Capacity analysis and estimation method for time-sensitive network traffic group
By defining the capacity indicators of the TSN network and building a capacity analysis model, the problem of being unable to evaluate the TSN network capacity in existing technologies is solved, rapid judgment of traffic schedulability is achieved, and the efficiency and accuracy of network deployment are improved. It is suitable for time-sensitive network traffic groups in the Industrial Internet of Things.
Patent Information
- Application Number
- CN202510940092.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-08
- Publication Date
- 2025-09-23
AI Technical Summary
Existing capacity analysis methods cannot effectively evaluate the capacity of TSN networks, resulting in an inability to quickly determine the schedulability of traffic, affecting the solution of TSN scheduling problems and network deployment efficiency.
Define the capacity indicators of the TSN network, build single-link and multi-link capacity analysis models, and combine the characteristics of traffic to provide a quick judgment capability for traffic schedulability, including single-link capacity estimation and expansion of the overall network capacity.
It improves the accuracy and efficiency of TSN network capacity assessment, improves the schedulability of traffic scheduling, realizes the schedulability of traffic, improves the efficiency and accuracy of assessment, solves the problem of being unable to effectively judge the schedulability of traffic in existing technologies, improves the flexibility of network deployment and reduces costs.
Smart Images
Figure SMS_4 
Figure FDA0005489552230000011 
Figure FDA0005489552230000012
Abstract
Description
Technical Field
[0001] The present invention relates to the field of industrial Internet of Things, and in particular to a capacity analysis and estimation method for time-sensitive network traffic groups. Background Art
[0002] With the rapid development of the Industrial Internet of Things (IIoT), the widespread connectivity of people, machines, materials, and other elements in industrial sites has become an inevitable trend. However, industrial applications present certain time-sensitive requirements, including business traffic that requires low latency and high-quality service delivery. Traditional industrial Ethernet networks struggle to meet these demands. To address this challenge, the IEEE established a working group and proposed a suite of protocols for Time Sensitive Networking (TSN). TSN provides deterministic traffic transmission through refined scheduling, but its configuration process also becomes more complex, requiring certain pre-orchestrations and scheduling of traffic to implement network applications. This process is called TSN traffic scheduling.
[0003] Scheduling traffic requires extensive computation, which typically consumes considerable time. Furthermore, as demand changes, it's often difficult to find a feasible solution. Being able to determine whether traffic is schedulable before scheduling begins can save significant time. From a network perspective, if a set of traffic flows can be deterministically transmitted within a TSN network, it means the network can accommodate this set of traffic flows—that is, the flow rate is within the network's capacity. Therefore, defining and estimating the capacity of a TSN network is crucial for determining the schedulability of TSN traffic flows, accelerating the solution to TSN scheduling problems, and improving network deployment efficiency.
[0004] Most existing analyses of network capacity focus on other types of networks and are not applicable to TSN networks. Therefore, designing capacity metrics and estimation methods tailored to the characteristics of TSN networks and traffic is a highly innovative endeavor.
[0005] Therefore, researchers in this field have dedicated themselves to developing a capacity analysis and estimation method for time-sensitive network traffic groups. By leveraging the characteristics of TSN networks and constructing capacity analysis models applicable to both single and multi-links, they provide the ability to quickly determine the schedulability of traffic flows. This method not only accurately estimates the capacity of a single link but also extends this capacity from a single link to the entire network, providing technical support for comprehensive assessment of TSN network capacity. This analytical framework, spanning the link layer to the network layer, significantly improves the efficiency and accuracy of assessments, providing a solid foundation for the feasibility of traffic scheduling in complex industrial network environments. Summary of the Invention
[0006] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is to define and estimate the capacity of the TSN network, determine the schedulability of TSN traffic, accelerate the solution of TSN scheduling problems, and improve network deployment efficiency.
[0007] To achieve the above objectives, the present invention provides a capacity analysis and estimation method for time-sensitive network traffic groups, including determining traffic schedulability by defining network capacity indicators.
[0008] Furthermore, the network capacity is defined as: TSN network N, fixed flow group F, if N can simultaneously allow a maximum of c groups of flow groups F to be transmitted while meeting traffic requirements, then the capacity of network N for flow group F is c, denoted as C(N,F)=c.
[0009] Furthermore, the method for judging the schedulability of the traffic is that if the number of flow groups in the network exceeds the network capacity, the traffic cannot be scheduled; otherwise, the traffic can be scheduled.
[0010] Furthermore, a single-link capacity estimation method for TSN flow groups is established.
[0011] Furthermore, the single link capacity is obtained through the packet length, period, number, network bandwidth and CQF time slot length parameters of each flow in the flow group.
[0012] Furthermore, the traffic is grouped into uniform flows to determine TSN time slot occupancy.
[0013] Furthermore, any flow is reduced to a uniform flow.
[0014] Furthermore, the single link capacity calculation includes the following steps:
[0015] The flow numbered i in flow group F has a length of l i , period is p i , the number is n i , the network bandwidth is Γ, and the CQF time slot length parameter is T;
[0016] Step 1: Initialize the capacity occupancy ζ←0. The arrow indicates that the value on the right is assigned to the value on the left.
[0017] Step 2: Calculate the number of the stream to be reduced: γ←argmin j∈F n j / p j ;
[0018] Step 3: Calculate the time slot occupied by the current uniform flow:
[0019] Step 4: Remove the flow f that needs to be reduced from the flow group F. γ , and other flows in the flow group fj The number n j Reset to
[0020] Step 5: Repeat steps 2 to 4 until there are no more flows in flow group F.
[0021] Step 6: Calculate the TSN network capacity C = ΓT / ζ, which is the single-link capacity.
[0022] Furthermore, the overall TSN capacity of the network is obtained based on the single link capacity.
[0023] Furthermore, the overall TSN capacity of the network is the minimum value of the single link capacity.
[0024] Existing capacity metrics are incapable of assessing the capacity of TSN networks, and therefore, are unable to quickly determine traffic schedulability. This paper, combining the characteristics of the TSN network and traffic, proposes a capacity analysis method for time-sensitive network traffic groups. Specifically, it defines a network capacity metric to determine traffic schedulability. This paper analyzes the characteristics of the TSN mechanism, network, and traffic, explores the relationship between the TSN network and traffic groups, and defines a metric that can measure TSN capacity. Capacity is defined as follows: given a TSN network N and a fixed flow group F, if N can simultaneously allow at most c flow groups F to transmit while meeting traffic requirements, then the capacity of network N for flow group F is c, denoted as C(N,F) = c. Under this definition, if the number of flow groups in the network exceeds the capacity, they are unschedulable; otherwise, they are schedulable. This paper can quantitatively determine the capacity of a TSN network, providing a basis for determining whether TSN applications can be deployed. This helps quickly eliminate unschedulable scenarios, improves the flexibility of TSN application deployment, and reduces costs.
[0025] Existing capacity estimation methods cannot estimate the capacity of a single-link TSN. The present invention establishes a single-link network capacity estimation method for TSN flow groups. Given the packet length, period, number of each flow in the flow group and the network bandwidth and CQF time slot length parameters, the value of the single-link capacity can be calculated through a set of iterative calculations. By analyzing the TSN network mechanism, the present invention finds that grouping traffic into uniform flows can quickly determine TSN time slot occupancy, and designs a method that can reduce any traffic to uniform flows. Suppose the flow numbered i in the flow group has a length of l i , period is p i , the number is n i , the network bandwidth is Γ, and the CQF time slot length parameter is T. Then the single-link network capacity is calculated by the following steps:
[0026] 1. Initialize capacity occupancy ζ←0;
[0027] 2. Calculate the number of streams that need to be reduced: γ←argmin j∈F n j / p j ;
[0028] 3. Calculate the time slot occupied by the current uniform flow:
[0029] 4. Remove the flow f that needs to be reduced in flow group F γ , and other flows in the flow group f j The number n j Reset to
[0030] 5. Repeat steps 2-4 above until there are no flows in flow group F.
[0031] 6. Calculate the TSN network capacity C = ΓT / ζ.
[0032] According to this step, the C calculated in the last step is the single-link network capacity. The present invention provides a practical TSN capacity estimation method that can quickly estimate the TSN network capacity of a single link.
[0033] Existing capacity estimation methods cannot verify the relationship between single-link capacity and overall network capacity in the TSN network capacity for flow groups. Based on the single-link network capacity estimation method for TSN flow groups, the present invention derives a method for analyzing the overall TSN capacity of the network based on the local TSN capacity. Under the premise of routing determination, given the single-link capacity of each link, the overall TSN capacity of the network is the minimum value of the single-link capacity. Under the premise of given flow routing behavior, the present invention derives through mathematical deduction that the overall TSN capacity of the network is equal to the minimum value of the capacity of each single link in the network. Specifically, let the capacity of the network be C and the minimum link capacity be C. m , the corresponding link is L m Assume that the initial number of flow groups in the network is C m Obviously, the number of flow groups assigned to each link is also C m Since the capacity group number of each link is not less than C m , so all links in the network can carry their corresponding flow groups. In other words, the network N can accommodate the flow group F. From the definition of network capacity, we can get C ≥ C m On this basis, if a new set of flows F is added, since the routing is already determined, the link L m The number of flows on will become C m+1 >C m Since link L m The capacity is C m , cannot carry the newly added traffic, so the entire network N cannot carry the newly added flow group F, that is, C <Cm+1 From the above, we can get that C=C m , that is, when the flow routing is given, the network capacity is equal to the minimum value of the capacity of each single link. This invention expands the estimation of TSN network capacity from single link to multiple links, which can quickly estimate the overall capacity of the TSN network and thus quickly determine the schedulability of the TSN.
[0034] Compared with the prior art, the present invention has the following obvious substantial features and significant advantages:
[0035] The present invention proposes a capacity indicator definition and estimation method for TSN networks, which fills the gap in the field of TSN network capacity analysis in existing research and has significant technical advantages. This method effectively combines the characteristics of TSN networks and provides a rapid judgment capability for traffic schedulability by constructing a capacity analysis model suitable for single links and multiple links. Compared with traditional methods, this solution can not only accurately estimate the capacity of a single link, but also achieves the expansion from a single link to the entire network, providing technical support for the comprehensive evaluation of TSN network capacity. This analysis framework from the link layer to the network layer greatly improves the efficiency and accuracy of the evaluation, and provides a solid guarantee for the feasibility of traffic scheduling in complex industrial network environments.
[0036] In terms of performance metrics, the capacity estimation method of this invention can quantitatively analyze the relationship between TSN network capacity and traffic characteristics, providing a scientific basis for determining the feasibility of traffic scheduling. By accurately defining capacity metrics, this method can quickly identify unschedulable scenarios, thereby avoiding inefficient resource consumption. This feature is particularly suitable for industrial scenarios with frequent dynamic load changes, significantly improving network configuration efficiency and response speed.
[0037] In terms of production implementation, this technical solution has broad prospects for industrial application. Its core advantage lies in combining the latency-sensitive characteristics of TSN traffic to provide efficient capacity analysis tools for high-precision communication scenarios such as the Industrial Internet and intelligent manufacturing. This capability not only reduces the cumbersome scheduling requirements and unnecessary resource overhead during network deployment, but also optimizes resource allocation and system performance by quickly estimating the matching degree between network capacity and traffic, providing a theoretical analysis tool for quickly adjusting the network according to actual application needs. Therefore, the present invention can reduce the cost of TSN deployment and provide strong technical support for the large-scale implementation of TSN in the field of Industrial Internet of Things.
[0038] The concept, specific structure and technical effects of the present invention will be further described below to fully understand the purpose, characteristics and effects of the present invention. DETAILED DESCRIPTION
[0039] The following describes several preferred embodiments of the present invention to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.
[0040] The present invention aims to solve the problem of capacity analysis and estimation in TSN networks. To this end, the present invention proposes a TSN capacity analysis method and estimation algorithm.
[0041] In the capacity method, the TSN network is configured using the Cyclic Queuing and Forwarding (CQF) method specified in IEEE 802.1Qch. The complete analysis steps are as follows:
[0042] Step 1-1: Analyze the relevant parameter characteristics of the TSN network and establish parameters such as bandwidth Γ, queue buffer, network topology, and time slot length T in the TSN network N.
[0043] Step 1-2: Analyze the relevant parameter characteristics of the TSN traffic group and establish flow f with label i in traffic group F. i The number n i , period p i , length l i and other parameters.
[0044] Step 1-3: Establish the capacity C(N,F) of the TSN network N for traffic group F. This means the number of traffic groups F that can be transmitted as required within N, which is the TSN network capacity.
[0045] In the capacity estimation method, the single link estimation method is as follows:
[0046] Step 2-1: Initialize capacity occupancy ζ←0 and obtain flow attributes and network attributes. Suppose there are m types of flows in the flow group F, where each flow f is labeled i. i The properties include the number of streams n i , flow period p i , flow length l i Assume that in the network N, the bandwidth is Γ and the CQF time slot capacity is T.
[0047] Step 2-2: When there is traffic in the flow group, execute steps 2-3 to 2-5:
[0048] Step 2-3: Calculate the stream number γ that needs to be reduced: γ←argmin j∈F n j / p j ;
[0049] Step 2-4: Calculate the time slot ζ occupied by the current uniform flow:
[0050] Step 2-5: Remove the flow f that needs to be reduced from the flow group F γ , and other flows in the flow group f j The number n j Reset to
[0051] Step 2-6: Repeat the above steps until there are no flows in flow group F.
[0052] Step 2-7: Calculate the TSN network capacity C = ΓT / ζ.
[0053] In the capacity estimation method, the overall capacity estimation method of the TSN network is as follows:
[0054] Step 3-1: Repeat the above single-link capacity estimation method for each link in the TSN network to estimate the capacity of each link;
[0055] Step 3-2: Find the minimum value of the single-link capacity in the TSN network, which is the overall capacity of the network.
[0056] According to the capacity estimation method, the method for judging the schedulability of traffic in the TSN network is as follows:
[0057] Step 4-1: If the number of flow groups in the TSN network is greater than the capacity, the group of flows cannot be scheduled; otherwise, it can be scheduled.
[0058] The preferred embodiments of the present invention have been described in detail above. It should be understood that numerous modifications and variations based on the concepts of the present invention are possible without inventive effort by those skilled in the art. Therefore, any technical solution that can be derived by one skilled in the art through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.
Claims
1. A capacity analysis and estimation method for time-sensitive network traffic groups, characterized in that: This includes determining traffic schedulability by defining network capacity indicators.
2. The method for capacity analysis and estimation of time-sensitive network traffic groups according to claim 1, wherein: The network capacity is defined as: TSN network N, fixed flow group F, if N can simultaneously allow a maximum of c groups of flow groups F to be transmitted while meeting traffic requirements, then the capacity of network N for flow group F is c, denoted as C(N,F)=c.
3. The method for capacity analysis and estimation of time-sensitive network traffic groups according to claim 2, wherein: The method for judging the schedulability of the traffic is that if the number of flow groups in the network exceeds the network capacity, the traffic cannot be scheduled; otherwise, the traffic can be scheduled.
4. The method for capacity analysis and estimation of time-sensitive network traffic groups according to claim 1, wherein: A single-link capacity estimation method for TSN flow groups is established.
5. The capacity analysis and estimation method for time-sensitive network traffic groups according to claim 4, characterized in that: The single link capacity is obtained by the packet length, period, number, network bandwidth and CQF time slot length parameters of each flow in the flow group.
6. The method for capacity analysis and estimation of time-sensitive network traffic groups according to claim 4, wherein: The traffic is grouped into uniform flows to determine TSN time slot occupancy.
7. The method for capacity analysis and estimation of time-sensitive network traffic groups according to claim 4, wherein: Reduce any flow to a uniform flow.
8. The method for capacity analysis and estimation of time-sensitive network traffic groups according to claim 4, wherein: The single link capacity calculation comprises the following steps: The flow numbered i in flow group F has a length of l i , period is p i , the number is n i , the network bandwidth is Γ, and the CQF time slot length parameter is T; Step 1: Initialize the capacity occupancy ζ←0. The arrow indicates that the value on the right is assigned to the value on the left. Step 2: Calculate the number of the stream to be reduced: γ←argmin j∈F n j / p j ; Step 3: Calculate the time slot occupied by the current uniform flow: Step 4: Remove the flow f that needs to be reduced from the flow group F. γ , and other flows in the flow group f j The number n j Reset to Step 5: Repeat steps 2 to 4 until there are no more flows in flow group F. Step 6: Calculate the TSN network capacity C = ΓT / ζ, which is the single-link capacity.
9. The method for capacity analysis and estimation of time-sensitive network traffic groups according to claim 4, wherein: Based on the single link capacity, the overall TSN capacity of the network is obtained.
10. The capacity analysis and estimation method for time-sensitive network traffic groups according to claim 9, characterized in that: The overall TSN capacity of the network is the minimum value of the single link capacity.