18 results about "Virtual network mapping" patented technology

The invention provides a source-load-storage communication virtual network mapping method and device, and belongs to the field of communication. The method comprises the following steps: constructing a virtual network mapping model of source-load-storage communication; obtaining a region service request of a target region; wherein the regional service request comprises a plurality of virtual nodes and a plurality of virtual links, the virtual nodes comprise a starting node and a target node, and the virtual links are links between the corresponding starting node and the target node; according to the virtual network mapping model and the regional service request, establishing a target function for mapping the virtual nodes to the physical nodes by taking the minimum sum of propagation delays of all the physical nodes as a target; solving the objective function to obtain an optimal node mapping scheme; and based on the virtual network mapping model and the optimal node mapping scheme, mapping the plurality of virtual links to obtain a virtual network mapping scheme. The communication efficiency can be improved, and safe and economical operation of a power system is guaranteed.

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 delay-aware dual-time scale resource allocation method, which belongs to the technical field of virtual network mapping, and optimizes on different time scales by decomposing a delay optimization task into two levels, namely a short time scale and a long time scale. Within a short time scale, a deep reinforcement learning algorithm is used to output a mapping decision in time, network changes are quickly responded, and resource allocation is carried out in real time; and on a long-time scale, long-term network load and resource requirements are optimized through a remapping mechanism, and balance and stability of global resource allocation are ensured.

The present invention discloses a multi-threaded virtual network mapping and performance adjustment method based on entropy weight method, which belongs to the field of network mapping, including: setting differentiated service quality relaxation coefficients for virtual network requests of multiple levels, taking QoS indicators into account in virtual network requests, and providing different levels of service quality assurance for VNRs with different QoS requirements, thereby achieving the service quality of VNRs on the basis of ensuring the request success rate. Multiple levels of virtual network requests are input into a trained node mapping strategy evaluation network for node and link mapping, and preliminary mapping results of virtual network requests of multiple levels are obtained. The preliminary mapping results are analyzed by entropy weight method to achieve dynamic adjustment of service acceptance rate, resource utilization rate and QoS loss rate. The multi-threaded asynchronous dominant actor critic algorithm is used to train the network, and the model is used to complete the mapping, which optimizes the link mapping while optimizing the node mapping.

The invention relates to multi-domain virtual network security mapping for cross matching and optical path separation fault tolerance, and belongs to the technical field of optical fiber communication. According to the method, a double-layer block chain network architecture is adopted, a main chain is used for inter-domain resource collaborative scheduling and information synchronization of an optical network, and a secondary chain realizes intra-domain resource autonomous management and trust maintenance of the optical network; in the virtual node mapping stage, mapping preference values between virtual nodes and physical nodes are introduced to construct a mapping bigraph, and virtual node mapping is optimized through a cross matching mechanism; in the virtual link mapping stage, the security and reliability of quantum key distribution are guaranteed by adopting multi-path routing and a security redundant path, and an encrypted data transmission optical path based on minimum spectrum consumption is adopted; punishing malicious nodes according to a virtual network mapping result, and tracing and updating credibility values of non-malicious nodes participating in virtual network mapping through block information. According to the method, the security and the resource utilization rate of virtual network mapping in the optical network multi-domain environment can be improved.

The present application provides a virtual network mapping method, apparatus, and device based on an unmanned aerial vehicle (UAV) cluster scenario. When a virtual network embedding request is received, a topological graph sequence of the UAV cluster network is obtained; spatio-temporal dynamic features between each topological graph in the topological graph sequence are extracted based on a pre-constructed feature extraction network; the spatio-temporal dynamic features are input into a pre-trained node policy generation network to generate a node mapping policy; the virtual nodes to be embedded are mapped to the UAV nodes of the UAV cluster network based on the node mapping policy; after all the virtual nodes are successfully mapped, the virtual links to be embedded are mapped to the shortest physical links. The embodiments of the present invention can meet the virtual mapping requirements in a network dynamic scenario; and a two-stage mapping method is adopted to map the virtual nodes and virtual links in sequence, which can ensure that both the nodes and the links are successfully mapped and improve the mapping success rate.

The invention relates to multi-domain virtual network security mapping based on key balance and minimum consumption, and belongs to the technical field of optical fiber communication. According to the method, a quantum key distribution network and a block chain are integrated into a multi-domain elastic optical network, a block chain assisted multi-domain virtual network security mapping architecture is constructed, and security coordination management of multi-domain resources is realized. In the virtual node mapping process, multi-dimensional resource attributes of nodes are considered, and a node importance evaluation formula is designed to optimize virtual node mapping; the virtual link mapping process is divided into two stages: in the quantum key distribution path mapping stage, a key weight formula selection path is designed by adopting a quantum key distribution path mapping method based on key balance; in an encrypted data transmission path mapping stage, a routing spectrum allocation method for minimizing spectrum consumption is adopted, and a spectrum consumption calculation formula is designed to measure spectrum resource consumption. The method can reduce the virtual network mapping request blocking rate and improve the key utilization rate.

This invention discloses a virtual network mapping method based on deep reinforcement learning. This method characterizes the physical network and virtual network as weighted undirected graphs, encodes them into a network topology vector using a graph neural convolutional network, and then inputs the network topology vector and the network resource state vector into a policy neural network in a scheduler, which is trained using a proximal policy optimization algorithm. Compared with existing technologies, this method dynamically adjusts scheduling decisions based on the current environmental state, coordinates node mapping and link mapping, improves the mapping benefit-cost ratio while improving physical network resource utilization, and reduces physical network cluster energy consumption. The model is simple and easy to train, significantly improving user experience and providing technical support for the field of virtual network mapping.

The application discloses a method, device and electronic equipment for virtual network mapping. The method comprises the following steps: receiving a virtual network mapping request for virtual network resource deployment, wherein the virtual network mapping request comprises resource requirement information of a virtual network; determining first physical resources matched with the resource requirement information from physical resources of a physical network; determining mapping costs corresponding to the first physical resources, wherein the mapping costs are used for quantitatively representing costs of mapping the virtual network to the first physical resources; determining second physical resources from the first physical resources according to the mapping costs, taking the second physical resources as target mapping points of the virtual network; and mapping the virtual network to the target mapping points. The application solves the technical problem of low resource utilization rate of an underlying device in a virtual network mapping request processing process.

The present invention proposes a satellite virtual network mapping method based on deep reinforcement learning to solve the technical problem that the existing heuristic mapping methods for virtual network requests are difficult to adapt to the dynamics of satellite networks. The steps of the present invention are as follows: both the underlying satellite network and virtual network requests are modeled as undirected graphs. For the incoming virtual network requests, a sliding window is used to divide the mapping area of the entire satellite network, and the mapping area with the smallest load factor is selected; node mapping is performed through the DRL method; if the node mapping is successful, the link mapping process is entered; if the link mapping is successful, the reward value is calculated using the reward function; it is judged whether it is time for batch update rounds, and the loss function is calculated through the reward value to update the network parameters and gradients of the DRL agent. The present invention continuously updates the network through reinforcement learning, selects appropriate satellite nodes and inter-satellite links to be allocated to different virtual network requests, thereby improving the acceptance rate of virtual network requests and the utilization rate of network resources, and reducing the average delay.

The invention discloses an intelligent efficient virtual network mapping algorithm based on an attention mechanism, and relates to the technical field of network virtualization, and the algorithm comprises the following steps: extracting feature vectors of physical nodes by using a GCN model, and enhancing node feature representation by aggregating neighbor node features; a Transfomer architecture is adopted to form an encoder system, a GRU model forms a decoder system, and an encoder model comprises an embedded layer, a multi-head attention layer, a connection layer, a feedforward network layer, a connection layer and a linear transformation layer; according to the method, a Transfomer model is used for forming an encoder, more comprehensive feature vectors are generated for all physical nodes through multiple composition structures in the Transfomer model, in the GRU decoding process, in combination with the attention mechanism thought, attention weight is calculated, node mapping is selected, in the link mapping stage, link pruning is adopted firstly, links which do not meet the requirement are removed, and then a GRU decoding process is carried out. And then link mapping is carried out, so that the success rate of link mapping is improved.

The present invention discloses a method for integrating measurement, operation and control resources of low-orbit satellites, which relates to the technical field of measurement, operation and control resource integration. The present invention adopts a spatiotemporal attention module to comprehensively process continuous network status snapshots of low-orbit satellites, aerial platforms and ground stations, integrates data information of multiple time points, and forms a spatiotemporal feature representation of node status, so that mapping decisions can refer to network changes in continuous time slots; through the cooperation of a graph attention network and a self-attention mechanism, the problem of invalid mapping results of a single time snapshot is solved. In virtual network mapping, a two-stage mapping scheme is adopted; in the first stage, a graph convolutional network is used to map nodes in a virtual network request to physical nodes, and the normalized adjacency matrix and weight matrix are fully utilized to fuse neighborhood information; in the second stage, the link path between candidate nodes is calculated by the Floyd algorithm, and the node mapping results and the link mapping are uniformly planned.

The cloud network system and the interaction method of the cloud network system provided by the application belong to the technical field of communication network. The cloud network system comprises a cloud virtual machine, an Internet data center and a controller. The cloud virtual machine comprises a layer 2 gateway, a virtual gateway and a point-of-presence switch. The Internet data center comprises a tunnel switch, a router and a virtual network switch. The controller is used for mapping the virtual network of the Internet data center to a virtual network tunnel, so as to allocate tunnel configurations for each virtual network tunnel in the cloud network system. A layer 2 virtual network tunnel is constructed between the tunnel switch and the layer 2 gateway based on the tunnel configurations. The layer 2 virtual network tunnel comprises an inner layer virtual network tunnel between the tunnel switch and the layer 2 gateway, an outer layer virtual network tunnel between the layer 2 gateway and the virtual gateway, an outer layer virtual network tunnel between the point-of-presence switch and the virtual gateway, and a dedicated line virtual network tunnel between the router and the point-of-presence switch.

The application belongs to the field of virtual network mapping of space-ground integration, discloses a virtual network mapping method based on request pre-segmentation in a space-ground integrated network, designs a hierarchical mapping architecture based on a space-time resource tree, and completes the mapping process of the request based on the same; when the request arrives, if the resource amount of the sub-domain where the request access point is located cannot meet the required resource of the request, an improved body mapping label propagation algorithm is used to extract the high-order characteristics of the request, the low-order characteristics are combined for label propagation, so that the request is pre-segmented, and a suitable sub-domain for the pre-segmentation result is found for mapping. The application realizes the space controllability of the request mapping result, and can also reduce the remapping of the mapping result along the time dimension as much as possible, improve the resource utilization rate and the mapping success rate.

There is provided a network apparatus comprising: means for storing a mapping, the mapping comprising an indication of a first virtual network and at least one other virtual network configured to provide a service at a lower quality than the first virtual network; means for receiving, from another network apparatus, a request to provide a service using the first virtual network; and means for using the mapping and the indication of the first virtual network to select a virtual network to provide the requested service.

The invention discloses a low-earth-orbit satellite measurement, operation and control resource fusion method, and relates to the technical field of measurement, operation and control resource fusion. A space-time attention module is adopted to comprehensively process continuous network state snapshots of a low-earth-orbit satellite, an air platform and a ground station, and data information of multiple time points is fused into a whole; the spatial-temporal characteristic representation of the node state is formed, so that the mapping decision can refer to the network change in the continuous time slot; through cooperation of a graph attention network and a self-attention mechanism, the problem of failure of a single-time snapshot mapping result is solved, and a double-stage mapping scheme is adopted in virtual network mapping; in the first stage, nodes in a virtual network request are mapped to physical nodes by utilizing a graph convolutional network, and neighborhood information is fused by fully utilizing a normalized adjacency matrix and a weight matrix; and in the second stage, a link path between candidate nodes is calculated through a Floyd algorithm, and a node mapping result and link mapping are uniformly planned.

This invention discloses a cloud-edge-device virtualized zero-configuration network interconnection method and system for heterogeneous links, relating to the field of cloud-edge-device network interconnection technology. The method includes the following steps: acquiring state parameters and traffic characteristics of heterogeneous links between cloud-edge-device nodes and constructing a first link set; sending probe payloads to the first link set and performing monitoring and identification to obtain an address change set, and performing network topology migration based on the address change set to obtain a virtual network mapping set; identifying the initial scope of the address change set based on the virtual network mapping set to obtain an address disturbance set; performing adjacency expansion on the address disturbance set within a preset network topology connectivity domain to obtain an address update domain set; generating virtual network configuration instructions and distributing the virtual network configuration instructions to the corresponding cloud-edge-device virtual network elements. This solves the problem that in a heterogeneous link environment between cloud-edge-device nodes, the lack of real-time and accurate identification of address changes makes it difficult to automatically restore virtual network connectivity.

The present invention relates to a virtual network mapping method based on node delay perception, including: obtaining virtual node sorting values according to a node sorting algorithm, and sorting virtual nodes based on the virtual node sorting values; introducing the physical node delay importance into the node sorting algorithm, weighting the deployment delay of physical nodes, node topology attributes, and global network resources, and then sorting physical nodes through a physical node sorting algorithm based on delay perception to obtain physical node sorting values; performing virtual node mapping according to the virtual node sorting values and physical node sorting values; and using the K - shortest path algorithm to find K shortest physical paths between the two physical nodes according to the physical nodes mapped by the end nodes of the virtual link. The present invention reduces the deployment delay of the virtual network and improves the virtual network mapping revenue - expenditure ratio on the premise of ensuring the mapping success rate of the virtual network, achieving the purpose of balancing the performance and delay of the virtual network mapping algorithm.