2336 results about "Router" patented technology

The invention discloses a multi-door-lock cooperative processing method and system, and the method comprises the steps: generating an assistance request IGMP multicast message according to a preset triggering operation detected by a first door lock B and read RFID identification information, and transmitting the assistance request IGMP multicast message to a router; according to the assistance request IGMP multicast message, recording interface information and validity period associated with the RFID identification information in a multicast forwarding table entry, and caching the multicast message; according to the preset triggering operation detected by the second door lock A and the same read RFID identification information, an assisted request multicast message containing the RFID identification information is generated, and the assisted request multicast message is sent to the router; according to the RFID identification information and the multicast forwarding table item, after the validity period is verified to be not expired, the message is directionally forwarded to the first door lock B, the first door lock B sends an assistance confirmation message to the second door lock A, and a robot calling instruction is triggered. According to the embodiment of the invention, low-delay and high-reliability door lock cooperative control can be realized.

The systems and methods disclosed herein orchestrate task execution among autonomous (or semi-autonomous) AI agentic models (“agents”) using a gateway router that dynamically coordinates the agents based on prompt characteristics, user context, and / or real-time operational factors. Received inputs (e.g., prompts) are segmented into subcomponents (e.g., sub-queries), which are routed / mapped to candidate agents based on the output parameters of the subcomponent (e.g., performance thresholds, cost thresholds) and operational parameters (e.g., cost, performance metric values, user access restrictions, timing restrictions) of each agent. The gateway router maintains dynamic routing data structures for each agent that are continuously updated based on environmental stimuli (e.g., geo-political stimuli, sensor stimuli, agent stimuli). For example, the gateway router causes agents to dynamically switch between rule engines identified by the routing tables in response to detecting environmental stimuli. Responses from the candidate agents are aggregated into an output that is responsive to the input.

The invention relates to the technical field of computer network communication, and relates to a reconfigurable router architecture based on buffer sharing and a dynamic routing method. The router architecture comprises a demultiplexer used for distributing data packets to centralized buffer areas of other ports according to port load states; the virtual channel distribution module is configured to determine an output port of the data packet according to a routing algorithm and generate a corresponding virtual channel ID; the centralized buffer area supports multi-port sharing and is divided into a plurality of logic sub-channels according to virtual channel IDs, and each sub-channel manages a data packet queue through a head pointer and a tail pointer; the credit management module is used for monitoring the idle capacity of the centralized buffer area and transmitting congestion information through a reserved bit of a data packet; and the congestion register dynamically records the congestion state of the neighbor node through the congestion information table. According to the invention, the problems of unbalanced load, low utilization rate of a buffer area and high congestion control cost of a traditional router are solved.

The invention belongs to the technical field of natural language processing, and discloses a large language model fine tuning method based on a hybrid routing policy, which comprises the following steps: acquiring a pre-trained large language model; constructing an expert model of the hybrid fine tuning module; an expert model of the hybrid fine tuning module comprises a pre-trained large language model, an expert network and a gating network; the expert network comprises a plurality of experts enhanced by the fine tuning module; inputting the training data into an expert model of a hybrid fine tuning module, obtaining information entropy of an expert network, and determining expert distribution based on a hybrid routing strategy of the entropy; obtaining the total loss of the expert model of the hybrid fine tuning module; and optimizing the expert network and the gating network according to the total loss. The invention provides a new hybrid routing strategy, the uncertainty of routers is reduced, the stability of the model is enhanced, and the fair participation of experts is promoted, so that the convergence speed of the model is accelerated, and the performance is improved.

The invention discloses a TSN scheduling optimization method and device based on flow sensing autonomous learning, equipment and a medium. The method comprises the following steps: deploying a lightweight flow detection module in a switch or a router, and after a controller receives a data request, automatically identifying a newly arrived unknown service flow by using the lightweight flow detection module, and judging whether the newly arrived unknown service flow is a periodic TT flow or an unpredictable burst flow; the controller performs classification management on the identified service flow types, collects topological information and flow requirements of the whole network and issues the topological information and the flow requirements to the terminal nodes through a network management interface; the controller constructs an intelligent queue scheduling task based on the collected network topology information and traffic demand and converts the task into a Markov decision process MDP, network resources, queue states and priorities are used as a state space, a scheduling strategy is used as an action space, a reward function is designed in combination with throughput and delay indexes, and an intelligent queue scheduling task is obtained. Driving a dynamic environment through real-time data and training a DRL model; an enhanced queue scheduling mechanism Pro-CQF is adopted, different priority labels are configured according to classified flow types, and then mixed flow scheduling is carried out; the controller periodically collects time delay, packet loss rate and end-to-end transmission delay indexes and feeds back the indexes to the DRL model, and a scheduling strategy is updated online. According to the method, the traffic sensing and scheduling efficiency is greatly improved in a network environment in which multiple service flows coexist and end-side equipment functions are different, and the reliability and the expandability of the TSN in industrial Internet of Things, edge computing and other high-real-time application scenes are remarkably enhanced.

The invention discloses a door lock and card cooperative processing method and system, and the method comprises the steps: triggering a robot calling process according to preset times of doorbell button operation detected by a first door lock and read neighbor RFID card information, verifying a matching result of an RFID card and a resident registration form, and generating an authority verification passing signal; calculating a target multicast address based on the ID number of the RFID card, and converting the ID number of the RFID card into a false source IP address to obtain a communication parameter set; sending an IGMPv3 report message to the router according to the communication parameter set; according to the triggering operation of the same RFID card detected by the assisted second door lock, an assistance multicast request message is generated, the (S, G) forwarding table item is matched through the router and forwarded to the first door lock, the first door lock sends an assistance confirmation message to the second door lock, and robot calling authorization is completed. By utilizing the embodiment of the invention, efficient, safe and low-delay cross-door-lock cooperative operation can be realized.

The invention discloses a multi-communication group fusion holding method and system for an intelligent door lock, and the method comprises the steps: building an independent multicast tree taking a feature vector as an identifier according to a communication group request initiated by each main control door lock; generating a multicast message containing a target communication group feature vector and a digital signature according to a merging request sent by the first master control door lock; according to feature vectors exchanged by the master control door locks of the two parties through a secure channel, a target master control door lock initiates feature interception registration, and negotiates and establishes a boundary feature tag rewriting rule; and according to the established forwarding path and the rewriting rule, the audio and video streams of the communication groups of the two parties are subjected to feature tag rewriting through the network boundary router, and then bidirectional intercommunication is realized. By utilizing the embodiment of the invention, the dynamic security isolation and seamless fusion of multiple communication groups can be realized, the network load and transmission delay are reduced, meanwhile, the terminal door lock can realize audio and video intercommunication without sensing the network address of the opposite end or changing the configuration, and the expansibility, the security and the user experience of the system are improved.

The invention discloses a door lock communication group decomposition method and system, and the method comprises the steps: encrypting and issuing a sub-feature vector to a local storage of a corresponding member door lock through a secure channel according to the sub-feature vector generated by a main control door lock and a binding relation between the sub-feature vector and the member door lock; generating a multicast message containing a target subgroup feature vector and a digital signature according to a splitting request sent by a main control door lock; according to the result of querying the local ARP table and the subgroup registry by the multicast router, determining the subgroup to which the door lock connected with each interface belongs, and registering the corresponding sub-feature vector for the corresponding interface in the multicast forwarding table; and according to a detection message sent by each sub-group master control door lock and a confirmation message of a member door lock, the multicast router identifies and removes sub-feature vector registration of an interface which is not associated with an effective door lock, and communication group decomposition is completed. By utilizing the embodiment of the invention, safe, reliable and interface-level refined door lock group decomposition can be realized, and the splitting efficiency and the communication safety are improved.

The invention relates to the field of equipment safety verification, in particular to a wifi router on-line upgrading safety verification method and a wifi router on-line upgrading safety verification system. A wifi router online upgrade security verification system comprises a firmware package construction module, a firmware package verification module and a guarantee activation module. According to the method, a firmware packet header protection logic closed loop is formed through an innovatively designed multi-order composite verification structural body and an anti-offset signature positioning framework; the system can detect the nanoscale tampering of an attacker to the length identification field in real time, cuts off the power supply of the flash memory controller and erases the data buffer area at the hardware fusing level response speed when verification fails, and compared with a traditional scheme, the response efficiency is improved.

The invention discloses an optimal service discovery method based on explicit multicast assistance, a medium and equipment. The method comprises the steps that a user side obtains an IP address list of a target server through DNS query and caches the IP address list; based on the real-time network topology, generating a shortest path tree from a user side to all target servers by using a Dijkstra algorithm; an explicit multicast message with a head carrying a target IP list is constructed, and the intermediate router splits the message into unicast or sub-multicast messages according to a next-hop interface; and the user side determines the optimal server by monitoring the first effective response message. According to the method, traditional multicast addresses are replaced by explicit multicast, address resources are saved, network repeated traffic is reduced, meanwhile, dynamic path tree optimization and an intelligent message splitting mechanism are combined, service discovery delay is remarkably shortened, the method is suitable for high-concurrency scenes such as a distributed CDN, the Internet of Things and video live broadcast, bandwidth consumption is effectively reduced, and service reliability is improved.

A method and apparatus comprising computer code configured to cause a processor or processors to form one or more routers of a mixture of experts (MoE) model by computing router parameters based on a plurality of MoE weights, implement at least one of the routers of the MoE model to compute one or more outputs by using a number of the MoE weights less than that of the plurality of MoE weights, and derive an MoE expert based on iteratively updating the router parameters according to the one or more outputs.

The invention provides a communication method and a communication device. The method comprises: a transmitting end device receives a first message from a receiving end device through a router, the first message comprising a first field for representing the signal strength of the receiving end device, and further comprising a second field for representing the signal strength of the router. And the sending end equipment determines the distance between the receiving end equipment and the router according to the first field, determines the distance between the router and the sending end equipment according to the second field, and establishes signaling connection with the receiving end equipment through the local area network when the two distances meet a certain condition. And when the distance between the transmitting end equipment and the receiving end equipment meets a certain condition, establishing data connection with the receiving end equipment in a point-to-point manner. According to the communication method and the communication device, the success rate of establishing the signaling connection by the electronic equipment through the local area network can be ensured as much as possible, and the success rate of establishing the data connection in a point-to-point manner can be ensured as much as possible.

The invention discloses a method for increasing and suppressing the number of alarms through an alarm source, and relates to the technical field of communication network operation and maintenance, and the method specifically comprises the steps: constructing a network real-time topology, building a set membership graph between a router and a downlink device, and when the heartbeat overtime alarm of the downlink device occurs in the network, starting the router according to the set membership graph. And counting the proportion of the lost equipment in a sliding time window taking the parent router as the dimension, when the proportion of the lost equipment exceeds a threshold value, judging that the router is a root cause, sending out an early warning of the router meeting the condition of a root cause determiner, and shielding the overtime alarm of the router. According to the method, the root cause event of'all the downlink devices of the same route are disconnected 'is dynamically identified in the real-time alarm flow, and a high-severity router fault alarm is used for replacing a plurality of device heartbeat overtime alarms, so that the alarm noise is remarkably reduced, and the operation and maintenance efficiency is improved.

The invention belongs to the technical field of artificial intelligence, and particularly relates to a dialogue message queue management system based on a large model cluster. The system comprises a message preprocessing module, a message router, a plurality of independent priority queues, a resource scheduler, a dialogue state manager and a large model cluster. The message preprocessing module receives an external dialogue request, constructs a metadata object, allocates a priority queue tag, calculates a static score and records a timestamp; the message router distributes the request to a corresponding priority queue according to the label; the resource scheduler executes a scheduling strategy, calculates a dynamic priority score, realizes load balancing, retrieves historical context information from the dialogue state manager, combines a request and a context into a processing task and dispatches the processing task to a large model cluster; the large model cluster returns model response to the resource scheduler, and the dialogue state manager is responsible for context storage and updating. All the components of the system can efficiently process the dialogue message request through close cooperation.

A computer-implemented system and computer instructions stored on non-transitory computer readable medium for bilingual medical inquiry in both Arabic and English, including multiple-choice question answering, open-ended question answering, and multi-turn question answering. The system and instructions use a mixture of experts large language model (MOE LLM) having a router network connected to multiple expert networks. The MOE LLM is trained with medical domain data and is used to receive the input bilingual text in a format for a medical inquiry, and output text in a format of a response to the medical inquiry, in sequence. The system and instructions incorporate an English-to-Arabic translation pipeline having a language translation model to generate Arabic language medical instruction sets from English language medical instructions, for large scale use in Arabic and English medical inquiry.

Methods and systems are provided for cooperating routers in communication networks. The cooperating routers conduct a handshake to exchange information with respect to “cooperation types” which they are capable of performing and / or are configured to perform. In an exemplary “emergency connection” cooperation type, one cooperating router may use the ISP connection of another cooperating router to send and receive packets. In an exemplary “bandwidth sharing” cooperation type, one cooperating router may make excess bandwidth available for use by other cooperating routers. In an exemplary “latency optimization” cooperation type, one cooperating router may use another cooperating router to transmit duplicates of packets or to implement suppression techniques.

The present application provides an endogenous-security network method and architecture, a medium, and a device. The method comprises: extracting forwarding characteristic information of normal service data flows; delivering the forwarding characteristic information of the normal service data flows to a transport network element, so as to form a forwarding table entry, a flow table, and a forwarding white list, wherein forwarding modes in a method for binding service data flow forwarding characteristics to forwarding table entries of transport network elements comprises: an authentication and authorization mode and an automatic learning mode; the transport network element performing packet forwarding according to the forwarding characteristic information of the normal service data flows; and discarding packets of abnormal service data flows, wherein packet identification is performed for forwarding operations on the basis of a whitelist automatically generated from configuration files of switches and routers. The present application solves the technical issue of an IP network having low inherent security protection capabilities due to the openness thereof and thus requiring the deployment of a large number of external security protection facilities. The issue results in poor effectiveness, high costs, and difficulty in establishing low-cost security protection capabilities, and ultimately makes the IP network easy to attack but difficult to defend.

A method and apparatus comprising computer code configured to cause a processor or processors to receive an input question to the LLM, the LLM comprising a pretrained model and an aligned model, determine whether a portion of an unfinished response to the input question is factual, and answer the input question by at least adding to the unfinished response by a token-level router routing either to the pretrained model, based on determining that the portion is factual, or routing to the aligned model, based on determining that the portion is not factual.

The invention discloses a forestry scene target detection method and system based on multi-modal continuous learning, and the method comprises the five steps: building a multi-modal data set, preprocessing data, constructing an incremental hybrid expert model, training a model, and carrying out the detection and reasoning of a target. In combination with an adaptive expert extension module, a frequency domain router module and a weight fusion method based on spherical linear interpolation, efficient and accurate detection of a multi-modal forestry scene target is realized. According to the method, the expert module can be dynamically generated and adjusted in the process of continuously introducing new tasks so as to adapt to target detection requirements of different fields and categories, the disastrous forgetting phenomenon is effectively prevented, and robust representation of the model on the previous tasks is kept.

The invention is suitable for the technical field of wireless communication, and relates to a dynamic sensing Mesh network equipment upgrading method, a medium and a terminal.The dynamic sensing Mesh network equipment upgrading method comprises the steps that firstly, a boundary router inquires all wireless terminal node equipment lists in a network by broadcasting an inquiry instruction, and a complete network topology structure is constructed; dividing nodes into a plurality of groups according to the network topology based on the hop count and the signal strength threshold; the boundary router only pushes a complete software update package to the seed node of each group, and the seed node locally caches the update package and triggers to send a polling notification to the adjacent node; after receiving the notice, the adjacent node actively pulls the firmware update packet from the seed node, the updated node becomes a new source node, diffusion and update are continued to the next hop node, and if node pulling fails, the update packet is retransmitted from the superior node or the server; and the node sends a state report to the border router after finishing firmware downloading and updating. According to the invention, the upgrading efficiency and reliability of the Mesh network equipment are effectively improved.

The utility model relates to the field of energy routers, and discloses an energy router charging management device. The energy router charging management device comprises an energy router assembly; the heat dissipation mechanism is arranged in the box body in the energy router assembly and is used for realizing heat dissipation of the interior of the energy router charging management equipment; and the anti-falling mechanism is arranged on one side of the wire passing groove in the energy router assembly and is used for enhancing the connection strength between the network cable and the energy router charging management equipment. Liquid in a cooling water tank is cooled through a cooler in the heat dissipation mechanism, then the cooled liquid is conveyed into a circulating pipe and a ventilation pipe through a circulating pump, the cooling liquid is subjected to ventilation through the ventilation pipe, and finally the interior of the energy router assembly is cooled and dissipated through a nozzle. The cooling water tank, the circulating pump, the cooler, the circulating pipe and the ventilation pipe form a cooling circulation system, so that the cooling and heat dissipation effects can be achieved, and the service life of the energy router charging management equipment is effectively prolonged.

The invention discloses a door lock collaborative identity recognition method and system, and the method comprises the steps: obtaining a locally stored MAC-face binding relation table according to face recognition verification completed by a resident through an intelligent door lock; according to the moving path of the resident from the floor entrance to the home door lock, identity recognition is completed through cooperation of multicast communication and the home door lock, an identity confirmation result is obtained, and a directional multicast network is constructed; according to a leaving event that a resident opens a door lock from a room, triggering a passing door lock to collect leaving person information, and obtaining a leaving person identity confirmation result; and selecting to send a unicast or multicast confirmation message by the own door lock according to the completion degree of identity confirmation of the leaving person, and triggering the passing door lock to send an IGMP leaving message to the router to obtain a simplified multicast forwarding table. According to the embodiment of the invention, seamless identity cooperation of the whole process from home returning to home leaving can be realized, and the automation degree of home security and protection and the user experience are improved.

A router is configured to send an advertisement of selective Ethernet Virtual Private Network (VPN) (EVPN) routes based on the Route Types. The router is configured to determine a requirement for a route refresh from a Border Gateway Protocol (BGP) peer for a subset of Route Types of all of the Route Types, send a route refresh message to the BGP peer with an indication of the subset of Route Types, and receive updated routes from the BGP peer only for the subset of Route Types. The indication can be included in a reserved field and payload of the route refresh message. The route refresh message can be defined in RFC 2918.

The invention discloses a Mesh routing selection method based on a graph neural network and reinforcement learning. The Mesh routing selection method comprises the following steps: step S1, constructing a Mesh network graph structure model; s2, extracting an embedded representation of each router node by using a graph neural network; s3, designing a routing decision mechanism based on Q-learning, and generating a Q value in combination with the current router node state and neighbor information; s4, introducing a Masking mechanism to process the dynamic action space, and shielding illegal next-hop action; and step S5, defining a multi-target reward function which is used for guiding a Q-learning training process. The invention provides a Mesh routing selection method based on a graph neural network and reinforcement learning, and aims to solve the problems of poor adaptive capacity to a dynamic network environment, low routing efficiency and lack of intelligent decision support in the prior art and improve the data transmission performance of a Mesh network.

The disclosure regards a spiking neural network device comprising a plurality of cores, wherein each core is configured to generate and process a sequence of spike packets, wherein each of the cores is assigned to a router, said router being configured to transmit sequences of spike packets to one or more other routers or to receive sequences of spike packets from one or more other routers, and a plurality of transmission paths, wherein each transmission path is configured to connect two routers so that sequences of spike packets can be transmitted between the two routers, wherein the topology of the spiking neural network device is represented by a grid, in which the plurality of routers are arranged in a plurality of rows and a plurality of columns, wherein a substantially diamond-shaped or rhomboidal outline of the grid is obtained by letting the number of routers arranged in each row increase gradually by one or two routers between neighbouring rows from a minimum row length, such as one or two routers, in the two outermost rows in the grid, to a maximum row length in one or more central rows in the grid, and by letting the number of routers arranged in each column increase gradually by one or two routers between neighbouring columns from a minimum column height, such as one or two routers, in the two outermost columns in the grid, to a maximum column height in one or more central columns in the grid.

The invention relates to the field of routers, in particular to a combined router with signal expansion, which comprises a mounting plate, a connecting seat and four partition plates are fixedly mounted on the upper surface of the mounting plate, and the connecting seat and the four partition plates divide the top space of the mounting plate into a first placement cavity and three second placement cavities. A router body is fixedly mounted in the first placement cavity, supporting plates are fixedly connected to the interiors of the three second placement cavities, repeaters are arranged on the upper surfaces of the three supporting plates, connecting plugs are arranged on the four repeaters, and through cooperation of sliding blocks, first sliding grooves and self-locking mechanisms, a router body is fixedly mounted in the first placement cavity, and the supporting plates are fixedly connected to the interiors of the three second placement cavities; when the repeater is pushed to slide towards the direction of the connecting seat, the inclined surface of the inclined groove on the sliding block pushes the limiting rod to automatically retract, manual lifting operation is not needed, single-hand pushing installation of the repeater is achieved, after the repeater is in place, the limiting rod is inserted into the limiting hole under the action of the first spring, automatic locking is completed, and therefore the installation efficiency is remarkably improved.

A method and system for updating a topology on router nodes in a distributed storage system are described. The method can include obtaining, by a first service of a control plane service, topology information from an inventory data store by querying a second service of the control plane service that manages the inventory data store for the topology information. The method can further include generating, by the first service of the control plane service, a topology payload based on the obtained topology information. The method can also include sending, by the first service of the control plane service, a request comprising the topology payload to a router node.

The invention provides a processor chip with adjustable power consumption, which is integrated with a plurality of functional cores and a plurality of routers, all the routers are mutually connected to form an on-chip network, all the functional cores are mounted on the on-chip network to realize mutual communication, and the chip further comprises a power consumption management module, all the functional cores and all the routers are configured to be subjected to clock frequency adjustment by a power consumption management module, and the power consumption management module is used for adjusting the clock frequency of the processor chip based on a received frequency raising request from the functional cores and / or the routers; the clock frequency of the processor chip is adjusted based on a self-configured adjusting mechanism; wherein the power consumption management module is configured to support unified adjustment of clock frequencies of all the functional cores and / or all the routers, and independently adjust the clock frequencies of each functional core and each router; and the power consumption management module is mounted on the network-on-chip to realize communication with each router and each functional core.

Embodiments of the disclosure relate to a mesh network. The mesh network includes a plurality of drones and a border router. The plurality of drones is configured to communicate with each other and with the border router over a first protocol. The first protocol uses IPV6 mesh connectivity at a first bandwidth. At least one drone of the plurality of drones operates as a router, and at least one drone of the plurality of drones operates as an end device. The router is configured to communicate with the border router and with the end device, and the end device is configured to communicate only with the router. The border router can be configured to communicate with an external network over a second protocol using a second bandwidth, and the second bandwidth can be greater than the first bandwidth.

The invention discloses a door lock cooperative communication group establishment method and system, and the method comprises the steps: transmitting an IGMP report message carrying a feature vector to a home router according to the same feature vector part obtained by a door lock and an associated media terminal in advance; according to a communication group request initiated by the main control door lock, an invitation message containing a feature vector is sent to the multicast address, and the invitation message is forwarded to the associated door lock and the media terminal at the same time; according to the received invitation message, generating mapping table items respectively corresponding to the control stream and the media stream; according to NAT mapping table item information issued by the home router, the door lock sends a communication group response message to the main control door lock, and the home router forwards the data flow to the corresponding terminal according to the mapping table item. By utilizing the embodiment of the invention, the rapid and automatic networking of the door lock and the media terminal in the local network can be realized, the efficient separation and reliable forwarding of the control stream and the media stream can be realized, and the communication response speed, the security and the resource utilization efficiency can be improved.