92 results about "Dynamic network topology" patented technology

The invention is directed towards routing a packet using both IPSec and common routing protocols within dynamic network topologies in a VPN. The routing of IPSec packets employs Open System Interconnection (OSI) layer three information. In one embodiment, a tree mechanism is used for looking up layer three information that may be associated with a protected subnetwork. When a packet is identified as being associated with a protected subnetwork, the packet may be encrypted and encapsulated, including the original destination and source IP address header information within another packet employing the IP Encapsulating Security Payload (ESP) protocol. New source and destination IP addresses are provided for the new packet using IP addresses associated with an entry gateway and an exit gateway to the VPN. The new packet may then be routed through the VPN using traditional routing protocols.

The present invention provides a method and system for establishing a highly mobile, long range secure wireless network with dynamic topologies and near full connectivity with acceptable latency using low cost, low power, compact and lightweight devices. One aspect of the system deals with a highly mobile network with dynamic network topologies and a time varying wireless medium that has neither absolute nor readily observable boundaries outside of which radio nodes are known to be unable to receive network frames, although the desirable open field boundary is 1 mile in radius from a base station node. A synchronous frequency hopping technique is used with mobile nodes that can become slave base station nodes to a master base station node to increase the effective range of the master base station without increasing the transmit power. Furthermore, the use of adjustable sleep times for the mobile nodes, as well as a novel clock calibration method, provides a substantial range increase with acceptable battery size and system latency.

Systems and methods are disclosed for the operation of a network of cognitive radios in a dynamic network topology. The systems and methods make use cross-layer communications to input parameters required by network services into a reasoning engine, which in turns processes the parameter the parameter information and reports linguistic results to the network services. The cross-layer communications are leveraged to shift network service functionality to the MAC or PHY Layers. Route discovery functionality is shifted from routing services down to the MAC or PHY Layers. The route discovery functionality is implemented by a process to elect a relay node based on the number of available channels that each of the source node neighbors has available to communicate with other neighbors.

The invention provides a node connected island merge based network topology analysis method. In a network topology map, a connected region formed by a node i and adjacent nodes j and k thereof is called as a connected island of the node I, and a minimum node number of the nodes i, j and k included in the island serves as an island number of the connected island. A two-dimensional array data storage structure capable of storing part of intermediate operation results is adopted, so that the method is adaptable to ring network type and tree network connectivity analysis, and is particularly suitable for plant station configuration parallel topology analysis. When the topology analysis method uses a computer multi-core parallel processing technology for achieving plant station bus configuration topology analysis, data collision due to the fact that multiple processors read and write a same node connected island number simultaneously is avoided, topology analysis operational speed is increased, parallel processing efficiency is improved, and operational speed is high while real-time performance is good.

The invention discloses a method for establishing a dynamic network model using a deep convolutional neural network. On the basis of properties of the time sequence and sociality of node movement in a dynamic network, a method for extracting network structure features by using a convolution neural network based on digging of a relationship between network topology and a link state. According to the method disclosed by the invention, slicing up is carried out on dynamic time sequence data based on a chaotic time series theory; segmented network topology information is expressed by a state matrix; the state matrix is transformed into an observation map; and then high-order features that can affect the link change in the observation map are extracted by using a deep convolution neural network. Therefore, with full consideration of a complicated relationship between nodes in the dynamic network, a rule of a network structure with a time change is learned from lots of time sequence data by using a deep learning model, so that the potential mode under the dynamic network topological change is extracted effectively; and the certain support is provided for related studies like the structure evolution analysis of the dynamic network.

The invention relates to a clustering time synchronization method combining topological maintenance. The method comprises the steps that in the election stage, the system selects a node of the maximum connectivity to act as a network root node; in the clustering stage, network clustering is completed with the network root node acting as the start under the scene of network event asynchronization and equipping a directional antenna, and network initial synchronization is also completed; in the synchronization stage, the synchronization effect is stabilized through intra-cluster and inter-cluster synchronization; and in the maintenance stage, maintenance strategy analysis is performed on mobile network accessing and network withdrawing of the node. Compared with the methods in the prior art, the clustering time synchronization method has excellent synchronization performance and can greatly solve the problem of clustering synchronization based on a beam forming antenna, the synchronization efficiency can be effectively enhanced by adopting an inter-cluster synchronization priority policy, and the influence caused by dynamic network topology can be greatly adopted by combining topological maintenance.

The invention discloses a distributed multi-robot dynamic network connectivity control method, and belongs to the technical field of multi-robot dynamic network control. The method comprises minimum spanning tree distributed dynamic network topology control based on event drive and optimal neighbor communication link connectivity maintaining motion control based on a local artificial potential field. Each robot node periodically broadcasts the location thereof and updates neighbor information, the local minimum spanning trees of the neighbor networks of the nodes are established in a distributed manner, an optimal neighbor is selected, and the communication link with the optimal neighbor is maintained through a local artificial potential field method, so that global information connectivity is optimized and optimal network communication cost is achieved on the basis of maintaining the connectivity of the multi-robot dynamic network. There is no need for a center node to perform calculation in a centralized way, and distributed and parallel implementation can be carried out, so that the communication load is reduced, and the scalability and robustness of the multi-robot network are enhanced. The problem of communication link interrupt caused by robot motion can be overcome.

The invention discloses a multi-unmanned-aerial-vehicle network cooperative communication method. The multi-unmanned-aerial-vehicle network cooperative communication method comprises the steps that amaster unmanned aerial vehicle and slave unmanned aerial vehicles start monitoring after flying to a preset flight track according to the preset flight track to collect fire disaster data; according to the distributed link flow, link power and information transmission rate, the slave unmanned aerial vehicles summarize and send the collected data to the master unmanned aerial vehicle in an ad-hoc method; and the master unmanned aerial vehicle sends the collected data to a ground receiving station. According to the communication requirements in fire monitoring, in the unmanned aerial vehicle network with dynamic network topology changes, the communication capability of the network is improved by jointly considering the data acquisition rate control, channel transmission rate arrangement andresource allocation of the unmanned aerial vehicles. The joint information source and channel rate control scheme in the unmanned aerial vehicle fire monitoring network provided by the multi-unmanned-aerial-vehicle network cooperative communication method effectively improves the communication performance of the network, so that technical guarantee is provided in the communication field for the network to be applied to actual fire monitoring, the communication capability of the network is ensured, and the multi-unmanned-aerial-vehicle network cooperative communication method has a great significance for fire rescue.

The invention provides a road network congestion area dynamic identification method and system based on community discovery. The road network congestion area dynamic identification method comprises the steps that a dynamic network topology model is built for an urban road network; congestion source points in the road network are searched; diffusion of a congestion area is described with the minimum congestion heterogeneity as a target, and community structures are extracted; the superimposed community structures in the road network are combined; and the boundary range of the congestion area isdetermined finally, and dynamic identification for the congestion area under different congestion levels is achieved. According to the road network congestion area dynamic identification method and system based on community discovery, rich traffic state information provided by a bayonet data source can be effectively utilized to excavate the traffic congestion time and spatial variation law, namely the whole process of congestion occurring, diffusion and dispersing so as to achieve dynamic extracting within the range of the congestion area under the different congestion levels.

The invention discloses an industrial heterogeneous network information interaction method based on software definition. The industrial heterogeneous network information interaction method comprises the following steps: 1, building a multifunctional composite production line integrating USB flash disk machining, carving pendant machining, product packaging and the like according to the requirementfor small-batch multi-variety personalized customization; 2, designing and building a multi-protocol data gateway, and standardizing data generated by equipment to facilitate communication between the equipment and between the equipment and an upper computer; 3, establishing a software-defined network architecture, and performing initialization configuration of a data transmission flow table on the switch; step 4, formulating a data flow grading and network service quality optimization mechanism, realizing dynamic network topology structure optimization by a controller, and realizing load balancing of data flow; and step 5, realizing data interaction between the industrial Internet of Things and the cloud through the private cloud, uploading the computing task which cannot be processed locally to the cloud, and acquiring the user customized information through the client of the cloud.

The invention provides a dynamic link predication method of a network structure. The method comprises the following steps: step one, inputting a network structure corresponding to a service object; step two, performing Jaccard distance conversion on the input network structure to obtain a processed network structure; step three, calculating the distance between every two nodes in the network structure; step four, obtaining a network structure with a marked priority at current time; step five, repeatedly executing the step one to the step four at next time to obtain a network structure with a marked priority at the next time, wherein the priority of each link at the next time is postponed to the priority at the current time, and the priorities of the links are successively marked from high to low according to time; and step fix, taking a network structure with a marked priority at each time as a predication result of one network structure for a user to perform analysis processing on the service object. According to the invention, based on a dynamic network topology structure, a dynamic evolution mechanism of a complex network is taken into consideration, the calculation complexity is quite low, and the method provided by the invention is applied to link prediction of a large-scale network.

The invention discloses a network coding control method for a multi-hop-multi-data stream network, belonging to the technical field of network coding. The method of the present invention improves the network coding conditions of the multi-hop-multi-data stream network, so that the data streams after network coding can be decoded by nodes within the downstream limited hop range to recover the data streams before respective coding as network coding Conditions, instead of the traditional condition that each data stream participating in this network coding can obtain the original data stream through network decoding in its downstream data nodes as the network coding condition. Therefore, the defect of high computational complexity in the prior art is overcome; and because only the local network topology is considered, when the network conditions outside the local network topology change, it does not affect the currently considered network coding conditions, Therefore, it is suitable for networks with dynamic topology. The invention also discloses a data transmission method of the multi-hop-multi-data flow wireless network.

The embodiment of the invention provides a method and a device for interaction between a user interface (UI) and model layer data in an application program. The method comprises: a UI thread receives message disposition notification sent from a worker thread; the UI thread acquires message data according to the message disposition notification; and the UI thread updates the UI by use of the message data. Through the adoption of the method and the device provided by the embodiment of the invention, the UI thread can update the interface when a data model changes, without using a cross-thread complex synchronization mechanism to realize interface update. When the model layer data changes, the change condition can be forwarded to the UI thread for interface update through an internally maintained message disposition queue, so that not only is logic realization simplified, but also the UI is decoupled from a data model layer, and a scalable and more dynamic network topology can be adapted more easily accordingly.

The embodiment of the invention discloses a control method, a device and a communication system of a direct tunnel. The embodiment of the invention adopts obtained network topology information and then determines whether the direct tunnel is put into service according to the network topology information, for instance, when the connection condition between RN and GGSN is determined to be normal, the direct tunnel is put into service. As the scheme uses relative factors of establishing the direct tunnel, such as the dynamic network topology information as the condition for putting the direct tunnel into service, compared with the prior art, the method can more flexibly and effectively control whether the direct tunnel is put into service.

The invention discloses an airborne passenger cabin core system of a dynamic daisy chain looped network and a dynamic positioning method. The system includes a passenger cabin interface unit and a plurality of passenger control units, and the passenger cabin interface unit and the passenger control units are mutually connected in daisy chain looped network mode. The passenger cabin interface unitand each passenger control unit obtain equipment information, the obtained equipment information is provided to an external network management system, thereby dynamically generating a network topological graph of connection of the passenger cabin interface unit and the passenger control units, and a mapping table of the passenger control units and seat positions is calculated according to the topological graph, thereby realizing dynamic positioning. The system provided by the invention adopts a looped network architecture, network wiring greatly shrinks, and system reliability is improved. Inaddition, the dynamic positioning method adopts a dynamic network topological graph mode to realize positioning of the passenger control units and seats, thus all wiring is the same, and the workloadof maintenance personnel is greatly lightened.

The invention discloses a device for dynamically identifying the network equipment topology of an intelligent substation based on MAC address matching. The device comprises a dynamic network topology identifier; the dynamic network topology identifier comprises a dynamic network topology configurator, a dynamic network topology analyzer, a dynamic network topology graphical device and a dynamic network topology updater; the dynamic network topology identifier is connected with a control hall central host of the substation; a control hall multi-media display screen displays the topological state of the whole substation network system; the control hall central host is connected with a mobile packet wireless communication device through a control hall wireless communication device; and real-time dynamic graphical query of the network equipment topology can be realized through a mobile packet. By means of the device disclosed by the invention, the topological graph of an online switch and the states of various ports in an intelligent substation network can be monitored intuitively in real time; the topological relationship of equipment can be queried on the scene in real time through the mobile packet; therefore, remote judgement and prediction on equipment failure can be realized; and the efficiency is increased.

The invention discloses a distribution automation master station system based on cloud service. The system comprises a production control large area module which comprises a plurality of sub distribution master station systems which are used for managing a distribution circuit and terminal equipment of the station, obtaining the network topology structure data of the station and the geographic information of the terminal equipment, and collecting operation data of the terminal equipment; and a management information large area module which is erected at the cloud and is used for synchronizingthe network topology structure data, the geographic information and the operation data sent by the sub distribution master stations based on a message queue service mechanism, constructing a distribution network connection relationship, carrying out real-time dynamic network topology analysis, monitoring distribution data in real time, monitoring security constraints and carrying out load transfer. Data of the sub distribution master station system is received through the management information large area, the monitoring management capability of the distribution network is enhanced, the powersupply reliability and the electric energy quality are improved, equipment operation is guaranteed, and the distribution automation benefit is improved.

The invention discloses a passive dynamic network topology chromatography method based on network coding, which includes the following steps: intermediate nodes of a network select local random coding vectors respectively in a common limited domain, and the local random coding vectors are collected into a code book through an information source to be spread to information sink nodes; the information sink nodes utilize error correction codes to calculate candidate initialization vector collection of the whole network topology according to received information matrixes; and the intermediate nodes of the network utilize the code book and the candidate initialization vector collection to calculate and judge connectivity of an upstream link until the information source layer by layer to obtain a network topology structure. The method can achieve passive network topology chromatography without additional increase of network throughput overhead, furthermore can detect dynamic variation of the network topology, and achieves the topology chromatography of dynamic network environment.

The invention discloses a topological perception routing method for unmanned aerial vehicle cluster formation, and the method divides a formation flight process into two stages: a formation keeping stage and a formation reconstruction stage based on the characteristics of the unmanned aerial vehicle cluster formation. In the formation keeping stage, the network topology configuration is relativelystable; in the formation reconstruction stage, the network topology relatively changes frequently. An unmanned aerial vehicle formation control strategy and unmanned aerial vehicle formation topologychange are combined, and when unmanned aerial vehicle nodes execute a formation reconstruction task, the nodes actively change the sending period of HELLO detection frames, so that neighbor nodes ofthe nodes can quickly sense the topology change. The method can effectively adapt to the characteristic of high dynamic network topology of the unmanned aerial vehicle cluster formation, dynamically changes the sensing frequency, reduces the network overhead in the flight process of the unmanned aerial vehicle formation, and can be applied to the communication network of the unmanned aerial vehicle formation.

The invention provides an unmanned system-oriented starling-group-type intelligent group dynamic network topology construction method. Network topology construction is carried out to realize agent group cooperation, and initialization of an agent group and corresponding parameters is included; a network status transition mechanism is set, corresponding network state operation is judged and executed for a current frame according to physical locations of the agent group and obstacles, a labeled small-world network is constructed if no obstacle exists, a 6-neighbor network is constructed if the group is about to approach an obstacle influence area, and network topology splitting is carried out if the group is located in the obstacle influence area; and next moment speed and location updatingof each agent in the agent group is carried out until the agent group reaches a destination or a maximum iteration number is reached. Agent group system synchronization capability of the method can behigher than that of traditional regular networks, random networks and the like, and the method has scalability and stability, and can have a very broad application prospect in the aspect of unmannedsystems.

The invention provides an SDN-based satellite network management and control protocol semi-physical test device, which constructs a test device oriented to an SDN satellite network scene and used forwhole system simulation from user application to data physical transmission, and can provide an optimized network management and control scheme for a network decision maker. According to the device, an SDN centralized hierarchical management and control architecture is adopted, and a self-designed satellite network simulation system hardware deployment scheme and network management and control software are applied. The method has a reasonable human-computer interface, wherein a satellite transmission simulation system is constructed by using the SDN physical switch, and the designed virtual topology mechanism can quickly generate the satellite dynamic network topology in the wired network, so that the method has the advantages of being friendly to visualization, real and reliable in simulation result, quick in network scene construction and the like.

The invention provides an unequal protection erasure coding method for distributed dynamic network topology of space communication. The session transmission process comprises two stages. At the first stage, information source S1 and S2, original information m1 with the length of k1 and original information m2 with the length of k2 generate C1' and C2' through fountain coding, and C1' and C2' are sent to D. At the second stage, information source S1 and S2 and original information m1 and m2 generate C1 and C2 through fountain coding, and C1 and C2 are sent to R; R carries out the network coding rule P of a relay node on received C1 and C2; and C1 and C2 are synthesized to CR, and CR is sent to D. According to the invention, the transmission efficiency of a system is improved, and a flexible unequal protection scheme is realized.

The invention discloses a stochastic gradient descent based spectral hashing method in distributed storage. According to the method, the algorithm training time is shortened by utilizing stochastic gradient descent based on a spectral hashing algorithm with semantic consistency, a consistent hashing algorithm based on Cauchy distribution is further proposed, and each data item is compressed into a one-dimensional real-value by utilizing the algorithm. Therefore, the distributed storage can be realized in dynamic network topology by utilizing a consistent hashing thought, and similar data items can be stored in same or similar storage server nodes. The method enables each storage server node to only need to maintain information of a small amount of neighboring nodes; and when the server nodes are added into or exit from a system, only a small amount of related nodes participate in topology maintenance, so that the convergence speed is increased and the storage accuracy is improved.

The invention discloses an unmanned aerial vehicle group bandwidth resource allocation method under high dynamic network topology, and belongs to the technical field of wireless communication. The method comprises the following steps: establishing a user base station matching preference list based on the instantaneous QoS quality of a link and the time delay requirement of a service; executing a fine matching process, and sequentially utilizing a many-to-one matching algorithm and a dynamic many-to-many matching algorithm based on a three-layer auction framework to perform unmanned aerial vehicle link matching for the user; and executing a simple matching process based on a fine matching result, and dynamically adjusting bandwidth allocation of the unmanned aerial vehicle group. According to the invention, the fine matching and simple matching processes can be executed circularly, and bandwidth resource allocation of the unmanned aerial vehicle group is carried out. According to the invention, efficient on-demand allocation of bandwidth resources between unmanned aerial vehicle base stations and flexible conversion of a user single-connection / multi-connection mode can be realized, and the influence of an unstable non-line-of-sight link and user movement in a network on the stability of a network topology structure can be overcome. According to the method, the system throughput is improved, and the calculation complexity is reduced.

The invention discloses an NFV resources scheduling method, device and system. The NFV resources scheduling method comprises the following steps: acquiring original request data and dynamic network topology information; obtaining service request features according to the original request data; obtaining topological features according to the dynamic network topological information; fusing the service request features and the topological features to form fused features; obtaining a trained resources allocation scheduling model; and inputting the fusion features into a resources allocation scheduling model so as to obtain scheduling processing action information. The NFV resources scheduling method has the following advantages: in an NFV resources allocation problem, the three sub-problems are solved in a coordinated scheme for the first time, and a new thought is provided for NFV-RA problem research. In the NFV-RA problem, the graph convolutional neural network is used for the first time and is improved, feature extraction is carried out on node and link states at the same time, and topology changes are effectively sensed.

The invention discloses a dynamic network topological graph generation method based on logs and graphs, and the method comprises the steps: S1, data collection: collecting the traffic of each server, ssh, ftp and web system logs in a streaming manner; S2, data analysis: analyzing feature information of each server from each type of logs collected in the step S1; S3, graph creating: constructing a directed graph by using the feature information analyzed in the step S2; wherein a connection initiator points to an accessed server in the direction of the edge of the directed graph; and S4, network topology state querying: querying the network topology state by using the directed graph created in the step S3. ,According to logs such as connection and flow between the servers, the relation between the servers is analyzed, the servers with the relation are inevitably in the active state of data exchange, the graph is created after the connection relation is analyzed, network changes can be found at any time, and by utilizing the concept of the graph and related algorithms, network active condition display in any time period and analysis of intranet user behaviors can be carried out according to the edge.

The invention discloses an intelligent substation network topology step-by-step sniffing method based on an LLDP protocol. The LLDP protocol is expanded, a switch transmits the detected primary neighbor topology information to the primary neighbor and receives neighbor topology information diffused by the primary neighbor so as to update the local network topology which can be detected by itself and then continues to diffuse the local network topology detected by itself to the primary neighbor through the extended LLDP protocol and the rest can be done as the same manner until the obtained primary neighbor topology information is not updated any more. Finally, each switch in the intelligent substation network can obtain the network topology information of the whole station and the positionof the switch in the intelligent substation network. The method provides an effective operation and maintenance means for dynamic network topology detection of the intelligent substation, network fault positioning by taking the switch as an interface and the like.