26 results about "Mobility prediction" patented technology

The invention discloses an asset management information processing method, device and equipment based on a big data platform, and relates to the technical field of data processing. The method comprises the steps that an asset management data platform carries out data specification processing to obtain index data, the index data are sent to an information big data platform, an asset management graph neural network dynamic relation graph is constructed based on the index data, and asset early warning index prediction is carried out based on the asset management graph neural network dynamic relation graph and an asset management behavior prediction model. And based on the mobility prediction model, carrying out mobility index prediction to obtain response information. Data management is performed through the two data management platforms, repeated processing is avoided, risk early warning information is generated based on the model method and the information management graph neural network dynamic relation graph, risk early warning is realized, and the real-time performance of risk early warning is ensured.

The invention relates to a method and a device for predicting the fluidity of alkali-activated mortar. The method comprises the following steps: determining a training set of the alkali-activated mortar; determining hyper-parameters of the deep belief network; inputting the training sample into the deep belief network, outputting a mobility degree predicted value corresponding to the training sample, optimizing hyper-parameters by adopting a grey wolf algorithm based on the mobility degree predicted value, and updating the deep belief network based on the optimized hyper-parameters, using the trained deep belief network as a mobility degree prediction model to predict the mobility degree of the target data; according to the method, the hyper-parameters of the deep belief network are optimized through the grey wolf algorithm, the high-precision prediction model is constructed, the complex nonlinear relation between the multiple parameters of the alkali-activated mortar and the fluidity of the alkali-activated mortar can be accurately extracted, the prediction precision of the fluidity degree of the alkali-activated mortar is improved, and the prediction efficiency is improved. And a theoretical basis and practical guidance are provided for mix proportion optimization and engineering application of the alkali-activated mortar.

This invention relates to the field of routing technology, and more particularly to an on-demand predictive routing method for unmanned aerial vehicle (UAV) networks. The invention determines the state information of neighboring UAVs based on an improved Kalman filter model, predicts the latest state information of neighboring UAVs based on this state information, determines the communication status between the UAV and its neighbors and the message coverage rate of the neighboring UAVs for flooding commands based on the latest state information, sets replay delays and replay probabilities for the neighboring UAVs, determines the distance correlation factor and traffic load factor of the UAVs in response to the communication status between the UAV and its neighbors, calculates the routing cost value of candidate links, determines the optimal path to forward data, and for the failure of the optimal path, finds alternative links to replace the failed optimal path or performs path repair. This invention combines node mobility prediction and a neighbor coverage flooding mechanism to achieve fast addressing and dynamic optimization of routes.

The present disclosure describes method for Non Terrestrial Networks (NTN) conditional and RACH-less handover comprising the steps: performing UE mobility prediction and grouping UEs, whereby a determination of different suitable conditional handover configurations and / or of a sequence of multiple conditional handover candidates is done, followed by pre-allocating uplink grants.

The invention relates to the technical field of unmanned aerial vehicle ad hoc networks of ultraviolet light communication, and provides a dynamic routing method and device for an ultraviolet light flight ad hoc network, a medium and a product, and the method comprises the steps: setting a communication mode of the ultraviolet light flight ad hoc network; determining a motion model; detecting network neighbor nodes, and generating a neighbor node information table; based on the motion model, updating position information of each node in the neighbor node information table; and dynamically forwarding a data packet by adopting a data packet forwarding mode based on the updated neighbor node information table. By setting a communication mode, determining a motion model, predicting mobility, and selecting and recovering a dynamic route, the problems of frequent disconnection of the route and decision error caused by high-speed motion of the node are effectively alleviated, the end-to-end time delay and energy consumption are reduced, the delivery rate of data packets is improved, and the robustness of the route is enhanced.

The invention provides user equipment and a method for performing mobility prediction by using artificial intelligence. In one novel aspect, UE mobility prediction is performed through machine learning techniques based on UE measurement data. In an embodiment, the UE obtains a set of mobility-related data, inputs the set of mobility-related data into a mobility AI model for UE mobility prediction, and obtains UE mobility prediction based on the mobility AI model. In an embodiment, UE mobility prediction is range based. In another embodiment, two independent AI models are applied to predict UE mobility in different cases, such as within and not within the service area. In an embodiment, the UE obtains mobility feedback from one or more UE applications and fine-tunes the mobility AI model based on the mobility feedback.

The embodiments of the present disclosure relate to measurement methods, devices, and a storage medium. The method comprises: a terminal executing mobility prediction on the basis of a measurement event, wherein the mobility prediction can be used for performing prediction to obtain mobility information of the terminal, and the mobility information can be used by the terminal to perform a mobility operation. In this way, the mobility prediction of the terminal can be flexibly controlled and managed, thereby improving the communication reliability.

The embodiment of the invention relates to a measurement method, equipment and a storage medium. The method comprises the steps that the terminal executes mobility prediction based on a measurement event, the mobility prediction can be used for predicting mobility information of the terminal, and the mobility information can be used for the terminal to carry out mobility operation. Therefore, the mobility prediction of the terminal can be flexibly controlled and managed, and the communication reliability is improved.

This invention discloses an adaptive application deployment method in an edge computing network based on mobility prediction. The deployment framework implemented by this method is based on a general adaptive mobility prediction framework, achieving good accuracy for different movement trajectories while saving on model design and training costs. An adaptive application deployment method is proposed, using mobile user mobility predictions over multiple future time periods as input. The goal is to optimize and maximize user service quality while minimizing deployment costs at edge nodes, dynamically deploying mobile user services to the most suitable edge nodes. This invention constructs a joint optimization mechanism to determine the deployment location and duration of user applications.

The embodiment of the invention provides a time domain mobility prediction method and device, communication equipment, a chip and a storage medium, and the method comprises the steps that a terminal obtains K measurement instances corresponding to a first downlink reference signal set, the K measurement instances are obtained by measuring downlink reference signals in the first downlink reference signal set at K measurement moments based on the terminal, and the first downlink reference signal set comprises first downlink reference signal subsets of N cells; wherein the K measurement instances are used for acquiring F prediction instances corresponding to a second downlink reference signal set based on a first model, different prediction instances correspond to different prediction moments, and the second downlink reference signal set comprises second downlink reference signal subsets of P cells; k, N, F and P are integers greater than or equal to 1.

The invention discloses an unmanned aerial vehicle MEC network data caching and trajectory optimization method, and relates to the technical field of information and communication engineering, and the method comprises the following steps: deploying a semantic analysis platform and a mobility prediction platform to a plurality of unmanned aerial vehicles, setting flight parameters, caching capacity and an initial strategy of a deep reinforcement learning agent, and forming a system operation basis; receiving a ground user data stream, performing fragment-level analysis on the data content by using a semantic analysis platform, generating a semantic value sequence, and determining a semantic value integral based on the semantic value sequence; according to historical user position information, a position prediction track of each user is output through a mobility prediction platform, and a future link quality change trend is evaluated; fusing the state of the unmanned aerial vehicle, the user prediction position, the semantic value integral and the link quality evaluation result to construct an environment state vector; and inputting the environment state vector into a deep reinforcement learning agent to generate a joint action vector.

Various aspects of the present disclosure generally relate to wireless communication. In some aspects, a first network node in a radio access network (RAN) may transmit mobility history data for a user equipment (UE) to a second network node in a core network associated with the RAN. The first network node may receive a UE mobility prediction model that is based at least in part on the mobility history data from the second network node. Numerous other aspects are described.

This technology provides an improved prediction accuracy for coal expansion pressure information. [Solution] The coal expansion pressure information prediction method includes an acquisition step of obtaining the average maximum reflectance, inert ratio, and maximum fluidity of the coal to be predicted, and a prediction step of predicting expansion pressure clusters corresponding to the expansion pressure based on the acquired average maximum reflectance, inert ratio, and maximum fluidity of the coal to be predicted.

The present disclosure describes a method for non-terrestrial network (NTN) conditional handover and RACH-free handover, the method comprising the steps of performing UE mobility prediction and grouping UEs, thereby determining different appropriate conditional handover configurations and / or a sequence of a plurality of conditional handover candidates, followed by pre-allocation of uplink grants.

This invention discloses a method for UAV access capacity in the 1090 MHz band based on variable channels and mobility prediction. It introduces S-mode and A / C-mode messages in the real-world 1090 MHz band; and allows for the setting of corresponding channel propagation models according to different environments. All aircraft are modeled as mobile devices; and the ADS-B message update cycle can be adjusted according to the UAV's speed. Finally, the continuous packet loss situation and corresponding packet loss rate of civil aircraft and UAVs can be obtained. Under the condition of meeting the minimum message update rate of civil aircraft, i.e., maintaining the normal operation of the civil aviation surveillance system, the method analyzes how many UAVs equipped with ADS-B OUT devices can be added to this frequency band.

The present disclosure describes a method for non-terrestrial network (NTN) conditional handover and RACH-free handover, the method comprising the steps of performing UE mobility prediction and grouping UEs, thereby making a determination of different suitable CHO configurations and / or a sequence of a plurality of CHO candidates, thereby pre-allocating UL grants; qoS requirements and mobility characteristics of the UE are considered.

This invention discloses a method for three-dimensional fine modeling and stability probability evaluation of complex rock mass blocks, comprising three steps: first, three-dimensional modeling of complex blocks based on three-dimensional laser scanning, integrating real high-precision terrain and random boundaries; second, introducing full-space stereographic projection for kinematic mobility prediction, inputting the extracted random geometric parameters into a rigid body limit equilibrium vector model considering water-rock coupling effects such as pore water pressure reduction, and calculating the probability distribution of the safety factor; and third, based on a multi-source collaborative support inversion model of target reliability, dynamically calculating and outputting the probability distribution of anchor cable reinforcement force required to maintain the target's stable state. This invention overcomes the inherent defects of traditional deterministic analysis and simplified probabilistic models, and overcomes the shortcomings of existing technologies that often use fixed attitude parameters and are difficult to detect potential risks.

The invention discloses a V2E authentication load balancing method based on multi-agent reinforcement learning, and the method comprises the steps: firstly constructing a V2E authentication model, constructing a mobility prediction model based on an attention mechanism space-time diagram convolutional network, and predicting an authentication task demand based on the traffic data and road network topology of the V2E authentication model. Secondly, the optimization problem of the V2E authentication model is converted into a Markov decision process, the Markov decision process is solved, and an optimal dynamic adjustment strategy is learned in a collaborative mode; finally, the intelligent agent outputs a decision instruction in real time according to the learned optimal dynamic adjustment strategy, and dynamic balance of the global authentication load is achieved. According to the method, the contradiction between authentication task domain dependency and system load balancing is effectively solved, the limitation that a traditional task unloading scheme cannot be applied to an authentication scene is fundamentally overcome, and the average authentication delay of the system is remarkably reduced.

The present application relates to a kind of intelligent internet of things multi-modal data sensing method, belong to internet of things field.First, the weight of space-time attention mechanism is constructed when sensing platform to capture the influence of space-time relationship on worker mobility prediction, the mobility of sensing worker is predicted using GRU prediction model with space-time attention mechanism.Then based on the prediction result, using the multi-modal sensing task utility maximization task allocation method based on mobility prediction to the sensing worker in different time period appears in multi-modal sensing POI task area is allocated multi-modal sensing task.Effectively improve the prediction accuracy of sensing worker in different time period located in different multi-modal sensing task area, assign different sensing worker suitable multi-modal sensing task, improve the utility of sensing platform and the completion rate of multi-modal sensing task.

The embodiment of the invention discloses a cross-city mobility prediction method fusing a large language model and comparative learning. The method comprises the steps of obtaining spatial feature vectors corresponding to a source city and a target city, and performing dimension reduction processing on the spatial feature vectors to obtain semantic feature vectors corresponding to the source city and the target city; performing cross-city semantic alignment on the semantic feature vector corresponding to the source city and the semantic feature vector corresponding to the target city to obtain a semantic feature alignment result; determining an origin-destination prediction result between the source city and the target city based on the semantic feature alignment result; and based on the spatial feature vector corresponding to the target city and the time information, determining a space-time grid unit of the target city, and based on the space-time grid unit and the origin-destination prediction result, determining an origin-destination traffic matrix corresponding to the target city. According to the embodiment of the invention, comprehensive and accurate data support can be provided for urban planning, traffic management, resource allocation and the like, and improvement of urban operation efficiency and rationality of resource utilization is facilitated.

A human flow prediction method and device based on a large language model cooperates with a small model plug-in, the method comprises the following steps: preprocessing original user check-in data, constructing a time sequence knowledge graph containing the global interaction relationship of users and places in each period, which is used to represent the check-in behavior of the user. Meanwhile, the long-term historical check-in trajectory and the recent check-in trajectory of the user are constructed; the user memory vector is constructed, and a plug-in small model composed of multiple encoders is trained. The plug-in small model is not limited to the form described in the application and can be freely replaced with other plug-in small models; the user's partial historical check-in trajectory and recent check-in trajectory are extracted, and the small model is used to generate the top K prediction labels, confidence scores, benchmark true values and other data to generate prompt words; the large language model is fine-tuned by using historical data supervision, the user's whereabouts are predicted according to the prompt without benchmark true value, and the final prediction result is obtained. The application can provide more accurate prediction results.

Apparatus and methods are provided for UE mobility prediction with AI. In one novel aspect, UE mobility prediction is performed based on UE measurement data through machine learning techniques. In one embodiment, the UE obtains a set of mobility-related data, feeds the set of mobility-related data to a mobility AI model for UE mobility prediction and obtains a UE mobility prediction based on the mobility AI model. In one embodiment, the UE mobility prediction is range-based. In another embodiment, two independent AI models are applied to predict the UE mobility under different situations, such as in-service and out-of-service. In one embodiment, the UE obtains mobility feedback from one or more UE applications and performs fine turning for the mobility AI model based on the mobility feedback.

Provided are a temporal domain mobility prediction method and apparatus, a communication device, a chip, and a storage medium. The method includes: a terminal obtaining K measurement instance(s) corresponding to a first downlink reference signal set, the K measurement instance(s) being obtained based on measurement performed by the terminal on downlink reference signals in the first downlink reference signal set at K measurement time instance(s), the first downlink reference signal set including first downlink reference signal subsets of N cell(s), the K measurement instance(s) being used to obtain F prediction instance(s) corresponding to a second downlink reference signal set based on a first model, different prediction instances corresponding to different prediction time instances, and the second downlink reference signal set including second downlink reference signal subsets of P cell(s), K, N, F, and P being integers greater than or equal to 1.

Various aspects of the present disclosure relate to cause detection of a problem related to mobility. In one aspect, a UE transmits, to a first base station, first information related to mobility prediction based on AI or ML. Then, the UE determines occurrence of a problem related to mobility of the UE. In turn, the UE transmits a report related to the occurrence of the problem related to mobility of the UE to the first base station or a second base station. The report comprises second information related to a problem of the mobility prediction.

The invention provides a fund data flow control method and system and electronic equipment, and relates to the field of fund data control, and the method comprises the steps: constructing a closed-loop management system of global fund flow data acquisition-mobility prediction, fund flow threshold dynamic generation-fund allocation strategy updating; a management control mode of capital data flow from passive bookkeeping to active operation is realized, the use efficiency of capital is remarkably improved, and the anti-risk capability of enterprise capital is improved.