123 results about "Large networks" patented technology

Systems, methods, and computer readable media related to: training an encoder model that can be utilized to determine semantic similarity of a natural language textual string to each of one or more additional natural language textual strings (directly and / or indirectly); and / or using a trained encoder model to determine one or more responsive actions to perform in response to a natural language query. The encoder model is a machine learning model, such as a neural network model. In some implementations of training the encoder model, the encoder model is trained as part of a larger network architecture trained based on one or more tasks that are distinct from a “semantic textual similarity” task for which the encoder model can be used.

The invention provides a communication method and a communication device, and the method relates to the technical field of communication, and the method comprises the steps: a first UPF receives a first data message, the first data message comes from a first terminal, the first data message is a data message of a first group of services, the target IP address of the first data message is a first IP address, and the target IP address is a second IP address; the first IP address is a large network IP address of the second terminal, and both the first terminal and the second terminal sign a first group of services; and a second data message is sent to a second user plane function UPF, the second UPF is the UPF of the park where the second terminal is located, the second data message is generated based on the first data message, the target IP address of the second data message is a second IP address, and the second IP address is obtained based on the first IP address. The second IP address is a private network IP address applied by the second terminal based on the first group of services. Based on the method, the service access delay of the terminal can be reduced.

A network operation and maintenance metrics question and answer method and system based on a knowledge graph and a large model, relating to the technical field of computers. According to the method, a large network model may be triggered to extract at least one type of target operation and maintenance data from entity objects, entity relationships, entity attributes, and operation and maintenance events in network operation and maintenance data, so as to construct a knowledge graph on the basis of the target operation and maintenance data; the knowledge graph can assist the large network model in responding to knowledge question and answer, and the knowledge question and answer result can further assist in enhancing a modeling process of the knowledge graph, thereby improving the response precision of the large network model to knowledge question and answer in the field of operation and maintenance; and the knowledge graph can model and represent operation and maintenance actions, thereby achieving the standardization and automation of the operation and maintenance actions. The large network model can further optimize these operation procedures by learning a large amount of operation and maintenance action data, thereby improving the efficiency of operation and maintenance.

The invention discloses an agricultural land boundary extraction method and system fusing SAM and an improved Segform model, and relates to the technical field of image processing, and the method comprises the steps: carrying out the polygonal ground feature extraction of a preprocessed image tile data set through employing the SAM model; inputting the agricultural land sample data set into an improved Segform model for optimization, and then performing agricultural land boundary extraction; the improved Segform model takes the Segform model as a basic network structure, a feature fusion module fuses high-level semantic information, a DySample ultra-light dynamic up-sampling module is adopted to generate sampling points, and a DANet dual attention mechanism and an ECA channel attention mechanism are added for feature optimization. According to the method, on the basis of high-resolution image data, refined extraction is carried out on the boundary of the remote sensing image by fusing an SAM large network model and an improved Segform model.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for determining neural network architectures. One of the methods includes generating, using a controller neural network, a batch of output sequences, each output sequence in the batch specifying a respective subset of a plurality of components of a large neural network that should be active during the processing of inputs by the large neural network; for each output sequence in the batch: determining a performance metric of the large neural network on the particular neural network task (i) in accordance with current values of the large network parameters and (ii) with only the subset of components specified by the output sequences active; and using the performance metrics for the output sequences in the batch to adjust the current values of the controller parameters of the controller neural network.

The invention discloses a time sequence and congestion collaborative optimization-oriented GPU (Graphics Processing Unit) parallel acceleration global wiring method, which comprises the following steps of: according to a given netlist, dividing a super-large network by adopting a Kruskal algorithm in combination with a lookup set, and constructing a time sequence propagation path of the super-large network; performing network decomposition based on the timing margin estimation of the pins; calculating the time sequence weight of the two-pin network by adopting a time sequence weight calculation method based on an Elmore delay model; performing path cost calculation by adopting a cost function which comprehensively considers the time sequence and the congestion cost; executing two-level GPU parallel kernel acceleration mode wiring; optimizing the network topology by adopting a delay-aware pin connection improvement technology; the invention relates to a congestion-driven GPU (Graphics Processing Unit) accelerated routing and routing strategy for executing non-critical networks. According to the method, a high-quality wiring result with balanced time sequence congestion can be quickly obtained, the time sequence performance is effectively improved, and the requirement of a current super-large-scale high-performance circuit design wiring stage can be met.

The invention relates to the technical field of multi-element network equipment topological optimization, in particular to a multi-element network equipment topological optimization method based on clustering analysis. Comprising the following steps: setting a plurality of network nodes based on historical network topology data, and generating a communication evaluation value of each network node; setting a plurality of center nodes according to all the communication evaluation values, and establishing a communication topological structure according to all the center nodes; the method comprises the following steps: acquiring a feedback data packet according to a preset update time node, judging whether to update a communication topological structure according to the feedback data packet, establishing a plurality of network nodes based on historical network topological data, analyzing communication path information among the network nodes, effectively clustering the network nodes, and updating the communication topological structure according to the network nodes. According to the method, a large-scale network is divided into a plurality of small areas and clusters, and each cluster is iteratively optimized to construct a communication topological structure, so that an optimal communication path between network nodes is generated, and the communication efficiency between the network nodes is improved.

Systems, methods, and computer readable media related to: training an encoder model that can be utilized to determine semantic similarity of a natural language textual string to each of one or more additional natural language textual strings (directly and / or indirectly); and / or using a trained encoder model to determine one or more responsive actions to perform in response to a natural language query. The encoder model is a machine learning model, such as a neural network model. In some implementations of training the encoder model, the encoder model is trained as part of a larger network architecture trained based on one or more tasks that are distinct from a “semantic textual similarity” task for which the encoder model can be used.

A personal area network that includes a wearable electronic device, a system and methods of using the personal area network that includes a wearable electronic device. The wearable electronic device can act as an aggregator of the data that is being acquired by the one or more sensors and from other devices that are within wireless signal range of the personal area network in order to send some or all of the data over a wireless low power wide area network to remote locations within a larger network for subsequent processing, user notification, analysis of location-determination, contact tracing or the like. Data may flow in a bidirectional manner between the wearable electronic device and at least some of the other devices within the personal area network. In one form, the aggregated data may be used to provide information related to one or more operational parameters of an industrial asset and, if necessary, take control-based action in order to adjust one or more such operational parameters. In one form, a communication network formed by the wearable electronic device is used within an industrial setting in order to perform such data acquisition, processing and related control.

The invention discloses a forest fire smoke monitoring method and related device based on false detection feedback and model self-evolution, and the method comprises the steps: obtaining a video stream of a monitoring region, carrying out the preprocessing, inputting a target detection model and a scene understanding model, and judging whether a candidate alarm event is generated or not through consistency analysis; when a candidate alarm event is generated, a work order is generated for manual judgment, a real alarm executes a disposal process, and a false alarm is stored in a database; screening high-value false detection samples based on an information entropy priority strategy, storing the samples into a dynamic memory bank, selecting the samples to form an incremental training set, performing incremental training on a pre-training teacher model, and introducing low-rank constraint stability parameters; migrating the discrimination ability of the teacher model to the student model through knowledge distillation, wherein alignment of a channel dimension and a multi-scale space dimension is carried out; and redeploying the updated student model to execute target detection. The invention aims to solve the problems that the current monitoring system depends on manpower, the coverage capability is limited, the response is slow and the cost is high, and the traditional method is high in false alarm rate, easy to leak and misjudge, difficult in large-scale network edge deployment, limited in light-weight model discrimination capability, lack of linkage between alarm processing and model updating and the like, and more accurate and efficient forest fire smoke monitoring is realized.

The invention discloses a robot group real-time control method based on edge computing in the field of robots, and the method comprises the following steps: 1), deploying an edge computing node on each robot through employing a distributed edge computing architecture, and carrying out the real-time processing of local sensor data and the generation of a control instruction, communication and collaboration are carried out among the edge computing nodes through a wireless network, so that distributed decision and control are realized; and step 2), enabling each robot to autonomously learn a cooperative control strategy according to self sensor data and state information of the surrounding robots by using a cooperative control algorithm based on deep reinforcement learning. Through a distributed edge computing architecture, a deep reinforcement learning algorithm and a data compression transmission optimization technology, the problems of poor real-time performance, large network bandwidth pressure, low system reliability and the like existing in a traditional method are effectively solved, and the method has important application value and development prospect.

The invention provides a topological structure, a routing method and a routing device of a multi-dimensional hypercube interconnection network, and the multi-dimensional hypercube interconnection network comprises 2N network nodes. Target network nodes included by each sub-network in 2N-3 sub-networks of the multi-dimensional hypercube interconnection network as a target sub-network are divided into a first node set and a second node set, and four target network nodes in the first node set are connected with four target network nodes in the second node set in a one-to-one correspondence manner; the four target network nodes in the first node set are connected into a ring, the first network node in the second node set is connected with the third network node and the fourth network node, and the second network node is connected with the third network node and the fourth network node. According to the method and the device, the problem that the network communication time delay of the multi-dimensional hypercube interconnection network is relatively large is solved, and the effect of reducing the network communication time delay of the multi-dimensional hypercube interconnection network is achieved.

Efficient scaling of in-network compute operations to large numbers of compute nodes is disclosed. Each compute node is connected to a same plurality of network compute nodes, such as compute-enabled network switches. Compute processes at the compute nodes generate local gradients or other vectors by, for instance, performing a forward pass on a neural network. Each vector comprises values for a same set of vector elements. Each network compute node is assigned to, based on the local vectors, reduce vector data for a different a subset of the vector elements. Each network compute node returns a result chunk for the elements it processed back to each of the compute nodes, whereby each compute node receives the full result vector. This configuration may, in some embodiments, reduce buffering, processing, and / or other resource requirements for the network compute node or network at large.

The invention provides a network router management method and system. The method comprises the following steps: collecting a target routing address of a target router through an address collector in a routing interface circuit; when the target router exits the network, the routing interface circuit is triggered to send routing exit information to other routers, and the routing exit information comprises the target routing address; and searching other routing tables of other routers based on the target routing address, and marking and / or deleting entries including the target routing address in the other routing tables, so that when the router exits the network, routing exit information can be sent through the routing interface circuit. The problem that the accuracy of the routing table of each router is ensured by sending the routing table at present, resulting in a large network load and affecting the network transmission speed can be solved, the network load is reduced, and the network transmission speed is improved.

The application relates to the technical field of network communication, and discloses a network self-healing device based on node state data, which aims to improve the stability and self-healing ability of network communication. By monitoring the network state in real time, combining the extended Kalman filtering algorithm to perform state estimation and filtering, and combining the ant colony algorithm and the simulated annealing algorithm to perform fault prediction, path selection and resource allocation, the potential fault can be predicted and quickly recovered. The device comprises five modules of state monitoring, state estimation and filtering, fault prediction and prevention, fault recovery and resource reallocation, forms a closed-loop optimization mechanism, and is especially suitable for complex large network environments, such as data center networks.

The present application relates to a method for individual identification of electromagnetic signals using open set semi-supervised learning technology, belonging to the field of electromagnetic signal identification processing. The technical scheme of the present application trains a large classification network as a first model through self-supervised pre-training, so that it can more fully learn the internal relationship of unlabeled data and improve the network feature extraction capability; a small-scale classification network model closer to the characteristics of the identification sample is used as a second model, and in semi-supervised learning, the feature extraction capability of the pre-trained model is used to filter out unlabeled data other than the target classification, reducing the negative impact of out-of-distribution unlabeled data, and at the same time, model distillation is used to enable the small-scale classification network model to learn the generalization and robustness of the large network model for data features. Through the above three aspects, the influence of data other than the target classification on the semi-supervised learning network recognition accuracy in real situations is reduced, and robust identification of electromagnetic signals is achieved.

The present invention discloses a method and device for controlling false information on a social network, and relates to the technical field of network information analysis. Large networks can be decomposed quickly and efficiently, thereby controlling the spread of false information in a timely manner and reducing the losses caused by it. The method comprises: determining the first sequence with the smallest network resilience among multiple first sequences as the optimal sequence; deleting multiple third nodes with the highest probability from the coarse-grained network and adding them to a second set of removed nodes; placing the newly occupied nodes at the beginning of the removed sequence according to the sequence number, placing the unoccupied nodes included in the unoccupied node set at the middle of the removed sequence according to the node sequence number, and placing the removed nodes at the end of the removed sequence to form a removed sequence, and connecting the optimal sequence and the removed sequence to obtain a final sequence.

The application belongs to the field of FPGA-based hardware accelerator design, and proposes a point cloud target detection neural network accelerator and an acceleration method based on FPGA. The convolution accelerator comprises an instruction decoding module, a convolution module, an accumulation module, a truncation module and an activation function. The intermediate value cache only temporarily stores the channels of one feature map, consumes less BRAM, and is suitable for accelerating the point cloud target detection neural network with large feature map size, large number of channels and large network size. The application has flexible configurability and excellent reusability, so the application can accelerate different convolution layers of the point cloud target detection neural network, and is suitable for being integrated as an IP into a large design, reducing FPGA resource consumption and improving resource utilization. Moreover, the weight and the input feature map share one data input channel, and the convolution intermediate value is temporarily stored on-chip BRAM, reducing the feature map carrying frequency and saving the FPGA interface bandwidth.

The present application is a kind of lightweight target detection method for embedded platform, which overcomes the problems of large network parameter quantity, slow detection speed and poor precision index in the prior art. The present application not only solves the problems of slow speed and low accuracy of the current detection method in the deployment of embedded devices, but also optimizes the computation graph for specific operators from the hardware level, enabling fast and accurate target detection on limited resource devices. The present application comprises the following steps: step 1: obtaining basic data and making data set; step 2: data enhancement preprocessing; step 3: constructing a benchmark network model and pre-training; step 4: building a lightweight detection network model; step 5: reparameterization operation and pruning on the overall network structure; step 6: knowledge distillation to restore accuracy; step 7: deployment and acceleration of embedded platform.

The invention discloses a sensor network time synchronization method based on local bidirectional correction and related equipment, relates to the technical field of sensor networks, and solves the problem of high overhead of a bidirectional message time synchronization network. According to the method, the communication range is reduced through local bidirectional correction, and the synchronization hop count is reduced through hierarchical division of a topological structure, so that the network overhead is reduced while the overall time synchronization of the sensor network is realized: the reference nodes are dynamically elected to adapt to the network communication quality change, and the time synchronization accuracy is ensured while the network communication quality is improved. And the robustness of the time synchronization method is improved.

The application provides a photonic neural network system based on a multi-task neural network, which comprises: an optical computing module, taking a photonic artificial intelligence chip as a core, modulating an optical signal, and completing multiplication, summation and nonlinear operation of high-precision and high-speed optical analog quantities; a multi-task neural network module, having a main branch neural network and multiple secondary branch neural networks, and capable of simultaneously processing multiple classification tasks and regression tasks; and an optical computing communication and control module based on FPGA, connected with the optical computing module and the multi-task neural network module, and used for realizing real-time and high-speed communication of the optical computing module and the multi-task neural network module. The system has the characteristics of high bandwidth and low energy consumption, utilizes high-dimension parallel computing to greatly reduce the time of original neural network serial operation, has the advantages of large network scale and fast parallel operation speed, and can improve recognition accuracy and shorten training time on the basis of a single task.

The application relates to the field of electric communication technology, in particular to a multi-thread processing method based on a core surface target IP address, which comprises the following steps: establishing a signaling surface data packet according to a service request of a user, wherein the signaling surface data packet comprises a base station IP of a user terminal, signaling data and a target IP, the signaling data comprises a TEID of the user terminal; sending the signaling surface data packet to the base station; sending the signaling surface data packet to an edge distribution gateway by the base station; sending the signaling surface data packet to a signaling collection node for identification by the edge distribution gateway, and selecting a corresponding thread according to an identification result and transmitting the thread to a core network for data transparent transmission of a large network user; the application adopts a scheme of resource allocation by using a core network IP in a signaling learning library, ensures that a same user terminal is only processed in a same transmission thread, and avoids the problem of efficiency reduction caused by switching and data calling of different transmission threads.

The invention discloses a service shunting processing method and device and a storage medium, relates to the field of communication, and is used for realizing the purpose of shunting a service flow for a user terminal. The method comprises the following steps: receiving first shunting rule request information from a large network SMF; the first shunting rule request information comprises a first tracking area code TAC of the user terminal and an access network type of the user terminal; determining a target offload user plane function (UPF) based on the first offload rule request information; the target offload UPF comprises a first offload UPF or a second offload UPF, the first offload UPF is used for offload private network services, and the second offload UPF is used for offload public network services; determining a target shunting rule based on a shunting rule corresponding to the target shunting UPF; and sending the target shunting rule and first indication information to the large network SMF, wherein the first indication information is used for indicating the large network SMF to configure the target shunting UPF based on the target shunting rule. The method and the device are used in a user terminal service shunting process.

In a transmission scene with large network delay or large out-of-order distance, in order to solve the problems that the bit width of a packet receiving bitmap of an RDMA data receiving end is too large and storage resources are occupied too much, the invention provides a packet receiving bitmap management method of out-of-order RDMA. According to the method, an RDMA queue does not apply for a storage space in advance for an annular packet receiving bitmap of Q), the annular packet receiving bitmap is segmented into a plurality of sub-bitmaps, and a sub-bitmap space is dynamically applied or released from an available sub-bitmap address pool according to needs; the available sub-bitmap address pool is a queue, all sub-bitmap addresses are queued into the address pool during initialization, when the QP needs a sub-bitmap, one sub-bitmap address is dequeued from the address pool, and after a certain sub-bitmap of the QP is used, the used sub-bitmap address is queued back into the address pool; when a certain sub-bitmap of the QP is 1, the sub-bitmap space can be released in advance. According to the invention, the RDMA data receiving end supports the packet receiving bitmap with a large bit width, the occupied bitmap storage resources are small, and the utilization rate is high.

The invention provides for testing large-scale networks such as data centres by providing a test environment comprising network emulation platforms (NEPs) emulating switching fabric and testing physic

The present application relates to the technical field of pavement crack detection, and particularly relates to a lightweight pavement crack detection method, wherein a lightweight feature extraction module SandGlass and a coordinated attention mechanism are integrated into an encoding-decoding structure to construct a lightweight pixel-level crack detection network, a pavement image database is used to train the network model, so that the network model can realize the same detection effect as a larger network model under less computing power, the coordinated attention mechanism module suitable for mobile terminals is introduced to reduce the influence of noise and improve the accuracy of crack detection, and finally the encoding-decoding structure is improved to fuse feature maps of different levels, so that the final prediction result is closer to the actual pavement condition. Further, the improved method can be effectively applied to mobile devices to realize real-time and effective end-to-end detection.

Embodiments of the present application provide a device control method, apparatus, device, and storage medium, relating to the technical field of the Internet of Things. Among them, the device control method includes: obtaining a scenario control instruction for a plurality of devices; the scenario control instruction is used to instruct each of the devices to perform corresponding scenario actions respectively; identifying a first device among each of the devices; the first device is a device configured with a scenario action record; if at least one of the first devices is identified, then in accordance with the multicast transmission mode, sending the scenario control instruction to each of the first devices, so as to instruct each of the first devices to synchronously perform the scenario actions recorded for the first device in the scenario action record in response to the scenario control instruction. Embodiments of the present application solve the problem of relatively large network latency in the related art when multiple devices perform multiple actions.

The present application relates to the technical field of image processing, and more particularly to a water surface garbage identification method based on deep learning, comprising cutting pictures of a water surface garbage identification dataset; after adding a residual connection between a feature map and an input graph, an inhibition weight is obtained; each channel of the input graph is multiplied by the inhibition weight after subtraction to obtain a weighted feature map; two deep separable convolution groups are used in the deep layer to form a DSC module, and a Res module is used in the shallow layer; a deep separable convolution combined with a channel attention mechanism is used to obtain a one-dimensional feature vector, and a one-dimensional convolution is used to generate a channel weight, and finally the channel weight is multiplied by the original feature map; an ECA module is added after the Res module and the DSC module of the encoder; and a deep separable convolution module is used in the decoder. The present application solves the problems of noise interference such as reflection and reflection in the water surface scene, large network structure model size, large storage space occupation and high consumption of computing resources.

The application discloses a similar network behavior user identification method based on an unsupervised variational autoencoder, and relates to the technical field of network security. The method is based on super-large network communication metadata, uses an unsupervised variational autoencoder model, learns deep law features of user network behaviors from massive user network flow metadata, abstracts network behaviors of users into high-dimensional vector representations of essential behavior modes, and finally identifies users by using the similarity between vectors. Specifically, the method comprises network flow metadata, network behavior statistical feature extraction, behavior feature image processing, deep learning model training, deep feature extraction and similar behavior user identification.

This application belongs to the field of interconnection network reliability and fault diagnosis technology, and discloses a method for intermittent fault diagnosis of low-bandwidth long interconnection networks based on graph attention mechanism. Targeting the hierarchical recursive structure and high connectivity of the network, under the PMC fault diagnosis model, a multi-round testing strategy is used to obtain test symptoms within the node's neighborhood. A feature vector is constructed for each node using local statistical feature extraction methods, and preprocessed using zero-padding and noise enhancement techniques. Finally, a graph attention network model is constructed, and the importance weights of neighboring node test results are dynamically learned using the attention mechanism to achieve accurate diagnosis of node fault states in the network. This application leverages the powerful local information aggregation capability of graph attention networks to overcome the diagnostic limitations of traditional algorithms, maintaining high diagnostic accuracy and robustness even with high fault rates, incomplete test symptoms, and large network sizes.