5g-tsn cross-domain cooperative scheduling method oriented to deterministic latency guarantee

Abstract

The present application relates to a kind of 5G-TSN cross-domain collaborative scheduling method for deterministic latency guarantee, belong to wireless communication field.The method includes: S1: build 5G-TSN cross-domain collaborative optimization communication system: S2: according to the service requirement of business flow, build 5G wireless transmission latency and TSN processing latency Weighted optimization problem;S3: device transmit power allocation: under the satisfaction of the reliability transmission requirement of terminal device, deduce the closed expression of interruption probability and the solution condition of optimal power;S4: TSN gateway resource pre-allocation: based on greedy algorithm to obtain gateway resource pre-allocation, provide optimal TSN gateway placement and processing resource allocation for task;S5: using the heuristic algorithm based on preemption to realize cross-domain collaborative scheduling.The present application under the condition of meeting business deadline, both reduce system total latency, also improve gateway resource utilization, further guarantee the timely processing of high priority business.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication and relates to the problem of providing deterministic latency guarantees for Internet of Things (IoT) devices. Specifically, it relates to a 5G-TSN cross-domain collaborative scheduling method for deterministic latency guarantees. Background Technology

[0002] 5G has made new technological improvements in low latency and high reliability, and "5G+ empowering industries" has become a common demand in the industrial and communications sectors. The emergence of a large number of computationally intensive and time-sensitive services poses challenges to the future wireless cellular networks in providing large-scale connectivity and ensuring Quality of Service (QoS). In industrial IoT systems, IoT devices often only provide simple data upload functions. In this environment, there is an urgent need for solutions that can alleviate the pressure on the core network while addressing computational bottlenecks. However, industrial services place extremely stringent performance requirements on the network, demanding deterministic latency guarantees in addition to a bearer network with low latency, low jitter, and high reliability. Therefore, a key technical challenge in the core aspects of 5G empowerment is how to improve the deterministic latency guarantee capability of 5G networks. Time-Sensitive Networking (TSN) technology can provide guaranteed services for Ethernet-based communications, offering traffic shaping scheduling, resource reservation, and packet-free transmission for periodic services, thus meeting the requirements of time-sensitive applications. Traditional industrial automation systems are mostly based on wired real-time networks. Combining wired and wireless technologies can bring benefits such as deployment flexibility and reduced maintenance costs. Industrial applications have very strict requirements for real-time services and deterministic transmission. In time-sensitive networks, analyzing the factors affecting various network latency is beneficial for coordinated scheduling.

[0003] Large-scale communication networks are a crucial aspect of smart factories, supporting not only communication between controllers and sensors but also integrating massive computing resources such as edge cloud and factory cloud systems. Traditional wired TSN networks struggled to meet terminal access and data transmission requirements, while wireless communication offers high flexibility, dynamic interaction, and mobility support within the system. Furthermore, ensuring that TSN networks provide deterministic services to most businesses within their deadlines with limited resources is challenging. Fifth-generation mobile communication technology (5G) offers wide coverage, security, and reliability, while edge networks can assist in extending the computing and storage resources of communication intermediary devices such as switches and routers. Therefore, the combination of 5G and TSN is not only a requirement for 5G to achieve vertical integration services from the application layer to the physical layer in the industrial sector but also an inherent requirement of smart factories. The realization of Industry 4.0 relies on the convergence of operational and communication technologies (OT & IT), but cloud computing cannot guarantee that the stringent QoS requirements of services cannot be used for industrial OT applications. Therefore, the Industrial Internet of Things (IIoT) may become a reality based on the edge computing paradigm, and the combination of 5G and TSN will drive the development of inherent needs in the future industrial sector. Few studies have taken into account the uncertainties introduced by 5G wireless channels, which could negatively impact network schedulability.

[0004] Therefore, there is an urgent need for an optimized scheduling method to provide deterministic latency guarantees for equipment. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a 5G-TSN cross-domain collaborative scheduling method for deterministic latency assurance. This method comprehensively considers reliable transmission and available system resources to design an efficient scheduling strategy, providing deterministic latency assurance capabilities for both time-sensitive and non-time-sensitive service flows. Specifically, latency is optimized from the perspectives of device transmit power, gateway resources, and service scheduling. First, to ensure service reliability, the interruption probability expression for service transmission in the 5G wireless channel and the solution conditions for the optimal transmit power are derived. Second, to achieve interconnection between 5G and TSN, a greedy gateway resource pre-allocation method is designed to optimize scheduling. Finally, considering low latency requirements, a preemptive service scheduling approach is adopted, preempting low-priority services to ensure timely service for high-priority services.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A 5G-TSN cross-domain cooperative scheduling method for deterministic latency assurance is proposed. First, a 5G-TSN cross-domain cooperative model is constructed under a centralized management framework of edge-assisted SDN (Software-Defined Networking). Then, the latency of 5G wireless transmission and TSN processing is analyzed to construct a latency-weighted optimization problem. Finally, a cross-domain cooperative scheduling method is designed to optimize this model. Specifically, the method includes the following steps: S1: Construct a 5G-TSN cross-domain collaborative optimization communication system: Consider the joint scheduling of 5G and TSN, inject time-sensitive and non-time-sensitive services of sensor devices from 5G into TSN. TSN gateways and 5G micro base stations are deployed at the edge layer. The TSN gateway is configured with a standard 5G interface to assist the TSN gateway in communicating with sensor devices under the centralized management of the 5G network. The 5G micro base station acts as an edge computing node, while the entire edge 5G system acts as a TSN logical bridge. S2: Constructing a weighted optimization model for 5G wireless transmission latency and TSN processing latency: In order to find an optimal cross-domain scheduling between 5G and TSN to optimize the latency of the overall service flow within the deadline, a weighted optimization model for 5G wireless transmission latency and TSN processing latency is constructed based on the service requirements of the service flow. S3: Device Transmit Power Allocation: Reliable transmission of wireless channels is closely related to transmit power. Under the premise of meeting the reliable transmission requirements of terminal devices, we derive the closed expression of the interruption probability and the solution conditions for the optimal power to provide reliable data delivery. S4: TSN Gateway Resource Pre-allocation: To ensure that most services in the TSN network provide deterministic services with limited resource capacity within their deadline, pre-allocation of gateway resources can effectively improve the system's service capacity. Based on a greedy algorithm, gateway resource pre-allocation is obtained to provide optimal TSN gateway placement and processing resource allocation for tasks. S5: A preemptive heuristic algorithm is used to achieve cross-domain collaborative scheduling.

[0007] Furthermore, in step S1, the construction of a 5G-TSN cross-domain collaborative optimization communication system specifically includes the following steps: S11: To ensure deterministic latency, the TSN gateway provides a time-aware shaper that provides guaranteed time slots for specific traffic categories through a pre-calculated gate control list, enabling time-sensitive services to complete task delivery within the deadline. TSN scheduling is calculated in a centralized manner, adopting a fully centralized architecture defined in the IEEE 802.1 Qcc standard, and is centrally managed through a Software Defined Network (SDN) controller, including centralized user configuration and centralized network configuration. S12: The Business Composed of a series of ordered sets of subtasks, often referred to as a task function chain, it has a strict execution order and is represented as... , any Indicates business Subtasks Indicates business The number of subtasks, and each subtask With parameter set , This indicates the CPU resources required for the execution of the subtask. This indicates the amount of cache space required for the execution of the subtask. This refers to the size of the subtask data that needs to be uploaded; S13: In order to ensure that latency-sensitive services are served in a timely manner, the application needs to retain priority tags for services to coordinate the processing of services with different priorities; S14: TSN networks can consist of incomplete connection diagrams. Let represent, where the set It is a TSN gateway node that assists in communication between the 5G network and sensor nodes, and integrates... These are sets of wired links used for communication between TSN gateways. For the first... gateway With CPU resource size The cache size is and the set of neighboring nodes , Directly connected to the node The network consists of neighboring TSN gateway nodes. The latency budget for services in a 5G network is mainly related to the air interface channel, while the air interface rate... It is closely related to the quality of the wireless channel and is accompanied by a certain degree of uncertainty.

[0008] Furthermore, in step S2, a weighted optimization model for 5G wireless transmission latency and TSN processing latency is constructed, specifically including the following steps: S21: The device injects data into the TSN network through the 5G wireless system. The power configuration of the device needs to ensure the reliable transmission of data, while also optimizing the wireless transmission latency. S22: In the TSN network, the processing time of the subtask at the gateway is calculated based on the arrival time and completion time of the service and subtask, and a binary variable is defined to indicate whether the TSN gateway serves the subtask. The processing delay of the subtask at the TSN gateway is also defined as the time difference between the arrival time and completion time of the service. S23: Different subtasks may represent different functions, so it is necessary to process the set of subtasks that make up the same service in sequence. For time-sensitive constraints, the flow isolation and frame isolation characteristics of TSN must be followed. S24: Service requests generated by the device are injected into the TSN network for processing by the 5G system as a TSN logical bridge. Therefore, we need to analyze the wireless transmission latency of the service in the 5G band and the processing latency on the TSN gateway. S25: The total delay of service coordination and scheduling consists of two parts: the 5G segment and the TSN segment. Since the TSN gateway may serve another subtask when a new subtask arrives, the TSN segment includes processing delay and gateway fixed queuing forwarding delay. S26: Optimize the latency of the overall service flow between 5G and TSN within the deadline, and provide deterministic latency guarantees for both time-sensitive and non-time-sensitive services.

[0009] Furthermore, in step S26, the latency optimization problem is constructed as follows:

[0010] Among them, constraints C1 and C2 represent the limitations on CPU resources and cache capacity of the TSN gateway, respectively; constraint C3 ensures that a TSN gateway can only execute one subtask at a time; constraint C4 indicates that the TSN gateway has sufficient resources to schedule the subtask set; constraint C5 represents the interruption probability threshold condition; constraint C6 represents the flow isolation and frame isolation requirements of TSNs that execute the same service flow subtasks sequentially; constraint C7 is the service flow latency threshold constraint; binary variables Indicates TSN gateway Does it serve the business? The Sub-tasks , It is a sub-task of TSN The execution start time, It is the transmit power at the sensor device end. Indicates business Coordinated scheduling latency, Represents a set of business functions; Indicates gateway The cached set of subtasks from different business flows; Indicates the probability of interruption; and These are the interrupt threshold and the delay threshold, respectively. Subtasks Processing latency in TSN gateways and Subtasks Arrival time and completion time.

[0011] Furthermore, step S3 specifically includes: under certain reliability requirements, [the following steps will be performed]. To achieve optimal power allocation and optimize 5G wireless transmission latency.

[0012] Furthermore, step S4 specifically includes the following steps: S41: The network not only optimizes latency to deliver traffic as quickly as possible, but also aims to deliver traffic in a timely manner with limited computing resources. Different gateways have different CPU sizes and buffer capacities. All gateways in the candidate gateway set for subtasks are sorted in descending order based on weighted resources. S42: Although each TSN gateway can communicate with all gateways in the gateway set, considering the impact of distance between gateways, the optimal solution is to choose the gateway directly connected to it for communication; S43: When there are insufficient gateway resources to allocate to subtasks, the subtask is assigned to the candidate gateway set of the previous subtask. That is, the gateway to which the subtask is executed is the neighboring gateway directly connected to the gateway to which the previous subtask was executed.

[0013] Furthermore, step S5 specifically includes the following steps: S51: Due to the robustness and real-time requirements of critical services, in order to enable TSN gateway devices to provide wait-free transmission services for time-sensitive critical services, calculate subtasks. The processing start time is:

[0014] in, Indicates business The arrival time (i.e., the arrival time of the first subtask). Subtasks Processing latency in TSN gateways Subtasks Uploaded to the gateway Fixed queuing forwarding latency for port gating; S52: Obtain the subtask according to step S51. Execution start time in the gateway Based on the 5G wireless transmission latency obtained in step S3, gateway resources are allocated to the task to be executed in step S4. S53: If allocated gateway If a task cannot be occupied, a preemptive scheduling mechanism will be triggered, assigning a higher priority subtask. It can only preempt subtasks of other low-priority services that are already cached. Specifically, by finding subtasks with lower priority Other subtasks Forcefully terminate the execution of the subtask ,until After the initial execution is complete, continue processing the remaining unprocessed data. S54: Repeat steps S4 and S5 until all tasks have been completed.

[0015] The beneficial effects of this invention are as follows: This invention analyzes the wireless transmission latency of services in the 5G network and the processing latency of services in the TSN network for 5G and TSN collaborative transmission systems, and proposes a 5G-TSN collaborative optimization architecture. It also considers the impact of wireless channel quality on the reliability of service transmission. To ensure the reliability and determinism of service transmission, this invention designs a cross-domain collaborative scheduling method for 5G-TSN with deterministic latency guarantees, analyzing the transmit power at the sensor end, gateway resource allocation, and service scheduling. This invention saves service completion latency costs under service deadline constraints while improving the utilization efficiency of TSN gateway resources and coordinating the processing of services with different priorities, demonstrating broad application prospects.

[0016] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0017] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein: Figure 1 This is an architecture diagram of the 5G-TSN cross-domain collaborative optimization communication system of the present invention; Figure 2 This refers to the network configuration and user configuration involved in this invention; Figure 3 Setting the output port gating list for this invention; Figure 4 This describes the synchronization process of the network in this invention. Detailed Implementation

[0018] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0019] Please see Figures 1-4This invention provides a 5G-TSN cross-domain collaborative scheduling method for deterministic latency guarantees. First, a 5G-TSN integrated cross-domain collaborative framework is proposed, deploying 5G and TSN gateways together at the edge. Second, the problem of providing deterministic latency guarantees for time-sensitive and non-time-sensitive services is investigated. Finally, considering the channel uncertainty and limited gateway resources of the 5G wireless system, a joint 5G and TSN cross-domain collaborative scheduling method is designed to optimize service processing latency. The proposed method consists of three heuristic algorithms. To ensure the reliability of service transmission, a power allocation strategy is designed based on the device's packet loss rate to obtain the optimal feasible solution. Then, based on limited gateway resources and the service quality requirements of the services, gateway resources are pre-allocated. To further optimize the processing latency of higher-priority services, a preemptive task scheduling method is designed. The method specifically includes the following steps: Step 1: Construct a 5G-TSN cross-domain collaborative optimization communication system: Considering the collaborative scheduling of 5G and TSN, time-sensitive and non-time-sensitive services of sensor devices are injected from 5G to TSN. The TSN gateway and 5G micro base stations are deployed at the edge layer. The TSN gateway is configured with a standard 5G interface to assist the TSN gateway in communicating with sensor devices under the centralized management of the 5G network. The 5G micro base stations act as edge computing nodes, and the entire edge 5G system acts as a TSN logical bridge. Specifically, this includes the following steps: Step 1.1: To guarantee deterministic latency, the TSN gateway provides a time-aware shaper, which provides guaranteed time slots for specific traffic categories through a pre-calculated gate control list, enabling time-sensitive services to complete task delivery within the deadline. Since TSN scheduling needs to be calculated in a centralized manner, a fully centralized architecture defined in the IEEE 802.1 Qcc standard is adopted, and centralized management is achieved through a software-defined networking (SDN) controller, including centralized user configuration and centralized network configuration. Step 1.2: The Business Composed of a series of ordered subtasks, often referred to as a task function chain, it has a strict execution order and can be represented as... , any , indicating business Subtasks Indicates business The number of subtasks, and each subtask With parameter set , This indicates the CPU resources required for the execution of the subtask. This indicates the amount of cache space required for the execution of the subtask. This refers to the size of the subtask data that needs to be uploaded.

[0020] Step 1.3: To ensure timely service for latency-sensitive services, the application needs to retain priority tags for these services to coordinate the processing of services with different priorities. Assume that service requests with different QoS requirements are processed according to their arrival rate... The Poisson distribution arrives at the network in an orderly manner, and the services... Has priority .

[0021] Step 1.4: A TSN network can consist of an incomplete connection graph. Let represent, where the set It is a TSN gateway node that assists in communication between the 5G network and sensor nodes, and integrates... These are sets of wired links used for communication between TSN gateways. For the first... gateway With CPU resource size The cache size is and the set of neighboring nodes , Directly connected to the node The network consists of neighboring TSN gateway nodes. The latency budget for services in a 5G network is mainly related to the air interface channel, while the air interface rate... It is closely related to the quality of the wireless channel and involves a certain degree of uncertainty. The 5G air interface rate can be expressed as:

[0022] Wherein, the signal-to-interference ratio is expressed as , It's bandwidth. It is the transmit power at the sensor device end. This represents the straight-line distance between the sensor and the 5G micro base station, and is a parameter in the path loss model. Represents the loss exponent and channel gain. It follows a Rayleigh distribution with mean 0 and variance 1, and additive white Gaussian noise is introduced, with a power spectral density of... .

[0023] Step 2: Weighted Model of 5G Wireless Transmission Latency and TSN Processing Latency: To find an optimal cross-domain scheduling mechanism between 5G and TSN to optimize the overall service flow latency within the deadline, a weighted model of 5G wireless transmission latency and TSN processing latency was designed. This includes the following steps: Step 2.1: The device injects data into the TSN network via the 5G wireless system. The power configuration of the device ensures reliable data transmission. Without loss of generality, the interruption probability can be expressed as:

[0024] in, This is the signal-to-noise ratio threshold; communication is interrupted if the value falls below this threshold. Furthermore, power configuration also optimizes wireless transmission latency and subtasks. Transmission latency on 5G wireless channels It can be represented as follows:

[0025] Step 2.2: In the TSN network, services The arrival time and completion time are defined as follows: and Similarly, the arrival and completion times of subtasks are defined as follows: and Defined binary variables Indicates TSN gateway Does it serve the business? The Sub-tasks The subtask is located on the TSN gateway. The processing latency is defined as the time difference between the arrival time and the completion time of the service, as follows:

[0026] Step 2.3: Different subtasks may represent different functions, therefore it is necessary to process sets of subtasks that make up the same service sequentially. For time-sensitive constraints, the flow isolation and frame isolation characteristics of TSN must be followed. The isolation constraints between subtasks are then expressed as:

[0027] in, It is a sub-task of TSN The execution start time.

[0028] Step 2.4: Service requests generated on the device side are injected into the TSN network for processing by the 5G system as a TSN logical bridge. Therefore, service analysis is required. The wireless transmission latency in the 5G band and the processing latency on the TSN gateway are expressed by the following expressions:

[0029]

[0030] Step 2.5: Business The collaborative scheduling latency consists of two parts: the 5G segment and the TSN segment, and can be obtained through the following methods:

[0031] Because when a new subtask arrives, the TSN gateway may be serving another subtask. Therefore, the business... The waiting latency is the weighted sum of the waiting latency of all subtasks, defined as:

[0032] Among them, parameters Represented as subtask Uploaded to the gateway Fixed queuing forwarding latency for port gating; Step 2.6: Find an optimal cross-domain scheduler between 5G and TSN to optimize the latency of the overall service flow within the deadline, providing deterministic latency guarantees for both time-sensitive and non-time-sensitive services. The specific optimization objectives are constructed as follows:

[0033] Among them, parameters and These are the interrupt threshold and the latency threshold, respectively. Indicates gateway The cached set of subtasks for different service flows. Constraints C1 and C2 represent the limitations on CPU resources and cache capacity of the TSN gateway, respectively; constraint C3 ensures that a TSN gateway can only execute one subtask at a time; constraint C4 indicates that the TSN gateway has sufficient resources to schedule the set of subtasks; constraint C5 represents the interruption probability threshold condition; constraint C6 represents the flow isolation and frame isolation requirements of TSNs that execute the same service flow subtasks sequentially; constraint C7 is the service flow latency threshold constraint.

[0034] Step 3: Device Transmit Power Allocation: Reliable wireless channel transmission is closely related to power. The optimal power allocation is derived while meeting the packet loss rate requirements of the terminal device to ensure reliable data transmission. This includes the following steps: Step 3.1: The expression for the interruption probability is strongly correlated with the transmission power. The wireless transmission latency of the 5G network decreases as the power increases, and the interruption probability also shows the same trend. Step 3.2: Based on the analysis in step S31, under certain reliability requirements, [the following will be implemented] To achieve optimal power allocation and optimize 5G wireless transmission latency.

[0035] Step 4: TSN Gateway Resource Pre-allocation: To ensure that most services in the TSN network provide deterministic service within limited resource capacity during their deadlines, pre-allocating gateway resources can effectively improve the system's service capacity. A greedy algorithm is used to obtain the pre-allocated gateway resources, providing optimal TSN gateway placement and processing resource allocation for tasks. Specifically, this includes the following steps: Step 4.1: The network not only optimizes latency to deliver traffic as quickly as possible, but also aims to deliver traffic promptly with limited computing resources. Different gateways have different CPU sizes and buffer capacities, depending on the weighted resources. Subtasks in descending order Candidate gateway set Sort all gateways in the system. It is just a weighting factor used to balance the proportion of gateway CPU resources and buffer resources; Step 4.2: Although each TSN gateway can communicate with all gateways in the gateway set, considering the impact of distance between gateways, the optimal solution is to select the gateway that is directly connected to it. Step 4.3: When gateway resources are insufficient, allocate resources to subtasks. In this case, subtasks Assigned to the previous subtask Candidate gateway set Subtask The gateway being executed is the previous subtask. The gateway being executed is the neighboring gateway directly connected to.

[0036] Step 5: Preemptive 5G-TSN Cross-Domain Cooperative Scheduling Method: To provide a feasible solution, a preemptive heuristic algorithm is proposed. To further ensure low latency and real-time performance for high-priority services, a preemptive cross-domain cooperative scheduling scheme is studied. Specifically, it includes the following steps: Step 5.1: Due to the robustness and real-time requirements of critical services, in order to enable TSN gateway devices to provide wait-free transmission services for time-sensitive critical services, calculate subtasks. Processing start time:

[0037] Step 5.2: Obtain the execution start time in the gateway where the subtask is located, based on Step 5.1. Based on the 5G wireless transmission latency obtained in step 3, allocate gateway resources for the task to be executed according to step 4. Step 5.3: If assigned to gateway If a task is not available for preemption, a preemptive scheduling mechanism will be triggered, assigning higher priority subtasks to it. It can only preempt other low-priority business subtasks that are already cached. Specifically, by finding subtasks with lower priority Other subtasks Forcefully terminate the execution of the subtask ,until After the initial execution is complete, continue processing the remaining unprocessed data. Step 5.4: Repeat steps 4 and 5 until all tasks have been completed.

[0038] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A 5G-TSN cross-domain collaborative scheduling method for deterministic latency assurance, characterized in that, The method specifically includes the following steps: S1: Construct a 5G-TSN cross-domain collaborative optimization communication system: Consider the joint scheduling of 5G and TSN, inject time-sensitive and non-time-sensitive services of sensor devices from 5G into TSN. TSN gateways and 5G micro base stations are deployed at the edge layer. The TSN gateway is configured with a standard 5G interface to assist the TSN gateway in communicating with sensor devices under the centralized management of the 5G network. The 5G micro base station acts as an edge computing node, while the entire edge 5G system acts as a TSN logical bridge. S2: Based on the service requirements of the business flow, construct a weighted optimization model for 5G wireless transmission latency and TSN processing latency; S3: Equipment transmit power allocation: Derive the closed expression for the interruption probability and the solution conditions for the optimal power while meeting the reliability transmission requirements of the terminal equipment; S4: TSN Gateway Resource Pre-allocation: Based on a greedy algorithm, gateway resources are pre-allocated to provide optimal TSN gateway placement and processing resource allocation for tasks; S5: A preemptive heuristic algorithm is used to achieve cross-domain collaborative scheduling; In step S1, the 5G-TSN cross-domain collaborative optimization communication system is constructed, which specifically includes the following steps: S11: The TSN gateway provides a time-aware shaper that provides guaranteed time slots for specific traffic categories through a pre-computed gate control list. TSN scheduling is calculated in a centralized manner, using a fully centralized architecture defined in the IEEE 802.1 Qcc standard, and is centrally managed through a software-defined network controller, including centralized user configuration and centralized network configuration. S12: The Business Composed of a series of ordered sets of subtasks, often referred to as a task function chain, it has a strict execution order and is represented as... , any Indicates business Subtasks Indicates business The number of subtasks, and each subtask With parameter set , This indicates the CPU resources required for the execution of the subtask. This indicates the amount of cache space required for the execution of the subtask. This refers to the size of the subtask data that needs to be uploaded; S13: In order to ensure that latency-sensitive services are served in a timely manner, the application retains priority tags for services to coordinate the processing of services with different priorities; S14: The TSN network has an incomplete connection diagram. Let represent, where the set It is a TSN gateway node that assists in communication between the 5G network and sensor nodes, and integrates... These are sets of wired links used for communication between TSN gateways. For the first... gateway With CPU resource size The cache size is and the set of neighboring nodes , Directly connected to the node It consists of neighboring TSN gateway nodes. The latency budget of services in the 5G network is related to the air interface channel, while the air interface rate... Related to wireless channel quality; In step S2, a weighted optimization model for 5G wireless transmission latency and TSN processing latency is constructed, which specifically includes the following steps: S21: The device injects data into the TSN network through the 5G wireless system. The power configuration of the device needs to ensure the reliable transmission of data, while also optimizing the wireless transmission latency. S22: In a TSN network, calculate the processing time of a subtask at the gateway based on the arrival and completion times of the service and subtask. S23: Process sub-task sets that make up the same service in sequence. For time-sensitive constraints, the flow isolation and frame isolation characteristics of TSN must be followed. S24: Service requests generated by the device are injected into the TSN network for processing by the 5G system as a TSN logical bridge, and the wireless transmission latency of the service in the 5G band and the processing latency on the TSN gateway are analyzed. S25: The total delay of service coordination and scheduling consists of two parts: the 5G segment and the TSN segment. Since the TSN gateway may serve another subtask when a new subtask arrives, the TSN segment includes processing delay and gateway fixed queuing forwarding delay. S26: Optimize the latency of the overall service flow between 5G and TSN within the deadline, and provide deterministic latency guarantees for both time-sensitive and non-time-sensitive services; In step S26, the latency optimization model constructed is as follows: Among them, constraints C1 and C2 represent the limitations on CPU resources and cache capacity of the TSN gateway, respectively; constraint C3 ensures that a TSN gateway can only execute one subtask at a time; constraint C4 indicates that the TSN gateway has sufficient resources to schedule the subtask set; constraint C5 represents the interruption probability threshold condition; constraint C6 represents the flow isolation and frame isolation requirements of TSNs that execute the same service flow subtasks sequentially; constraint C7 is the service flow latency threshold constraint; binary variables Indicates TSN gateway Does it serve the business? The Sub-tasks , It is a sub-task of TSN The execution start time, It is the transmit power at the sensor device end. Indicates business Coordinated scheduling latency, Represents a set of business functions; Indicates gateway The cached set of subtasks from different business flows; Indicates the probability of interruption; and These are the interrupt threshold and the delay threshold, respectively. Subtasks Processing latency in TSN gateways and Subtasks Arrival time and completion time.

2. The 5G-TSN cross-domain collaborative scheduling method according to claim 1, characterized in that, Step S3 specifically includes: under reliability requirements, in To achieve optimal power allocation and optimize 5G wireless transmission latency.

3. The 5G-TSN cross-domain collaborative scheduling method according to claim 1, characterized in that, Step S4 specifically includes the following steps: S41: Sort all gateways in the candidate gateway set of the subtask in descending order according to the weighted resources; S42: Each TSN gateway selects the gateway directly connected to it for communication; S43: When there are insufficient gateway resources to allocate to subtasks, the subtask is assigned to the candidate gateway set of the previous subtask. That is, the gateway to which the subtask is executed is the neighboring gateway directly connected to the gateway to which the previous subtask was executed.

4. The 5G-TSN cross-domain collaborative scheduling method according to claim 1, characterized in that, Step S5 specifically includes the following steps: S51: Computation Subtask The processing start time is: in, Indicates business The arrival time, i.e., the arrival time of the first subtask; Subtasks Processing latency in TSN gateways Subtasks Uploaded to the gateway Fixed queuing forwarding latency for port gating; S52: Obtain the subtask according to step S51. Execution start time in the gateway Based on the 5G wireless transmission latency obtained in step S3, gateway resources are allocated to the task to be executed in step S4. S53: If allocated gateway If a task cannot be occupied, a preemptive scheduling mechanism will be triggered, assigning a higher priority subtask. It can only preempt subtasks of other low-priority services that are already cached. Specifically, by finding subtasks with lower priority Other subtasks Forcefully terminate the execution of the subtask ,until After the initial execution is complete, continue processing the remaining unprocessed data. S54: Repeat steps S4 and S5 until all tasks have been completed.

Images