High-bandwidth cluster control method and system based on digital communication network

Abstract

The present invention relates to a high-bandwidth cluster control method and system based on a digital communication network, comprising the following steps: performing timestamp marking on the real-time status data to obtain timestamp-marked status data; performing distributed status estimation on the timestamp-marked status data to obtain cluster coordination situation information; performing adaptive routing calculation on each device in the cluster based on the cluster coordination situation information to obtain a network resource scheduling strategy; performing channel multiplexing mapping on a preset digital communication network based on the network resource scheduling strategy through a preset space-time coding technology to obtain a channel multiplexing mapping matrix; performing resource scheduling, allocation and control on each device in the cluster based on the channel multiplexing mapping matrix to obtain a cluster control strategy, thereby solving the technical problem that the prior art is usually unable to provide sufficient time accuracy when processing these large-scale data, thereby affecting the overall performance of the cluster.

Description

Technical Field

[0001] The present invention relates to the technical field of digital communication networks, and in particular to a high-bandwidth cluster control method and system based on a digital communication network. Background Art

[0002] With the rapid development of information technology, digital communication networks play an increasingly important role in modern society, especially in the field of cluster control. Traditional cluster control systems often rely on preset, static network architectures and scheduling strategies, which makes it difficult for them to adapt to rapidly changing environments and growing data transmission needs. In order to improve the response speed and resource utilization efficiency of cluster systems, it is necessary to explore a high-bandwidth cluster control method that can sense and adjust dynamically in real time. In addition, with the explosive growth in the number of Internet of Things (IoT) devices, the massive data generated by these devices has put higher requirements on the bandwidth of existing networks and brought more complex management and control challenges.

[0003] Current cluster control methods have several limitations, mainly reflected in inaccurate state estimation, unreasonable network resource allocation, and low channel utilization. First, due to the lack of an effective timestamp mechanism to mark the device status, the time synchronization problem seriously affects the accuracy of system decisions when multiple devices work together. Secondly, the existing routing calculation method fails to fully consider the intentions and real-time collaboration needs of devices within the cluster, thereby limiting adaptability and flexibility. Furthermore, traditional channel reuse technology cannot well meet the needs of modern cluster systems for efficient and flexible resource scheduling, resulting in a waste of spectrum resources. Therefore, new technical means need to be introduced to optimize these problems in order to achieve more efficient cluster control.

[0004] In order to solve the above problems, this study proposed a high-bandwidth cluster control method based on digital communication network, aiming to build an intelligent and efficient cluster control system by integrating advanced technologies such as SDN controller, intent-driven network and space-time coding. This method can not only accurately obtain and process the real-time status information of each device in the cluster, but also perform distributed state estimation according to the cluster coordination situation to ensure the timeliness and accuracy of system decision-making. At the same time, through intent-driven adaptive routing calculation and advanced channel reuse mapping technology, network resources can be more reasonably scheduled, channel utilization can be improved, and ultimately precise control of each device in the cluster can be achieved. This method is expected to become an important part of future high-bandwidth, low-latency cluster control solutions, providing strong technical support for large-scale Internet of Things applications. Summary of the invention

[0005] The main purpose of the present invention is to provide a high-bandwidth cluster control method and system based on a digital communication network, which solves the technical problem that the existing technology usually cannot provide sufficient time accuracy when processing these large-scale data, thereby affecting the overall performance of the cluster.

[0006] To achieve the above object, the present invention provides a high-bandwidth cluster control method based on a digital communication network, comprising the following steps:

[0007] Acquire real-time status data of each device in the cluster, and timestamp the real-time status data to obtain timestamp status data;

[0008] By means of a preset SDN controller, distributed state estimation is performed on the timestamp-marked state data to obtain cluster coordination situation information;

[0009] Through a preset intent-driven network, adaptive routing calculation is performed on each device in the cluster based on the cluster collaborative situation information to obtain a network resource scheduling strategy;

[0010] By using a preset space-time coding technology, channel multiplexing mapping is performed on a preset digital communication network based on the network resource scheduling strategy to obtain a channel multiplexing mapping matrix;

[0011] Based on the channel reuse mapping matrix, resource scheduling, allocation and control are performed on each device in the cluster to obtain a cluster control strategy.

[0012] Further, the acquiring of real-time status data of each device in the cluster and timestamping the real-time status data to obtain timestamped status data includes:

[0013] Perform multi-dimensional state acquisition on each device in the cluster through a distributed sensor network to obtain an original state data stream;

[0014] Performing spectrum analysis on the original state data stream to obtain spectrum state data;

[0015] Performing multi-resolution analysis on the spectrum state data through a preset data processing module to obtain multi-resolution state data;

[0016] By using a preset state matching algorithm, based on the multi-resolution state data, the state of each device in the cluster is matched to obtain matching state data, and the matching state data is classified to obtain classified state data, and the classified state data is used as real-time state data; wherein the classified state data includes network signal classification and data transmission classification;

[0017] The classified state data is timestamped to obtain preliminary time-stamped state data, and the preliminary time-stamped state data is visualized based on graph theory state visualization technology to obtain time-stamped state data.

[0018] Furthermore, the distributed state estimation is performed on the timestamp-marked state data by a preset SDN controller to obtain cluster coordination status information, including:

[0019] Using a preset SDN controller, performing cluster state analysis on the timestamp-marked state data to obtain a device state feature matrix;

[0020] Through a preset graph neural network, a topological relationship model is performed on each device in the cluster based on the device state feature matrix to obtain a device state relationship graph, and the nodes in the device state relationship graph are marked with importance to obtain a marked device state relationship graph; wherein the nodes are used as each device in the cluster;

[0021] Based on the device state relationship diagram with the mark, high-bandwidth stability analysis is performed on each device in the cluster to obtain stability analysis data;

[0022] If the stability analysis data is not within the preset stability range, the device state relationship diagram is dynamically adjusted through the adaptive topology reconstruction technology to obtain the adjusted device state relationship diagram;

[0023] If the stability analysis data is within a preset stability range, the local state information of each device is securely aggregated through a federated learning mechanism to obtain a global state vector, and a state prediction is performed based on the global state vector to obtain a predicted state vector;

[0024] Based on the predicted state vector and the device state relationship diagram or the adjusted device state relationship diagram, a dynamic evolution simulation is performed on each device in the cluster to obtain a future device state relationship diagram, and a community structure analysis is performed on the future device state relationship diagram to obtain a community division result; wherein the community division result is used to reflect the collaborative relationship between each device in the future cluster and the organizational structure of the cluster;

[0025] The global state vector, the predicted state vector and the community division result are estimated and recorded in a distributed manner through blockchain technology to obtain cluster collaborative situation information.

[0026] Furthermore, the preset intent-driven network performs adaptive routing calculation on each device in the cluster based on the cluster collaborative situation information to obtain a network resource scheduling strategy, including:

[0027] Through a preset intent-driven network, the cluster collaborative situation information is analyzed for intent to obtain a network demand representation, and ontology-based semantic reasoning is performed on the network demand representation to obtain an enhanced semantic network demand representation;

[0028] The enhanced semantic network demand representation is used to perform network topology virtualization mapping to obtain a virtual network topology, and the virtual network topology is divided into resources based on the network slicing technology to obtain a network slice resource pool; wherein the network slice resource pool includes multiple virtual network slices, each slice corresponding to a specific network demand;

[0029] By using a preset ant colony algorithm, initial routing planning is performed on each device in the cluster based on the network slice resource pool to obtain an optimized routing solution;

[0030] Performing path selection on the optimized routing solution through a software defined network controller to obtain an adaptive routing path;

[0031] Acquire network bandwidth resources, and allocate the network bandwidth resources based on the adaptive routing path to obtain a bandwidth allocation scheme;

[0032] The bandwidth allocation scheme, the adaptive routing path and the network slice resource pool are integrated to obtain a network resource scheduling strategy.

[0033] Furthermore, the channel multiplexing mapping is performed on the preset digital communication network based on the network resource scheduling strategy by using the preset space-time coding technology to obtain a channel multiplexing mapping matrix, including:

[0034] By using a preset space-time coding technology, the network resource scheduling strategy is subjected to space-time resource coding to obtain a space-time coding sequence, and a multi-antenna channel characteristic analysis is performed on the space-time coding sequence to obtain a multi-antenna channel characteristic matrix;

[0035] Performing frequency domain decomposition on the multi-antenna channel characteristic matrix through MIMO-OFDM technology to obtain a subcarrier resource matrix, and performing orthogonal multiplexing calculation on the subcarrier resource matrix to obtain an orthogonal multiplexing strategy;

[0036] Based on the beamforming technology, spatial resources are allocated to the preset digital communication network based on the orthogonal multiplexing strategy to obtain a beam multiplexing scheme;

[0037] Based on the beam multiplexing scheme, the channel capacity of the digital communication network is estimated to obtain a channel capacity parameter; wherein the channel capacity parameter includes a signal-to-noise ratio parameter and a throughput parameter;

[0038] Through spectrum correlation analysis, channel multiplexing mapping is performed on the digital communication network based on the channel capacity parameters to obtain a channel multiplexing mapping matrix.

[0039] Further, the estimating the channel capacity of the digital communication network based on the beam multiplexing scheme to obtain a channel capacity parameter includes:

[0040] Based on the beam multiplexing scheme, performing multipath channel analysis on the digital communication network to obtain a multipath channel feature vector, and performing Hilbert transform processing on the multipath channel feature vector to obtain a multidimensional channel phase feature set;

[0041] Performing nonlinear Vandermonde matrix decomposition on the multidimensional channel phase feature set to obtain channel state estimation parameters;

[0042] The channel state estimation parameters are orthogonalized by a preset Walsh transform algorithm to obtain an orthogonal channel eigenvector, and information entropy analysis is performed on the orthogonal channel eigenvector to obtain a channel entropy value matrix;

[0043] Based on the channel entropy matrix, the capacity constraint analysis of the digital communication network is performed to obtain the channel capacity constraint condition, and the channel capacity constraint condition is subjected to Lagrangian dual decomposition to obtain the channel capacity boundary parameter; wherein the channel capacity boundary parameter is the channel maximum capacity threshold and the minimum capacity guarantee value;

[0044] Performing Gauss-Markov chain analysis on the channel capacity boundary parameters to obtain a channel capacity transfer matrix;

[0045] The channel capacity transfer matrix is ​​solved by Poisson's equation through a preset Green's function algorithm to obtain a channel capacity distribution function, and the channel capacity is estimated based on the channel capacity distribution function and the channel capacity boundary parameters to obtain channel capacity parameters.

[0046] Furthermore, the resource scheduling, allocation and control of each device in the cluster based on the channel reuse mapping matrix to obtain a cluster control strategy includes:

[0047] Based on the channel multiplexing mapping matrix, a channel quality assessment is performed on the communication channels in the digital communication network to obtain a channel quality parameter matrix, and a channel state classification is performed on the channel quality parameter matrix to obtain a channel classification state; wherein the channel classification state includes a high-quality channel, a medium-quality channel and a low-quality channel;

[0048] Predicting the future channel state of the channel classification state through a preset long short-term memory network to obtain a predicted channel state table;

[0049] Performing initial resource allocation on each device based on the predicted channel state table to obtain an initial resource allocation scheme;

[0050] Performing dynamic resource scheduling on the initial resource allocation scheme through a multi-agent reinforcement learning algorithm to obtain a dynamic resource scheduling strategy;

[0051] Based on the dynamic resource scheduling strategy, control instructions are generated and compressed for each device to obtain a device control instruction set;

[0052] The device control instruction set is distributed and controlled through a software-defined network controller to obtain a cluster control strategy.

[0053] The present invention also provides a high-bandwidth cluster control system based on a digital communication network, comprising:

[0054] An acquisition module is used to acquire real-time status data of each device in the cluster, and timestamp the real-time status data to obtain timestamp status data;

[0055] An estimation module, used to perform distributed state estimation on the timestamp-marked state data through a preset SDN controller to obtain cluster coordination situation information;

[0056] A computing module, configured to perform adaptive routing calculations on each device in the cluster based on the cluster collaborative situation information through a preset intent-driven network to obtain a network resource scheduling strategy;

[0057] A mapping module, used to perform channel multiplexing mapping on a preset digital communication network based on the network resource scheduling strategy by using a preset space-time coding technology to obtain a channel multiplexing mapping matrix;

[0058] The control module is used to perform resource scheduling, allocation and control on each device in the cluster based on the channel reuse mapping matrix to obtain a cluster control strategy.

[0059] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of any one of the above methods when executing the computer program.

[0060] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods are implemented.

[0061] The high-bandwidth cluster control method based on a digital communication network provided by the present invention comprises the following steps: collecting real-time power consumption data of a target chip through a preset on-chip sensor array to obtain a target chip power consumption distribution diagram; wherein a storage and computing integrated chip is used as a target chip; collecting temperature data of the target chip through a temperature sensor network to obtain a target chip temperature distribution diagram; performing power consumption-temperature correlation analysis on the target chip based on the target chip power consumption distribution diagram and the target chip temperature distribution diagram to obtain a power consumption-temperature correlation matrix; performing power consumption management planning on the target chip based on the power consumption-temperature correlation matrix through a multi-objective optimization algorithm to obtain a dynamic power consumption management strategy; adjusting the voltage and frequency of the target chip based on the dynamic power consumption management strategy to obtain voltage and frequency control parameters; performing power consumption management on the target chip based on the voltage and frequency control parameters through an adaptive clock gating technology to obtain a chip power consumption management solution, thereby solving the technical problem that the prior art is usually unable to provide sufficient time accuracy when processing these large-scale data, thereby affecting the overall performance of the cluster, and realizing that by adopting space-time coding technology, the method can significantly improve spectrum efficiency without increasing additional hardware costs. The application of space-time coding technology enables channels in digital communication networks to be reused more efficiently. The resource allocation guided by the generated channel reuse mapping matrix can maximize the potential of the existing communication infrastructure, while enhancing the technical effects of system stability and anti-interference performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a schematic diagram of the steps of a high-bandwidth cluster control method based on a digital communication network in one embodiment of the present invention;

[0063] Figure 2 is a structural block diagram of a high-bandwidth cluster control system based on a digital communication network in one embodiment of the present invention;

[0064] Figure 3 It is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention.

[0065] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0067] like Figure 1 As shown, Figure 1It is a schematic diagram of the steps of a high-bandwidth cluster control method based on a digital communication network in one embodiment of the present invention;

[0068] In one embodiment of the present invention, a high-bandwidth cluster control method based on a digital communication network is provided, comprising the following steps:

[0069] Step S1, obtaining real-time status data of each device in the cluster, and timestamping the real-time status data to obtain timestamped status data.

[0070] Specifically, when obtaining the real-time status data of each device in the cluster and timestamping the real-time status data to obtain the timestamped status data, it is first necessary to ensure that each device can generate and send its own status information. This status information may include but is not limited to the operating parameters, working mode, health status, etc. of the device. For smart city or unmanned driving application scenarios in the Internet of Things (IoT) environment, these data are crucial. For example, in the application scenario of a smart city, street lights can adjust the brightness according to traffic flow to save energy; and in a cluster of unmanned vehicles, vehicles share location and speed information to maintain a safe distance. In order to ensure that all collected status data are comparable and synchronized, the system will add an accurate timestamp to each received data. This timestamp comes from a unified and high-precision time source, such as a network time protocol (NTP) server or a GPS satellite timing system, to ensure that the time base of all devices is consistent even in a distributed environment. In this way, even if there are slight time differences between different devices, they can be calibrated to the same timeline through timestamp marking, thereby accurately reflecting the state changes at each moment in the cluster. For example, in a swarm of driverless vehicles, when vehicle A detects an obstacle ahead and slows down, it immediately sends a timestamped status update to nearby vehicles B and C, allowing them to respond in a timely manner based on the latest situation, such as adjusting speed or changing driving paths. These timestamped status data are then transmitted to the central processing unit or distributed SDN controller, where they are further analyzed and processed. Since each piece of status data is accompanied by a timestamp, this provides a solid foundation for subsequent state estimation, allowing the system to accurately track the historical behavior patterns of devices and predict future trends. For example, in a smart city, by analyzing the timestamped status data returned by multiple sensors over a period of time, city managers can better understand the laws of traffic flow and optimize the control logic of traffic lights accordingly, thereby improving road traffic efficiency. In short, the process of timestamping real-time status data is not only the basis for cluster collaboration, but also an indispensable part of ensuring the efficient operation of the entire system.

[0071] Step S2: performing distributed state estimation on the timestamp-marked state data through a preset SDN controller to obtain cluster coordination status information.

[0072] Specifically, in the process of implementing the distributed state estimation of the timestamp-marked state data through the preset SDN controller to obtain the cluster collaborative situation information, it is first necessary to understand the role of the SDN (Software Defined Network) controller in this step. As the core component of network management and control, the SDN controller can not only centrally manage network resources, but also dynamically adjust and optimize network configuration to adapt to different application requirements. When it comes to devices in the cluster, the SDN controller is responsible for receiving timestamp-marked state data from each node, which contains the real-time operating status of the device and is marked with precise time tags. Next, in order to accurately obtain the cluster collaborative situation information, the SDN controller will use a distributed state estimation algorithm to process these timestamped data. Distributed state estimation is a method based on mathematical models and technologies for inferring the current state of the system from data collected from multiple sources. In this case, it allows the system to maintain efficiency and accuracy even in the face of complex and massive state updates generated by a large number of heterogeneous devices. For example, in the application scenario of smart cities, the SDN controller can use timestamp-marked state data from different sensors to build a traffic flow map of the entire city, and then analyze traffic bottlenecks and predict future trends. Similarly, in a cluster of unmanned vehicles, the controller can evaluate the safety and coordination of the cluster based on the timestamp-marked state data such as position and speed shared between vehicles to ensure that all vehicles can travel safely and smoothly. Furthermore, the distributed state estimation process performed by the SDN controller does not occur in isolation, but is completed in close cooperation with other nodes in the cluster. Each node may participate in the local state estimation, and then summarize the results to the SDN controller, which then integrates the information of all nodes to form a global view. This approach not only improves the speed of estimation, but also enhances the robustness of the system, because even if some nodes fail or data is lost, other nodes can still provide enough information to maintain the overall estimation accuracy. For example, in a cluster of unmanned vehicles, even if one vehicle temporarily loses its communication capability, the remaining vehicles can still maintain the normal operation of the cluster by exchanging timestamp-marked state data between them. Therefore, in this way, the SDN controller successfully achieves the effective acquisition of cluster collaborative situation information, providing a solid foundation for subsequent adaptive routing calculation and resource scheduling.

[0073] Step S3, through a preset intent-driven network, adaptive routing calculation is performed on each device in the cluster based on the cluster collaborative situation information to obtain a network resource scheduling strategy.

[0074] Specifically, in the process of implementing adaptive routing calculation for each device in the cluster based on the cluster collaborative situation information through a preset intent-driven network to obtain a network resource scheduling strategy, it is first necessary to clarify the core concept of intent-driven networking (IDN). Intent-driven networking aims to allow users to express their network needs and goals in a high-level, abstract manner without having to care about the underlying technical details. This network architecture enables the system to automatically adjust the network configuration and optimize the data flow path according to the intention of the user or application, thereby improving network performance and service quality. After the SDN controller has completed the distributed state estimation of the timestamp-marked state data and obtained the cluster collaborative situation information, the next step is to use this information to guide the operation of the intent-driven network. In this step, the system analyzes the optimal route selection under the current network environment based on the cluster collaborative situation information. For example, in the application scenario of a smart city, it is assumed that the travel experience of citizens is improved by reducing traffic congestion in a specific area. At this time, the intent-driven network will receive this high-level intent and automatically calculate the most effective vehicle guidance path in combination with the real-time traffic flow reflected by the timestamp-marked state data collected from each traffic sensor. For a cluster of unmanned vehicles, the intent-driven network can also dynamically plan the best route for each vehicle based on the cluster collaborative situation information, such as vehicle location, speed, and destination, to ensure the safe and efficient operation of the entire fleet. Furthermore, the adaptive routing calculation not only takes into account the current network status, but also predicts future trends so that corresponding adjustments can be made in advance. The intent-driven network uses machine learning algorithms or other advanced data analysis technologies to model factors such as network load and device status changes in the future. The benefit of this is that preventive measures can be taken before problems occur to avoid potential risks. For example, in an unmanned vehicle cluster, if it is predicted that a traffic accident is about to occur on the road ahead or the road is closed due to construction, the intent-driven network can re-plan a detour route for the relevant vehicles in advance to reduce waiting time and the possibility of secondary accidents. Ultimately, the network resource scheduling strategy formed through the above process will ensure smoother and more efficient communication between all devices, while maximizing the use of existing network resources. Whether it is traffic management in smart cities or the coordinated operation of unmanned vehicle clusters, the intent-driven network can respond quickly and flexibly based on the cluster collaborative situation information, ensuring the stability and reliability of the system, and also providing strong technical support for more intelligent application scenarios in the future.

[0075] Step S4, performing channel multiplexing mapping on the preset digital communication network based on the network resource scheduling strategy by using the preset space-time coding technology to obtain a channel multiplexing mapping matrix.

[0076] Specifically, in the process of implementing channel reuse mapping of a preset digital communication network based on the network resource scheduling strategy through a preset space-time coding technology to obtain a channel reuse mapping matrix, it is first necessary to understand the role of space-time coding (STC) technology and how it is combined with a digital communication network. Space-time coding is a technology used in wireless communication that uses multiple antennas to send information simultaneously, dispersing data streams in both time and space dimensions to improve transmission efficiency and reliability. When this technology is applied to cluster control methods, its purpose is to more efficiently manage and optimize channel resources to ensure smooth communication between all devices. Once the intent-driven network has formulated a network resource scheduling strategy based on the cluster collaborative situation information, the next step is to convert these strategies into specific action guidelines, that is, to perform channel reuse mapping through space-time coding technology. In this step, the system will determine which channels can be reused and how to best allocate these channels to each device in the cluster based on the network resource scheduling strategy. For example, in the application scenario of a smart city, assuming that the city management department wants to optimize the coordination of traffic lights, the system will use space-time coding technology to plan the most effective communication path between sensors in different sections. For a swarm of autonomous vehicles, this means selecting the optimal frequency band or time slot for each vehicle to exchange data, thereby avoiding interference and ensuring communication quality. Further, channel reuse mapping involves building a mapping matrix that defines which specific channel combination each device should use for communication. This process not only takes into account current network conditions, such as signal-to-noise ratio, interference level, etc., but also predicts possible future changes. In this way, the system can maintain good performance even when the network environment changes dynamically. For example, in a swarm of autonomous vehicles, if there is a temporary high-density communication demand in certain areas, such as a large number of vehicles concentrated near an intersection, space-time coding technology can adjust channel allocation in real time according to the latest network resource scheduling strategy to ensure that each vehicle can obtain enough bandwidth to transmit critical information, such as location updates or emergency brake warnings. Ultimately, a carefully designed channel reuse mapping matrix can significantly improve the utilization and stability of the entire digital communication network. Space-time coding technology plays a vital role in managing data transmission from numerous sensors in the complex environment of smart cities and coordinating safe driving between multiple vehicles in a swarm of autonomous vehicles. It not only enhances the system's anti-interference ability and reduces delays and packet loss, but also provides a solid technical foundation for future intelligent applications, enabling each member in the cluster to work together more efficiently and achieve higher-level goals together.

[0077] Step S5: performing resource scheduling, allocation and control on each device in the cluster based on the channel reuse mapping matrix to obtain a cluster control strategy.

[0078] Specifically, the process of scheduling and allocating resources and controlling each device in the cluster based on the channel reuse mapping matrix to obtain the cluster control strategy is a key link in the entire high-bandwidth cluster control system. This process first relies on the previously constructed channel reuse mapping matrix, which defines in detail the specific channel combination and corresponding communication time window that each device should use. Through such mapping, the system can ensure that when multiple devices communicate simultaneously, the signals between the devices will not interfere with each other, and the existing communication resources can be maximized. After the channel reuse mapping matrix is ​​established, the next step is to apply this information to the actual resource scheduling and control. At this stage, the system will allocate appropriate network resources, such as frequency bands, time slots, etc., to each device in the cluster according to the instructions in the matrix to ensure that they can operate under optimal conditions. For example, in the application scenario of smart cities, assuming that the data transmission of multiple traffic sensors needs to be coordinated, the system will determine which sensors should use which frequency to send data in what time period based on the channel reuse mapping matrix, thereby avoiding data loss or delay problems caused by frequency band conflicts. Similarly, in a cluster of unmanned vehicles, each vehicle exchanges information with surrounding vehicles or other infrastructure (such as traffic lights) according to a pre-set time and frequency arrangement to ensure driving safety and efficiency. Furthermore, in order to achieve effective resource scheduling, allocation and control, the system also needs to have a certain degree of flexibility and adaptability. This is because the cluster environment is often dynamically changing. For example, as new devices are added or existing devices move, the original channel reuse mapping may no longer be the optimal choice. Therefore, the system must be able to monitor changes in network status in real time and update the channel reuse mapping matrix in a timely manner to adjust the resource allocation strategy. For example, in a cluster of unmanned vehicles, if a vehicle suddenly changes its route and enters a busy intersection, the system needs to quickly re-evaluate and optimize the channel allocation in the area to cope with the increased communication needs and ensure that all vehicles can continue to communicate safely and smoothly. Finally, the cluster control strategy formed through the above steps not only ensures efficient collaboration between all devices in the cluster, but also enhances the overall stability and reliability of the system. Whether it is managing the data transmission of numerous sensors in the complex environment of smart cities, or coordinating the safe driving of multiple vehicles in a cluster of unmanned vehicles, this resource scheduling allocation and control method based on the channel reuse mapping matrix can provide strong support. It enables each device to use communication resources in an orderly manner according to predetermined rules, reducing unnecessary waiting time and energy consumption, while also laying a solid foundation for future intelligent applications and promoting more extensive technological progress and service innovation.

[0079] In a specific embodiment, the acquiring of real-time status data of each device in the cluster and timestamping the real-time status data to obtain timestamped status data includes:

[0080] Perform multi-dimensional state acquisition on each device in the cluster through a distributed sensor network to obtain an original state data stream;

[0081] Performing spectrum analysis on the original state data stream to obtain spectrum state data;

[0082] Performing multi-resolution analysis on the spectrum state data through a preset data processing module to obtain multi-resolution state data;

[0083] By using a preset state matching algorithm, based on the multi-resolution state data, the state of each device in the cluster is matched to obtain matching state data, and the matching state data is classified to obtain classified state data, and the classified state data is used as real-time state data; wherein the classified state data includes network signal classification and data transmission classification;

[0084] The classified state data is timestamped to obtain preliminary time-stamped state data, and the preliminary time-stamped state data is visualized based on graph theory state visualization technology to obtain time-stamped state data.

[0085] Specifically, in order to achieve the above-described "obtaining the real-time status data of each device in the cluster, and timestamping the real-time status data to obtain the timestamped status data", this process involves a series of closely connected technical steps. First, the multi-dimensional state acquisition of each device in the cluster is performed through a distributed sensor network to obtain the original state data stream. The distributed sensor network consists of multiple sensors deployed on each device, which can simultaneously monitor multiple physical parameters such as temperature, humidity, vibration, etc., thereby providing comprehensive state information. For example, in an autonomous vehicle, the FPC sensor module and its related components constitute a complex device cluster, and the working state of each component is crucial to the stable operation of the entire system. Through this multi-dimensional state acquisition, we can ensure that we capture every subtle change. Once the original state data stream is obtained, the next step is to perform spectral analysis on these data to obtain spectral state data. Spectral analysis is a signal processing technique that converts time domain signals into frequency domain representations, which helps to identify periodic and non-periodic components hidden in complex waveforms. This is very useful for understanding the dynamic behavior inside the device because it can reveal which frequency components have the greatest impact on device performance. For example, in an FPC module, vibration or electromagnetic interference at certain specific frequencies may cause abnormal temperature fluctuations or signal distortion. Through spectrum analysis, we can detect these problems in advance and take corresponding measures to solve them. Then, the spectrum state data is analyzed by multi-resolution using the preset data processing module to obtain multi-resolution state data. Multi-resolution analysis is a method that can simultaneously process features of different scales, which can simplify the data structure without losing details. This method is particularly suitable for processing large-scale data sets from high-density sensor networks because it can effectively filter out noise and redundant information and highlight key patterns. For example, in an autonomous driving scenario, when a vehicle is driving on different road conditions, the FPC module may face a variety of environmental challenges, such as sudden temperature changes, mechanical shocks, etc. Through multi-resolution analysis, we can see more clearly the impact of these changes on the device state, thereby providing a basis for subsequent state matching. With the multi-resolution state data, the next step is to process it through a preset state matching algorithm to obtain matching state data, classify the data into states, and finally obtain classified state data. The state matching algorithm aims to compare the currently observed state with a known pattern library to find the most similar result. Classification further refines these matching results and classifies them into different categories, such as network signal classification and data transmission classification. For example, in the FPC module, if the network signal strength in a certain area is detected to drop suddenly, or the data transmission rate fluctuates abnormally, this may be caused by hardware failure or external interference. By accurately classifying these situations, engineers can quickly locate the root cause of the problem and take appropriate maintenance measures.Finally, the classified state data is timestamped to obtain preliminary timestamped state data. The timestamp not only records the exact moment when each state occurs, but also provides a time reference framework for subsequent data analysis. Based on the graph theory state visualization technology, these preliminary timestamped state data can be presented in an intuitive way to form timestamped state data. Graph theory state visualization technology constructs a graphical model of nodes and edges to show the relationship between devices and the trend of state changes over time. For example, in the monitoring system of an autonomous driving vehicle, operators can use such a visual interface to track the working status of the FPC module and its related components in real time, discover potential risk points in time, and respond quickly. In this way, even in the face of complex and changing working environments, the entire cluster can be ensured to be in the best operating state at all times. In summary, the above steps together constitute a complete process from raw state data acquisition to timestamped state data generation. Each step is closely linked and interlocked, ensuring a comprehensive understanding and precise control of the status of each device in the cluster. Through this systematic analysis method, even in the face of complex and changing application scenarios, reliable monitoring support can be provided for the autonomous driving platform to ensure its stability and safety.

[0086] In a specific embodiment, the distributed state estimation is performed on the timestamp-marked state data by a preset SDN controller to obtain cluster coordination status information, including:

[0087] Using a preset SDN controller, performing cluster state analysis on the timestamp-marked state data to obtain a device state feature matrix;

[0088] Through a preset graph neural network, a topological relationship model is performed on each device in the cluster based on the device state feature matrix to obtain a device state relationship graph, and the nodes in the device state relationship graph are marked with importance to obtain a marked device state relationship graph; wherein the nodes are used as each device in the cluster;

[0089] Based on the device state relationship diagram with the mark, high-bandwidth stability analysis is performed on each device in the cluster to obtain stability analysis data;

[0090] If the stability analysis data is not within the preset stability range, the device state relationship diagram is dynamically adjusted through the adaptive topology reconstruction technology to obtain the adjusted device state relationship diagram;

[0091] If the stability analysis data is within a preset stability range, the local state information of each device is securely aggregated through a federated learning mechanism to obtain a global state vector, and a state prediction is performed based on the global state vector to obtain a predicted state vector;

[0092] Based on the predicted state vector and the device state relationship diagram or the adjusted device state relationship diagram, a dynamic evolution simulation is performed on each device in the cluster to obtain a future device state relationship diagram, and a community structure analysis is performed on the future device state relationship diagram to obtain a community division result; wherein the community division result is used to reflect the collaborative relationship between each device in the future cluster and the organizational structure of the cluster;

[0093] The global state vector, the predicted state vector and the community division result are estimated and recorded in a distributed manner through blockchain technology to obtain cluster collaborative situation information.

[0094] Specifically, in the process of implementing the distributed state estimation of the timestamp-marked state data through the preset SDN controller to obtain the cluster collaborative situation information, it is first necessary to use the preset SDN controller to process the timestamp-marked state data collected from each device. This process begins with cluster state parsing of these data, with the purpose of extracting information that can reflect the current status of each device to form a device state feature matrix. This matrix not only records the state parameters of a single device, such as working mode, health status, etc., but also includes the relative position and connection relationship between devices. For example, in the application scenario of smart cities, the SDN controller can analyze data from different sensors to build a comprehensive city infrastructure operation status diagram; in a cluster composed of unmanned vehicles, it integrates the speed, direction, distance and other information of each vehicle to generate a real-time dynamic model of the fleet. Next, in order to more deeply understand the internal structure of the cluster and the interaction mode between its members, the system will further process the device state feature matrix through a preset graph neural network (GNN). Graph neural network is an artificial intelligence algorithm specially designed for processing graph data, which can discover hidden patterns in complex relationship networks. Based on the device state feature matrix, GNN will model the topological relationship of each device in the cluster and obtain a device state relationship graph. In this process, each node represents a device in the cluster, and the edge represents the communication or physical connection between devices. In addition, the system will mark the nodes in the device state relationship graph with importance, and obtain a device state relationship graph with labels. This step is crucial for identifying key nodes because it can help determine which devices play a more important role in the cluster. For example, in the traffic management system of a smart city, some traffic lights located at major intersections may be more important than those in other places; similarly, in a cluster of unmanned vehicles, vehicles in leadership positions or responsible for special tasks will also be given higher weights. Once the device state relationship graph with labels is available, the next step is to perform high-bandwidth stability analysis on each device in the cluster to evaluate the robustness and reliability of the entire system. This analysis aims to check whether there are factors that may cause performance degradation, such as network congestion, signal interference, etc., and generate stability analysis data. If it is found that the stability analysis data is not within the preset stability range, that is, the system exhibits unstable behavior, then measures need to be taken to improve it. At this time, adaptive topology reconstruction technology will come into play. It can automatically adjust the device status relationship diagram according to the current situation to ensure that all devices can maintain an efficient communication environment.For example, in a smart city, when a wireless sensor network in a certain area fails, the smooth data transmission can be restored by changing the connection mode of some nodes; in a cluster of unmanned vehicles, if the communication link quality between two vehicles is detected to be poor, the system can replan their driving paths to keep them within a better communication range. On the contrary, if the stability analysis data shows that the system is already within the preset stability range, then it can enter the next stage - securely aggregating the local state information of each device through the federated learning mechanism. Federated learning is a privacy-preserving machine learning method that allows multiple participants to jointly train models without sharing original data. In this case, it is used to collect and fuse the local state information of each device in the cluster to form a global state vector. This vector comprehensively reflects the overall status of the entire cluster at a certain moment, providing a basis for further state prediction. For example, in the power distribution network of a smart city, federated learning can help substations in different regions to safely exchange power demand information, thereby optimizing power dispatch across the city; and in a cluster of unmanned vehicles, it allows vehicles to share road condition information while protecting passenger privacy, improving overall driving safety. Afterwards, based on the global state vector, the system will perform state prediction and obtain a predicted state vector. This prediction not only takes into account the current state, but also combines historical trends and external factors to provide guidance for future device behavior. For example, in smart cities, by analyzing factors such as weather forecasts and holiday arrangements, changes in traffic flow can be predicted in advance and corresponding preparations can be made; in unmanned vehicle clusters, the prediction function can help drivers foresee potential road hazards and take evasive measures in advance. Finally, based on the predicted state vector and the device state relationship diagram (or the adjusted device state relationship diagram), the system will simulate the dynamic evolution of each device in the cluster to obtain the future device state relationship diagram, and perform community structure analysis on this diagram to obtain the community division result. The community division result here is used to reflect the collaborative relationship between each device in the future cluster and the organizational structure of the cluster. For example, in the public security monitoring system of smart cities, community structure analysis can reveal which areas of cameras form a close cooperation group, which helps to respond quickly to emergencies; and in unmanned vehicle clusters, it can help identify which vehicles form temporary teams to complete specific tasks together. In order to ensure the security and credibility of all the above information, the system will use blockchain technology to perform distributed state calculations and consensus records on the global state vector, predicted state vector, and community division results, and finally obtain cluster coordination situation information. As a decentralized ledger technology, blockchain ensures that all participants can verify and trust the recorded information, while preventing the possibility of data tampering.For example, in smart cities, blockchain can be used to track every status update of public service facilities to ensure that citizens can obtain the most accurate service information; in driverless vehicle clusters, it can serve as a reliable witness for inter-vehicle communication, enhancing users' confidence in autonomous driving technology. In summary, through this series of complex steps and technical means, the system not only achieves accurate monitoring and efficient management of cluster status, but also lays a solid foundation for future intelligent applications.

[0095] In a specific embodiment, the preset intent-driven network performs adaptive routing calculation on each device in the cluster based on the cluster collaborative situation information to obtain a network resource scheduling strategy, including:

[0096] Through a preset intent-driven network, the cluster collaborative situation information is analyzed for intent to obtain a network demand representation, and ontology-based semantic reasoning is performed on the network demand representation to obtain an enhanced semantic network demand representation;

[0097] The enhanced semantic network demand representation is used to perform network topology virtualization mapping to obtain a virtual network topology, and the virtual network topology is divided into resources based on the network slicing technology to obtain a network slice resource pool; wherein the network slice resource pool includes multiple virtual network slices, each slice corresponding to a specific network demand;

[0098] By using a preset ant colony algorithm, initial routing planning is performed on each device in the cluster based on the network slice resource pool to obtain an optimized routing solution;

[0099] Performing path selection on the optimized routing solution through a software defined network controller to obtain an adaptive routing path;

[0100] Acquire network bandwidth resources, and allocate the network bandwidth resources based on the adaptive routing path to obtain a bandwidth allocation scheme;

[0101] The bandwidth allocation scheme, the adaptive routing path and the network slice resource pool are integrated to obtain a network resource scheduling strategy.

[0102] Specifically, in order to realize the process of adaptively calculating the routing of each device in the cluster based on the cluster collaborative situation information through the preset intent-driven network and obtaining the network resource scheduling strategy, it is first necessary to understand how to use the intent-driven network to process and parse this information. The intent-driven network is a technical architecture that can understand and respond to the user's high-level intentions, which allows the system to automatically adjust the network configuration according to the user's business needs. In this case, the system will perform intent parsing based on the cluster collaborative situation information obtained in the previous step, that is, information about the future collaborative relationship and organizational structure of the devices in the cluster. This parsing process aims to extract a clear network demand representation from complex situation information. For example, in the application scenario of smart cities, it may be necessary to identify which areas of infrastructure will face higher service requests during a specific period of time; in the unmanned vehicle cluster, it is to determine the type of tasks that the fleet will perform and the communication quality required. Once the network demand representation is obtained, the next step is to perform semantic reasoning based on ontology to generate an enhanced semantic network demand representation. Ontology here refers to a conceptual model within a domain, which describes the entities in the domain, their relationships, and attributes. Semantic reasoning is the process of deriving new information from known information using logical rules. In this example, by applying ontology, the meaning behind each network requirement can be understood more accurately and converted into a form that the system can understand and operate. For example, in a smart city, if it is found that traffic flow in a certain area is expected to increase significantly, the system not only knows that the bandwidth capacity of the area needs to be increased, but also can further infer the impact that this may have on neighboring areas, so as to prepare in advance; in a cluster of unmanned vehicles, if it is known that a group of vehicles will perform emergency rescue missions, the system can realize that these vehicles will need higher priority communication links to ensure the smooth completion of the mission. Subsequently, the network topology virtualization mapping is performed using the enhanced semantic network requirement representation to obtain the virtual network topology. The so-called network topology virtualization here refers to the creation of one or more virtual network environments that correspond to the physical network but operate independently. The virtual network topology can change dynamically according to different business needs without affecting the underlying physical network structure. Afterwards, the virtual network topology is divided into resources based on the network slicing technology to obtain the network slice resource pool. Network slicing is a 5G network technology that allows multiple logically isolated virtual networks to share the same physical network infrastructure. Each virtual network slice can be customized to meet specific business needs, such as low latency, high bandwidth or large-scale connections. For example, in the power distribution network of a smart city, exclusive network slices can be created for different types of loads (residential electricity, industrial electricity) to ensure their respective service quality; in a cluster of unmanned vehicles, different slices are set up for real-time data transmission and non-real-time data updates to ensure that the data flow of critical tasks is not disturbed.Next, the preset ant colony algorithm is used to perform initial routing planning for each device in the cluster based on the network slice resource pool to obtain an optimized routing solution. The ant colony algorithm is a heuristic search algorithm that imitates the foraging behavior of ants. It is particularly suitable for solving path optimization problems. In this process, the algorithm will find the most effective path combination among all possible routing options, considering factors including but not limited to distance, cost, and latency. For example, in the logistics distribution system of a smart city, the ant colony algorithm can help plan the optimal cargo transportation route, reduce traffic congestion and energy consumption; in a cluster of unmanned vehicles, it is to find a driving path that allows the members of the fleet to maintain the best communication efficiency. Then, the software-defined network (SDN) controller intervenes to select the path for the above-mentioned optimized routing solution and finally determine the adaptive routing path. As the brain of the entire network, the SDN controller can centrally manage and dynamically adjust the allocation of network resources. After receiving multiple candidate routing solutions provided by the ant colony algorithm, the SDN controller will make decisions based on factors such as the current network status and historical performance data to select the most suitable routing path. This process is critical to ensure efficient operation of the network, especially in the face of rapidly changing workloads, such as during large-scale events in smart cities or when unmanned vehicle clusters respond to emergencies. Obtain network bandwidth resources and allocate these resources based on adaptive routing paths to form a bandwidth allocation plan. Since network resources are limited, they must be allocated reasonably to maximize overall benefits. On this basis, game theory techniques are also used to compete and negotiate for resources on bandwidth allocation plans to ensure that all demand parties can accept the results. Game theory provides a set of methods for analyzing strategic interactions between participants, which are applied here to resolve potential competition conflicts between different network slices. For example, in the public safety monitoring system of a smart city, when an emergency occurs, it may be necessary to temporarily increase the bandwidth of certain cameras. At this time, it is necessary to negotiate with other regular monitoring points to readjust the bandwidth allocation; similarly, in a cluster of unmanned vehicles, if some vehicles are performing special tasks, they need to coordinate resources with other ordinary vehicles to ensure that important tasks are completed first. Finally, the bandwidth allocation plan, adaptive routing path and network slice resource pool are integrated to form a complete network resource scheduling strategy. This strategy should not only take into account current needs, but also predict future trends to ensure that the network can continue to operate stably in a changing environment. For example, in smart cities, such scheduling strategies can help city managers better plan the layout and operation of public service facilities; in unmanned vehicle clusters, they can improve the overall efficiency of the fleet and promote a more intelligent and efficient transportation experience. In summary, through a series of sophisticated designs and technical means, the system has achieved effective utilization of cluster collaborative situation information, thereby achieving flexible and efficient network resource scheduling.

[0103] In a specific embodiment, the channel multiplexing mapping is performed on the preset digital communication network based on the network resource scheduling strategy by using the preset space-time coding technology to obtain a channel multiplexing mapping matrix, including:

[0104] By using a preset space-time coding technology, the network resource scheduling strategy is subjected to space-time resource coding to obtain a space-time coding sequence, and a multi-antenna channel characteristic analysis is performed on the space-time coding sequence to obtain a multi-antenna channel characteristic matrix;

[0105] Performing frequency domain decomposition on the multi-antenna channel characteristic matrix through MIMO-OFDM technology to obtain a subcarrier resource matrix, and performing orthogonal multiplexing calculation on the subcarrier resource matrix to obtain an orthogonal multiplexing strategy;

[0106] Based on the beamforming technology, spatial resources are allocated to the preset digital communication network based on the orthogonal multiplexing strategy to obtain a beam multiplexing scheme;

[0107] Based on the beam multiplexing scheme, the channel capacity of the digital communication network is estimated to obtain a channel capacity parameter; wherein the channel capacity parameter includes a signal-to-noise ratio parameter and a throughput parameter;

[0108] Through spectrum correlation analysis, channel multiplexing mapping is performed on the digital communication network based on the channel capacity parameters to obtain a channel multiplexing mapping matrix.

[0109] Specifically, in order to realize the process of performing channel multiplexing mapping on the preset digital communication network based on the network resource scheduling strategy through the preset space-time coding technology and obtaining the channel multiplexing mapping matrix, it is first necessary to understand how to use space-time coding technology to process and optimize the network resource scheduling strategy. Space-time coding (STC) is a technology used in wireless communication that can disperse data streams in two dimensions of time and space to improve transmission efficiency and reliability. In this step, the system will perform space-time resource coding according to the previously determined network resource scheduling strategy, that is, information about how to optimally allocate resources such as bandwidth and routing, so as to generate a space-time coding sequence. This process aims to convert abstract scheduling strategies into specific operation instructions that can guide actual communication behaviors. For example, in the application scenario of smart cities, it is assumed that the optimal communication paths between various sensors and between them and the data center have been planned; in the unmanned vehicle cluster, the efficient communication scheme between each vehicle and other vehicles or infrastructure is determined. Next, the system will perform multi-antenna channel feature analysis on these space-time coding sequences to obtain a multi-antenna channel feature matrix. The multi-antenna channel feature analysis is to better understand the specific characteristics of each communication link, such as signal strength, propagation delay, etc. This is because using multiple antennas to send and receive signals can significantly improve communication performance, especially in complex environments. In this way, the system can identify which antenna combinations are best suited for specific mission requirements. For example, in a smart city's traffic monitoring system, some cameras may be installed in special locations, resulting in different channel conditions between them and the base station than other devices; in a cluster of unmanned vehicles, the channel conditions will change in different time periods due to the constant movement of vehicles. Therefore, a detailed analysis of these channel characteristics is essential to ensure high-quality communication. Then, the multi-antenna channel characteristic matrix is ​​decomposed in the frequency domain through MIMO-OFDM (Multiple Input Multiple Output - Orthogonal Frequency Division Multiplexing) technology to obtain a subcarrier resource matrix, and this matrix is ​​orthogonally multiplexed to obtain an orthogonal multiplexing strategy. MIMO-OFDM combines two advanced wireless communication technologies: MIMO allows multiple antennas to work simultaneously, while OFDM divides the available spectrum into multiple independent subcarriers, so that the data on each subcarrier can be transmitted independently. This combination of technologies not only improves spectrum utilization, but also effectively combats frequency selective fading. By performing orthogonal multiplexing calculations on the subcarrier resource matrix, the system can ensure that different data streams do not interfere with each other at the same time. For example, in the broadband network of a smart city, MIMO-OFDM can help multiple users share the same spectrum without reducing the quality of service; in a cluster of driverless vehicles, it can ensure clear communication between fleet members and maintain a stable connection even at high speeds.Furthermore, based on the beamforming technology, the system will allocate spatial resources for the preset digital communication network according to the orthogonal multiplexing strategy obtained above, and obtain the beam multiplexing scheme. Beamforming is a signal processing method that allows the transmitter to adjust the directionality and strength of the signal in order to more accurately aim at the receiver. This not only increases the communication distance, but also reduces sidelobe interference, improving the overall performance of the system. For example, in large-scale public events in smart cities, beamforming allows temporarily deployed wireless access points to provide services to the audience area instead of wasting energy in unmanned areas; in unmanned vehicle clusters, when the fleet passes through tunnels or other areas with severe signal shielding, beamforming technology can enhance the signal strength in a specific direction to ensure that the in-vehicle equipment continues to work normally. Afterwards, based on the beam multiplexing scheme, the system will estimate the channel capacity of the digital communication network and obtain the channel capacity parameters. Channel capacity refers to the maximum information transmission rate that a communication channel can support under given conditions, and is usually measured by two key indicators: signal-to-noise ratio (SNR) and throughput. Accurate channel capacity estimation is very important for evaluating the effectiveness of existing network configurations and planning future expansion. For example, in IoT applications in smart cities, channel capacity parameters can help operators decide whether to add new base stations or optimize existing facilities; in unmanned vehicle clusters, it provides a basis for the fleet to set reasonable driving speeds and spacing to avoid safety hazards caused by communication bottlenecks. Finally, through spectral correlation analysis, the system performs channel reuse mapping on the digital communication network based on the channel capacity parameters to obtain a channel reuse mapping matrix. Spectral correlation analysis is a mathematical tool that can reveal the internal structure of a data set and its interrelationships. In this case, by analyzing the correlation patterns between channel capacity parameters, the system can find the best way to allocate different types of traffic to the corresponding channels to form a channel reuse mapping matrix. This matrix defines which specific channel combination each device should use and the corresponding communication time window to ensure smooth operation of the entire network. For example, in smart grid management in smart cities, the channel reuse mapping matrix can ensure that the communication between power metering equipment and the control center is both fast and secure; in unmanned vehicle clusters, it allows each vehicle to obtain sufficient bandwidth to transmit key information, such as location updates or emergency brake warnings, thereby promoting a more intelligent and efficient transportation experience. In summary, through a series of precise designs and technical means, the system has realized the transformation from network resource scheduling strategy to specific channel multiplexing mapping, which not only improves communication efficiency and reliability, but also lays a solid technical foundation for future intelligent application scenarios.

[0110] In a specific embodiment, the estimating the channel capacity of the digital communication network based on the beam multiplexing scheme to obtain a channel capacity parameter includes:

[0111] Based on the beam multiplexing scheme, performing multipath channel analysis on the digital communication network to obtain a multipath channel feature vector, and performing Hilbert transform processing on the multipath channel feature vector to obtain a multidimensional channel phase feature set;

[0112] Performing nonlinear Vandermonde matrix decomposition on the multidimensional channel phase feature set to obtain channel state estimation parameters;

[0113] The channel state estimation parameters are orthogonalized by a preset Walsh transform algorithm to obtain an orthogonal channel eigenvector, and information entropy analysis is performed on the orthogonal channel eigenvector to obtain a channel entropy value matrix;

[0114] Based on the channel entropy matrix, the capacity constraint analysis of the digital communication network is performed to obtain the channel capacity constraint condition, and the channel capacity constraint condition is subjected to Lagrangian dual decomposition to obtain the channel capacity boundary parameter; wherein the channel capacity boundary parameter is the channel maximum capacity threshold and the minimum capacity guarantee value;

[0115] Performing Gauss-Markov chain analysis on the channel capacity boundary parameters to obtain a channel capacity transfer matrix;

[0116] The channel capacity transfer matrix is ​​solved by Poisson's equation through a preset Green's function algorithm to obtain a channel capacity distribution function, and the channel capacity is estimated based on the channel capacity distribution function and the channel capacity boundary parameters to obtain channel capacity parameters.

[0117] Specifically, in order to realize the process of estimating the channel capacity of the digital communication network based on the beam multiplexing scheme and obtaining the channel capacity parameters, it is first necessary to understand how to perform multipath channel analysis on the network according to the beam multiplexing scheme. Multipath channels refer to multiple propagation paths formed by wireless signals during transmission due to reflection, refraction, etc. These paths will cause delay spread and phase change in the signal received by the receiving end, thereby affecting the communication quality. In this step, the system will analyze the multipath channel characteristics in the digital communication network based on the previously determined beam multiplexing scheme, that is, information about how to optimally allocate spatial resources. Specifically, by performing detailed measurements and calculations on each communication link, a multipath channel feature vector describing the multipath effect can be obtained. For example, in the traffic monitoring system of a smart city, the multipath effect between the camera and the base station may become complicated due to the presence of surrounding buildings; and in a cluster of unmanned vehicles, the relative positions between vehicles are constantly changing, which will also lead to dynamic adjustment of the multipath channel characteristics. Next, the system will perform Hilbert transform processing on the multipath channel feature vector to extract a multidimensional channel phase feature set. The Hilbert transform is a mathematical tool that can transform a real-valued signal into an analytical signal, thereby revealing the instantaneous amplitude and phase information of the signal. Through this transformation, the system can obtain a richer description of the channel characteristics, especially the capture of phase information, which is crucial for the subsequent channel state estimation. For example, in the application scenario of smart cities, the multidimensional channel phase feature set after the Hilbert transform can help more accurately identify which factors affect the communication quality between a specific sensor and a data center; in a cluster of unmanned vehicles, it helps to better predict the possible interference on the communication link between vehicles. Then, the multidimensional channel phase feature set is subjected to nonlinear Vandermonde matrix decomposition to obtain the channel state estimation parameters. Vandermonde matrix decomposition is a technique for solving polynomial interpolation problems, which is used here to extract key channel state information from complex multidimensional data. This process aims to simplify the data structure while retaining the part that best reflects the channel characteristics. For example, in the broadband network of smart cities, through nonlinear Vandermonde matrix decomposition, it is possible to more clearly understand which frequency bands are severely affected by multipath fading; in a cluster of unmanned vehicles, it can accurately evaluate the status of each communication link under different driving conditions, providing a basis for optimizing communication strategies. Furthermore, the system orthogonalizes the channel state estimation parameters through the preset Walsh transform algorithm to obtain orthogonal channel eigenvectors, and performs information entropy analysis on these eigenvectors to obtain the channel entropy value matrix. Walsh transform is a discrete transform technology that can decompose a set of signals into a set of orthogonal basis functions. Through this transformation, the system can ensure the independence and non-interference between different channels, which is very important for improving communication efficiency.Information entropy analysis is used to measure the degree of uncertainty of the system. A higher entropy value means more randomness and unpredictability. For example, in the IoT application of smart cities, the channel entropy matrix can help identify which areas have a relatively stable communication environment and which areas may have potential risks; in the unmanned vehicle cluster, it provides a scientific basis for the fleet to formulate reasonable driving speeds and spacing to avoid safety hazards caused by communication bottlenecks. After that, based on the channel entropy matrix, the system performs capacity constraint analysis on the digital communication network to obtain the channel capacity constraint conditions. The capacity constraint analysis here is to determine the maximum data transmission rate that the network can support under given channel conditions. This step not only takes into account the current channel status, but also combines historical data and future trend forecasts to ensure the accuracy of the estimation results. Next, the channel capacity constraint conditions are Lagrangian dual decomposition to obtain the channel capacity boundary parameters. Lagrangian dual decomposition is an optimization method that allows us to decompose a complex constraint problem into several sub-problems that are easier to solve. In this way, the system can find the channel maximum capacity threshold and the minimum capacity guarantee value, which together define the boundary range of the channel capacity. For example, in the power distribution network of a smart city, the channel capacity boundary parameters can help operators decide whether to add new base stations or optimize existing facilities; in a cluster of unmanned vehicles, it provides a guarantee for efficient communication between fleet members, ensuring that good service quality can be maintained even during peak network load periods. Next, the system performs a Gauss-Markov chain analysis on the channel capacity boundary parameters to obtain the channel capacity transfer matrix. The Gauss-Markov chain is a probabilistic model that assumes that the future state depends only on the current state and is not affected by the past state. Through this analysis, the system can predict the changing trend of the channel capacity in the future and make corresponding adjustments in advance. For example, in the public safety monitoring system of a smart city, when an emergency occurs, it may be necessary to temporarily increase the bandwidth of certain cameras. At this time, it is necessary to negotiate with other regular monitoring points to readjust the bandwidth allocation; similarly, in a cluster of unmanned vehicles, if some vehicles are performing special tasks, they need to coordinate resources with other ordinary vehicles to ensure that important tasks are completed first. Finally, the channel capacity transfer matrix is ​​solved by the Poisson equation using the preset Green's function algorithm to obtain the channel capacity distribution function. The Green's function is a method used to solve partial differential equations, which is used here to simulate the changing law of channel capacity. The purpose of solving the Poisson equation is to find a solution that satisfies specific boundary conditions. In this way, the system can obtain a complete channel capacity distribution diagram, which reflects the probability distribution of channel capacity in different time periods. Based on the channel capacity distribution function and channel capacity boundary parameters, the channel capacity is estimated, and the channel capacity parameters are finally obtained.For example, in smart grid management in smart cities, channel capacity parameters can ensure that communication between power metering equipment and the control center is fast and secure; in driverless vehicle clusters, each vehicle can obtain sufficient bandwidth to transmit key information, such as location updates or emergency brake warnings, thereby promoting a more intelligent and efficient transportation experience. In summary, through a series of sophisticated designs and technical means, the system has achieved a comprehensive estimation from beam reuse schemes to specific channel capacity parameters, which not only improves communication efficiency and reliability, but also lays a solid technical foundation for future intelligent application scenarios.

[0118] In a specific embodiment, the resource scheduling, allocation and control of each device in the cluster based on the channel reuse mapping matrix to obtain a cluster control strategy includes:

[0119] Based on the channel multiplexing mapping matrix, a channel quality assessment is performed on the communication channels in the digital communication network to obtain a channel quality parameter matrix, and a channel state classification is performed on the channel quality parameter matrix to obtain a channel classification state; wherein the channel classification state includes a high-quality channel, a medium-quality channel and a low-quality channel;

[0120] Predicting the future channel state of the channel classification state through a preset long short-term memory network to obtain a predicted channel state table;

[0121] Performing initial resource allocation on each device based on the predicted channel state table to obtain an initial resource allocation scheme;

[0122] Performing dynamic resource scheduling on the initial resource allocation scheme through a multi-agent reinforcement learning algorithm to obtain a dynamic resource scheduling strategy;

[0123] Based on the dynamic resource scheduling strategy, control instructions are generated and compressed for each device to obtain a device control instruction set;

[0124] The device control instruction set is distributed and controlled through a software-defined network controller to obtain a cluster control strategy.

[0125] Specifically, in order to implement the resource scheduling, allocation and control of each device in the cluster based on the channel reuse mapping matrix and obtain the cluster control strategy, it is first necessary to understand how to use the channel reuse mapping matrix to evaluate the communication channel quality in the digital communication network. The channel reuse mapping matrix defines which specific channel combination each device should use and the corresponding communication time window, so it provides a basis for channel quality evaluation. The system collects performance indicators about each channel according to this matrix, such as bit error rate, signal strength, etc., to form a channel quality parameter matrix. Through comprehensive analysis of these parameters, the channels can be classified into high-quality channels, medium-quality channels and low-quality channels, that is, channel classification states. For example, in the traffic monitoring system of a smart city, the channel between some cameras and base stations may have poor quality due to building obstruction; and in a cluster of unmanned vehicles, the relative position changes between vehicles may cause channel quality fluctuations in different time periods. Next, the system predicts the future channel state of the channel classification state through a preset long short-term memory network (Long Short-Term Memory, LSTM) to obtain a predicted channel state table. LSTM is a neural network architecture that is particularly suitable for processing sequence data. It can capture long-term dependencies and is very suitable for predicting future channel states. In this step, the LSTM model learns the time series pattern of the historical channel classification state and infers the channel status in the future. For example, in the application scenario of smart cities, if the wireless sensor network in a certain area experiences a short signal weakening at the same time every day, LSTM can predict this phenomenon in advance to help the system prepare; in the unmanned vehicle cluster, it can foresee the possible changes in the communication link between the members of the fleet to ensure safe driving. Then, the initial resource allocation is performed on each device based on the predicted channel state table to obtain the initial resource allocation plan. This step aims to reasonably plan the resource requirements of each device, including bandwidth, time slots, etc., according to the predicted future channel state. The initial resource allocation plan should not only consider the current demand, but also combine the expectation of future trends to avoid potential waste or shortage of resources. For example, during the public activities of smart cities, it is predicted that the communication demand in some areas will increase significantly. The system can adjust the resource allocation in advance to ensure the smooth progress of the activities; in the unmanned vehicle cluster, if it is expected that the road ahead will be congested, the system can pre-optimize the communication path between vehicles to reduce delays. Further, the initial resource allocation plan is dynamically scheduled through the multi-agent reinforcement learning algorithm to obtain a dynamic resource scheduling strategy. Multi-agent reinforcement learning is a method that allows multiple agents to collaborate and learn optimal behaviors through trial and error. In this process, each device is regarded as an agent, and they constantly adjust their behavior strategies according to their own task requirements and changes in the surrounding environment.For example, in the power distribution network of a smart city, multiple substations can find the most efficient way to transmit power through reinforcement learning; in a cluster of unmanned vehicles, fleet members can negotiate the best driving speed and spacing in real time to improve overall efficiency and ensure safety. After that, control instructions are generated and compressed for each device based on the dynamic resource scheduling strategy to obtain a device control instruction set. The control instruction generation here refers to formulating specific action guidelines for each device based on the optimized resource allocation plan, such as adjusting the antenna direction, changing the transmission power, etc. The purpose of control instruction compression is to reduce the overhead during instruction transmission and ensure efficient execution. For example, in the IoT application of a smart city, compressed control instructions can help save precious spectrum resources; in a cluster of unmanned vehicles, it can speed up the response speed and enhance the flexibility and reliability of the system. Finally, the device control instruction set is distributed and controlled by the Software-Defined Networking Controller (SDN Controller) to obtain the cluster control strategy. As the brain of the entire network, the SDN controller is responsible for centralized management and dynamic adjustment of network resource allocation. After receiving the control instruction set generated by the above steps, the SDN controller ensures that each device can accurately execute according to the predetermined plan, while monitoring the network operation and making necessary adjustments in time. For example, in the smart grid management of smart cities, the SDN controller can ensure that the communication between power metering equipment and the control center is fast and secure; in the unmanned vehicle cluster, each vehicle can obtain enough bandwidth to transmit key information, such as location updates or emergency brake warnings, thereby promoting a more intelligent and efficient transportation experience. In summary, through a series of sophisticated designs and technical means, the system realizes the transformation from the channel reuse mapping matrix to the specific cluster control strategy, which not only improves the communication efficiency and reliability, but also lays a solid technical foundation for future intelligent application scenarios. This process ensures that all devices can work together under optimal conditions to achieve higher-level goals together.

[0126] The above describes the high-bandwidth cluster control method based on a digital communication network in an embodiment of the present invention. The following describes the high-bandwidth cluster control system based on a digital communication network in an embodiment of the present invention. Figure 2 In one embodiment of the present invention, a high-bandwidth cluster control system based on a digital communication network includes:

[0127] The acquisition module 21 is used to acquire the real-time status data of each device in the cluster, and timestamp the real-time status data to obtain timestamp status data;

[0128] An estimation module 22, configured to perform distributed state estimation on the timestamp-marked state data through a preset SDN controller to obtain cluster coordination situation information;

[0129] A calculation module 23 is used to perform adaptive routing calculation on each device in the cluster based on the cluster collaborative situation information through a preset intent-driven network to obtain a network resource scheduling strategy;

[0130] A mapping module 24 is used to perform channel multiplexing mapping on a preset digital communication network based on the network resource scheduling strategy by using a preset space-time coding technology to obtain a channel multiplexing mapping matrix;

[0131] The control module 25 is used to perform resource scheduling, allocation and control on each device in the cluster based on the channel reuse mapping matrix to obtain a cluster control strategy.

[0132] In this embodiment, for the specific implementation of each unit in the above system embodiment, please refer to the above method embodiment, which will not be repeated here.

[0133] Reference Figure 3 The present invention also provides a computer device in an embodiment, wherein the internal structure of the computer device can be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0134] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0135] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0136] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0137] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0138] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A high-bandwidth cluster control method based on a digital communication network, characterized in that: The following steps are involved: Acquire real-time status data of each device in the cluster, and timestamp the real-time status data to obtain timestamp status data; By means of a preset SDN controller, distributed state estimation is performed on the timestamp-marked state data to obtain cluster coordination situation information; Through a preset intent-driven network, adaptive routing calculation is performed on each device in the cluster based on the cluster collaborative situation information to obtain a network resource scheduling strategy; By using a preset space-time coding technology, channel multiplexing mapping is performed on a preset digital communication network based on the network resource scheduling strategy to obtain a channel multiplexing mapping matrix; Based on the channel reuse mapping matrix, resource scheduling, allocation and control are performed on each device in the cluster to obtain a cluster control strategy; The method of performing adaptive routing calculation on each device in the cluster based on the cluster collaborative situation information through a preset intent-driven network to obtain a network resource scheduling strategy includes: Through a preset intent-driven network, the cluster collaborative situation information is analyzed for intent to obtain a network demand representation, and ontology-based semantic reasoning is performed on the network demand representation to obtain an enhanced semantic network demand representation; The enhanced semantic network demand representation is used to perform network topology virtualization mapping to obtain a virtual network topology, and the virtual network topology is divided into resources based on the network slicing technology to obtain a network slice resource pool; wherein the network slice resource pool includes multiple virtual network slices, each slice corresponding to a specific network demand; By using a preset ant colony algorithm, initial routing planning is performed on each device in the cluster based on the network slice resource pool to obtain an optimized routing solution; Performing path selection on the optimized routing solution through a software defined network controller to obtain an adaptive routing path; Acquire network bandwidth resources, and allocate the network bandwidth resources based on the adaptive routing path to obtain a bandwidth allocation scheme; The bandwidth allocation scheme, the adaptive routing path and the network slice resource pool are integrated to obtain a network resource scheduling strategy.

2. The high-bandwidth cluster control method based on a digital communication network according to claim 1, characterized in that: The acquiring of real-time status data of each device in the cluster and timestamping the real-time status data to obtain timestamped status data includes: Perform multi-dimensional state acquisition on each device in the cluster through a distributed sensor network to obtain an original state data stream; Performing spectrum analysis on the original state data stream to obtain spectrum state data; Performing multi-resolution analysis on the spectrum state data through a preset data processing module to obtain multi-resolution state data; By using a preset state matching algorithm, based on the multi-resolution state data, the state of each device in the cluster is matched to obtain matching state data, and the matching state data is classified to obtain classified state data, and the classified state data is used as real-time state data; wherein the classified state data includes network signal classification and data transmission classification; The classified state data is timestamped to obtain preliminary time-stamped state data, and the preliminary time-stamped state data is visualized based on graph theory state visualization technology to obtain time-stamped state data.

3. The high-bandwidth cluster control method based on a digital communication network according to claim 1, characterized in that: The method of performing distributed state estimation on the timestamp-marked state data through a preset SDN controller to obtain cluster coordination status information includes: Using a preset SDN controller, performing cluster state analysis on the timestamp-marked state data to obtain a device state feature matrix; Through a preset graph neural network, a topological relationship model is performed on each device in the cluster based on the device state feature matrix to obtain a device state relationship graph, and the nodes in the device state relationship graph are marked with importance to obtain a marked device state relationship graph; wherein the nodes are used as each device in the cluster; Based on the device state relationship diagram with the mark, high-bandwidth stability analysis is performed on each device in the cluster to obtain stability analysis data; If the stability analysis data is not within the preset stability range, the device state relationship diagram is dynamically adjusted through the adaptive topology reconstruction technology to obtain the adjusted device state relationship diagram; If the stability analysis data is within a preset stability range, the local state information of each device is securely aggregated through a federated learning mechanism to obtain a global state vector, and a state prediction is performed based on the global state vector to obtain a predicted state vector; Based on the predicted state vector and the device state relationship diagram or the adjusted device state relationship diagram, a dynamic evolution simulation is performed on each device in the cluster to obtain a future device state relationship diagram, and a community structure analysis is performed on the future device state relationship diagram to obtain a community division result; wherein the community division result is used to reflect the collaborative relationship between each device in the future cluster and the organizational structure of the cluster; The global state vector, the predicted state vector and the community division result are estimated and recorded in a distributed manner through blockchain technology to obtain cluster collaborative situation information.

4. The high-bandwidth cluster control method based on a digital communication network according to claim 1, characterized in that: The method of performing channel multiplexing mapping on a preset digital communication network based on the network resource scheduling strategy by using a preset space-time coding technology to obtain a channel multiplexing mapping matrix includes: By using a preset space-time coding technology, the network resource scheduling strategy is subjected to space-time resource coding to obtain a space-time coding sequence, and a multi-antenna channel characteristic analysis is performed on the space-time coding sequence to obtain a multi-antenna channel characteristic matrix; Performing frequency domain decomposition on the multi-antenna channel characteristic matrix through MIMO-OFDM technology to obtain a subcarrier resource matrix, and performing orthogonal multiplexing calculation on the subcarrier resource matrix to obtain an orthogonal multiplexing strategy; Based on the beamforming technology, spatial resources are allocated to the preset digital communication network based on the orthogonal multiplexing strategy to obtain a beam multiplexing scheme; Based on the beam multiplexing scheme, the channel capacity of the digital communication network is estimated to obtain a channel capacity parameter; wherein the channel capacity parameter includes a signal-to-noise ratio parameter and a throughput parameter; Through spectrum correlation analysis, channel multiplexing mapping is performed on the digital communication network based on the channel capacity parameters to obtain a channel multiplexing mapping matrix.

5. The high-bandwidth cluster control method based on digital communication network according to claim 4, characterized in that: The estimating the channel capacity of the digital communication network based on the beam multiplexing scheme to obtain a channel capacity parameter includes: Based on the beam multiplexing scheme, performing multipath channel analysis on the digital communication network to obtain a multipath channel feature vector, and performing Hilbert transform processing on the multipath channel feature vector to obtain a multidimensional channel phase feature set; Performing nonlinear Vandermonde matrix decomposition on the multidimensional channel phase feature set to obtain channel state estimation parameters; The channel state estimation parameters are orthogonalized by a preset Walsh transform algorithm to obtain an orthogonal channel eigenvector, and information entropy analysis is performed on the orthogonal channel eigenvector to obtain a channel entropy value matrix; Based on the channel entropy matrix, the capacity constraint analysis of the digital communication network is performed to obtain the channel capacity constraint condition, and the channel capacity constraint condition is subjected to Lagrangian dual decomposition to obtain the channel capacity boundary parameter; wherein the channel capacity boundary parameter is the channel maximum capacity threshold and the minimum capacity guarantee value; Performing Gauss-Markov chain analysis on the channel capacity boundary parameters to obtain a channel capacity transfer matrix; The channel capacity transfer matrix is ​​solved by Poisson's equation through a preset Green's function algorithm to obtain a channel capacity distribution function, and the channel capacity is estimated based on the channel capacity distribution function and the channel capacity boundary parameters to obtain channel capacity parameters.

6. The high-bandwidth cluster control method based on a digital communication network according to claim 1, characterized in that: The resource scheduling, allocation and control of each device in the cluster based on the channel reuse mapping matrix to obtain a cluster control strategy includes: Based on the channel multiplexing mapping matrix, a channel quality assessment is performed on the communication channels in the digital communication network to obtain a channel quality parameter matrix, and a channel state classification is performed on the channel quality parameter matrix to obtain a channel classification state; wherein the channel classification state includes a high-quality channel, a medium-quality channel and a low-quality channel; Predicting the future channel state of the channel classification state through a preset long short-term memory network to obtain a predicted channel state table; Performing initial resource allocation on each device based on the predicted channel state table to obtain an initial resource allocation scheme; Performing dynamic resource scheduling on the initial resource allocation scheme through a multi-agent reinforcement learning algorithm to obtain a dynamic resource scheduling strategy; Based on the dynamic resource scheduling strategy, control instructions are generated and compressed for each device to obtain a device control instruction set; The device control instruction set is distributed and controlled through a software-defined network controller to obtain a cluster control strategy.

7. A high-bandwidth cluster control system based on a digital communication network, characterized in that: include: An acquisition module is used to acquire real-time status data of each device in the cluster, and timestamp the real-time status data to obtain timestamp status data; An estimation module, used to perform distributed state estimation on the timestamp-marked state data through a preset SDN controller to obtain cluster coordination situation information; A computing module, configured to perform adaptive routing calculations on each device in the cluster based on the cluster collaborative situation information through a preset intent-driven network to obtain a network resource scheduling strategy; A mapping module, used to perform channel multiplexing mapping on a preset digital communication network based on the network resource scheduling strategy by using a preset space-time coding technology to obtain a channel multiplexing mapping matrix; A control module, used to perform resource scheduling, allocation and control on each device in the cluster based on the channel reuse mapping matrix to obtain a cluster control strategy; The method of performing adaptive routing calculation on each device in the cluster based on the cluster collaborative situation information through a preset intent-driven network to obtain a network resource scheduling strategy includes: Through a preset intent-driven network, the cluster collaborative situation information is analyzed for intent to obtain a network demand representation, and ontology-based semantic reasoning is performed on the network demand representation to obtain an enhanced semantic network demand representation; The enhanced semantic network demand representation is used to perform network topology virtualization mapping to obtain a virtual network topology, and the virtual network topology is divided into resources based on the network slicing technology to obtain a network slice resource pool; wherein the network slice resource pool includes multiple virtual network slices, each slice corresponding to a specific network demand; By using a preset ant colony algorithm, initial routing planning is performed on each device in the cluster based on the network slice resource pool to obtain an optimized routing solution; Performing path selection on the optimized routing solution through a software defined network controller to obtain an adaptive routing path; Acquire network bandwidth resources, and allocate the network bandwidth resources based on the adaptive routing path to obtain a bandwidth allocation scheme; The bandwidth allocation scheme, the adaptive routing path and the network slice resource pool are integrated to obtain a network resource scheduling strategy.

8. A computer device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images