Hierarchical control network architecture and multi-domain resource collaborative management and control method based on industrial SDN (Software Defined Network)

Abstract

The invention discloses a hierarchical control network architecture based on an industrial software defined network (SDN), provides a dual-time scale network slicing method and a multi-domain resource collaborative management and control scheme, and shows a remarkable industrial application prospect. The hierarchical control architecture allows flexible configuration of the SDN controller according to a specific industrial environment, and realizes multi-level resource management and control. And the SDN controllers of different levels can be dynamically adjusted according to actual requirements to adapt to complex and changeable industrial scenes. According to the dual-time-scale network slicing method, resource reservation and arrangement are decoupled, and the resource scheduling efficiency is improved. By monitoring the network state in real time, the resource configuration is optimized, and the heterogeneous QoS requirement is ensured to be met, so that the utilization rate of resources is remarkably improved.

Description

Technical Field

[0001] The present invention mainly relates to the field of communication technology, and in particular to a hierarchical control network architecture based on industrial SDN, a dual-time scale network slicing method and a multi-domain resource collaborative management and control method. Background Art

[0002] Network systems in modern industrial environments are becoming increasingly complex, placing higher demands on the management, control and allocation of network resources. In order to ensure the heterogeneous quality of service (QoS) requirements of different types of industrial applications, network slicing technology has gradually become a research hotspot in academia and industry. Network slicing divides physical network resources into multiple virtual networks on demand to achieve flexible resource scheduling and control to meet the needs of different applications. However, existing technologies still have some key problems when dealing with dynamic and changing industrial environments.

[0003] In the industrial Internet scenario, the network is large in scale, with many nodes, and the business requirements of each node vary significantly. The traditional single-controller architecture faces many challenges, such as high control latency, high signaling overhead, and difficulty in adapting to large-scale network dynamic changes. In addition, existing network slicing research mainly focuses on the independent optimization of resource slicing and scheduling, but fails to explore the joint optimization between them. Due to the interdependence between resource slicing and scheduling decisions, ignoring their joint optimization will lead to inefficient resource utilization and make it difficult to fully meet the heterogeneous QoS requirements in industrial scenarios.

[0004] At the same time, most existing network slicing solutions use static or historical data-based optimization methods, which are difficult to adapt to the spatiotemporal changes in network conditions in industrial scenarios. For example, the communication links in a factory may be limited by factors such as the physical environment and equipment movement, and the changes in resource demand and service quality are highly uncertain. In this context, how to achieve efficient allocation and scheduling of resources while ensuring service quality has become an urgent problem to be solved in the Industrial Internet.

[0005] In response to the above problems, software-defined networking (SDN), as an emerging network architecture, provides a more flexible means of network management and control by separating the control plane and the data plane. In industrial environments, the SDN architecture can centrally control network resources, making network management and control more efficient. However, existing SDN technologies usually use a single controller, which is difficult to adapt to the resource coordination needs in large-scale industrial environments. In addition, the scalability and flexibility of traditional SDN architectures are also limited when facing multi-domain and multi-level resource management and control. Therefore, the hierarchical control network architecture based on industrial SDN has become an important direction for solving resource management and control problems in large-scale network environments.

[0006] In order to solve the above problems, the present invention proposes a hierarchical control network architecture based on industrial SDN, and on this basis proposes a dual-time-scale network slicing method and a multi-domain resource collaborative management and control scheme. Through this hierarchical architecture, SDN controllers at different levels can be responsible for resource management in different ranges respectively, so as to achieve efficient allocation and scheduling of resources. The dual-time-scale slicing method improves the flexibility and response speed of resource scheduling by decoupling resource reservation and resource scheduling. In addition, the multi-domain resource collaborative management and control scheme further optimizes the coordination and control of cross-domain resources, and enhances the scalability and stability of the system. Summary of the invention

[0007] In view of the above-mentioned defects of the prior art, the technical problems to be solved by the present invention include:

[0008] How to design a hierarchical control network architecture, network slicing method and resource coordination method to solve the above technical problems.

[0009] To achieve the above object, the present invention provides a hierarchical control network architecture based on industrial SDN, including a terminal-side SDN controller, an edge-side SDN controller and an industrial cloud-side SDN controller;

[0010] The terminal-side SDN controller is deployed at the access terminal, switch and gateway, has a small coverage area, and is responsible for resource scheduling of terminal users;

[0011] The edge-side SDN controller is responsible for coordinating multiple terminal-side SDN controllers to achieve higher-level network operations;

[0012] The industrial cloud-side SDN controller, as the highest-level controller, obtains network status information provided by the underlying controller to formulate global rules and perform cross-domain coordination for the entire network.

[0013] A dual-time-scale network slicing method based on the aforementioned network architecture divides resource allocation into different time domains for processing, including a resource reservation method and a resource orchestration method.

[0014] Furthermore, the resource reservation method is specifically as follows: at the beginning of each slice window w, the industrial cloud side SDN controller makes a resource reservation decision based on the network status feedback information from the edge side SDN controller and the global business needs. Where ε represents the edge-side SDN controller and s represents the index of the network slice.

[0015] Furthermore, the resource reservation method also includes: during the entire network slicing window, the reservation decision remains unchanged; after each network slicing window ends, the industrial cloud-side SDN controller evaluates the system performance through feedback and adjusts the resource allocation decision of the next network slicing window based on changes in business needs.

[0016] Furthermore, the resource orchestration method is specifically as follows: within each network slice window, the edge-side SDN controller schedules resources in real time according to the spatiotemporal changes of network topology, link status and business needs.

[0017] Furthermore, the resource scheduling method is specifically as follows: the real-time scheduling of resources is specifically as follows: allocating network resources to the terminal-side SDN controller, and flexibly adjusting the resources between different time slots.

[0018] Furthermore, the resource orchestration method mainly focuses on the association between the edge-side SDN controller and the terminal-side SDN controller and the efficiency of resource allocation.

[0019] A multi-domain resource collaborative management and control method based on the aforementioned dual-time-scale network slicing method comprises the following steps:

[0020] Step 1: According to the current network status and resource requirements, the dual-time-scale network slicing problem is decomposed into a resource reservation sub-problem and a resource scheduling sub-problem, and the related constraints are also decomposed accordingly;

[0021] Step 2: At the resource orchestration layer, due to the isolation of slices, a resource orchestration algorithm based on multi-backpack matching is adopted;

[0022] Step 3: At the resource reservation layer, the resource scheduling strategy is substituted into the resource reservation sub-problem as an incentive, and it is transformed into a closed expression related only to the reservation ratio, and then solved using the iterative block coordinate gradient descent algorithm.

[0023] Furthermore, the step 2 is specifically as follows:

[0024] Step 2.1: Construct a bipartite graph G = (Γ t ,Ε t ,Ξ t ), where Γ t is the set of sampling terminals, Ε t It is a collection of edge-side SDN controllers. t is the connection Γ t and t The set of edges with vertices in the graph; according to the expected service delay and related network resource constraints, the multi-domain network resources required to meet the requirements are given under the given association decision <γ,ε>

[0025]

[0026] in, are the required communication, computing and storage resources respectively. Further, γ represents the terminal-side SDN controller, ε represents the edge-side SDN controller, b represents communication resources, c represents computing resources, m represents storage resources, and t represents time slot;

[0027] Step 2.2: For each terminal-side SDN controller, the resources required under different associations The edge-side SDN controllers are sorted in ascending order to build a preference list, that is, the terminal-side SDN controller prefers the edge-side SDN controller with higher resource utilization;

[0028] Step 2.3, each terminal-side SDN controller sends a request to its currently favorite edge-side SDN controller, and then deletes this edge-side SDN controller from the preference list;

[0029] Step 2.4: Each edge-side SDN controller checks all requests received, including new requests and requests accepted in previous iterations; given a set of proposals, each request has a required resource quantity and a priority weight p γ , each edge-side SDN controller needs to determine which requests to accept so as not to violate the total weight constraint Maximize the total value under the circumstances;

[0030] Step 2.5: For all unqualified terminal-side SDN controllers, go to step 2.3; when all terminal-side SDN controllers are matched to edge-side SDN controllers or the available resources of all edge-side SDN controllers are allocated, the matching process ends.

[0031] Furthermore, the step 3 is specifically as follows:

[0032] Step 3.1: In the initialization phase, randomly generate an initial resource reservation value that meets the requirements for each slice. And set the learning factor α and the number of iterations n = 0;

[0033] Step 3.2: Calculate the gradient of the objective function based on the current resource reservation decision and network status And update the reserved variables Make sure the updated reservation value is within the valid range;

[0034] Step 3.3: Calculate the difference in the objective function value of the current iteration If it is less than the set convergence threshold or the current number of iterations reaches the maximum number of iterations, the iteration is stopped; otherwise, return to step 3.2 for the next iteration.

[0035] Compared with the prior art solutions, the technical effects of the present invention are:

[0036] The present invention effectively improves the scalability and response speed of SDN networks in industrial environments by introducing a hierarchical SDN controller architecture. Multi-level controllers can flexibly schedule resources according to different communication scenarios and QoS requirements, reduce signaling overhead and optimize the efficiency of network status acquisition. In addition, the architecture ensures efficient collaborative management and control of cross-domain resources, thereby improving the overall performance and service quality of the network. This hierarchical control method solves the problem of inefficiency in the prior art when a single SDN controller controls a wide range of resources, and significantly improves the network operation performance in complex industrial scenarios.

[0037] The present invention realizes the joint optimization of resource slicing and scheduling by introducing a dual-time-scale network slicing method, which brings the following technical effects:

[0038] Improve resource utilization: Due to global resource reservation on a large time scale and refined resource orchestration on a small time scale, the system can use available resources more efficiently and reduce resource waste.

[0039] Meeting heterogeneous QoS requirements: Through a flexible resource allocation mechanism, the present invention can dynamically adjust to different business scenarios to ensure that the QoS requirements of each slice are guaranteed, thereby improving the service quality of the network.

[0040] Enhanced system scalability: The dual-time-scale optimization approach enables the architecture to cope with changes in resource requirements under different business models and maintain stable performance when network load and business requirements change dynamically.

[0041] Reducing signaling overhead: By reducing frequent decision adjustments at the resource reservation layer, the present invention significantly reduces the signaling overhead of the system and improves the overall efficiency of network operation.

[0042] This innovative network slicing method solves the inefficiency caused by the separation of resource slicing and scheduling in existing technologies, and enhances the system's adaptability to complex industrial network requirements.

[0043] By introducing a multi-domain resource collaborative management and control solution, the present invention brings the following technical effects in the implementation of the dual-time-scale network slicing method:

[0044] Improve optimization efficiency: Decomposing the problem into simpler sub-problems reduces the computational complexity of resource scheduling and reservation decisions from exponential to linear, greatly improving the execution efficiency of the algorithm.

[0045] Improve convergence: The iterative block coordinate gradient descent algorithm is used to optimize the convergence speed of resource reservation decisions, thereby improving the stability and reliability of the system.

[0046] Enhanced resource utilization: Through precise resource allocation strategies, network resources can be more efficiently utilized to meet heterogeneous QoS requirements.

[0047] Strong adaptability: The solution can dynamically adjust resource allocation strategies to adapt to changing network conditions and business needs, improving the flexibility and responsiveness of industrial network systems.

[0048] In general, the invention provides an innovative and effective solution to the complex dual-time-scale network slicing problem. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of a hierarchical control network architecture based on an industrial software defined network (SDN) provided in an embodiment of the present invention;

[0050] Figure 2 It is a schematic diagram of collaborative management and control of multi-domain resources under dual-time-scale network slicing provided by an embodiment of the present invention;

[0051] Figure 3 It is a schematic diagram of the working principle of the multi-domain resource collaborative management and control solution provided by an embodiment of the present invention;

[0052] Figure 4 is a flow chart of a resource scheduling algorithm based on multi-backpack matching provided by an embodiment of the present invention;

[0053] Figure 5 It is a flow chart of a resource reservation algorithm based on iterative block coordinate gradient descent provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following describes several preferred embodiments of the present invention with reference to the drawings in the specification, so that the technical content is clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the protection scope of the present invention is not limited to the embodiments mentioned in the text.

[0055] Figure 1 A hierarchical control network architecture based on industrial software-defined networking (SDN) is demonstrated for resource management and scheduling in complex industrial scenarios. Taking the measurement of the angle of the active surface during aircraft assembly as an example, the architecture aims to improve the overall performance of the system and meet the quality of service (QoS) requirements of different industrial applications through hierarchical control and resource collaboration. In this SDN-based hierarchical control network architecture, there are mainly three layers: field perception layer, edge computing layer, and industrial cloud decision layer. Each layer has a clear functional division to achieve efficient control and scheduling of cross-domain resources.

[0056] On-site perception layer: The on-site perception layer consists of multiple terminal clusters and terminal-side SDN controllers (TSCs). The main task of this layer is to perform real-time perception and data collection of equipment status in industrial sites. Taking the measurement of the angle of the active surface of aircraft assembly as an example, multiple sensors are deployed on the active surface to collect key data such as angle changes and position deviations in real time. These data are locally pre-processed and preliminarily analyzed by the terminal-side SDN controller to ensure timely and reliable data transmission. At the same time, TSCs are responsible for interacting the perceived device information with the upper-layer network to ensure immediate response to on-site tasks.

[0057] Edge computing layer: The edge computing layer consists of multiple edge-side SDN controllers (ESCs) and edge servers. The main function of this layer is to further process and analyze the data from the perception layer and provide real-time computing, storage, and network resource allocation capabilities. In the scenario of aircraft assembly, the edge computing layer can complete a large amount of data processing work in a local data center close to the production site, reducing the system's dependence on cloud resources and reducing transmission delays. In addition, ESCs are also responsible for optimizing the scheduling of network resources based on network topology and business needs in the case of dynamic changes in time and space, so as to achieve real-time response and resource allocation for on-site tasks.

[0058] Industrial cloud decision layer: The industrial cloud decision layer consists of an industrial cloud-side SDN controller (CSC) and cloud servers, and is responsible for the global resource scheduling and optimization decisions of the entire system. At the beginning of each network slice window, the CSC determines the proportion of communication, computing, and storage resources that need to be reserved for each edge server based on the QoS requirements of different applications. In the scenario of aircraft assembly, the CSC can dynamically adjust the resource allocation strategy by integrating feedback information from multiple edge servers to ensure the execution quality of key tasks and the optimization of the overall system performance. At the same time, the industrial cloud decision layer is also responsible for long-term analysis of historical data and formulating global strategies and optimization plans.

[0059] Through this hierarchical control network architecture based on industrial SDN, the system can flexibly allocate communication, computing and storage resources according to the needs of different application scenarios, realize efficient resource utilization, reduce system response time, and improve resource utilization and task completion quality. In addition, the hierarchical control strategy of this architecture ensures the independence and coordination of each layer, supports resource scheduling and optimization at different time scales, and helps to cope with the dynamic changes of network conditions in industrial environments.

[0060] Figure 2 The figure shows the collaborative management and control of multi-domain resources under dual-time scale network slicing. The left side of the figure shows the backbone diagram of the SDN-based hierarchical control network architecture, and the right side shows the dual-time scale network slicing framework.

[0061] Left: Backbone diagram of the hierarchical control network architecture

[0062] This figure shows the closed-loop interaction mechanism of data flow and decision flow in the SDN architecture. The upward data flow is responsible for feeding back real-time device status and network information from the perception layer and edge computing layer; the downward decision flow is issued by the SDN controller on the industrial cloud side and edge side to issue resource reservation instructions and orchestration decisions to ensure that the system operates as expected. Through the closed loop of data flow and decision flow, the system realizes global perception and intelligent scheduling of network status, ensuring efficient use of resources.

[0063] Right: Dual-time-scale network slicing framework

[0064] The dual time scale is divided into large time scale and small time scale, corresponding to different levels of control and optimization:

[0065] ①Large time scale resource reservation

[0066] On a large time scale, resource pre-allocation is mainly managed by the industrial cloud-side SDN controller (CSC) and edge-side SDN controllers (ESCs). CSC predicts future resource requirements based on long-term historical data and business needs, and reserves corresponding communication, computing, and storage resources for each network slice. At the beginning of each slice window, CSC makes a global resource reservation decision and determines the resource allocation ratio of different slices to ensure that the system achieves global optimization while meeting QoS requirements.

[0067] ② Small time scale resource scheduling

[0068] On a small time scale, the dynamic orchestration of resources is completed by edge-side SDN controllers (ESCs) and terminal-side SDN controllers (TSCs). ESC performs resource scheduling in each time slot based on actual business needs and network conditions, and is responsible for allocating reserved network resources to each terminal. In this process, ESC dynamically adjusts the association between each terminal and allocates bandwidth, computing resources, and storage resources in a fine-grained manner to ensure the real-time performance of the system and efficient use of resources.

[0069] Resource reservation and orchestration decisions are interdependent on both large and small time scales, forming a tightly coupled management and control mechanism. This dual-time scale network slicing design not only ensures global resource reservation optimization, but also achieves rapid response to dynamically changing services, with strong adaptability and flexibility. Through this architecture, we can effectively cope with the complex spatiotemporal dynamic changes in industrial scenarios, achieve collaborative optimization of cross-domain resources, and improve network performance and resource utilization in the industrial Internet environment.

[0070] Figure 3The working principle of the multi-domain resource collaborative control scheme is demonstrated. The scheme decomposes the complex dual-time-scale network slicing problem into resource reservation sub-problems and resource orchestration sub-problems, realizes the collaborative control of resources between different domains, and optimizes the efficiency of resource allocation. The core of the scheme contains two key algorithms: a resource orchestration algorithm based on multi-knapsack matching and a resource reservation algorithm based on iterative block coordinate gradient descent, which work together. The resource orchestration algorithm dynamically adjusts the association between edge-side SDN controllers (ESCs) and terminal-side SDN controllers (TSCs) and the allocation of network resource blocks through four steps: list construction, request sending, reverse selection, and cyclic request, to meet real-time business needs. The resource reservation algorithm converges to the optimal resource reservation ratio through partial derivative calculation and iterative update to ensure that the resource allocation of each slice can meet long-term needs in each slice window. The results of resource orchestration will be fed back to the resource reservation algorithm to provide reference data for the next global resource reservation, forming a closed-loop optimization mechanism. Through this decomposition and coordination mechanism, this solution can effectively handle the complexity of resource scheduling in complex industrial Internet scenarios, significantly improve the efficiency of resource scheduling, and support efficient configuration and optimization of resources to meet the needs of multi-time scale and multi-domain resource coordination in industrial scenarios.

[0071] Figure 4 This is a flow chart of the resource scheduling algorithm based on multi-backpack matching, which includes the following steps:

[0072] Step 1: Initialization phase, obtain network status, terminal resource requirements and edge server resource status;

[0073] Step 2: Construct a bipartite graph G = (Γ t ,Ε t ,Ξ t ), where Γ t is the set of sampling terminals, Ε t is a collection of ESCs, t is the connection Γ t and t The set of edges of the vertices in ;

[0074] Step 3: Resource requirement evaluation: Based on the expected service delay and related network resource constraints, calculate the network resources required to meet the requirements given the associated decision <γ,ε>

[0075] Step 4: Terminal preference sorting, generate a preference list for each TSC, and sort the resources required under different associations Sort ESCs in ascending order;

[0076] Step 5: Request sending: Each TSC sends an association request to its favorite ESC and deletes the ESC from the preference list;

[0077] Step 6: Request Check: Each ESC checks all requests received, including new requests and requests accepted in previous iterations, and performs resource availability assessment, including the number of resources required by each TSC. and priority weight p γ ;

[0078] Step 7: Matching decision: Each ESC decides the list of TCSs to accept the request based on resource constraints and request weights so as not to violate the total weight constraint. Maximize the total value under the circumstances;

[0079] Step 8: loop processing. If a TCS request is rejected and ECSs still have available resources, return to step 5 to continue processing; otherwise, end the matching process;

[0080] Step 9: Output the results, output the resource scheduling results, and update the network status information.

[0081] Figure 5 It is a flow chart of the resource reservation algorithm based on iterative block coordinate gradient descent, which includes the following steps:

[0082] Step 1: Initialization phase: randomly generate an initial resource reservation value that meets the requirements for each slice And set the learning factor α and the number of iterations n = 0;

[0083] Step 2: Calculate the gradient. Calculate the gradient of the objective function based on the current resource reservation decision and network status.

[0084] Step 3: Update the reserved value by formula Update the reserved variable to ensure that the updated reserved value is within the valid range;

[0085] Step 4: Calculate the difference of the objective function value, and calculate the difference of the objective function value of the current iteration

[0086] Step 5: Convergence detection, judging whether the difference of the objective function value is less than the set convergence threshold or whether the current number of iterations reaches the maximum number of iterations;

[0087] Step 6: Determine the end of the iteration. If the stop condition is met, the algorithm ends and the final resource reservation plan is output; otherwise, return to step 2 and proceed to the next iteration.

[0088] Step 7: Output the results and the updated resource reservation value

[0089] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. The description and examples are to be considered exemplary only, and the true scope and spirit of the present invention is indicated by the following claims.

[0090] It should be understood that the present invention is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A hierarchical control network architecture based on industrial SDN, characterized by: Including terminal-side SDN controller, edge-side SDN controller and industrial cloud-side SDN controller; The terminal-side SDN controller is deployed at the access terminal, switch and gateway, has a small coverage area, and is responsible for resource scheduling of terminal users; The edge-side SDN controller is responsible for coordinating multiple terminal-side SDN controllers to achieve higher-level network operations; The industrial cloud-side SDN controller, as the highest-level controller, obtains network status information provided by the underlying controller to formulate global rules and perform cross-domain coordination for the entire network.

2. A dual-time-scale network slicing method based on the network architecture of claim 1, characterized in that: The resource allocation is divided into different time domains for processing, including resource reservation method and resource scheduling method.

3. The network slicing method according to claim 2, characterized in that: The resource reservation method is specifically as follows: at the beginning of each slice window w, the industrial cloud side SDN controller makes a resource reservation decision based on the network status feedback information from the edge side SDN controller and the global business needs. Where ε represents the edge-side SDN controller and s represents the index of the network slice.

4. The network slicing method according to claim 3, characterized in that: The resource reservation method also includes: the reservation decision remains unchanged during the entire network slicing window; after each network slicing window ends, the industrial cloud-side SDN controller evaluates system performance through feedback and adjusts the resource allocation decision of the next network slicing window based on changes in business needs.

5. The network slicing method according to claim 2, characterized in that: The resource orchestration method is specifically as follows: within each network slice window, the edge-side SDN controller schedules resources in real time according to the temporal and spatial changes of network topology, link status and business needs.

6. The network slicing method according to claim 5, characterized in that: The resource scheduling method is specifically as follows: the real-time scheduling of resources is specifically to allocate network resources to the terminal side SDN controller and flexibly adjust the resources between different time slots.

7. The network slicing method according to claim 5, characterized in that: The resource orchestration method mainly focuses on the association between the edge-side SDN controller and the terminal-side SDN controller and the efficiency of resource allocation.

8. A multi-domain resource collaborative management and control method based on the dual-time-scale network slicing method of claim 2, characterized in that: Includes steps: Step 1: According to the current network status and resource requirements, the dual-time-scale network slicing problem is decomposed into a resource reservation sub-problem and a resource scheduling sub-problem, and the related constraints are also decomposed accordingly; Step 2: At the resource orchestration layer, due to the isolation of slices, a resource orchestration algorithm based on multi-backpack matching is adopted; Step 3: At the resource reservation layer, the resource scheduling strategy is substituted into the resource reservation sub-problem as an incentive, and it is transformed into a closed expression related only to the reservation ratio, and then solved using the iterative block coordinate gradient descent algorithm.

9. The multi-domain resource collaborative management and control method according to claim 8, characterized in that: The step 2 is specifically as follows: Step 2.1: Construct a bipartite graph G = (Γ t ,Ε t ,Ξ t ), where Γ t is the set of sampling terminals, Ε t It is a collection of edge-side SDN controllers. t is the connection Γ t and t The set of edges with vertices in the graph; according to the expected service delay and related network resource constraints, the multi-domain network resources required to meet the requirements are given under the given association decision <γ,ε> in, are the required communication, computing and storage resources respectively. Further, γ represents the terminal-side SDN controller, ε represents the edge-side SDN controller, b represents communication resources, c represents computing resources, m represents storage resources, and t represents time slot; Step 2.2: For each terminal-side SDN controller, the resources required under different associations The edge-side SDN controllers are sorted in ascending order to build a preference list, that is, the terminal-side SDN controller prefers the edge-side SDN controller with higher resource utilization; Step 2.3, each terminal-side SDN controller sends a request to its currently favorite edge-side SDN controller, and then deletes this edge-side SDN controller from the preference list; Step 2.4: Each edge-side SDN controller checks all requests received, including new requests and requests accepted in previous iterations; given a set of proposals, each request has a required resource quantity and a priority weight p γ , each edge-side SDN controller needs to determine which requests to accept so as not to violate the total weight constraint Maximize the total value under the circumstances; Step 2.5: For all unqualified terminal-side SDN controllers, go to step 2.3; when all terminal-side SDN controllers are matched to edge-side SDN controllers or the available resources of all edge-side SDN controllers are allocated, the matching process ends.

10. The multi-domain resource collaborative management and control method according to claim 8, characterized in that: The step 3 is specifically as follows: Step 3.1: In the initialization phase, randomly generate an initial resource reservation value that meets the requirements for each slice. And set the learning factor α and the number of iterations n = 0; Step 3.2: Calculate the gradient of the objective function based on the current resource reservation decision and network status And update the reserved variables Make sure the updated reservation value is within the valid range; Step 3.3: Calculate the difference in the objective function value of the current iteration If it is less than the set convergence threshold or the current number of iterations reaches the maximum number of iterations, the iteration is stopped; otherwise, return to step 3.2 for the next iteration.

Images