66 results about "Bandwidth distribution" patented technology

The present application relates to the technical field of large-scale data processing, and particularly relates to a parallel computing method and system suitable for large-scale data processing. The system comprises a task management unit, a distributed load balancing module, an elastic expansion architecture, an intelligent communication optimization module, and a resource monitoring unit, wherein the task management unit divides large-scale data into a plurality of sub-tasks by means of a task decomposer, and distributes the sub-tasks to computing nodes by means of a task scheduler and a priority distributor; the distributed load balancing module achieves global load balancing by means of a load sensing unit, a dynamic adjustment unit and a balance optimization unit; the elastic expansion architecture dynamically adjusts system resources by means of a node manager, a resource pool controller and an expansion decision-making device; the intelligent communication optimization module optimizes inter-node communication by means of a communication path planning unit, a bandwidth distribution unit and a delay compensation unit; and the resource monitoring unit monitors the system performance in real time by means of a performance collector, a state analyzer and an anomaly detector.

The invention discloses a GPU Box hardware decoupling system based on a Retimer card and a PCIe Switch chip, and belongs to the technical field of computer hardware architecture and high-speed interconnection. According to the system, a Retimer card and a PCIe Switch chip are integrated in an independent GPU Box, and a decoupling link of a CPU server and a GPU acceleration card is constructed; the Retimer card realizes 30-meter long-distance PCIe signal transmission and breaks through physical distance limitation; the PCIe Switch chip pools GPU resources through a dynamic routing and MRIOV technology, supports flexible allocation of computing power by multiple servers, and realizes Peer-to-Peer direct connection communication between GPUs. Aiming at a large model reasoning scene, the system optimizes KV cache bandwidth allocation and video memory and memory cooperative scheduling, so that the 100B parameter model reasoning throughput is greatly improved; and meanwhile, the usability of the system is greatly improved through fault isolation and hot plug design. According to the method, the problems of physical binding of the CPU and the GPU, limited transmission distance, rigid resource allocation and the like in a traditional architecture are solved, and the method is suitable for large-scale AI calculation and distributed GPU cluster deployment.

The embodiment of the invention provides a server bandwidth allocation method and device, equipment and a storage medium. The method comprises the following steps: acquiring state data of an expansion board card; determining a power supply level signal according to the state data; determining an output enabling state according to the power supply level signal; combining the state data, the power supply level signal and the output enabling state to determine hardware feature information; determining configuration information corresponding to the hardware feature information; determining bandwidth identification information matched with the configuration information based on a preset bandwidth allocation strategy; and based on the bandwidth identification information, bandwidth allocation is carried out on the expansion board card, the bandwidth can be automatically identified and reallocated, the same expansion board card can adapt to various configuration combinations, and the utilization rate and the flexibility of the server are remarkably improved.

The invention discloses a multilayer space dynamic topology adaptive communication network construction method, and belongs to the technical field of communication networks. The method is realized through five steps of constructing a multi-layer fusion architecture of space-based, air-based and ground-based networks, instantiating network slices of power service perception, performing dynamic topology abstraction and resource pre-mapping based on virtual cell division, performing power and bandwidth collaborative allocation based on reinforcement learning, and deploying multi-domain fault diagnosis and predicting a closed loop. According to the invention, power and bandwidth collaborative allocation based on deep reinforcement learning not only can optimize the total capacity of the system, but also can realize dynamic balance of the network load. The system can automatically adjust the power and bandwidth distribution according to the real-time network state in the operation process, and the self-adaptive mechanism enables the network to efficiently cope with the demand changes of different power businesses.

The invention relates to the technical field of industrial automation control and robots, in particular to a heterogeneous high-real-time multi-axis robot cooperative motion control system and method.The system comprises an SoC main control layer, an FPGA driving layer and a hardware interconnection layer, the SoC main control layer adopts a heterogeneous SoC chip, an ARM processor is responsible for track planning and task scheduling, an FPGA logic deployment reconfigurable PID calculation engine is responsible for FPGA logic deployment, and the FPGA driving layer is responsible for FPGA logic deployment. A multi-axis position ring and speed ring algorithm is realized through the time division multiplexing processing unit; the FPGA driving layer comprises a plurality of FPGA coprocessors, and an FOC current loop IP core is instantiated to execute bottom layer vector control related operation; the hardware interconnection layer adopts an EPPI protocol supporting topology awareness, dynamic bandwidth allocation and a ping-pong buffering mechanism, and efficient transmission of a control instruction and feedback data is achieved. Through hierarchical control and hardware collaborative design, the problems that an existing multi-axis control system is large in calculation delay, insufficient in synchronization precision and low in hardware resource utilization rate are effectively solved.

The present disclosure provides a bandwidth allocation method, an electronic device, and a computer readable storage medium. The bandwidth allocation method comprises: determining an overall wide area network (WAN) interface bandwidth of a network according to capability information of a WAN interface of each node currently added to the network, wherein nodes comprise a master node and a slave node, and the WAN interface is an interface where the node is connected to the WAN; and allocating traffic to at least one egress node according to the current overall WAN interface bandwidth of the network and a bandwidth distribution condition of each node.

The application discloses a network card management system and method and electronic equipment, and relates to the technical field of network card management. The system comprises a mainboard, a network card and a preset cable. The mainboard is provided with a mainboard power supply circuit, a logic controller and a connector. The connector comprises a power supply pin connected to the mainboard power supply circuit and at least one bandwidth identification pin. The inside of the mainboard side plug of the preset cable is provided with a target circuit connection structure, and the power supply pin and the bandwidth identification pin are directly electrically connected when the cable is connected. The logic controller automatically determines the network card bandwidth allocation mode by collecting the level signal of the bandwidth identification pin and generates a configuration signal to control the network card work. The technical problem that the network card bandwidth configuration in the related art depends on software setting is solved. The automatic identification and configuration of the network card bandwidth are realized through the hardware connection topology, and the system deployment efficiency and reliability are significantly improved.

Disclosed in the present application are a bandwidth construction method, a bandwidth prediction method, and a related apparatus. The bandwidth construction method comprises: first, acquiring first bandwidth data and task scheduling information, wherein the first bandwidth data comprises n bandwidth sampling values of a first memory access component of a vehicle in a first driving scenario, m operation tasks associated with the first driving scenario are run on the first memory access component, and the task scheduling information is used for indicating the runtime sequence of the m operation tasks; and then, on the basis of the first bandwidth data and the task scheduling information, obtaining second bandwidth data, wherein the second bandwidth data is used for indicating time-domain distribution information of the bandwidth of each operation task among the m operation tasks in the first driving scenario. In this way, the construction of fine-grained, operation-task-level bandwidth distribution on a memory access component is realized, such that the bandwidth of the memory access component can be predicted more accurately, thereby not only facilitating the utilization rate of bandwidth resources, but also improving the performance of an autonomous driving system.

The application relates to the technical field of Internet of Vehicles, and discloses a car-machine multi-screen cooperative display and interaction method, which comprises the following steps: maintaining a global session and a member view through a centralized session arranger, realizing session access and dynamic member management of multi-screen devices, and determining multi-screen related parameters; constructing a full-order execution rule based on a logical clock, the multi-screen related parameters and an operation identifier, and assigning a global serial number to each multi-screen operation; obtaining a focus of an operation request, and determining feedback parameters through a soft lock and a priority preemption strategy; periodically hashing and checking based on a shadow state and a domain Merkle tree, positioning state differences according to the feedback parameters, and performing incremental repair through an incremental patch; and adopting WebRTC layered coding and a weighted bandwidth allocation strategy, preferentially guaranteeing control signaling and a main driving screen experience, and dynamically adapting in a bandwidth-limited scene. The application can solve the problems of multi-screen interaction inconsistency and insufficient real-time performance, thereby improving the multi-screen cooperative experience in a vehicle.

The present application relates to the field of communications, provides a communication method and a communication apparatus, facilitating the reduction of the signaling overhead of a trigger frame. The method comprises: receiving a trigger frame, wherein the trigger frame comprises first information and second information; the first information is used for indicating a discrete bandwidth distribution mode in a first frequency domain range when the first frequency domain range comprises a discrete RU; the second information is used for indicating a discrete bandwidth distribution mode in a second frequency domain range when the second frequency domain range comprises a discrete RU; and an indication granularity of the discrete bandwidth distribution mode in the first frequency domain range is different from an indication granularity of the discrete bandwidth distribution mode in the second frequency domain range, thereby reducing the indication overhead.

This utility model discloses a communication device with dynamic bandwidth physical isolation for Ethernet, including a main control circuit and a communication circuit. The main control circuit and the communication circuit are connected to realize external network communication. The communication circuit is connected to an isolation circuit, which includes a photoelectric conversion chip M1. Its optical signal input pin is connected to an external fiber optic network to receive optical signals, and its electrical signal output pin is connected to the communication circuit to output electrical signals. This device constructs a physical isolation channel through the photoelectric conversion chip, and uses optical signal transmission to achieve physical layer isolation of Ethernet communication, effectively blocking network attacks and data leakage paths. The dynamic bandwidth characteristic can adaptively adjust bandwidth allocation according to transmission needs, improving network resource utilization. It is suitable for scenarios with high requirements for network security and bandwidth stability, such as finance and government, providing a physical isolation solution for Ethernet communication that combines security and flexibility.

The invention relates to a multi-source disturbance self-calibration method and system for a rotary filtering type monochromator, and belongs to the technical field of optical instrument filtering calibration. The method comprises the following steps: converting a multi-source disturbance signal response characteristic into a disturbance energy spectrum basis vector, and extracting spectrum bandwidth distribution to obtain a spectrum self-disturbance calibration parameter; luminous flux preposition constraint is carried out, the convergence state of a real-time spectrum observation value under the disturbance change condition is detected, confidence coefficient calibration is carried out on the disturbance contribution weight on which the self-calibration operation depends, and a filtering disturbance compensation control strategy is generated; constructing a filtering performance evolution model, identifying spectral wavelength shift and resolution data after disturbance self-calibration, verifying the filtering degradation period of the monochromator, and outputting a filtering calibration feedback report; and setting a critical interference stable threshold, establishing a filtering virtual regulation and control mechanism, enhancing the compensation calibration weight of a filtering disturbance compensation control strategy, and generating a filtering disturbance self-calibration scheme.

This disclosure provides methods, components, devices and systems for techniques for tone mapping for 60 MHz distribution bandwidth. Some aspects more specifically relate to techniques for tone mapping for the 60 MHz distribution bandwidth that support distributed transmissions. In some examples, the wireless device may transmit obtain data for transmission in a physical layer convergence protocol (PLCP) protocol data unit (PPDU) over a PPDU bandwidth that exceeds 80 megahertz (MHz) and comprises an 80 MHz distributed transmission subblock, and the 80 MHz distributed transmission subblock may be punctured providing a 60 MHz distribution bandwidth. Tones of the 60 MHz distribution bandwidth may be shifted by tone shift values to align with a boundary of valid tones defined in the punctured 80 MHz distributed transmission subblock of the PPDU bandwidth.

The invention discloses a bandwidth allocation and loss self-compensation cooperative control method, system and device and a medium, and the method comprises the steps: S1, obtaining electric power communication service data, optical fiber line state data and external environment data through multi-source data collection; s2, constructing and training an AI model according to the collected multi-source data, and constructing and training a multi-source data fusion model, a reinforcement learning bandwidth allocation model and a loss trend pre-judgment model; s3, outputting results based on each model constructed and trained by the AI model, synchronously generating a bandwidth allocation strategy and loss compensation parameters, and executing the bandwidth allocation strategy and the loss compensation parameters; and S4, feeding back and inputting the communication state data subjected to bandwidth-loss cooperative regulation and control into each model in construction and training, and updating each model parameter to optimize subsequent regulation and control. According to the invention, dynamic coordination of bandwidth and loss compensation is realized, the hysteresis of discrete regulation and control is solved, the service demand change and the line loss fluctuation are synchronously responded, and the communication link stability and the resource utilization rate are improved.

The invention relates to a high-voltage switch fault diagnosis system based on non-intrusive sound wave monitoring. The system comprises: a separation module for performing blind source separation on a flowing acoustic wave signal to obtain a plurality of acoustic wave source signals and corresponding energy ratios; the calculation module is used for calculating the kurtosis of the amplitude sequence of each sound wave source signal and the standard deviation of a first-order difference value, and determining a fundamental frequency component and a plurality of harmonic frequency components based on a spectral analysis result; the first determining module is used for determining an SF6 gas acoustic wave signal according to the kurtosis of the amplitude sequence of each acoustic wave source signal, the standard deviation of the first-order difference value, the spectrum distribution, the energy ratio, the fundamental frequency component and the multiple harmonic frequency components; the second determining module is used for determining a leakage fault coefficient according to the bandwidth distribution of the plurality of harmonic frequency components and the distribution of the high-frequency components; and the output module is used for outputting early warning information according to a comparison result of the leakage fault coefficient and a fault threshold value. According to the invention, accurate identification of the SF6 gas sound wave signal can be realized.

The application relates to the technical field of data processing, and discloses a gateway service comprehensive management method and system based on FTTR demand. The method comprises the following steps: obtaining a room-level FTTR topology mapping table through topology scanning of an optical time domain reflectometer test signal, extracting features of service traffic of each room based on deep packet detection to obtain a room function identification label library, compensating attenuation of corresponding room optical ports by using an optical power meter to obtain an FTTR link quality evaluation matrix, performing bandwidth allocation by using a reinforcement learning algorithm to obtain a room-level QoS strategy configuration set, and obtaining an FTTR gateway service comprehensive management scheme based on MQTT protocol building data interaction. The application improves the intelligent level and resource allocation efficiency of room-level management of the FTTR network. The application improves the real-time performance of optical fiber link quality monitoring and the collaboration of building network management.

The invention relates to the technical field of data analysis, can be applied to business system platforms of financial science and technology, medical treatment and health and the like, and discloses a dynamic bandwidth adjustment method, device and equipment based on bucket-level resources and a medium. Obtaining high-priority resource buckets and common resource buckets, collecting system resource states of the resource buckets, performing health degree analysis on the high-priority resource buckets and the common resource buckets, and generating health degree scores of the resource buckets; and performing weighted averaging on the health degree score of the high-priority resource bucket and the health degree score of the common resource bucket according to a preset weight to obtain a global health degree, and performing different dynamic bandwidth allocation and dynamic speed-limiting adjustment on the high-priority resource bucket and the common resource bucket according to the global health degree to obtain a target bandwidth adjustment result. According to the invention, the differentiation and adaptive management of the bandwidth can be realized according to the service importance and the load change.

The application discloses a signal frequency parameter estimation method and device, and relates to the technical field of signal processing. The method comprises the following steps: acquiring a first center frequency, a first bandwidth and a signal-to-noise ratio, and determining a center frequency distribution range and a bandwidth distribution range in combination with a frequency interval; performing band-pass filtering processing and autocorrelation processing on an original signal, and then performing Fourier transform and frequency domain correction processing to obtain a target frequency domain signal; determining a second center frequency in the center frequency distribution range based on the target frequency domain signal, and then determining a corresponding bandwidth lower / upper limit distribution range; and determining a bandwidth start / end frequency in the bandwidth lower / upper limit distribution range based on the target frequency domain signal to obtain a second bandwidth. The application reduces the influence of noise interference through filtering and autocorrelation processing, adaptively limits the distribution range to avoid the limitation of Fourier transform length, and realizes accurate positioning of the frequency parameter through a difference method, so that the accuracy of frequency parameter estimation is significantly improved.

Embodiments of the present application relate to the technical field of computer systems, and provide a computer system for bandwidth allocation, a bandwidth allocation method, and a product. When a motherboard of the computer system is powered on, a monitoring module, upon identifying a change in applications of extension units plugged into the motherboard, generates, on the basis of bandwidth configuration information corresponding to the extension units, a bandwidth allocation information table corresponding to a central processing unit; after the motherboard of the computer system is powered on, a baseboard management controller generates an option configuration file on the basis of the bandwidth allocation information table; after a main unit of the computer system is triggered, a basic input / output system is used for configuring any connection port of the central processing unit on the basis of the option configuration file during a boot process, so as to allocate bandwidth resources of the central processing unit to the extension units. The embodiments of the present application aim to automatically allocate bandwidth resources in a computer system.

This application discloses a sensor bandwidth allocation method, system, and device based on the Industrial Internet of Things (IIoT), relating to the technical field of IIoT. The method includes: acquiring the status information of each static sensor node, planning a movement path for a mobile relay device based on the status information, configuring a dedicated channel for the mobile relay device, and determining a reference service time window; during movement, generating a channel trigger command based on the real-time location and real-time movement speed of the mobile relay device and the reference service time window of the target service segment; generating a global network time slot scheduling strategy based on the activation timing of the dedicated channel; distributing the global time slot scheduling strategy to the static sensor nodes, and distributing the movement path and the reference service time window to the mobile relay device. This application improves the accuracy of response to spatially dynamic unbalanced loads.

This invention discloses a base station resource scheduling method based on complementary prediction of traffic and communication features. It proposes a traffic flow and data flow prediction optimization model based on dynamic time programming. Vehicle connection data and vehicle communication volume data are acquired between base stations. The prediction model is trained using DDPG to predict the number of vehicles connected to each base station and the communication volume in the future. Based on the prediction data, a bandwidth allocation optimization model considering overall system latency and the equality of resource allocation at different times is designed. The optimal solution is found through projected gradient descent, achieving resource scheduling for future traffic flow. This method overcomes the problem of base stations being unable to dynamically adjust bandwidth allocation strategies according to traffic flow changes, leading to resource waste or shortages. By predicting traffic flow and communication demand, the method allows base stations to allocate more resources when traffic flow is expected to increase and less resources when traffic flow is expected to decrease, achieving efficient resource utilization and improving bandwidth resource utilization in intelligent transportation systems.

This invention discloses a remote control system for port container cranes. The system comprises a network monitoring module, a controller, a sensor array, a video encoder, and a remote operation terminal. The system utilizes the sensor array to acquire physical parameters such as spreader height, trolley and spreader speed, and distance from the top of the target container. It calculates the bandwidth allocation ratio coefficient for the region of interest in the video and sets mandatory numerical range constraints to achieve precise mapping between visual resources and operational urgency. To address the mechanical inertia of the trolley mechanism, the system employs a combination of trend prediction and physical boundary limiting to smoothly compensate for control commands. The rate of change of the predicted commands is limited by the maximum permissible acceleration, resolving motion misalignment issues caused by network jitter. This invention improves the system's fault tolerance under extreme operating conditions, ensuring high-definition feedback and inherent safety during remote container operations.

Disclosed are techniques that provide efficient routing strategies for AllReduce transfers, which are the the dominant traffic in machine learning-centric datacenters, resulting in faster parameter synchronization in distributed machine learning and improving the average training time by over 9%. As compared with the prior art, our efficient route of AllReduce traffic advantageously maximizes bandwidth allocation while minimizing bandwidth tax, accelerates training speed of distributed machine learning models or large language models in optical circuit switching-based clouds, and more efficiently provisions indirect optical paths, by leveraging the unused ports or bandwidth resources from GPU servers that run single or standalone computing jobs.

The invention discloses a signal frequency parameter estimation method and device, and relates to the technical field of signal processing. The method comprises the following steps: acquiring a first center frequency, a first bandwidth and a signal-to-noise ratio, and determining a center frequency distribution range and a bandwidth distribution range in combination with a frequency interval; carrying out band-pass filtering processing and self-correlation processing on the original signal, and then carrying out Fourier transform and frequency domain correction processing to obtain a target frequency domain signal; based on the target frequency domain signal, determining a second center frequency in the center frequency distribution range, and then determining a corresponding bandwidth lower bound / upper bound distribution range; and based on the target frequency domain signal, respectively determining bandwidth starting / ending frequencies in the bandwidth lower bound / upper bound distribution range to obtain a second bandwidth. According to the method, the influence of noise interference is reduced through filtering and self-correlation processing, the distribution range is adaptively limited to avoid the limitation of the Fourier change length, accurate positioning of the frequency parameters is realized through the difference method, and the precision of frequency parameter estimation is remarkably improved.

The invention provides a communication method and a communication device, relates to the field of communication, and helps to reduce the signaling overhead of a trigger frame. The method comprises: receiving a trigger frame, the trigger frame comprising first information and second information, the first information being used for indicating a discrete bandwidth distribution mode in a first frequency domain range when a discrete RU is included in the first frequency domain range, and the second information being used for indicating a discrete bandwidth distribution mode in a second frequency domain range when the discrete RU is included in the second frequency domain range, and the indication granularity of the discrete bandwidth distribution mode in the first frequency domain range is different from the indication granularity of the discrete bandwidth distribution mode in the second frequency domain range, so that the indication overhead can be saved.

The invention discloses a dual-mode communication data transmission method for an electric energy meter and an electricity utilization information acquisition terminal, and relates to the technical field of electric power communication, and the method comprises the steps: collecting physical layer and transmission layer parameters of dual channels in real time, evaluating the communication quality, predicting the out-of-order probability, and generating channel state information; analyzing the priority according to the characteristics of the electricity consumption information data, and calling an out-of-order prevention transmission strategy; carrying out serial number marking on the data packets, and distributing channels based on channel states to carry out parallel transmission; a recombination buffer area is set at a receiving end, the recombination data packets are sorted according to the serial numbers, and the continuity of the serial numbers is detected; monitoring a channel load rate in real time, and if the channel load rate exceeds a threshold value, transferring low-priority data to relieve congestion; and monitoring effective throughput and end-to-end time delay, dynamically adjusting time slot and bandwidth allocation, and feeding back an adjustment result to update channel state information.

The invention discloses a bandwidth distribution fitting method, system and device for a power distribution network and a medium, and the method comprises the steps: obtaining the service platform data of a target power distribution network, and constructing a power distribution network communication network scene under the access of a multi-element main body according to the service platform data; based on the power distribution network communication network scene, carrying out clustering analysis on the service platform data to obtain a service high-density cluster; calculating bandwidth sample data of the service high-density cluster, and constructing a bandwidth distribution initial fitting model according to the bandwidth sample data; reconstructing the bandwidth distribution initial fitting model by adopting a dynamic local area kernel window width adjustment mechanism to obtain a bandwidth distribution reconstruction fitting model; solving the bandwidth distribution reconstruction fitting model to obtain bandwidth distribution fitting data of the target power distribution network; the model is flexibly optimized through a dynamic local area kernel window width adjustment mechanism, and the fitting precision and efficiency of the power distribution network under multi-element main body access are improved.

To provide an optical network management device in an optical communication network using wavelength division multiplexing that improve utilization efficiency of communication networks.SOLUTION: An optical network management device 40041 includes wavelength selection information creating means for creating wavelength selection information, which is information about a wavelength selection process through which an optical path 40002 that accommodates information signals passes, for each optical path, and wavelength selection information notifying means for notifying an optical node device through which the optical path passes of the wavelength selection information.SELECTED DRAWING: Figure 5B

This disclosure discloses a dynamic bandwidth allocation system, a dynamic bandwidth allocation method, and a server, relating to the field of server technology. The dynamic bandwidth allocation system includes a processing unit, a management controller, a configuration storage unit, and a selection unit. The processing unit has at least one high-speed interconnect bus interface. The configuration storage unit stores link bandwidth configuration information for the high-speed interconnect bus interface. In response to server power-on, the management controller retrieves the corresponding link bandwidth configuration information from the configuration storage unit based on a mode selection signal and provides this information to the processing unit. The processing unit configures the link bandwidth of the corresponding high-speed interconnect bus interface according to the link bandwidth configuration information. By decoupling bandwidth allocation information from traditional hardware circuitry, and implementing dynamic and flexible allocation of high-speed interconnect bus interface link bandwidth in a software-defined manner, the system solves the problem of maintenance and reconfiguration difficulties caused by fixed hardware bandwidth configuration.