to the technical field of computers. The method comprises the following steps: obtaining to-be-processed data; inputting the to-be-processed data into a pre-trained variational autoencoder to obtain dimension-reduced features corresponding to the to-be-processed data; wherein the pre-trained variational autoencoder is obtained according to a training method based on contrast learning; the training method based on contrast learning adjusts parameters of the variational autoencoder by using a loss function set based on contrast learning; and a clustering method is used for clustering analysis of the dimension-reduced features corresponding to the to-be-processed data. The pre-trained variational autoencoder is used for dimension reduction processing of the to-be-processed data, and then clustering analysis is performed. A variational autoencoder capable of considering changes in a dynamic time sequence network is proposed, the variational autoencoder can perform dimension reduction processing on data, effectively extract main features of the dynamic time sequence network, and perform clustering analysis.

The method comprises the following steps: S1, dividing a global scheduling period into a peak window, a flat peak window and a valley window according to a task density distribution; S2, assigning different fixed discrete steps to the peak window, the flat peak window and the valley window; S3, establishing node coupling and arc segment continuity constraints at the junctions of windows with different precisions; and S4, outputting a model and optimizing. The method has the advantage of effectively suppressing calculation redundancy in a non-event triggering mode.

The invention provides a time-varying routing method, time-varying routing equipment and a time-varying network, and relates to the technical field of networks. And the first network device receives a routing message issued by the second network device, wherein the routing message comprises time information and a routing plan. The time information comprises a message publishing moment, and the message publishing moment is a system moment of the second network equipment when the second network equipment publishes the routing message. And if the time delay between the message issuing moment and the message receiving moment exceeds a preset error range, the first network equipment determines the effective moment of the routing plan according to the message receiving moment and the delay effective duration corresponding to the routing plan. The message receiving moment is the system moment of the first network equipment when the first network equipment receives the routing message. In a clock asynchronous scene, the first network equipment determines the effective moment of the routing plan according to the message receiving moment, so that the first network equipment and the second network equipment are quickly and consistently converged in the actual transmission delay of the routing message, and the network reliability is improved.

This invention provides a wind power interval prediction method based on mode decomposition and reconstruction and attention-based temporal network, belonging to the field of wind power interval prediction technology. The method includes: acquiring historical monitoring data of wind farms and performing empirical mode decomposition to obtain intrinsic mode function (IMF) components, which are then superimposed and reconstructed to obtain reconstructed IMF components; constructing a wind power output prediction model based on a Transformer encoder by introducing a bidirectional long short-term memory (LSTM) neural network; dividing and normalizing the reconstructed IMF components, training and predicting using the wind power output prediction model, obtaining prediction results, and superimposing them; dynamically reshaping the model through adaptive kernel density estimation to obtain a dynamic power prediction interval, thus completing the temporal network wind power interval prediction. This invention achieves efficient data processing, effectively reduces the non-stationarity and volatility of wind power data, avoids computational redundancy, and improves prediction efficiency.

In an aspect, input data can be received, including at least time series data associated with purchases of at least one product and causal influencer data associated with the purchases. The causal influencer data can include at least non-stationary data, where lost shares associated with said at least one product are unobserved. An artificial neural network can be trained based on the received input data to predict a future global demand associated with at least one product and individual market shares associated with at least one product. The artificial neural network can include at least a first temporal network to predict the global demand and a second temporal network to predict each of the individual market shares. The first temporal network and the second temporal network can be trained simultaneously.