73 results about "Hybrid routing" patented technology

A hybrid routing-application network fabric apparatus is presented where a fabric apparatus has multiple apparatus components or resources that can be dedicated to one or more application topologies. The apparatus can receive a topology image definition file describing an application topology and the apparatus can dedicate its local components for use with the application topology. The apparatus can dedicate general purpose processing cores, dedicated routing cores, data channels, networking ports, memory or other local resources to the application topology. Contemplated application topologies include routing topologies, computation topologies, database topologies, storage topologies, or other types of application topologies. Furthermore, application topologies can be optimized by modeling or simulating the topologies on a network fabric.

A hybrid router for dynamical control systems is described. The mobile hybrid software router (MHSR) combines distinctive computational and mathematical techniques, including evolutionary computation (EC), probabilistic simulations (PS), machine learning and artificial neural networks (A-NNs), in order to solve unique problems encountered in an unknown environment in real time. Embodied in intelligent mobile software agents (IMSAs), the MHSR operates within a multi-agent system (MAS) to continually optimize system operation. The MHSR is applied to several major complex system categories. In one embodiment of the system, the MHSR is implemented in hardware, including continuously programmable field programmable gate arrays (CP-FPGAs), for perpetually reconfigurable evolvable hardware operation. Whether in application-specific or multi-functional mode, the MHSR is useful to groups of agents in intelligent systems for adaptation to uncertain environments in order to perform self-organization capabilities.

The invention provides a space-ground integrated hybrid routing method, which comprises the following steps of: calculating a link distance between satellites in an orbital plane and establishing a time-varying bidirectional Manhattan satellite network model based on a time-varying link distance between the satellites in the orbital plane; counting queue time delay between each satellite and a neighbor satellite node and queue time delay of a satellite-ground link between each satellite and a ground node; calculating a time interval which does not affect the network routing performance, and predicting link transmission time delay and queue time delay of the neighbor node and the next neighbor node in the time interval; calculating link time delay of satellite network routing according to asatellite routing algorithm based on predicted secondary neighbors; when the source terminal communicates with the destination terminal, predicting the queue delay of each link on the ground route based on a minimum mean square error prediction mechanism; under a 5G-satellite integrated architecture, selecting an upper satellite 5G-satellite node and a lower satellite 5G-satellite node with optimal paths by taking minimization of link total time delay of space-ground integrated hybrid routing as a target, and obtaining a set of space-ground integrated hybrid routing.

The invention discloses an ad hoc networking and routing method implemented by logical area coordinates for wireless Internet of things. The method comprises the following steps: building network connection architecture; marking network connection entities with the logical area coordinates in logical areas; building a routing path of new network connection entities; and maintaining routing tables between adjacent points. In the ad hoc networking and routing method implemented by the logical area coordinates for the wireless Internet of things, node position information is marked on the basis of dynamic logical abstract coordinates relevant to application semantics rather than physical static coordinates. The method can be used in a way of being bound with a plurality of different physical address coordinate systems. Meanwhile, a hybrid routing algorithm based on the logical area coordinates proposed in the method supports routing / forwarding of a hierarchical area, and supports a sparse Mesh topology among different areas at the same time, so that a system topology with a large number of nodes is effectively supported, and the effectiveness of a forwarding path is considered at the same time.

The invention discloses a space-ground integrated network dynamic-static hybrid routing method, and aims to provide a dynamic-static integrated hybrid routing method which adapts to the characteristics of continuous dynamic change of space-ground integrated network topology, limited on-satellite load capacity of nodes, long inter-satellite link transmission time delay, unbalanced distribution of carried data flow and the like. According to the technical scheme, the method comprises the steps that a static route sends motion trail information of a satellite-based node to a ground operation andmaintenance management and control center, a routing request is initiated, and the motion trail information of the satellite-based node is obtained; the operation and maintenance management and control center performs time slice division, virtual topological graph construction and original routing table creation by adopting a time virtualization routing algorithm; and after the satellite-based node responds to the routing request and accepts the task, a task route is constructed according to the static routing table, an on-satellite load task is obtained, whether the sending of the data packetsucceeds or not is judged, if the packet sending fails, the route is dynamically reconstructed by adopting a DSR routing algorithm, and if the sending of the data packet succeeds, the transmission ofthe data packet is finished.

The invention discloses intelligent network architecture integrated with perceptual computing and storage, and an implementation method of the same. The intelligent network architecture integrated with perceptual computing and storage utilizes a model combined with a content identifier (CID), an equipment identifier (GUID) and a security identifier (SecID), and is characterized in that the content is stored in the network in a distributed manner; a content caching strategy is deployed in LCRS; a clustering strategy is deployed in a router; through mutual coordination between nodes, sharing and reuse of resources can be realized; a hybrid routing mode of device identifier / address is utilized, and NRS provides two levels of name analysis service to enable the network to support seamless host and network mobility; and an active defense security authentication mechanism is utilized, and an edge router is taken as a root of trust, with combination of the device identifier and the security identifier, access control and authentication are performed on the user, and effective interception of the illegal content is performed through combination of the content identifier. The intelligent network architecture integrated with perceptual computing and storage can percept the content of network transmission, changes of the network environment, and the behavior of the network user, can support seamless host and network mobility, and has the expandable ability and the evolution capability, and can realize active defense, thus having network security.

A hybrid routing and forwarding solution for a mobile wireless sensor network is provided. This invention allows each node of the wireless sensor network to move in random directions in a three dimensional space. The invention seeks to minimize resource consumption while optimizing routing and forwarding in a wireless sensor network.

The invention relates to a hybrid interconnection structure for a network-on-chip. The hybrid interconnection structure comprises a global network and a plurality of local networks, wherein the global network is of a binary tree structure consisting of a plurality of network nodes, and each local network is of a two-dimensional Mesh structure consisting of a plurality of network nodes, and a central node of the global network is connected with the network nodes on the global network. In the hybrid interconnection structure provided by the invention, the global network adopts the binary tree structure, each local network adopts the two-dimensional Mesh structure, and a matched network node encoding method and a hybrid routing algorithm are designed, so that data flow between global and local can be balanced, an average path distance between the local nodes can be reduced, and a network diameter is reduced, to improve the average transmission delay of the network and the network data throughput and effectively save the connecting line resources at the same time, and thus more network nodes can be added on the basis of the same connecting line resource.

The invention discloses an aeronautical ad hoc hybrid routing method based on tactical coordinated formations. The method comprises 1, the method for establishing and maintaining routings in the same formation; 2, the method of a node for sending a routing request message to a gateway when information is interacted between the formations; 3, the processing method of the node after receiving a routing response message sent by a gateway node; 4, the method of the node for maintaining the on-demand routing; 5, the method for establishing the gateway; 6, the method of for adding the node into the formation; and 7, the method for allowing the node to quit the formation. The method combines two routing driving modes including table driving and on-demand driving, the table-driven routing is used in the same tactical formation, and thus a real-time business requirement is ensured; the on-demand driving routing is used between the formations, so that average routing table entries of the node are greatly reduced, routing requests are forwarded in a purposeful manner instead of a blind manner based on the gateway node, and network flooding is reduced.

The invention belongs to the field of delay tolerant network (DTN), and specifically relates to a delay tolerant network node cache management method facing epidemic and probabilistic hybrid routing. The method comprises the following steps: whether a node in a network needs to send a message is judged firstly; if the node in the network needs to send the message, a step (2) is executed; if the node in the network does not need to send the message, a step (8) is executed; neighbor nodes of the node needing to send the message is traversed; and, if the number L of the neighbor nodes is equal to 1, a step (4) is executed, otherwise, a step (5) and the like are executed. Characteristics of an epidemic routing method and a probabilistic routing method are integrated, and the number of duplicates of the same message is effectively controlled; and, compared with the probabilistic routing method, the delay tolerant network node cache management method enables a delay of the message arriving at a target node to be reduced.

The invention discloses a low-overhead hybrid routing protocol method for a narrowband ad hoc network, which comprises the following steps: IP (Internet Protocol) head bandwidth consumption caused bynetwork layer relay is reduced by simplifying a protocol stack by adopting a link layer relay transmission mode; an active and passive hybrid routing mode is adopted; a low-overhead topology control mode is adopted, neighbor information is periodically exchanged among nodes, and information compression and single-hop broadcasting are carried out by adopting compressed encoding, so that a two-hop topology table on which a passive routing mode depends is established and maintained; a piggybacked routing updating mode is adopted, and a node carries path information to a destination node through an extension field of a Hello message and diffuses to an upstream node along a reverse path for updating a routing table required in an active routing mode by the upstream node. According to the invention, the bandwidth efficiency of narrowband ad hoc network and multi-hop transmission can be effectively improved.

The invention belongs to the technical field of ocean communication and discloses a hybrid routing method based on clustering and reinforcement learning and an ocean communication system. The method is characterized by macroscopically learning the Q value of the whole cluster on line to determine the optimal next-hop grid; locally determining a specific node in the optimal grid; selecting a next hop node; realizing a route discovery process in combination with the thought of on-demand routing of an AODV algorithm and the greedy principle of GPSR; when the node needs to send the data packet toa destination node, if the corresponding routing information exists, directly sending the data packet, and if the corresponding routing information does not exist, searching for whether the destination node is in the adjacency list or not, and performing forwarding according to a corresponding node forwarding strategy until the destination node is reached. According to the invention, the floodingof broadcast is reduced, and the routing overhead is reduced; the routing void can be effectively avoided, the selected path is more suitable for the current network state, and the packet loss rate islower.

A hybrid routing—application network fabric apparatus is presented where a fabric apparatus has multiple apparatus components or resources that can be dedicated to one or more application topologies. The apparatus can receive a topology image definition file describing an application topology and the apparatus can dedicate its local components for use with the application topology. The apparatus can dedicate general purpose processing cores, dedicated routing cores, data channels, networking ports, memory or other local resources to the application topology. Contemplated application topologies include routing topologies, computation topologies, database topologies, storage topologies, or other types of application topologies. Furthermore, application topologies can be optimized by modeling or simulating the topologies on a network fabric.

The invention discloses a satellite delay tolerant network-based hybrid routing algorithm. According to the algorithm, global optimal routing and local optimal routing are combined; a next hop satellite node with a larger utility value is selected according to encounter interval time, a link connectivity rate and a link congestion possibility; and therefore, the reliability of routing can be improved, and the arrival rate of entire network messages can be improved.

The invention discloses a vehicle networking hybrid route forwarding strategy based on named data network. In a vehicle networking environment, with multiple message types, including security messages, entertainment news, the public service message and the auxiliary driving message, taking into account geographic location information, the impact of message classification and the need for real-timemessaging on message delivery, messages are categorized into location-dependent and location-independent, each type of message contains an emergency message and a normal message, and different forwarding strategies are adopted to carry out adaptive forwarding, so as to realize fast and reliable forwarding of messages in vehicle networking environment, reduce transmission delay, reduce packet lossrate, thus providing technical guarantee for fast and reliable message forwarding in vehicle networking, and improve the forwarding efficiency of messages.

The invention provides a network-on-chip design method and device based on a hybrid routing algorithm, and the method comprises the steps: dividing nodes into a controller node, a data processing node, a data exchange node and an equipment node according to the functional characteristics of the nodes in a network-on-chip; arranging the nodes from the central position of the network-on-chip to theperiphery layer by layer according to the sequence of the controller node, the data processing node, the data exchange node and the equipment node; and taking the controller node and the data processing node as a group, taking the data exchange node and the equipment node as another group, and adopting self-adaptive routing between each group, and adopting XY routing or YX routing between the groups. Based on data transmission characteristics and arrangement modes of different types of nodes, data transmission between different nodes is realized by using different routing algorithms, and the performance of the network-on-chip and the whole chip is improved.

The invention discloses a data processing method for a multi-modal network mixed type FIB storage structure, and the method comprises the following steps: performing parallel retrieval operation on name prefixes or IP addresses with different lengths in a memory according to an LPM index mechanism, and judging whether the name prefix or the IP address prefix of a data packet exists in an FIB of a mixed router or not, so as to obtain the forwarding information of the next-hop route or output a retrieval result of Not Matched. The hybrid router can process different modal data packets in a multi-modal network, and differentiated processing is performed on the data by judging whether the packet header of the data packet is name data or IP address data, so that forwarding information is finally indexed. In the invention, the hybrid index model for realizing the above functions forms a tower structure through a BP neural network, and then forms the hybrid index model. Rapid retrieval of forwarding information is realized through the hybrid index model in the on-chip memory and the off-chip memory.

The technology disclosed herein enables network traffic routing using a hybrid routing table. In a particular embodiment, a method provides adding a first portion of a network prefix to one or more tiers of one or more tries in the hybrid routing table. Upon reaching a lowest node of the one or more tries when adding the first portion of the network prefix, the method provides adding key information to the lowest node. The key information and at least a second portion of the network prefix comprise a key to locate a value in a hash table of the hybrid routing table. The value indicates routing information for the network prefix. Additionally, the method provides routing packets using the hybrid routing table.

The invention discloses a multi-bus hybrid routing method and a gateway device. The method comprises the following steps of: receiving a first data unit from a source port in a current bus; matching afirst protocol corresponding to the current bus through a first preset rule, and analyzing the first data unit by adopting the first protocol to obtain data information in the first data unit and anID of a target port; based on the ID of the target port, matching a target bus where the target port is located through a second preset rule, the type of the target bus being different from that of the current bus; matching a second protocol corresponding to the target bus through the first preset rule, and packaging the data information by adopting the second protocol to form a second data unit;and sending the second data unit to the target port through the target bus. Routing of data unit transmission between different buses is achieved, and the data exchange and transmission efficiency isimproved.

The invention provides a three-dimensional mine hybrid routing algorithm based on a greedy thought, and belongs to a data transmission routing algorithm of a wireless sensor network of an underground coal mine. The three-dimensional mine hybrid routing algorithm comprises the following steps: at first, using the space mosaic theory to optimize a node deployment strategy, setting a deployment scheme suitable for the underground coal mine, then, optimizing a clustering routing algorithm of the network by use of the greedy thought, and using the residual energy and a broadcast factor as a weight reference of cluster head election, determining a weight function to select an optimal next hop node, and finally, forming an optimal data forwarding path to reduce the energy cost and the delay time of the network. The routing algorithm of the wireless sensor network with efficient global energy, balanced and small time delay is designed by use of the greedy thought; a triangular prism is selected according to the space mosaic theory to fill a three-dimensional space, and the three-dimensional space is converted into a two-dimensional expression when the node position is determined. Clustering is carried out by sensing the residual energy and the broadcast factor of the node, and optimal path multi-hop communication of the cluster head and a base station is achieved by the greedy thought when inter-cluster routing is established.

The invention relates to the technical field of Bluetooth networks, and discloses a multi-mesh-ring topology control method based on a Bluetooth network and a storage medium. A multi-mesh-ring scattering network comprises more than two mesh-ring sub-networks, and each mesh-ring sub-network comprises a plurality of main devices and an internal network bridge; the topology control method comprises the following steps: selecting a coordinator from a master device; calculating the number of piconets through a coordinator; establishing a multi-mesh-ring scattering network; performing a hybrid routing policy, wherein the active route is used for internal data transmission of each mesh-ring subnet, and the self route is used for data transmission between different mesh-ring subnets. According to the invention, the multi-mesh-ring scattering network can be created, the shortest path routing in the mesh-ring sub-network and the self-routing algorithm of different mesh-ring layers are executed by starting the hybrid routing strategy, so that the load balance in the network is realized, and the data transmission performance of the network is ensured.

The invention discloses a hybrid routing protocol design optimization method of a high-altitude high-speed mobile node ad hoc network. The method comprises the following steps of step 1, constructinga local routing table; 2, establishing an optimal route: if the destination node is in the local routing table, directly finding the optimal route according to the routing table; otherwise, acquiringthe optimal route by adopting an A * algorithm-based optimal route acquisition algorithm; 3, in the process of obtaining the optimal route, detecting the occurrence of a competitive link, and if the competitive link occurs, going to step 4; otherwise, determining that the route obtained in the step 2 is the optimal route; and step 4, deleting the route established in the last one or more communication processes, setting related nodes as temporary fault nodes, and restarting the route establishment process from the previous hop node of the temporary fault nodes. According to the method, the search area obtained by routing is effectively reduced, the calculated amount is reduced, the end-to-end time delay is reduced, and the optimal routing is obtained more timely.

The invention discloses a method for identifying a communication signal based on a deep hybrid routing network, and the method comprises the steps of obtaining signal data, and extracting a corresponding signal feature; designing a hybrid routing network model which takes a CNN neural network as a basic model, adding multiple routing units, combining all the units in a cross-layer connection network mode, and designing more than one hybrid routing network model; training a network model by utilizing the acquired signal data, and selecting the network model of the hybrid multi-routing unit according to a training effect; and identifying the signal data by using the obtained hybrid multi-routing network model, and finally outputting an identification result. The hybrid routing network structure provided by the invention provides a simple form for a complex routing logic network design, and the used network can generate a plurality of routing modes, so that the performance of extracting signal features is enhanced, and the training speed is higher.

The invention provides a dual-channel hybrid routing algorithm based on intelligent control and application of a water heater, comprising the following steps of: S1, receiving data by means of power line carrier and wireless communication; and S2, switching a communication mode to the power line carrier mode, the wireless communication mode or dual mode according to the received data. By adoptingthe dual-channel hybrid routing mechanism algorithm, the communication mode is switched according to the data received through the power line carrier and the wireless communication. Thus, an appropriate communication mode can be selected to improve the anti-noise capability of the communication and enhance the long-distance communication. In the complex environment of the water heater, the hybridrouting algorithm can compensate for the defects of the power line carrier and the wireless communication, and can meet the environmental requirements of the current networking work for the water heater.

The invention provides a hybrid route establishment method of a dual-mode Mesh network, relates to the technical field of communication, and mainly solves the technical problem of unbalanced loads oftwo paths of a broadband power line carrier channel and a micro-power wireless channel on a dual-mode network node. The method comprises the following steps that a network center node sends a beacon to a network child node to complete networking, and counts neighbor node information of the network child node; the network center node calculates the shortest path to each network child node and issues routing information to the network child nodes; the network child node determines a channel quality grade with the neighbor node according to the SNR value or the RSSI value and a receiving threshold; the channel path weight of the neighbor node is calculated according to the channel quality grade and the weight factor of the message; a channel path weight list of the neighbor nodes is determined, and an optimal transmission channel is determined according to the channel path weight during data transmission. Therefore, the throughput and the channel utilization rate of the network are improved, and the characteristic of coexistence of two medium channels of the dual-mode network is brought into full play.

The invention provides an implementation method of a hybrid routing protocol based on a time pulse source, belongs to the technical field of communication, and mainly solves the problem of how to balance network routing overhead and delay overhead and provide a high-capacity, low-delay and high-dynamic ad hoc network routing protocol for a system. According to the scheme, a routing protocol actively and periodically broadcasts pulse messages through pulse nodes to complete establishment and maintenance of a path tree, and when non-pulse nodes have data requirements, path establishment, maintenance and path optimization are completed along the path tree according to needs. The method has the advantages that the problems that the overhead is large and the method is not suitable for large-scale nodes due to the fact that the active routing protocol and the passive routing protocol adopt a flooding mode to carry out path discovery and path maintenance are better solved, the problem that the end-to-end delay of the passive routing protocol is large is solved, and the scheme is more suitable for the conditions that the network extension structure is rapidly changed and the network node scale is large.

An embodiment of the invention discloses a communication method, device and system in a data center, provides novel photoelectric hybrid routing data center networking architecture, provides a distributed source routing technology based on a server in the data center, and shares the load of a switch. Besides, in the data center networking architecture, optical cross equipment is provided to be interconnected with an access switch or a convergence switch, and different routing strategies are carried out for different types of data streams, so that the network transmission efficiency is improved, and the energy consumption of the data center is reduced.