A TSN network bandwidth parameter pre-allocation method based on network calculus

Through a network calculus-based method, the problems of long calculation time and low bandwidth utilization during TSN network parameter allocation are solved, and efficient bandwidth resource allocation at the node level is achieved, which is suitable for network topologies of various sizes.

CN119561916BActive Publication Date: 2025-10-03BEIHANG UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411559341.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-11-04
Publication Date
2025-10-03
Estimated Expiration
2044-11-04

AI Technical Summary

Technical Problem

Traditional methods have problems with long calculation time and low bandwidth utilization when allocating TSN network parameters.

Method used

A network calculus-based method is used to obtain the parameter information set of the TSN network, perform traffic modeling, obtain the aggregate arrival curve and remaining service curve of AVB traffic, and solve them to obtain the bandwidth allocation parameters of the TSN network.

Benefits of technology

Under the premise of ensuring the traffic delay limit, it allocates the least bandwidth resources, has high efficiency and good scalability, and can be applied to network topologies of various sizes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119561916B_ABST
    Figure CN119561916B_ABST
Patent Text Reader

Abstract

This invention discloses a TSN network bandwidth parameter pre-allocation method based on network calculation, comprising: obtaining a set of TSN network parameter information; the TSN network is a time-sensitive network; performing traffic modeling on the set of parameter information to obtain an aggregate arrival curve and a residual service curve for AVB traffic; and solving the aggregate arrival curve and the residual service curve to obtain bandwidth allocation parameters for AVB traffic in the TSN network. This method can allocate the minimum feasible bandwidth resources at the node level while ensuring traffic delay bounds. It is highly efficient and scalable, and can be applied to network topologies of various sizes.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of industrial Internet, and in particular to a TSN network bandwidth parameter pre-allocation method based on network calculation. Background Art

[0002] With the rapid development of industrial automation, cloud-based automation has emerged as an architecture that seamlessly integrates communication, computing, and control. Time-Sensitive Networking (TSN) plays a key role in supporting deterministic communication in cloud-based automation. TSN is a set of standards developed by the Time-Sensitive Networking Task Group of the IEEE 802.1 Working Group. It ensures real-time performance for time-critical communications. TSN supports the integration of multiple traffic types within a single network. Traffic flows are categorized into three types based on real-time requirements: Time-Aware Streaming (TT) flows, AVB flows, and Best-Effort (BE) flows, with progressively lower real-time requirements. The Time-Aware Shaper (TAS) ensures low latency for TT flows by reserving completely independent transmission windows. The Credit-Based Shaper (CBS) aims to avoid blocking low-priority traffic flows. Whether an AVB frame in a queue can be transmitted is controlled not only by the queue gate, but also by the CBS shaper. An AVB frame can be transmitted only when its queue gate is open and the CBS shaper allows it to be transmitted. How to allocate appropriate bandwidth parameters to AVB traffic while taking into account the fixed time slots of TT traffic and saving bandwidth resources as much as possible while ensuring the end-to-end delay bound of the traffic remains an open problem.

[0003] Traditional approaches to allocating bandwidth parameters for AVB traffic under the TAS-CBS shaper typically use continuous worst-case delay analysis (WLTP) feedback to gradually adjust bandwidth allocation parameters. However, this approach requires significant computation time and often results in significant bandwidth waste. The present invention addresses the challenges of traditional methods for allocating TSN network parameters, such as long computation times and low bandwidth utilization. Summary of the Invention

[0004] The present invention mainly solves the defects of long calculation time and low bandwidth utilization in traditional methods when allocating TSN network parameters. The present invention discloses a TSN network bandwidth parameter pre-allocation method and device based on network calculation.

[0005] In a first aspect, an embodiment of the present application discloses a TSN network bandwidth parameter pre-allocation method based on network calculation, comprising:

[0006] S1, obtaining a set of parameter information of a TSN network; the TSN network is a time-sensitive network;

[0007] S2, performing traffic modeling processing on the parameter information set to obtain an aggregated arrival curve and a residual service curve of the AVB traffic;

[0008] S3: Solve the aggregated arrival curve and the remaining service curve to obtain bandwidth allocation parameters of the TSN network.

[0009] The TSN network includes a plurality of interconnected nodes; each node includes a plurality of ports;

[0010] The parameter information set includes node information of the TSN network, traffic information flowing through each node, link rate C, TT traffic information set and AVB traffic information set in the TSN network;

[0011] The TT traffic information set includes traffic sequence number, frame length, end-to-end delay limit The transmission path, minimum frame interval, and scheduling schedule of TT traffic at each port; the scheduling schedule includes the start and end time of the transmission window of the TT traffic at the port, and the traffic sequence number;

[0012] The AVB traffic information set includes traffic sequence number, frame length, and end-to-end delay limit Propagation path, minimum frame interval;

[0013] The traffic types in the TSN network include TT traffic and AVB traffic.

[0014] The performing traffic modeling on the parameter information set to obtain an aggregated arrival curve and a residual service curve of the AVB traffic includes:

[0015] S21, performing statistical processing on the parameter information set to obtain the TT traffic sequence number and AVB traffic sequence number flowing through each node, and the super period of the TT traffic; the super period is the least common multiple of the minimum frame interval of the TT traffic;

[0016] S22, performing local delay calculation processing on the parameter information set, decomposing the end-to-end delay bound of the traffic based on the traffic rate at each port, and obtaining a local delay bound of the traffic;

[0017] S23 , performing curve modeling processing on the parameter information set, the traffic rate at each port, and the traffic local delay limit to obtain an aggregated arrival curve and a residual service curve of the AVB traffic.

[0018] The performing local delay calculation processing on the parameter information set, decomposing the end-to-end delay bound of the traffic based on the traffic rate at each port, and obtaining the local delay bound of the traffic, includes:

[0019] S221, using a port traffic calculation model, calculating and processing the parameter information set to obtain a traffic rate at each port;

[0020] S222 , using a local delay bound calculation model, calculate and process the flow rate at each port to obtain a local delay bound for the flow at each port.

[0021] The expression of the port flow calculation model is:

[0022]

[0023] in, The priority of the signal flowing through port h is M i AVB traffic set, l f is the frame length of flow f, P f is the minimum frame interval of flow f, The priority is M i The AVB traffic rate of port h, is the flow rate of TT flowing through port h, F TT (h) is the aggregate TT flow rate flowing through port h.

[0024] The local delay bound calculation model includes:

[0025] A local delay bound equation is constructed; the expression of the local delay bound equation is:

[0026]

[0027] Where Ω is the local delay limit factor to be solved, represents the end-to-end delay bound of flow f, ρ ∑ is the sum of the traffic rates of all ports on the end-to-end path of flow f, and According to the port h, we can get the sum: The priority of the problem to be solved is M i The end-to-end delay bound of AVB traffic f at port h is r f represents the set of all ports that flow f passes through on its end-to-end path;

[0028] Solving the local delay bound equation, we obtain Ω and

[0029] The end-to-end delay limit is calculated to obtain the minimum delay, and the priority is M i The local delay bound of all AVB flows at port h.

[0030] The expression for the minimum calculation process of the delay is:

[0031]

[0032] in, Indicates the priority is M i The local delay bound of all AVB flows at port h.

[0033] The curve modeling process is performed on the parameter information set, the traffic rate at each port, and the traffic local delay limit to obtain the aggregated arrival curve and the residual service curve of the AVB traffic, including:

[0034] S231, construct a calculation expression for the arrival curve of the flow f at the port, the calculation expression is:

[0035]

[0036] in, is the port h before the AVB traffic f arrives at port h within time t - The first data volume function at the starting port h0, the first data volume of the AVB traffic f arriving at the port h0 within time t, that is, the arrival curve It can be expressed as: according to Deducing item by item, represents the burstiness of traffic f at port h0, Represents the flow rate of flow f at port h0. Represents the previous port h of port h on the flow path f - The arrival curve expression at represents the flow rate f at port h - The local delay bound of the flow at ; represents the arrival curve expression at port h on the flow path f;

[0037] S232, constructing a link shaping curve and a CBS shaping curve;

[0038] S233: constructing an aggregated arrival curve and a residual service curve of the AVB traffic based on the link shaping curve and the CBS shaping curve;

[0039] The link shaping curve includes the independent variable t and the dependent variable The dependent variable of the link shaping curve is The calculation expression is:

[0040]

[0041] Where C represents the link rate, Represents port h - To port h, the priority is M i The maximum frame length of the traffic, Indicates the AVB traffic f at port h during time t - The dependent variable value of the link shaping curve to port h; the horizontal and vertical coordinates of the points on the link shaping curve are the independent variable t and the dependent variable

[0042] The CBS shaping curve includes the independent variable t and the dependent variable The dependent variable of the link shaping curve is The calculation expression is:

[0043]

[0044] in, For port h - The priority is M i Bandwidth allocation parameters for AVB traffic, and Port h - The priority is M i The upper and lower limits of the credit value of AVB traffic, Indicates port h - The minimum service variable provided for TT traffic; the horizontal and vertical coordinates of the points on the CBS shaping curve are the independent variable t and the dependent variable, respectively.

[0045] The aggregate arrival curve of the AVB traffic includes the independent variable t and the aggregate arrival curve dependent variable The aggregate arrival curve dependent variable The calculation expression is:

[0046]

[0047] in, It represents the second data volume of AVB traffic f arriving at port h during time t, represents the set of all output ports in the network, Indicates the slave port h - Leaving port h with priority M i AVB traffic set, x∧y represents min{x,y}; the horizontal coordinate and vertical coordinate of the point on the aggregate arrival curve of the AVB traffic are the independent variable t and the aggregate arrival curve dependent variable, respectively

[0048] The residual service curve of the AVB traffic, including the independent variable t and the residual service curve dependent variable The priority at port h is M i The residual service curve of AVB traffic is the dependent variable The expression is

[0049]

[0050] in, represents the upper envelope of all possible arrival curves within a super period of the TT traffic schedule, The priority at port h is M i Bandwidth allocation parameters for AVB traffic, Indicates that the priority at port h is M i The upper limit of the credit value of AVB traffic; [] ↑ + represents a non-negative and monotonically non-decreasing function; the horizontal and vertical coordinates of the points on the residual service curve of the AVB traffic are the independent variable t and the dependent variable of the residual service curve, respectively.

[0051] The solving process of the aggregated arrival curve and the remaining service curve to obtain the bandwidth allocation parameters of the TSN network includes:

[0052] S31, initialize the inflection point sequence number N to 1;

[0053] S32, from all the inflection points of the aggregate arrival curve, find the Nth inflection point on the right side of the inflection point whose slope is less than the link rate, which is the key cusp;

[0054] S33, based on the aggregated arrival curve and the remaining service curve, determining the horizontal coordinate of the delay key point on the remaining service curve The value range is

[0055]

[0056] in, represents the abscissa of the point on the residual service curve that has the same ordinate as the critical cusp, Represents the horizontal axis value The abscissa of the end point of the horizontal line segment on the residual service curve within the range, Represents the horizontal coordinate of the point on the residual service curve that has the same vertical coordinate as the last inflection point of the aggregate arrival curve, P h represents the super period of TT traffic at port h.

[0057] S34, constructing a model for solving the minimum bandwidth allocation parameters of the port;

[0058] S35, solving the port minimum bandwidth allocation parameter solution model, starting from the highest priority, and calculating the minimum bandwidth allocation parameter of the AVB traffic of each priority level step by level;

[0059] S36, the priority M of each port obtained by the solution i The minimum bandwidth allocation parameter of the AVB traffic is used to determine whether the corresponding point in the aggregate arrival curve meets the slope judgment condition. If so, the priority M of the port is determined. i The minimum bandwidth allocation parameter for AVB traffic is the priority M of the TSN network port. i AVB traffic bandwidth allocation parameters, execute S31 to S35, calculate the lower priority M i+1 If the minimum bandwidth allocation parameter of the AVB traffic is not satisfied, the inflection point number N is increased by 1 for the corresponding priority of the port, and S32 to S35 are executed;

[0060] S37, after calculating the minimum bandwidth allocation parameter for AVB traffic of all priorities on each port, confirming the calculated minimum bandwidth allocation parameter for AVB traffic of all priorities on each port as the AVB traffic bandwidth allocation parameter of the TSN network;

[0061] The aggregate arrival curve is composed of several line segments with gradually increasing slopes, and the connection points of adjacent line segments are inflection points; the aggregate arrival curve is composed of several oblique lines and horizontal line segments with constant slopes, and the connection points between the oblique lines and horizontal line segments are inflection points.

[0062] The port minimum bandwidth allocation parameter solution model includes:

[0063] An initial solution model is constructed; the expression of the initial solution model is:

[0064]

[0065] in, The priority is M i The set of delay key points on the remaining service curve, The priority at port h is M i The residual service curve of AVB traffic is The corresponding dependent variable is The priority is higher than M i The AVB traffic priority is M i The delay caused by AVB traffic, priority M j Higher than priority M i ,j <i, is higher than priority M i Bandwidth allocation parameters for high AVB traffic, Indicates that the priority of port h is M j The maximum frame length of the traffic, Indicates that the priority of port h is less than M i The maximum frame length of the traffic.

[0066] Solve the initial solution model to obtain the initial solution

[0067] The initial solution Optimize the process and get

[0068] The expression of the optimization process is:

[0069]

[0070] Among them, OperCycleTime represents the duration of a supercycle of the TT schedule table, and GateOpenTime represents the duration not occupied by TT and guard interval in a supercycle. The port priority is M i The sum of the AVB traffic rates. * indicates a multiplication sign.

[0071] The beneficial effects of the present invention are:

[0072] To address the long computation time and low bandwidth utilization associated with traditional TSN network parameter allocation methods, this paper proposes a TAS-CBS-based bandwidth parameter pre-allocation method for AVB traffic in TSN. This method allocates the minimum feasible bandwidth resources at the node level while ensuring traffic latency bounds. It is highly efficient and scalable, and can be applied to network topologies of various sizes.

[0073] This invention is targeted at TSN networks using TAS-CBS shapers. Based on network calculus theory, this invention analyzes the impact of TT traffic scheduling on AVB traffic and develops an analytical method for pre-allocating bandwidth parameters that can meet AVB traffic delay bounds. This method offers the following advantages: compared to traditional feedback-based bandwidth allocation methods, this method does not require incremental adjustments based on worst-case delay analysis results, resulting in extremely low runtime complexity. This method decomposes the end-to-end delay bound of traffic to derive local delay bounds for nodes, and uses analytical methods at the node level to allocate the minimum bandwidth parameters that meet local delay bounds. Compared to traditional methods, this method can conserve bandwidth resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 Flow chart for the implementation of the method of the present invention;

[0075] Figure 2 This is the network topology diagram in the example. DETAILED DESCRIPTION

[0076] In order to better understand the content of the present invention, two embodiments are given here.

[0077] Figure 1 Flow chart for the implementation of the method of the present invention; Figure 2 This is the network topology diagram in the example.

[0078] Example 1:

[0079] Aiming at the defects of long calculation time and low bandwidth utilization in traditional methods for TSN network parameter allocation, the present invention discloses a TSN network bandwidth parameter pre-allocation method based on network calculation.

[0080] In a first aspect, an embodiment of the present application discloses a TSN network bandwidth parameter pre-allocation method based on network calculation, comprising:

[0081] S1, obtaining a set of parameter information of a TSN network; the TSN network is a time-sensitive network;

[0082] S2, performing traffic modeling processing on the parameter information set to obtain an aggregated arrival curve and a residual service curve of the AVB traffic;

[0083] S3: Solve the aggregated arrival curve and the remaining service curve to obtain bandwidth allocation parameters of the TSN network.

[0084] The TSN network includes a plurality of interconnected nodes; each node includes a plurality of ports;

[0085] The parameter information set includes node information of the TSN network, traffic information flowing through each node, link rate C, TT traffic information set and AVB traffic information set in the TSN network;

[0086] The AVB (audio video bridging, AVB) traffic refers to audio and video bridging traffic; TT (time-triggered, TT) traffic refers to time-triggered traffic;

[0087] The TT traffic information set includes traffic sequence number, frame length, end-to-end delay limit The transmission path, minimum frame interval, and scheduling schedule of TT traffic at each port; the scheduling schedule includes the start and end time of the transmission window of the TT traffic at the port, and the traffic sequence number;

[0088] The AVB traffic information set includes traffic sequence number, frame length, and end-to-end delay limit Propagation path, minimum frame interval;

[0089] The flow information flowing through each node, including flow sequence number and flow value;

[0090] Traffic types in the TSN network include TT traffic and AVB traffic;

[0091] The performing traffic modeling on the parameter information set to obtain an aggregated arrival curve and a residual service curve of the AVB traffic includes:

[0092] S21, performing statistical processing on the parameter information set to obtain the TT traffic sequence number and AVB traffic sequence number flowing through each node, and the super period of the TT traffic; the super period is the least common multiple of the minimum frame interval of the TT traffic;

[0093] S22, performing local delay calculation processing on the parameter information set to obtain the flow rate and flow local delay limit at each port;

[0094] S23, performing curve modeling processing on the parameter information set, the traffic rate at each port, and the traffic local delay limit to obtain an aggregated arrival curve and a residual service curve of the AVB traffic;

[0095] The performing local delay calculation processing on the parameter information set to obtain the flow rate and flow local delay limit at each port includes:

[0096] S221, using a port traffic calculation model, calculating and processing the parameter information set to obtain a traffic rate at each port;

[0097] S222 , using a local delay bound calculation model, calculate and process the flow rate at each port to obtain a local delay bound for the flow at each port.

[0098] The expression of the port flow calculation model is:

[0099]

[0100]

[0101] in, The priority of the signal flowing through port h is M i AVB traffic set, l f is the frame length of flow f, P f is the minimum frame interval of flow f, The priority is M i The AVB traffic rate of port h, is the flow rate of TT flowing through port h, FTT (h) is the aggregate TT flow rate flowing through port h.

[0102] The local delay bound calculation model includes:

[0103] A local delay bound equation is constructed; the expression of the local delay bound equation is:

[0104]

[0105] Where Ω is the local delay limit factor to be solved, represents the end-to-end delay bound of flow f, ρ ∑ is the sum of the traffic rates of all ports on the end-to-end path of flow f, and According to the port h, we can get the sum: The priority of the problem to be solved is M i The end-to-end delay bound of AVB traffic f at port h is r f represents the set of all ports that flow f passes through on its end-to-end path;

[0106] Solving the local delay bound equation, we obtain Ω and

[0107] The end-to-end delay limit is calculated to obtain the minimum delay, and the priority is M i The local delay bound of all AVB flows at port h;

[0108] The expression for the minimum calculation process of the delay is:

[0109]

[0110] in, Indicates the priority is M i The local delay bound of all AVB flows at port h.

[0111] The curve modeling process is performed on the parameter information set, the traffic rate at each port, and the traffic local delay limit to obtain the aggregated arrival curve and the residual service curve of the AVB traffic, including:

[0112] S231, construct a calculation expression for the arrival curve of the flow f at the port, the calculation expression is:

[0113]

[0114] in, is the port h before the AVB traffic f arrives at port h within time t -The first data volume function at the starting port h0, the first data volume of the AVB traffic f arriving at the port h0 within time t, that is, the arrival curve It can be expressed as: according to Deducing item by item, represents the burstiness of traffic f at port h0 (the burstiness at the starting port is the traffic frame length, in bits). Represents the flow rate of flow f at port h0. Represents the previous port h of port h on the flow path f - The arrival curve expression at represents the flow rate f at port h - The local delay bound of the flow at ; represents the arrival curve expression at port h on the flow path f;

[0115] S232, constructing a link shaping curve and a CBS shaping curve;

[0116] S233: constructing an aggregated arrival curve and a residual service curve of the AVB traffic based on the link shaping curve and the CBS shaping curve;

[0117] The link shaping curve includes the independent variable t and the dependent variable The dependent variable of the link shaping curve is The calculation expression is:

[0118]

[0119] Where C represents the link rate, Represents port h - To port h, the priority is M i The maximum frame length of the traffic is obtained based on the frame length information statistics of the parameter information set, Indicates the AVB traffic f at port h during time t - The dependent variable value of the link shaping curve to port h; the horizontal and vertical coordinates of the points on the link shaping curve are the independent variable t and the dependent variable

[0120] The CBS shaping curve includes the independent variable t and the dependent variable The dependent variable of the link shaping curve is The calculation expression is:

[0121]

[0122] in, For port h -The priority is M i Bandwidth allocation parameters for AVB traffic, and Port h - The priority is M i The upper and lower limits of the credit value of AVB traffic, Indicates port h - The minimum service variable provided for TT traffic; the horizontal and vertical coordinates of the points on the CBS shaping curve are the independent variable t and the dependent variable, respectively. Indicates the priority is M i AVB traffic is sent from port h - The dependent variable value of the CBS shaping curve on the link to port h;

[0123] The aggregate arrival curve of the AVB traffic includes the independent variable t and the aggregate arrival curve dependent variable The aggregate arrival curve dependent variable The calculation expression is:

[0124]

[0125] in, It represents the second data volume of AVB traffic f arriving at port h during time t, represents the set of all output ports in the network, Indicates the slave port h - Leaving port h with priority M i AVB traffic set, x∧y represents min{x,y}; the horizontal coordinate and vertical coordinate of the point on the aggregate arrival curve of the AVB traffic are the independent variable t and the aggregate arrival curve dependent variable, respectively

[0126] The residual service curve of the AVB traffic, including the independent variable t and the residual service curve dependent variable The priority at port h is M i The residual service curve of AVB traffic is the dependent variable The expression is

[0127]

[0128] in, represents the upper envelope of all possible arrival curves within a super period of the TT traffic schedule, The priority at port h is M i Bandwidth allocation parameters for AVB traffic, Indicates that the priority at port h is M i The upper limit of the credit value of AVB traffic; [] ↑+ represents a non-negative and monotonically non-decreasing function; the horizontal and vertical coordinates of the points on the residual service curve of the AVB traffic are the independent variable t and the dependent variable of the residual service curve, respectively. in, It represents the upper envelope value of all possible arrival curves formed by using different transmission windows as the benchmark within a super period of the TT traffic scheduling schedule, and its expression is: in Represents the number of TT transmission windows at port h. h Represents the super period of the TT traffic transmission window at port h. in is the length of the mth transmission window of TT traffic, It is the length of the guard interval before the mth transmission window of TT traffic. Represents the offset between the start times of the nth and mth transmission windows of TT after adding the guard interval.

[0129] The expression is:

[0130]

[0131] in,[] + represents a non-negative function, Represents the number of TT transmission windows at port h. The delay of the minimum service curve for TT traffic is expressed as

[0132]

[0133] in, The rate of the minimum service curve for TT traffic is expressed as At port h, the nth transmission window of TT traffic is The transmission window starts at time Represents the offset between the start time of the nth and mth transmission windows of TT. h Represents the super period of the TT traffic transmission window at port h.

[0134] The solving process of the aggregated arrival curve and the remaining service curve to obtain the bandwidth allocation parameters of the TSN network includes:

[0135] S31, initialize the inflection point sequence number N to 1;

[0136] S32, from all the inflection points of the aggregate arrival curve, find the Nth inflection point on the right side of the inflection point whose slope is less than the link rate, which is the key cusp;

[0137] S33, based on the aggregated arrival curve and the remaining service curve, determining the horizontal coordinate of the delay key point on the remaining service curve The value range is

[0138]

[0139] in, represents the abscissa of the point on the residual service curve that has the same ordinate as the critical cusp, Represents the horizontal axis value The abscissa of the end point of the horizontal line segment on the residual service curve within the range, Represents the horizontal coordinate of the point on the residual service curve that has the same vertical coordinate as the last inflection point of the aggregate arrival curve, P h represents the super period of TT traffic at port h.

[0140] S34, constructing a model for solving the minimum bandwidth allocation parameters of the port;

[0141] S35, starting from the highest priority, calculate the minimum bandwidth allocation parameter of the AVB traffic of each priority level step by level;

[0142] S36, the priority M of each port obtained by the solution i The minimum bandwidth allocation parameter of the AVB traffic is used to determine whether the corresponding point in the aggregate arrival curve meets the slope judgment condition. If so, the priority M of the port is determined. i The minimum bandwidth allocation parameter for AVB traffic is the priority M of the TSN network port. i AVB traffic bandwidth allocation parameters, execute S31 to S35, calculate the lower priority M i+1 If the minimum bandwidth allocation parameter of the AVB traffic is not satisfied, the inflection point number N is increased by 1 for the corresponding priority of the port, and S32 to S35 are executed;

[0143] S37, after calculating the minimum bandwidth allocation parameter for AVB traffic of all priorities on each port, confirming the calculated minimum bandwidth allocation parameter for AVB traffic of all priorities on each port as the AVB traffic bandwidth allocation parameter of the TSN network;

[0144] The convergent arrival curve is composed of a number of line segments with gradually increasing slopes, and the connection points of adjacent line segments are inflection points; the convergent arrival curve is composed of a number of oblique lines and horizontal line segments with constant slopes, and the connection points between oblique lines and horizontal line segments are inflection points;

[0145] The port minimum bandwidth allocation parameter solution model includes:

[0146] An initial solution model is constructed; the expression of the initial solution model is:

[0147]

[0148] in, The priority is M i The set of delay key points on the remaining service curve, The priority at port h is M i The residual service curve of AVB traffic is The corresponding dependent variable is The priority is higher than M i The AVB traffic priority is M i The delay caused by AVB traffic, priority M j Higher than priority M i ,j <i, is higher than priority M i Bandwidth allocation parameters for high AVB traffic, Indicates that the priority of port h is M j The maximum frame length of the traffic, Indicates that the priority of port h is less than M i The maximum frame length of the traffic.

[0149] Solve the initial solution model to obtain the initial solution

[0150] The initial solution Optimize the process and get

[0151] The expression of the optimization process is:

[0152]

[0153] Among them, OperCycleTime represents the duration of a supercycle of the TT schedule table, and GateOpenTime represents the duration not occupied by TT and guard interval in a supercycle. The port priority is M i The sum of the AVB traffic rates. * indicates a multiplication sign.

[0154] The slope judgment condition is that the slope of all line segments on the left side of the aggregate arrival curve of the point is greater than At the same time, the slopes of all line segments on the right are less than

[0155] The initial solution model is solved to obtain an initial solution The vertical coordinates of all points on the aggregate arrival curve are judged to obtain a solution that satisfies the initial solution model.

[0156] Example 2:

[0157] In a first aspect, an embodiment of the present application discloses a TSN network bandwidth parameter pre-allocation method based on network calculation, comprising:

[0158] Step 1: Read traffic information and link parameters

[0159] Step 1-1: Read traffic information

[0160] According to the input txt file, read the information of TT traffic and AVB traffic, including traffic ID number, frame length (byte), end-to-end delay limit (unit: us), propagation path, minimum frame interval (unit: us), and scheduling schedule of TT traffic at each port.

[0161] The scheduling schedule includes the start time and end time of the transmission window of TT traffic at the port, as well as the traffic ID number;

[0162] Step 1-2: Read link parameters

[0163] Read the link rate C (link rate between switches and terminals) of the TSN network based on the input txt file. All links in the network have the same rate (in Mbit / s).

[0164] Steps 1-3: Preprocessing

[0165] Preprocess the read information, including:

[0166] Statistics and storage of traffic information flowing through each node in the network,

[0167] This includes the IDs of the AVB traffic and TT traffic flowing through the node, the scheduling schedule for each port, and the super period. The super period is the least common multiple of all minimum interframe intervals of TT traffic at a port.

[0168] Step 2: Calculate the local delay bound

[0169] Step 2-1: Calculate the traffic rate at each port

[0170] Using a port traffic calculation model, the information is calculated and processed to obtain a traffic rate at each port;

[0171] For any port h, the priority is M i The AVB traffic rate is in, The priority of the signal flowing through port h is M i AVB traffic set, l f is the frame length of flow f, P f is the minimum frame interval of flow f.

[0172] Similarly, the TT flow rate at port h is in, is the set of TT flows flowing through port h;

[0173] Step 2-2: Assign traffic local delay bounds based on traffic rate

[0174] The local delay bound calculation model is used to calculate the traffic rate at each port to obtain the local delay bound of the traffic at each port.

[0175] For priority M i For AVB traffic f, all ports on its propagation path can decompose the end-to-end delay bound of the traffic f according to the calculated traffic rate plus the TT traffic rate, and obtain the local delay bound, which is

[0176]

[0177] Because the priority of the traffic flowing through port h is M i There may be more than one AVB traffic. i The minimum value of the local delay bound obtained by decomposing all the traffic of is the local delay bound of the priority AVB traffic at this port, that is,

[0178]

[0179] Where Ω is the local delay limit factor, represents the end-to-end delay bound of flow f, ρ ∑ is the sum of the traffic rates of all ports on the end-to-end path of flow f, and According to the port h, we can get the sum: The priority is M i The end-to-end delay bound of AVB traffic f at port h is r f express, Indicates the priority is M i The local delay bound of all AVB flows at port h.

[0180] Step 3: Construct the AVB traffic model under the TAS-CBS shaper

[0181] Step 3-1: Construct AVB traffic arrival curve model

[0182] The priority at port h is M i The AVB flow aggregate arrival curve is obtained by summing the burstiness-rate arrival curves of individual flows. It also takes into account the link shaping curve and CBS shaping curve to improve the consistency of the analysis results.

[0183] At the starting port h0, the first data volume of AVB traffic f arriving at port h0 within time t is the arrival curve It can be expressed as: in, represents the burstiness of traffic f at port h0 (the burstiness at the starting port is the traffic frame length, in bits). Represents the flow rate of flow f at port h0.

[0184] The arrival curve of flow rate f at the subsequent port can be derived according to the following formula:

[0185]

[0186] in, Represents the previous port h of port h on the flow path f - The arrival curve expression at represents the flow rate f at port h - The local delay bound at .

[0187] The link shaping curve expression is calculated as follows:

[0188]

[0189] Where C represents the link rate, Represents port h - To port h, the priority is M i The maximum frame length of the traffic is obtained based on the frame length information statistics in step 1-1. Indicates the AVB traffic f at port h during time t - The dependent variable value of the link curve to port h.

[0190] The expression of the CBS shaping curve is constructed as follows:

[0191]

[0192] in, For port h - The priority is M i Bandwidth allocation parameters for AVB traffic, and Port h - The priority is M iThe upper and lower limits of the credit value of AVB traffic,

[0193] Indicates port h - The minimum service curve provided for TT traffic; in order to ensure that the aggregate arrival curve is a concave function, Zoomed into R[tT] + form.

[0194]

[0195] in,[] + represents a non-negative function, Represents the number of TT transmission windows at port h. The delay of the minimum service curve for TT traffic is expressed as

[0196]

[0197] in, The rate of the minimum service curve for TT traffic is expressed as At port h, the nth transmission window of TT traffic is The transmission window starts at time Represents the offset between the start time of the nth and mth transmission windows of TT. h Represents the super period of the TT traffic transmission window at port h.

[0198] In summary, the priority at port h is M i The expression of the aggregate arrival curve of AVB traffic is:

[0199]

[0200] in, It represents the second data volume of AVB traffic f arriving at port h during time t, represents the set of all output ports in the network, Indicates the slave port h - Leaving port h with priority M i AVB traffic set, x∧y represents min{x,y}.

[0201] Step 3-2: Construct AVB traffic residual service curve model

[0202] In the hybrid TSN / TAS-CBS architecture, based on the rate-delay classic service curve model of network calculation, and considering the impact of TT traffic and high-priority AVB traffic, it can be

[0203] The priority of port h is M.i The expression of the residual service curve of AVB traffic is

[0204]

[0205] in, represents the dependent variable value of the residual service curve, Represents the upper envelope value of all possible arrival curves formed using different transmission windows as the benchmark within a super period of the TT schedule, and is expressed as in Represents the number of TT transmission windows at port h. h Represents the super period of the TT traffic transmission window at port h. in is the length of the mth transmission window of TT traffic, It is the length of the guard interval before the mth transmission window of TT traffic. Represents the offset between the start time of the nth and mth transmission windows of TT after adding the guard interval. [] ↑ + represents a non-negative and monotonically non-decreasing function, Indicates that the priority at port h is M i The upper and lower limits of the credit value of AVB traffic, The priority at port h is M i The remaining service curve is the bandwidth resource available for the current AVB traffic after removing the bandwidth resource occupied by the TT traffic and higher priority AVB traffic from the port bandwidth resources.

[0206] Step 4: Calculate the minimum bandwidth allocation parameters for each port to meet the delay bound

[0207] Step 4-1: Determine the delay key points

[0208] The AVB flow aggregation arrival curve obtained in step 3-1 is a concave function, consisting of a series of line segments with gradually increasing slopes.

[0209] The residual service curve of the AVB stream obtained in step 3-2 consists of two parts. One part is a line segment with a constant slope. The other part is a horizontal line segment, whose starting point and end point on the time axis are fixed. Even if things change, these starting and ending points remain the same.

[0210] A turning point of the curve is regarded as a key cusp. The slope of the line segment to the left of the key cusp is greater than The slope of the line segments to the right of the key cusp is less than

[0211] By analyzing the AVB flow arrival curve and the residual service curve model, the key point position that can achieve the maximum delay can be determined, that is, the point on the residual service curve with the same vertical coordinate as the key point ( starting point) and within a certain range ( The endpoint of the horizontal line segment on the residual service curve (plus a super-period endpoint).

[0212] The starting point of this range is the point on the residual service curve with the same vertical coordinate as the critical cusp, and the end point is the point on the residual service curve with the same vertical coordinate as the last inflection point of the arrival curve plus one super period of the TT schedule.

[0213] Based on the AVB stream arrival curve (7) and the remaining service curve model (8), the horizontal coordinate of the delay key point with the maximum delay is determined The value range expression is as follows:

[0214]

[0215] in, Represents the horizontal coordinate of the point on the residual service curve that has the same vertical coordinate as the critical cusp (which changes with each iteration), represent The horizontal coordinate of the end point of the horizontal line segment within the range, Represents the horizontal coordinate of the point on the remaining service curve that has the same vertical coordinate as the last inflection point of the AVB flow arrival curve, P h Represents the super period of the TT schedule at port h.

[0216] Step 4-2: Calculate the minimum bandwidth allocation parameters for each port to meet the delay bound

[0217] AVB traffic bandwidth parameters on each port During pre-allocation, it is impossible to determine in advance which inflection point on the AVB stream aggregation arrival curve is the critical cusp.

[0218] Starting from the first inflection point where the slope of the right side of the AVB stream aggregation arrival curve is less than the link rate, assuming this point is the critical cusp, use expression (9) to determine the value range of the delay critical point, and solve the traffic bandwidth parameter at the delay critical point where the delay is equal to the local delay limit (Calculated using (10)), because the local delay bound requirement needs to be strictly met, the maximum value of the traffic bandwidth parameter obtained from a series of delay key points must be taken.

[0219] In summary, M that satisfies the local delay bound at port h i The minimum bandwidth allocation parameter expression of priority (solved quantity, known quantity) is:

[0220]

[0221] Among them, is the set of delay key points on the service curve.

[0222]

[0223] Among them, is the delay caused by the AVB traffic with a priority higher than M i to the AVB traffic with a priority of M i , and the priority M j is higher than the priority M i (j < i). Among them, is the bandwidth allocation parameter of the AVB traffic with a priority higher than M i .

[0224] To ensure the stability of network communication, it is necessary to ensure the lower limit of, that is,

[0225]

[0226] Among them, OperCycleTime represents the duration of one super cycle of the TT scheduling table, and GateOpenTime represents the duration not occupied by TT and the protection interval within one super cycle. is the sum of the AVB traffic rates with a port priority of M i , and * represents the multiplication sign.

[0227] Use the solved to verify whether this break point is a critical cusp, that is, the slope of the line segment on the left side of the break point is greater than and the slope of the line segment on the right side of the break point is less than

[0228] If it is satisfied, complete the pre-allocation of the AVB bandwidth parameters with a port priority of M i , otherwise, assume the next break point on the right as the critical cusp and repeat the above steps until the critical cusp is found.

[0229] According to formula (10), it is found that the bandwidth allocation parameters of high-priority AVB flows are not affected by low-priority AVB flows, and moreover, the smaller the bandwidth allocation parameters of high-priority AVB flows, the smaller the bandwidth allocation parameters of low-priority flows that satisfy the delay bound. Therefore, starting from the highest priority, complete the pre-allocation of the bandwidth parameters that satisfy the delay bound of AVB flows step by step from high to low according to the above steps.

[0230] In the second aspect of the embodiments of the present application, a method for pre-allocating TSN network bandwidth parameters based on network calculus is disclosed, including:

[0231] The network topology used in this example has three end systems and one switch. Figure 2 As shown:

[0232] A total of TT traffic and two priority types of AVB traffic are set. The traffic information is shown in Table 1. The link rate of all output ports is 100 Mbps.

[0233] Table 1 Flow information

[0234] Flow Size (byte) Period (us) Deadline (us) Class Path f1 200 1000 1000 TT [ES1,SW1]; [SW1,ES3] f2 400 2000 2000 TT [ES1,SW1]; [SW1,ES3] f3 200 1000 1000 TT [ES2,SW1]; [SW1,ES3] f4 400 2000 2000 TT [ES2,SW1]; [SW1,ES3] f5 200 1000 1000 A [ES1,SW1]; [SW1,ES3] f6 500 2000 2000 A [ES1,SW1]; [SW1,ES3] f7 300 2000 2000 B [ES1,SW1]; [SW1,ES3] f8 500 4000 4000 B [ES1,SW1]; [SW1,ES3] f9 200 1000 1000 A [ES2,SW1]; [SW1,ES3] f10 300 2000 2000 A [ES2,SW1]; [SW1,ES3] f11 500 2000 2000 B [ES2,SW1]; [SW1,ES3] f12 200 4000 4000 B [ES2,SW1]; [SW1,ES3]

[0235] Step 1: Read traffic information and link parameters

[0236] Step 1-1: Read traffic information

[0237] Read the flow information in Table 1.

[0238] Step 1-2: Read link parameters

[0239] The link rate is read from the input txt file and the link rate of the port is set to C = 100 Mbps.

[0240] Steps 1-3: Preprocessing

[0241] The read information is pre-processed, and the traffic information flowing through each node in the network is counted and stored.

[0242] Step 2: Calculate the local delay bound

[0243] Step 2-1: Calculate the traffic rate at each port

[0244] The rates of TT traffic and AVB traffic of each priority at each output port are calculated, as shown in Table 2.

[0245] Table 2 Flow rate calculation results

[0246] Port TT (Mbps) Class A (Mbps) Class B (Mbps) h=0 1.6 3.6 2.2 h=1 3.2 6.4 4.6 h=2 1.6 2.8 2.4

[0247] Step 2-2: Assign traffic local delay bounds based on traffic rate

[0248] For AVB traffic with priorities A and B, the end-to-end delay bound of the corresponding AVB traffic is decomposed into a local delay bound based on the calculated traffic rate plus the TT traffic rate. The minimum value of the local delay bounds obtained by decomposing all traffic of a certain priority is the local delay bound of the AVB traffic of that priority on this port. The results are shown in Table 3:

[0249] Table 3 Local deadline values ​​for each node

[0250] Class A (us) Class B (us) h = 0 ([ES1,SW1]) 423.33 828.43 h = 1 ([SW1,ES3]) 576.67 1161.9 h = 2 ([ES2,SW1]) 409.42 838.10

[0251] Step 3: Construct the AVB traffic model under the TAS-CBS shaper

[0252] Step 3-1: Construct AVB traffic arrival curve model

[0253] Combining formulas (4), (5), (6) and (7), the AVB traffic arrival curves of priority A and B of each output port can be obtained. The coordinates of the inflection points on each arrival curve are needed in subsequent calculations, so the arrival curve is expressed in the form of inflection points and slopes after the inflection points (i.e. ), the results are shown in Table 4:

[0254] Table 4 Coordinates and slopes of inflection points on each arrival curve

[0255]

[0256] Step 3-2: Construct AVB traffic residual service curve model

[0257] Based on formula (8), the AVB traffic residual service curve of priority A and B of each output port can be obtained. The starting point of the horizontal line segment of the residual service curve is needed in subsequent calculations, so the residual service curve is expressed as the horizontal coordinates of the starting and ending points of the horizontal line segment within a super period (i.e. ), the results are shown in Table 5:

[0258] Table 5 Starting points of horizontal segments on each residual service curve

[0259]

[0260] Step 4: Calculate the minimum bandwidth allocation parameters for each port to meet the delay bound

[0261] Step 4-1: Determine the delay key points

[0262] Based on formula (9) and the definition of the critical cusp, the coordinates of the critical cusp on the arrival curve of each output port's priority A and B and the horizontal coordinates of the delay critical point on the remaining service curve can be determined. The results are shown in Table 6:

[0263] Table 6 Coordinates of key cusp points and horizontal coordinates of delayed key points

[0264]

[0265] Step 4-2: Calculate the minimum bandwidth allocation parameters for each port to meet the delay bound

[0266] At the delay critical points in Table 6, based on the local delay bounds in Table 3, combined with formulas (10) and (11), the bandwidth parameters for each port that meet the local delay bounds are solved. To ensure the strict real-time requirements of traffic, the minimum bandwidth allocation parameter for each port to meet the delay bounds is the maximum bandwidth parameter obtained at each delay critical point. The results of the AVB stream bandwidth parameter pre-allocation at each port are shown in Table 7:

[0267] Table 7 Results of pre-allocation of bandwidth parameters for AVB streams at each node

[0268] Class A (Mbps) Class B (Mbps) h = 0 ([ES1,SW1]) 21.93 9.84 h = 1 ([SW1,ES3]) 39.86 19.49 h = 2 ([ES2,SW1]) 14.63 8.28

[0269] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A TSN network bandwidth parameter pre-allocation method based on network calculation, characterized in that: include: S1, obtaining a set of parameter information of a TSN network; the TSN network is a time-sensitive network; S2, performing traffic modeling processing on the parameter information set to obtain an aggregated arrival curve and a residual service curve of the AVB traffic; S3, solving the aggregated arrival curve and the remaining service curve to obtain bandwidth allocation parameters of the TSN network, including: S31, initialize the inflection point sequence number N to 1; S32, from all the inflection points of the aggregate arrival curve, find the Nth inflection point on the right side of the inflection point whose slope is less than the link rate, which is the key cusp; S33, based on the aggregated arrival curve and the remaining service curve, determining the horizontal coordinate of the delay key point on the remaining service curve The value range is in, represents the abscissa of the point on the residual service curve that has the same ordinate as the critical cusp, Represents the horizontal axis value The abscissa of the end point of the horizontal line segment on the residual service curve within the range, Represents the horizontal coordinate of the point on the residual service curve that has the same vertical coordinate as the last inflection point of the aggregate arrival curve, P h represents the super period of TT traffic at port h; S34, constructing a model for solving the minimum bandwidth allocation parameters of the port; S35, solving the port minimum bandwidth allocation parameter solution model, starting from the highest priority, and calculating the minimum bandwidth allocation parameter of the AVB traffic of each priority level step by level; S36, the priority M of each port obtained by the solution i The minimum bandwidth allocation parameter of the AVB traffic is used to determine whether the corresponding point in the aggregate arrival curve meets the slope judgment condition. If so, the priority M of the port is determined. i The minimum bandwidth allocation parameter for AVB traffic is the priority M of the TSN network port. i AVB traffic bandwidth allocation parameters, execute S31 to S35, calculate the lower priority M i+1 If the minimum bandwidth allocation parameter of the AVB traffic is not satisfied, the inflection point number N is increased by 1 for the corresponding priority of the port, and S32 to S35 are executed; S37, after calculating the minimum bandwidth allocation parameter for AVB traffic of all priorities on each port, confirming the calculated minimum bandwidth allocation parameter for AVB traffic of all priorities on each port as the AVB traffic bandwidth allocation parameter of the TSN network; The convergent arrival curve is composed of a number of line segments with gradually increasing slopes, and the connection points of adjacent line segments are inflection points; the convergent arrival curve is composed of a number of oblique lines and horizontal line segments with constant slopes, and the connection points between oblique lines and horizontal line segments are inflection points; The port minimum bandwidth allocation parameter solution model includes: An initial solution model is constructed; the expression of the initial solution model is: in, The priority is M i The set of delay key points on the remaining service curve, The priority at port h is M i The residual service curve of AVB traffic is The corresponding dependent variable is The priority is higher than M i The AVB traffic priority is M i The delay caused by AVB traffic, priority M j Above priority is higher than priority M i Bandwidth allocation parameters for high AVB traffic, Indicates that the priority of port h is M j The maximum frame length of the traffic, Indicates that the priority of port h is less than M i The maximum frame length of the traffic; Solve the initial solution model to obtain the initial solution The initial solution Optimize the process and get The expression of the optimization process is: Among them, OperCycleTime represents the duration of a supercycle of the TT schedule table, and GateOpenTime represents the duration not occupied by TT and guard interval in a supercycle. The port priority is M i The sum of the AVB traffic rates. * indicates a multiplication sign.

2. The TSN network bandwidth parameter pre-allocation method based on network calculation according to claim 1, characterized in that: The TSN network includes a plurality of interconnected nodes; each node includes a plurality of ports; The parameter information set includes node information of the TSN network, traffic information flowing through each node, link rate C, TT traffic information set and AVB traffic information set in the TSN network; The TT traffic information set includes traffic sequence number, frame length, end-to-end delay limit Propagation path, minimum frame interval, and scheduling schedule for TT traffic at each port; The scheduling schedule includes the start time and end time of the transmission window of the TT traffic at the port, and the traffic sequence number; The AVB traffic information set includes traffic sequence number, frame length, and end-to-end delay limit Propagation path, minimum frame interval; The traffic types in the TSN network include TT traffic and AVB traffic.

3. The TSN network bandwidth parameter pre-allocation method based on network calculation according to claim 2, characterized in that: The performing traffic modeling on the parameter information set to obtain an aggregated arrival curve and a residual service curve of the AVB traffic includes: S21, performing statistical processing on the parameter information set to obtain the TT traffic sequence number and AVB traffic sequence number flowing through each node, and the super period of the TT traffic; the super period is the least common multiple of the minimum frame interval of the TT traffic; S22, performing local delay calculation processing on the parameter information set, decomposing the end-to-end delay bound of the traffic based on the traffic rate at each port, and obtaining a local delay bound of the traffic; S23 , performing curve modeling processing on the parameter information set, the traffic rate at each port, and the traffic local delay limit to obtain an aggregated arrival curve and a residual service curve of the AVB traffic.

4. The TSN network bandwidth parameter pre-allocation method based on network calculation according to claim 3 is characterized in that: The performing local delay calculation processing on the parameter information set, decomposing the end-to-end delay bound of the traffic based on the traffic rate at each port, and obtaining the local delay bound of the traffic, includes: S221, using a port traffic calculation model, calculating and processing the parameter information set to obtain a traffic rate at each port; S222 , using a local delay bound calculation model, calculate and process the flow rate at each port to obtain a local delay bound for the flow at each port.

5. The TSN network bandwidth parameter pre-allocation method based on network calculation according to claim 4 is characterized in that: The expression of the port flow calculation model is: in, The priority of the signal flowing through port h is M i AVB traffic set, l f is the frame length of flow f, P f is the minimum frame interval of flow f, The priority is M i The AVB traffic rate of port h, is the flow rate of TT flowing through port h, F TT (h) is the aggregate TT flow rate flowing through port h.

6. The TSN network bandwidth parameter pre-allocation method based on network calculation according to claim 5, characterized in that: The local delay bound calculation model includes: A local delay bound equation is constructed; the expression of the local delay bound equation is: Where Ω is the local delay limit factor to be solved, represents the end-to-end delay bound of flow f, ρ ∑ is the sum of the traffic rates of all ports on the end-to-end path of flow f, and According to the port h, we can get the sum: The priority of the problem to be solved is M i The end-to-end delay bound of AVB traffic f at port h is r f represents the set of all ports that flow f passes through on its end-to-end path; Solving the local delay bound equation, we obtain Ω and The end-to-end delay limit is calculated to obtain the minimum delay, and the priority is M i The local delay bound of all AVB flows at port h.

7. The TSN network bandwidth parameter pre-allocation method based on network calculation according to claim 6, characterized in that: The expression for the minimum calculation process of the delay is: in, Indicates the priority is M i The local delay bound of all AVB flows at port h.

8. The TSN network bandwidth parameter pre-allocation method based on network calculation according to claim 5, characterized in that: The curve modeling process is performed on the parameter information set, the traffic rate at each port, and the traffic local delay limit to obtain the aggregated arrival curve and the residual service curve of the AVB traffic, including: S231, construct a calculation expression for the arrival curve of the flow f at the port, the calculation expression is: in, is the port h before the AVB traffic f arrives at port h within time t - The first data volume function at the starting port h0, the first data volume of the AVB traffic f arriving at the port h0 within time t, that is, the arrival curve It can be expressed as: according to Deducing item by item, represents the burstiness of traffic f at port h0, represents the flow rate of flow f at port h0; Represents the previous port h of port h on the flow path f - The arrival curve expression at represents the flow rate f at port h - The local delay bound of the flow at ; represents the arrival curve expression at port h on the flow path f; S232, constructing a link shaping curve and a CBS shaping curve; S233: constructing an aggregated arrival curve and a residual service curve of the AVB traffic based on the link shaping curve and the CBS shaping curve; The link shaping curve includes the independent variable t and the dependent variable The dependent variable of the link shaping curve is The calculation expression is: Where C represents the link rate, Represents port h - To port h, the priority is M i The maximum frame length of the traffic, Indicates the AVB traffic f at port h during time t - The dependent variable value of the link shaping curve to port h; the horizontal and vertical coordinates of the points on the link shaping curve are the independent variable t and the dependent variable The CBS shaping curve includes the independent variable t and the dependent variable The dependent variable of the link shaping curve is The calculation expression is: in, For port h - The priority is M i Bandwidth allocation parameters for AVB traffic, and Port h - The priority is M i The upper and lower limits of the credit value of AVB traffic, Indicates port h - The minimum service variable provided for TT traffic; the horizontal and vertical coordinates of the points on the CBS shaping curve are the independent variable t and the dependent variable, respectively. The aggregate arrival curve of the AVB traffic includes the independent variable t and the aggregate arrival curve dependent variable The aggregate arrival curve dependent variable The calculation expression is: in, represents the second data volume of AVB traffic f arriving at port h during time t, where H represents the set of all output ports in the network. Indicates the slave port h - Leaving port h with priority M i AVB traffic set, x∧y represents min{x,y}; the horizontal coordinate and vertical coordinate of the point on the aggregate arrival curve of the AVB traffic are the independent variable t and the aggregate arrival curve dependent variable, respectively The residual service curve of the AVB traffic, including the independent variable t and the residual service curve dependent variable The priority at port h is M i The residual service curve of AVB traffic is the dependent variable The expression is in, represents the upper envelope of all possible arrival curves within a super period of the TT traffic schedule, The priority at port h is M i Bandwidth allocation parameters for AVB traffic, Indicates that the priority at port h is M i The upper limit of the credit value of the AVB traffic; []↑+ represents a non-negative and monotonically non-decreasing function; the horizontal and vertical coordinates of the points on the residual service curve of the AVB traffic are the independent variable t and the dependent variable of the residual service curve, respectively.

Citation Information

Patent Citations

  • Time-sensitive network scheduling method based on bandwidth reservation

    CN116192651A

  • Mixed flow scheduling method for 5G-TSN network

    CN118748822A