141 results about "Smart network" patented technology

The invention relates to an optical data network (10-1) for connecting a plurality of data network nodes (N1-1, N2-1, N3-1, N4-1). The optical data network (10-1) comprises: i) the plurality of data network nodes (N1-1, N2-1, N3-1, N4-1), the data (99) comprising data packets (99p), wherein each data network node (N1-1, N2-1, N3-1, N4-1) is provided with a network interface (Nl); and ii) a data connection network (100-1) comprising at least one data switch (SW) and a plurality of optical connections (50) for allowing establishment of data links (DL1, DL2) between pairs of said plurality of data network nodes (N1-1, N2-1, N3-1, N4-1). At least one of the at least one data switch (SW) is a smart data switch (SSW) comprising at least one packet switch (PS, EPS) and at least one circuit switch (CS, OCS), At least two network interfaces are smart network interfaces (SN, SNC) comprising a data classifier (25) configured for classifying the data packets (99p) into at least two traffic classes including first traffic class (D1) and a second traffic class (D2) in accordance with a classification scheme. The smart data switch (SSW) and the network interfaces (SN, SNC) are configured to route the data packets (99p) of the first traffic class (D1) to the packet switch (PS, EPS) and to route the data packets (99p) of the second traffic class (D2) to the circuit switch (CS, OCS).

The invention belongs to the technical field of network scheduling, discloses an intelligent network scheduling method and system based on hybrid reinforcement learning, designs a multi-head attention strategy network based on Transform, can execute discrete protocol selection and continuous rate adjustment at the same time, and breaks through the limitation of a traditional scheduling model on action dimension division. Therefore, the scheduler can flexibly select the optimal strategy when facing a complex task, and the refinement and adaptability of the decision are remarkably improved. In order to overcome the problem of value over-estimation in reinforcement learning, a double-Q-Critic network structure is constructed, and double evaluation paths are introduced. In the training process, through minimizing the estimation value deviation of the two Critic values, the overhigh estimation is effectively inhibited, so that the convergence stability and the estimation value precision of the strategy are improved. In the reward mechanism, a structural causal model (SCM) is introduced to generate an anti-factual reward signal, and a real reward and an anti-factual reward are fused by constructing a causal intervention path for a specific action.

The invention relates to the technical field of network scheduling, and discloses an agricultural heterogeneous perception and intelligent network scheduling method and system, and the method comprises the following steps: S01, a system initialization stage; step S02, a resource scheduling and strategy learning stage; step S03, a sensing data fusion and processing stage; s04, a security mechanism and life cycle management stage: adopting an end-to-end encryption and identity authentication mechanism in a data transmission process, and collecting feedback parameters for updating a scheduling strategy and releasing resources after a task is completed; the system comprises a multi-mode agricultural sensing terminal, intelligent gateway equipment, a 5G and LPWAN hybrid access network, a resource scheduling module and a scheduling strategy learning engine. According to the method, the multi-dimensional state characteristics such as the network congestion level, the edge node load, the data timeliness and the task priority are sensed in real time through the deep reinforcement learning model, and the optimal allocation decision of communication and computing resources is dynamically generated.

The invention provides an encrypted traffic anomaly detection method based on unsupervised learning, and relates to the field of artificial intelligence network security. In order to solve the problems of dependence on a large number of labeled abnormal samples, poor model generalization ability and poor dynamic adaptability of the existing encrypted traffic anomaly detection method, the invention provides an encrypted traffic anomaly detection method based on unsupervised learning, which comprises the following steps: extracting effective features of an encrypted traffic data packet, generating a state sequence by using a KMeans clustering algorithm, and carrying out unsupervised learning on the state sequence; calculating the occurrence probability of the state sequence in combination with an n-order homogeneous Markov chain model; a dynamic adaptive threshold is constructed based on exponential weighted moving average (EWMA) and a sliding window mechanism, and abnormality judgment is realized by comparing the occurrence probability of a state sequence with the dynamic threshold. According to the method, only normal encrypted traffic is utilized for training, and effective detection of all encrypted traffic including abnormal traffic can be realized.

Systems and methods for intelligent network incident management, Root Cause Analysis (RCA), and automated remediation include receiving metrics, graphs, and historic logs associated with network performance of a tenant of the cloud system; identifying a network issue based on the received metrics, graphs, and historic logs; performing an automated RCA to determine a cause of the network issue; and remediating the identified network issue based on the determined cause. Various embodiments include training specialized Large Language Models (LLMs) for performing the automated incident identification, RCA, and remediation.

The invention provides an OCP intelligent network card test platform and a test method, which are applied to the technical field of OCP product test development, and the OCP intelligent network card test platform comprises a platform base, a switching test board assembly and a CPU board assembly. The switching test board assembly is fixedly arranged on the platform base, and the switching test board assembly is provided with a high-speed connector socket and an OCP slot; the CPU board assembly comprises a CPU, a memory and a high-speed connector plug; wherein the CPU board assembly is connected with the high-speed connector socket through the high-speed connector plug in a plugging manner; and the OCP slot is used for inserting an OCP intelligent network card to be tested, so that the CPU board assembly communicates with the OCP intelligent network card through the switching test board assembly, the test cost and the occupied space are remarkably reduced, the deployment and maintenance convenience is greatly improved, and the problems that a large number of servers are needed for testing in the OCP product development test, and the test efficiency is greatly improved are fundamentally solved. The test cost is high, and the test space demand is large.

This disclosure is directed to methods and systems that establish a secure data channel between a host and a disaggregated hardware device (“DHD”) of a data center. The system comprises an initiator host that runs objects, such as virtual machines and containers. The host includes an initiator smart network interface card (“SNIC”). The initiator SNIC includes a virtual device, a trust platform module (“TPM”) and a security engine. The system also comprises a target host equipped with a DHD and a target SNIC. The target SNIC includes a TPM and a security engine. The TPM and the security engine of the initiator SNIC and the TPM and the security engine of the target SNIC establish a secure data channel between an object running on the host and the DHD.

The invention discloses a vehicle-mounted network pressure testing method and device. The method comprises the steps that automatic testing software controls at least one hardware device to execute power-on operation of a vehicle to trigger awakening of the vehicle and execute power-off operation of the vehicle to trigger dormancy of the vehicle; injecting a message into a vehicle-mounted network through the automatic test software, wherein the message comprises a normal wake-up message or an abnormal message; acquiring vehicle-mounted network data of the vehicle through an intelligent network data acquisition instrument; and comparing the acquired vehicle-mounted network data with expected data, and judging whether the sleep time sequence or the wake-up time sequence is abnormal or not. According to the scheme, manual operation is reduced through automatic testing, testing cost and time cost can be reduced, and testing efficiency and accuracy are improved.

The application belongs to the technical field of network communication and relates to a smart network communication link design method, which designs a set of link quality maintenance mechanism: a joint feature vector containing a physical layer bit error rate and a logical layer queue depth is constructed, signal degradation acceleration and queue accumulation acceleration are captured based on second derivative calculation, adjacent link probability distribution differences are compared and layered sorted in a Pareto manner, and link quality fine evaluation and degradation judgment are realized; candidate backup links are screened based on physical disjoint constraints, real-time performance feedback of the candidate backup links is obtained through perturbation detection packets carrying traffic feature templates, service degradation risk probability is predicted in combination with quality scores of target links, resource margin and path cost information are interacted between network nodes based on a distributed negotiation mechanism, an optimal switching path is generated, and service flow is smoothly migrated, so that link maintenance is upgraded from passive response to active pre-inspection.

The application relates to the technical field of cloud computing, in particular to an intelligent network self-recovery method and system applied to a smart traffic system, which comprises the following steps: data collection, abnormality detection, fault diagnosis, automatic repair and self-learning optimization; the beneficial effects are that the intelligent network self-recovery method and system applied to the smart traffic system can immediately identify abnormal behaviors in the network and respond quickly through real-time monitoring and a quick abnormality detection mechanism, and the time for fault detection and response is greatly reduced; advanced fault diagnosis algorithms, including deep learning and pattern recognition technology, are used, and the system can accurately diagnose various complex faults, including unknown or novel fault modes.

The application discloses a kind of mobile network equipment forced authentication system and method based on multi-protocol perception, belong to wireless network portal authentication technical field, for the problem that non-HTTP protocol cannot trigger authentication in existing Captive Portal technology, system pop-up window can only trigger once and Wi-Fi connection is unstable and leads to authentication interruption, the present application works through delay authentication mark, dynamic DHCP lease management and intelligent network scoring protection cooperation, when the system detects the non-HTTP class protocol request of unauthenticated device, set delay authentication mark and block request, trigger authentication when its subsequent HTTP request or system detection;By allocating short lease for unauthenticated device, trigger system re-detection using lease expiration or IP address change, realize authentication pop-up window multiple trigger;And return normal network response to maintain high network score when user initial connection, prevent user equipment automatic switching network.

Smart Cybersecurity Computing Device for Real-Time Threat Analytics and Zero Trust Enforcement

Embodiments of the present application provide a cloud disk mounting method, a data processing method and related devices, and relate to the technical field of communication. The method comprises: obtaining configured cloud disk mounting information and cloud disk binding information; determining a target disk volume in a target storage server based on the cloud disk mounting information, and establishing a correspondence between the target disk volume and a smart network card. Based on the cloud disk binding information and the correspondence between the target disk volume and the smart network card, a correspondence between the target disk volume and a disk name in a host computer is established, so that the target disk volume is mounted to the smart network card. The present application thus realizes automatic mounting of the cloud disk, improves the accuracy of cloud disk mounting and the efficiency of cloud disk mounting.

The application discloses a collaborative office network security management method and system based on artificial intelligence and a medium, belongs to the technical field of artificial intelligence network security, and is used for solving the technical problem that when the existing collaborative office network has network abnormal conditions or network security events, the emergency mechanism mainly depends on manual intervention, and the response efficiency is low when data leakage or system paralysis is caused. The method comprises the following steps: through an edge computing node and based on an artificial intelligence engine architecture, performing edge preprocessing on initial office network data about adversarial samples to obtain a network feature matrix based on the initial office network data; through the artificial intelligence engine architecture, performing double-channel network anomaly detection processing on the network feature matrix to obtain network anomaly detection result data; and according to the network anomaly detection result data, performing corresponding network risk response actions and automatically generating a network anomaly solution strategy.

A network selection method includes connecting to a source 3GPP network, receiving a plurality of candidate 3GPP networks having ecological ratings, respectively, and selecting a target 3GPP network having a higher ecological rating from the candidate 3GPP networks when a power saving option is enabled. According to the method, the balance is sought between ensuring reliable network performance and promoting environmental protection practice through the optimal signal quality, so that the method becomes a feasible solution for reducing environmental influence while maintaining high-quality service of a network operator.

The present invention relates to an AI-driven cybersecurity system that provides adaptive protection by identifying, mitigating, and preventing cyber threats in real-time. The system leverages advanced machine learning algorithms to detect anomalies in network traffic, enabling the identification of potential threats, which are then neutralized through automatic adjustments to firewall rules and network configurations. The system includes a tracing module capable of locating the source of cyberattacks by analyzing IP addresses and other network metadata, allowing for comprehensive incident reporting. Additionally, the invention logs and monitors all devices connecting to the network, both via Wi-Fi and hardline access, ensuring security compliance and detecting unauthorized activity. The system's adaptive approach ensures continuous protection by automatically updating security measures and reporting detailed findings to security personnel for future prevention.

The application relates to the technical field of cloud computing and smart network cards, and provides a cloud disk hot plug method and device, a smart network card, a host and a storage medium, which are applied to the smart network card, the smart network card is in communication connection with the host through a bus, is in network connection with a storage server, a cloud disk created on the host establishes a logical corresponding relationship with a storage component of the storage server, and the method comprises the following steps: receiving a configuration instruction for hot plugging the cloud disk; based on the configuration instruction, the cloud disk is configured to be hot plugged; an interrupt signal is sent to the host to instruct the host to acquire current configuration information of the cloud disk from the smart network card based on the interrupt signal, and the hot plugging of the cloud disk is realized based on the current configuration information. The application realizes the hot plugging of the cloud disk based on the smart network card.

This application discloses a method for migrating smart NIC virtual machines based on libvirt and qemu, including the following steps: On the source smart NIC virtual machine to be migrated, configuration is performed: on the smart NIC side, a dirty page recording module and a first Migrate-agent module are configured; on the qemu side, a vfio migration agent module and a qmp interface are configured; on the destination virtual machine receiving the migration, configuration is performed: on the smart NIC side, a second Migrate-agent module is configured; on the qemu side, a vfio migration agent module and a qmp interface are configured. After configuration, the virtual machine migration is performed. Using the method of this application embodiment, the migration of virtual machines with smart NIC devices is realized based on libvirt and qemu.

The invention relates to the technical field of network security detection, in particular to an intelligent network security detection system and method based on big data analysis, and the method comprises the steps: continuously collecting network behavior data, and constructing a dynamic behavior portrait based on multi-dimensional feature analysis, the dynamic behavior portrait comprising a behavior pattern feature, a resource access feature and a time rule feature; performing behavior evolution analysis on the dynamic behavior portrait, and identifying an abnormal evolution behavior by quantifying a change track of a behavior mode; and generating a dynamic defense strategy based on the characteristics of the abnormal evolution behavior, and feeding back and optimizing the construction process of the dynamic behavior portrait according to a strategy execution result. According to the method, hidden and staged complex threat behaviors such as slow attacks, permission climbing and transverse movement can be recognized, the response capability of a detection system to advanced persistent threats and abnormal behavior drift is improved, and the method is effective reinforcement of an existing analysis technology based on a static strategy or a single cycle.

The invention provides a service providing method and device, electronic equipment, a storage medium and a program product, relates to the technical field of communication, and is used for improving the task execution efficiency of a communication system. The method is applied to a task management function network element, and comprises the following steps: receiving a task execution request sent by an intelligent network service proxy function network element; the task execution request is used for requesting the task management function network element to execute a to-be-executed task, and the to-be-executed task is determined based on a network service request intention of a user; decomposing the to-be-executed task into at least one subtask; obtaining network state information of network resources required by at least one subtask; and determining a task execution node corresponding to the at least one subtask based on the network state information, allocating a corresponding network resource to the at least one subtask, and issuing the at least one subtask to the corresponding task execution node.

The utility model discloses a large-scale intelligent network central control host with a touch screen, which relates to the technical field of intelligent networks, and comprises a box body, a mounting layer is arranged on the front surface of the box body close to the lower part, a host is arranged in the mounting layer, and a display layer is arranged on the front surface of the box body close to the top part. A partition plate is vertically arranged close to the interior of the display layer, a lower fixing claw is installed at the bottom of the partition plate in a penetrating mode, an upper push rod is installed at the top of the partition plate in a penetrating mode, and a touch screen is arranged between the lower fixing claw and the upper push rod. According to the utility model, the touch screen is arranged in the box body and is connected to a plurality of groups of hosts through lines, so that all hosts can be regulated and controlled in a unified manner, the touch screen can be placed in the box body and observed from the outside through the glass plate, and the glass plate can also be taken out for direct operation, and the whole use is more flexible and convenient.

A method of training multi-agent intelligent network providing distributed attention, comprising scheduler and K layers of nodes comprising plurality of nodes, first input layer of nodes, one or more intermediate layers of nodes and final layer of nodes. Each node is arranged to process input information through model to provide output information, nodes of first layer are arranged to receive input tokens as input information, nodes of one or more intermediate layers are arranged to receive input information from one or more nodes of a previous layer and send output information to one or more nodes of a subsequent layer, and nodes of the final layer are arranged to provide output information as output labels. The method includes processing input information in each node in current layer, share processed information with nodes of same layer, share processed tokens as input information to zero or more nodes of subsequent layer.

The invention relates to a host network protocol stack soft unloading method and device based on an intelligent network card. The method comprises the following steps: a software driver transmits user data between a user mode application program at a host side and a native system kernel socket interface in an embedded processor at an intelligent network card side by utilizing a hardware channel at the intelligent network card side based on an inter-process communication mode; a native system kernel socket interface in the embedded processor on the intelligent network card side replaces a native system kernel network protocol stack on the host side based on the native system kernel network protocol stack on the intelligent network card side to achieve the function of the network protocol stack. According to the embodiment of the invention, the network protocol stack function of the host CPU is unloaded by using the hardware channel at the intelligent network card side through the software driver, so that the computing resources of the host CPU are saved, the network processing pressure of the host CPU is reduced, the network transmission performance is improved, and meanwhile, the flexibility of a host network protocol stack unloading scheme and the convenience of deployment and implementation are improved.

The utility model relates to the technical field of smart home safety, and discloses a real-time video anti-shielding defense system based on smart AI home safety, which comprises a monitoring unit, a smart network camera unit, a signal shielding detection unit, a terminal control unit and an execution unit, the monitoring unit is used for detecting the surrounding environment of the detection unit in different dimensions or controlling the monitoring unit to change the environment through a terminal, and the intelligent network camera unit serves as a core unit of the system and is used for receiving data information collected by the monitoring unit and pushing the data information to the terminal control unit in an APP mode. The terminal control unit is used for receiving a control command sent by the intelligent network camera unit and displaying data of the intelligent network camera unit, the signal shielding detection unit is used for carrying out electromagnetic safety detection on the intelligent network camera unit in a family space environment, and an alarm is given if abnormity comprises malicious radio attack. And the execution unit is used for receiving instruction information sent by the terminal control unit through the intelligent network camera unit, sending data information to access the execution unit after accurate confirmation of audio and video, and executing security and fire-fighting tasks according to the instruction.

The invention provides an SIP VOIP phone intelligent network self-adaption method and device, and the method comprises the steps: a network monitoring module calls a network interface module to obtain a test message which is regularly transmitted to a communication opposite end by each network interface, and carries out the analysis to obtain a network statistical value, real-time audio and video quality data are obtained by analyzing a voice statistical value in an RTCP message or an RCPP-XR message received from an opposite terminal in the dialogue process; the adaptive module receives the network statistical value and the voice statistical value, calls an adaptive adjustment algorithm, and modifies corresponding video parameters and audio parameters; the intelligent network switching module receives the network statistical value and the voice statistical value and judges whether network switching and target switching are needed or not according to numerical values; the self-adaptive adjustment module and the intelligent network switching module call the SIP signaling module and send re-INVITE or UPDATE information to update IP port information of the RTP and new audio parameters to be used, and meanwhile, the intelligent network switching module calls the SIP signaling module to send a re-REGISTER command to perform re-registration with a new IP.

This invention relates to the field of artificial intelligence technology, specifically an intelligent network security detection system and method based on big data analytics. The system includes a network entity behavior modeling module, a threat-anomaly correlation analysis module, an anomaly narrative generation module, an intelligent threat response module, and a report generation module. The system dynamically models entity behavior using multi-source heterogeneous data, constructs behavioral baselines, and correlates external threats and anomaly events to generate a visualized knowledge graph. Based on the correlation of graph nodes, it identifies anomalous intents and forms narrative lines, completing threat assessment and confidence determination, and matching response strategies. Simultaneously, it continuously optimizes the knowledge base through attack chain feedback, automatically outputting threat reports, achieving a closed-loop protection system of monitoring, analysis, response, and knowledge accumulation. This invention significantly improves the speed of network security detection and response.

The application discloses a kind of intelligent network slice adaptive arrangement methods and systems for 6G communication, it is related to network slice technical field.For the same target slice in the same current period in the same target slice in the intersection area of vehicle-road cooperation, when control message transmission anomaly and target detection and tracking anomaly occur simultaneously, the existing method is difficult to determine the main local resource range of the current anomaly and the resource adjustment range is easy to enlarge problem, the application first unifies the two sides of the anomaly, then constructs the candidate resource domain covering wireless side resource and edge side resource, and compares the two sides of the anomaly associated results in the candidate resource domain, generates local anomaly type label, and then executes resource adjustment range mapping, extension, recovery and adaptive arrangement adjustment, to improve the stability of resource range determination and improve the consistency of range recovery.

The application provides a smart networked vehicle accident evidence storage and responsibility determination system and a retrieval method. The system comprises a blockchain network module, an intelligent contract module, a blockchain gateway module, a traffic management department evidence collection module and an SCMS link value query module. The blockchain gateway module is used for obtaining communication messages of a vehicle-mounted unit, forming structured evidence storage data and uploading the data to a blockchain intelligent contract. When an accident occurs, the accident is reported to the blockchain intelligent contract, and an accident evidence collection request is sent to surrounding vehicles. After receiving the request, the surrounding vehicles upload their link values to the intelligent contract. After receiving the reported accident and the link values, the traffic management department transmits the link values into the SCMS system, obtains all the link values of the vehicle being collected within the same time period, combines the link values and the accident time, calculates the retrieval key of all possible messages within the space-time range of the accident, and transmits the retrieval key into the blockchain network through a filter and an on-chain mapping query to retrieve relevant evidence.