17 results about "Virtual network topology" patented technology

The embodiment of the invention provides a network spoofing defense method and system based on AI virtual topology generation, and the method comprises the steps: continuously collecting multi-source heterogeneous data in a network, and constructing and dynamically updating a real network topology portrait based on the multi-source heterogeneous data; on the basis of a generative adversarial network model, adversarial training is carried out by taking the real network topology portrait as a training sample, the structure and characteristics of a real network are learned, and a dynamically changing simulation virtual network topology is generated; seamless fusion and dynamic deployment are carried out on the simulation virtual network topology and the real network, and a trapping environment is constructed; monitoring and recording an attack behavior of an attacker in the trapping environment, and performing deep interaction analysis on the attack behavior and generating threat intelligence; and based on the threat intelligence, dynamically optimizing a deception strategy and executing a closed-loop control response. According to the embodiment of the invention, the virtual topology which is highly similar to a real network but is completely imaginary can be dynamically generated by utilizing artificial intelligence.

The invention relates to a virtual network mapping method based on double-agent reinforcement learning in a space division multiplexing elastic optical network, belongs to the technical field of communication, and aims to solve the problems of multidimensional resource allocation complexity, fiber core crosstalk and spectrum fragmentation in the space division multiplexing elastic optical network. According to the technical scheme, collaborative optimization mapping is carried out on virtual network requests by setting a link fiber core mapping agent and a frequency spectrum mapping agent, and the collaborative optimization mapping comprises the steps of initializing an underlying optical network topology and a virtual network topology, sorting virtual network sets according to the total bandwidth of the requests, embedding virtual links in a descending order, and executing physical node embedding and path searching. And the intelligent agent selects a link fiber core and spectrum allocation based on the network state feature matrix, and a model is trained through a reward mechanism to realize efficient resource matching. The technical effects are that the real-time performance and efficiency of virtual network mapping are obviously improved, the service blocking rate is reduced, the spectrum resource utilization rate is optimized, and the overall load capacity of the network is enhanced.

The invention provides a smart park device collaborative data fusion processing method and system, and relates to the technical field of data processing, and the method comprises the steps: S2, based on a park multi-source fusion data field, extracting node state data of four fixed space positions; based on the space-time correlation of the node state data and the space coordinate relation between the nodes, the virtual connection relation between the nodes is dynamically judged so as to construct a dynamically changing virtual collaborative awareness network topology; and a space situation field intensity model is constructed based on the dynamic connection relationship between the nodes in the network topology and the data interaction intensity, and a multi-dimensional situation field intensity value is obtained. According to the method, through a closed-loop process of data fusion, virtual network topology construction, situation field intensity analysis and acquisition strategy dynamic optimization, the cooperative efficiency and situation awareness accuracy of the heterogeneous equipment in the smart park are improved, and continuous iterative optimization of dynamic risk identification and targeted management and control is realized.

The application provides a cloud resource dynamic scheduling method and system based on virtualization technology, comprising: collecting the running state of each resource node in the cloud platform and extracting multi-dimensional time sequence features; constructing a resource topology intention perception model based on a graph neural network, identifying the load dependency relationship and potential bottleneck intention between resource nodes; taking the load dependency relationship and bottleneck intention as input, combining a reinforcement learning strategy network, and pre-generating a resource migration and elastic scaling decision sequence for a future time window before scheduling triggering; according to the decision sequence, performing hot migration or in-place elastic scaling operation on the target resource node, and updating the corresponding traffic forwarding rules in the virtual network topology; collecting the actual resource utilization rate and service quality indicators after scheduling execution, and feeding back to the reinforcement learning strategy network to optimize the feature weight of the intention perception model and the generation strategy of the scheduling decision. The application can improve the scheduling timeliness and resource utilization efficiency.

The invention relates to a method and a system for realizing integrated accurate fault positioning of an optical switch and an optical link by a DM path based on bidirectional voltage constraint, and belongs to the technical field of optical networks. Comprising the following steps: constructing a dynamic traffic hotspot according to traffic matrix decomposition and a multi-stage packet scheduling model, including a traffic accumulation stage, a traffic scheduling stage and a traffic exchange stage, so as to reduce a monitoring range; an optical switch and an optical link are mapped into the virtual network topology, and monitoring priority labels are given to nodes and edges; a DM-Trals monitoring path based on bidirectional voltage constraint is designed, dense monitoring is carried out on a high-priority area, and conventional monitoring is carried out on a low-priority area; an efficient integer linear programming model is designed, rapid link fault positioning is realized through mathematical constraints, and a virtual network topology formed by dynamic traffic hotspots is adapted at the same time. According to the invention, the monitoring scale is greatly reduced, and rapid, efficient and low-cost fault location of the optical switch and the link is realized by deploying the minimum number of monitors.

This invention provides a user-autonomous allocation device and method for access layer channel resources in optical transmission networks. A user channel switching system is constructed at the edge of the optical transmission network. This system indicates that user nodes have at least one pre-set channel with other nodes through the autonomous allocation device. The autonomous allocation device performs data acquisition and processing to obtain data information from optical transmission layer devices and the local resources available from node access devices, automatically generating virtual network topology connections. Based on user channel requirements, route selection and calculation are performed. Based on the calculation results, channel resource allocation is autonomously completed. This method offers high flexibility and timeliness, meeting the flexible service requirements of users for on-demand access and channel allocation within the existing rigid transmission channels of the optical transmission network.

The present application relates to the field of network security, in particular to a data transmission method and system for device identity anonymous protection based on route hopping, the method comprising: generating a series of virtual network topologies based on real network nodes and false nodes; evaluating the nodes in the virtual network topologies to obtain a topology security level of the virtual network topologies, and taking the virtual network topologies with a topology security level greater than a preset threshold as a hopping pool; obtaining a hopping path satisfying a preset constraint condition in the hopping pool by using a backtracking method; screening the hopping path by using an evolutionary game model to obtain an optimal hopping strategy; and transmitting data based on the optimal hopping strategy. The present application firstly constructs multiple virtual network topologies to cause attack confusion, and simultaneously generates and selects a hopping path for the virtual network topologies to obtain a data transmission path with the highest security, thereby significantly improving the defense capability against network attacks.

The invention discloses a method for realizing cross-domain transmission of subnet information based on PCEP protocol extension, and belongs to the field of computing power networks, and the method can comprise the following steps: realizing cross-domain transmission of a global IP address pool, Clico node subnet information and Peer-Node SID by extending a PCEP protocol, synchronizing an intra-cloud virtual network topology comprising network information of a VPC, a subnet and a virtual machine to a network controller, and realizing the cross-domain transmission of the subnet information based on the PCEP protocol extension. A unified network topology is formed on the network controller, so that the network controller senses the Calico subnet, end-to-end deterministic path calculation and issuing are realized, and meanwhile, the dynamic response of the subnet and the real-time coordination of subnet distribution / recovery / capacity expansion events are supported.

The embodiment of the invention provides an optical network routing and resource allocation method and device, and the method comprises the steps: employing a path planning algorithm in a physical topology of an optical network, determining a first feasible path satisfying a constraint condition between any common node and any relay node, a second feasible path satisfying the constraint condition between any two relay nodes, and a third feasible path satisfying the constraint condition between any two common nodes; based on the first feasible path, the second feasible path and the third feasible path, constructing a virtual network topology comprising virtual links between the common nodes and the relay nodes, between the two relay nodes and between the two common nodes; for a virtual link in the virtual network topology, determining a dependent link which passes through at least one same optical fiber link with the virtual link; according to the available wavelength of the virtual link, updating the available wavelength of the virtual link and the dependent link thereof; and when the service arrives, the path corresponding to the virtual link with the available wavelength is allocated for the service based on the virtual network topology, so that the optical path can be quickly constructed and the number of relays can be minimized.

The application belongs to the field of multi-fork tree virtual network construction, and discloses a multi-fork tree virtual network construction method on a single host, comprising the following steps: S1: abstract definition of a multi-fork tree virtual network node, definition of a network attribute set of the multi-fork tree virtual network node based on a logical node model; S2: standard encapsulation of atomic actions, decomposition of a multi-fork tree virtual network creation process into minimum granularity indivisible operations, i.e. atomic actions, based on a preset atomic action library; S3: topology arrangement and heterogeneous node configuration generation based on layer sequence traversal, completion of multi-fork tree virtual network topology generation by inputting an atomic action list and executing atomic actions on corresponding nodes; and S4: instantiation execution of the multi-fork tree virtual network and mounting on underlying resources, conversion of the corresponding configuration of step S3 into entity resources in a host kernel by a virtual instantiation engine. Thus, the application realizes construction of a large-scale, high-fidelity and easily reconfigurable multi-level tree-shaped virtual network.

The invention discloses a security fixed value strategy configuration method, system and device based on a simulation test technology and a storage medium, and relates to the technical field of electric power information security, and the method comprises the steps: binding a standardized security fixed value file according with a preset specification with an appointed issuing application form through configuration application operation; performing content compliance verification on the standardized security constant value file according to the auditing rule; constructing a virtual network topology environment according to a target actual network structure, configuring virtual security node network parameters, loading a standardized security fixed value file to a target virtual security node, completing simulation link setting and executing a strategy simulation test; after simulation verification is passed, interfaces are called in a differentiated mode according to equipment types to conduct strategy remote configuration issuing, and standardized security and protection constant value files are archived according to version indexes; according to the method disclosed by the invention, standardized modeling, simulation verification and classified remote issuing of security fixed value strategy configuration are realized, and an automatic strategy management mechanism which can be traced in the whole process is constructed.

The application discloses a network resource optimization method, device, system, computer equipment and computer storage medium, and the method comprises the steps of acquiring physical network information, generating a virtual network topology graph based on the physical network information; the virtual network topology graph comprises nodes, links between the nodes and weights corresponding to the links; determining the shortest path information from a source node to each node in the virtual network topology graph; determining a first time period and a second time period contained in a basic time period according to monitoring requirements and / or network traffic characteristics; monitoring the performance indicators of the first link within the basic time period, performing fault detection on the first link based on the monitored performance indicators, the first time period and the second time period, and obtaining a fault detection result; determining a virtual routing flow table according to the fault detection result, and the virtual routing flow table is used for optimizing and configuring the network equipment and / or the connection between the network equipment.

This application discloses a target range platform for network security exercises, comprising: a heterogeneous virtualization environment construction module for constructing a locally virtualized target machine environment; a traffic recording and review module; and a console module. The heterogeneous virtualization environment construction module includes a construction submodule and a routing and forwarding submodule. The construction submodule is used to construct the target machine environment using hybrid virtualization technology. The target machine environment includes several target machines and several network devices. The target machines and network devices, as well as the target machines themselves, are connected via virtual communication links, forming a virtual network topology with an undirected graph structure. The routing and forwarding submodule is also used to achieve network isolation between the target machine environment and the external network by responding to outbound routing blocking policies. This enables localized deployment of the target machine environment and isolates external network connections during network security exercises, eliminating the need for traffic backhaul and reducing the risk of data leakage.

The invention belongs to the field of multi-way tree virtual network construction, and discloses a multi-way tree virtual network construction method on a single host, which comprises the following steps: S1, abstracted definition of multi-way tree virtual network nodes: completing network attribute set definition of the multi-way tree virtual network nodes based on a logic node model; s2, standardized packaging of atomic actions: based on a preset atomic action library, decomposing a multi-way tree virtual network creation process into inseparable operations with minimum granularity, namely atomic actions; s3, topology arrangement and heterogeneous node configuration generation based on layer sequence traversal: completing multi-way tree virtual network topology generation by inputting an atomic action list and executing atomic actions on corresponding nodes; s4, instantiation execution and bottom layer resource mounting of the multi-way tree virtual network are carried out, and corresponding configuration in the step S3 is converted into entity resources in a host machine kernel through a virtual instantiation engine. Therefore, the construction of a large-scale, high-fidelity and easy-to-reconstruct multi-stage tree-shaped virtual network is realized.

The application provides a smart park equipment coordination data fusion processing method and system, relates to the technical field of data processing, and the method comprises the following steps: in step S2, based on a park multi-source fusion data field, node state data of four fixed spatial positions is extracted; based on the space-time correlation of the node state data and the spatial coordinate relationship between the nodes, the virtual connection relationship between the nodes is dynamically determined to construct a dynamically changing virtual cooperative perception network topology; and based on the dynamic connection relationship between the nodes in the network topology and the data interaction intensity, a spatial situation field strength model is constructed to obtain a multi-dimensional situation field strength value. Through the closed-loop process of data fusion, virtual network topology construction, situation field strength analysis and dynamic optimization of collection strategy, the application improves the heterogeneous equipment coordination efficiency and situation awareness accuracy of the smart park, realizes dynamic identification of risks and continuous iteration optimization of targeted control.

The invention provides an IMS network embezzlement processing method and device, equipment, a medium and a product. The method comprises the steps of generating a bait user set based on feature information of a real user, and mapping the bait user set to an independently isolated virtual slice, so that the bait user is isolated from the real user; dynamically generating a virtual network topology, wherein the virtual network topology is used for simulating an IMS network environment; acquiring multi-dimensional data of the bait user, and performing feature extraction and feature fusion on the multi-dimensional data of the bait user to obtain fusion features of the bait user; and based on a pre-constructed embezzlement identification model, identifying the fusion features of the bait user and the state information of the virtual network topology to obtain an embezzlement identification result. Through the technical scheme of the invention, the bait user can be generated to detect the IMS network stealing behavior, and the loss caused in the process of passively detecting the IMS network stealing behavior in the call service process of a real user is reduced.

A virtual machine configuration and node allocation method oriented to a routing network comprises the steps that a virtual machine memory overhead prediction model is built based on a target physical machine platform, and the prediction model is used for quantifying resource consumption under different virtual machine configuration schemes; combining the memory overhead prediction model, the virtual network topology structure and the route update load of each virtual node, performing joint optimization on a feasible virtual machine configuration scheme, and generating an optimal virtual machine configuration and node allocation scheme considering the route update competition relieving effect and the memory resource overhead; and according to the optimal scheme, automatically deploying a multi-virtual machine environment on a single physical machine, and distributing each virtual node to the corresponding virtual machine according to a distribution result indicated by the optimal scheme. According to the method, the route updating competition intensity can be effectively reduced, the single-point bottleneck is avoided, and the simulation efficiency and fidelity are improved.