A method and system for joint scheduling of routing and time slots for a time-sensitive network

Abstract

This invention belongs to the technical field of Time-Sensitive Networking (TSN). To address the problems of existing TSNs' inability to adapt to changes in network topology or traffic, and the jitter issues in cyclic queuing forwarding, this invention proposes a joint scheduling method and system for TSNs, combining routing and time slots. An improved particle swarm optimization (PSO) algorithm is used to jointly optimize and solve the routing paths and initial transmission time slots of all time-triggered traffic to be scheduled, determining the routing path of each time-triggered traffic and the transmission time slots at each node along the route. The improved PSO algorithm introduces an external elite archive storing feasible solutions that meet the fitness requirements in the current iteration. The feasible solutions in the external elite archive guide the particle velocity update, effectively solving the scheduling problem under large-scale time-triggered traffic.

Description

Technical Field

[0001] This invention belongs to the technical field of time-sensitive networks, and particularly relates to a method and system for joint scheduling of routing and time slots in time-sensitive networks. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the rapid development of low-altitude logistics and the Industrial Internet, time-sensitive applications are placing increasingly higher demands on deterministic end-to-end network communication. Time-Sensitive Networking (TSN), as a solution providing precise traffic scheduling mechanisms on top of Ethernet, has become a key technology for meeting the requirements of low latency, low jitter, and high reliability transmission. Currently, TSN mainly provides two core traffic scheduling mechanisms: Time-Aware Shaper (TAS) and Cyclic Queuing Forwarding (CQF).

[0004] Time-aware shapers control traffic of different priorities by defining period-based, precise transmission windows, enabling deterministic latency and ensuring strict traffic isolation. However, existing time-aware shaper technologies suffer from the following major drawbacks: their configuration is extremely complex. In highly dynamic or large-scale networks, calculating and maintaining a network-wide consistent gated list (GCL) requires extremely high computational overhead and is difficult to adapt to changes in network topology or traffic.

[0005] Circular queuing forwarding uses a fixed-period queue to forward data packets between network hops. Its advantage lies in its simple queue management and ability to achieve low end-to-end jitter with low complexity. However, existing circular queuing forwarding techniques have the following main drawbacks: their periodic forwarding mechanism introduces non-negligible jitter. Typically, each hop can introduce up to two cycles of jitter, limiting its effectiveness in time-critical flows.

[0006] Furthermore, existing heuristic algorithms (such as the traditional particle swarm optimization algorithm PSO and genetic algorithm GA) also have limitations when solving the joint routing and time slot scheduling problem in time-sensitive networks. Since the scheduling problem in time-sensitive networks is a discrete-constrained optimization problem, and the traditional particle swarm optimization algorithm uses continuous encoding and fixed inertial weights, when directly applied to such problems, it often easily generates a large number of infeasible solutions (such as routing loops and time slot conflicts), or converges too early and gets trapped in local optima, making it difficult to find a high-quality scheduling scheme that satisfies all hard real-time constraints within a finite time.

[0007] In summary, the problems that need to be solved are how to address the difficulty in adapting to changes in network topology or traffic caused by the complex configuration of time-aware shapers in existing time-sensitive networks, the jitter problem in cyclic queuing forwarding, and how to find a high-quality scheduling scheme that satisfies all hard real-time constraints within a limited time. Summary of the Invention

[0008] To overcome the shortcomings of the prior art, the present invention provides a routing and time slot joint scheduling method and system for time-sensitive networks, which significantly improves the scheduling success rate, reduces end-to-end latency, and optimizes network load balancing.

[0009] To achieve the above objectives, the present invention adopts the following technical solution:

[0010] In a first aspect, the present invention provides a method for joint routing and time-slot scheduling in time-sensitive networks, comprising:

[0011] By treating switches and terminals as nodes, the time-sensitive network is modeled as a directed graph.

[0012] Obtain the set of time-triggered traffic to be scheduled. With the goal of successful scheduling and load balancing, use an improved particle swarm optimization algorithm to jointly optimize and solve the routing path and initial transmission time slot of all time-triggered traffic to be scheduled, and determine the routing path of each time-triggered traffic and the transmission time slot at each node port along the way.

[0013] The scheduling of time-triggered traffic is completed based on the routing path of each time-triggered traffic and the transmission time slots at each node port along the way;

[0014] The improved particle swarm optimization algorithm is as follows: an external elite archive is introduced, which stores feasible solutions that meet the fitness requirements in the current iteration, and the particle velocity update is guided by the feasible solutions in the external elite archive.

[0015] Secondly, the present invention provides a routing and time-slot joint scheduling system for time-sensitive networks, comprising:

[0016] The directed graph building module is configured to model time-sensitive networks as directed graphs, using switches and terminals as nodes.

[0017] The optimization solution module is configured to: acquire a set of time-triggered traffic to be scheduled; and, with the goal of successful scheduling and load balancing, jointly optimize the routing paths and initial transmission time slots of all time-triggered traffic using an improved particle swarm optimization algorithm to determine the routing path of each time-triggered traffic and the transmission time slots at each node port along the route. The improved particle swarm optimization algorithm involves: introducing an external elite archive, which stores feasible solutions that meet the fitness requirements in the current iteration; and using these feasible solutions to guide particle velocity updates.

[0018] The scheduling module is configured to schedule time-triggered traffic based on the routing path of each time-triggered traffic and the transmission time slots at each node port along the route.

[0019] Thirdly, the present invention provides an electronic device including a memory and a processor, and computer instructions stored in the memory and running on the processor, wherein the computer instructions, when executed by the processor, perform the method described in the first aspect.

[0020] Fourthly, the present invention provides a computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in the first aspect.

[0021] Fifthly, the present invention provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0022] The above one or more technical solutions have the following beneficial effects:

[0023] In this invention, an improved particle swarm optimization algorithm is used to jointly optimize and solve the routing paths and initial transmission time slots of all time-triggered traffic to be scheduled, thereby determining the routing path of each time-triggered traffic and the transmission time slots at each node port along the way. The improved particle swarm optimization algorithm introduces an external elite archive that stores feasible solutions that meet the fitness requirements in the current iteration. The particle velocity update is guided by feasible solutions in the external elite archive, which enables particles to jump out of local optima and move towards the global optimum, effectively solving the scheduling problem under large-scale time-triggered traffic.

[0024] In this invention, the uncertainty caused by queuing delay is eliminated from the physical mechanism by using a dual-circular queue alternation and a single-time-slot single-frame gating mechanism, thereby achieving microsecond-level zero-jitter transmission.

[0025] In this invention, a repair mechanism is introduced. If the updated routing index is invalid, instead of a simple random reset, a route is selected from the shortest path, the second shortest path, and the path with the least current network load according to a strategy. This ensures that the repaired solution still has high quality. If a time slot conflict is detected on a link, a feasible time slot farthest from the existing conflicting time slot is searched for and allocated. This strategy can effectively distribute traffic and improve the diversity of solutions.

[0026] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0027] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0028] Figure 1 This is a schematic diagram of cyclic queuing and time-aware scheduling in Embodiment 1 of the present invention. Detailed Implementation

[0029] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0030] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0031] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0032] Example 1

[0033] This embodiment discloses a method for joint routing and time slot scheduling in time-sensitive networks, including:

[0034] By treating switches and terminals as nodes, the time-sensitive network is modeled as a directed graph.

[0035] Obtain the set of time-triggered traffic to be scheduled. With the goal of successful scheduling and load balancing, use an improved particle swarm optimization algorithm to jointly optimize and solve the routing path and initial transmission time slot of all time-triggered traffic to be scheduled, and determine the routing path of each time-triggered traffic and the transmission time slot at each node port along the way.

[0036] The scheduling of time-triggered traffic is completed based on the routing path of each time-triggered traffic and the transmission time slots at each node port along the way;

[0037] The improved particle swarm optimization algorithm is as follows: an external elite archive is introduced, which stores feasible solutions that meet the fitness requirements in the current iteration, and the particle velocity update is guided by the feasible solutions in the external elite archive.

[0038] This embodiment uses an improved particle swarm optimization algorithm to jointly optimize and solve the routing paths and initial transmission time slots of all time-triggered traffic to be scheduled, determining the routing path of each time-triggered traffic and the transmission time slots at each node port along the way. The improved particle swarm optimization algorithm introduces an external elite archive that stores feasible solutions that meet the fitness requirements in the current iteration. By guiding particle velocity updates through feasible solutions in the external elite archive, particles can escape local optima and move towards the global optimum, effectively solving the scheduling problem under large-scale time-triggered traffic.

[0039] This embodiment first constructs a cyclic queuing and time-aware scheduling mechanism, utilizing a dual-cyclic queue alternating reception and transmission mechanism and a single-timeslot single-frame gating mechanism, combined with strict timeslot propagation rules, to achieve zero-jitter traffic transmission. Second, an improved particle swarm optimization algorithm is proposed, achieving joint optimization of TSN traffic routing paths and initial time slots through constraint-aware particle coding, an elite-file-guided multi-source speed update strategy, and a flexible conflict resolution mechanism. Experimental results show that this embodiment can significantly improve scheduling success rate, reduce end-to-end latency, and optimize network load balancing.

[0040] This embodiment constructs a Time-Sensitive Network (TSN) system model and defines the network topology and link parameters. The TSN system model adopts a circular queuing and time-aware (CQTA) scheduling mechanism, and configures a dual circular queue structure on the switch port to process time-triggered traffic.

[0041] This embodiment is based on a Time-Sensitive Network (TSN) environment, where the TSN topology is modeled as a directed graph. ,in, This represents the set of time-sensitive network nodes, namely switches and terminals. This represents a set of directed links.

[0042] Link parameters: Each directed link in a time-sensitive network It has a fixed bandwidth. (e.g., 1Gbps) and fixed propagation delay .

[0043] Node configuration: Each switch output port in the time-sensitive network is configured with 8 priority queues, labeled as follows. to The priority increases sequentially, among which It has the highest priority.

[0044] Time synchronization: All nodes in the time-sensitive network are precisely synchronized using the IEEE 802.1AS protocol to ensure a consistent time base across the entire network.

[0045] The circular queuing and time-aware (CQTA) scheduling mechanism proposed in this embodiment utilizes two highest-priority circular queues, namely... and It is specifically designed to handle time-triggered (TT) traffic to achieve zero-jitter transmission.

[0046] like Figure 1 As shown, the specific working logic is as follows:

[0047] Time slot allocation: Dividing the scheduling time into... A series of consecutive, equal-length time slots.

[0048] Odd-even alternation control:

[0049] In odd time slots: queues In the receive state (i.e., receive gate Rx-gate is open, transmit gate Tx-gate is closed), it is used to buffer data frames from the previous hop; at this time, the queue When in the sending state (i.e., the send gate Tx-gate is open and the receive gate Rx-gate is closed), it is used to send the data frames buffered in the queue to the next hop.

[0050] Among them, the receive gate (Rx-gate) and transmit gate (Tx-gate) are the core components of the IEEE 802.1Qch circular queuing and forwarding mechanism. They control the enqueueing and dequeueing of data packets respectively, and achieve deterministic forwarding through periodic alternating switching, ensuring low latency and low jitter for time-sensitive streams.

[0051] In even-numbered time slots: state inversion. Queue In receive state (i.e., receive gate Rx-gate is open, send gate Tx-gate is closed), queue It is in the sending state (i.e., the sending gate Tx-gate is open and the receiving gate Rx-gate is closed).

[0052] Gating Mechanism: Each queue is equipped with a receive gate (Rx-gate) and a transmit gate (Tx-gate) to control enqueue and dequeue operations. The gating state is strictly controlled by a Gating Control List (GCL). Through this mechanism, each queue is allowed to receive or transmit one data frame per time slot, achieving a per-flow scheduling effect similar to TAS, thereby eliminating jitter caused by queuing.

[0053] To ensure deterministic transmission, this embodiment imposes strict limitations on the time slot length and propagation rules:

[0054] Time slot length calculation: To ensure that data frames sent by the upstream switch can be completely received by the downstream switch within the same time slot, the time slot length is calculated. slot _{cycle} The following formula must be satisfied:

[0055]

[0056] in, The size of the largest data packet in the time-triggered traffic set. For network bandwidth, For link propagation delay, This refers to the processing latency of the TSN switch.

[0057] Time-slot propagation rule (zero jitter guarantee): Based on cyclic queuing and time-aware scheduling mechanism, time-triggered traffic. The allocation of time slots along the path follows a strict linear relationship. Assume time-triggered traffic. The initial timeslot for injecting the first link at the source node is... Then in the first Transmission slots on hop links Must meet:

[0058]

[0059] This rule ensures that time-triggered traffic seamlessly enters the transmission queue of the next time slot at each hop, thereby reducing end-to-end latency. Depends only on the number of hops and time slot length, i.e. This achieves precise deterministic delay.

[0060] This embodiment achieves strictly zero-jitter transmission through a "single-slot, single-frame constraint" mechanism: in the cyclic queuing and time-aware scheduling mechanism, each priority queue ( or Within each time slot, only one data frame is allowed to be accommodated and processed. Since there is at most one data frame to be sent in the queue at any given time, when the transmission gate of the queue opens in the corresponding time slot, the data frame can immediately obtain the right to use the transmission medium for transmission without waiting for the transmission of other frames.

[0061] This mechanism eliminates the variable queuing delay caused by multiple frames in traditional queuing. The dwell time of a data frame within the switch depends only on a fixed hardware processing latency and does not change with traffic load, thus achieving end-to-end zero-jitter transmission characteristics.

[0062] Based on the aforementioned zero-jitter characteristic, any time-triggered stream end-to-end delay It becomes a deterministic value, determined solely by the number of hops it has traversed. and time slot period Decide:

[0063]

[0064] Step S1: Obtain the set of time-triggered traffic to be scheduled. Each traffic item includes the source node, destination node, data packet size, period, and deadline constraint.

[0065] Step S2: Use the improved particle swarm optimization algorithm (TSN-PSO) to jointly optimize and solve the routing paths and initial transmission time slots of all time-triggered traffic.

[0066] To address the joint optimization problem of routing and time slot scheduling in time-sensitive networks, this embodiment employs an improved particle swarm optimization algorithm (TSN-PSO). This algorithm addresses the issues of traditional PSO's tendency to get trapped in local optima and generate infeasible solutions by introducing constraint-aware coding, elite profile guidance, and a flexible repair mechanism.

[0067] 1. Particle encoding and initialization.

[0068] The algorithm first processes each particle in the population. Encoding is performed using a two-layer vector structure. :

[0069] Routing decision vector : ,in The total number of streams. Each element It is an integer index pointing to the first... Set of feasible paths for a flow A specific path within it.

[0070] Time slot allocation vector : Each element Indicates the first The initial time slot for stream injection into the network refers to the time slot from the sender to the network; that is, transmission occurs within a specific time slot. A time slot has a duration, and transmission is complete as long as it lasts within that duration. To avoid searching for invalid space, during initialization... Limited to dynamic range Inside, among them, The number of hops in the selected path for this flow. The total number of time slots allocated for scheduling time.

[0071] 2. Path preprocessing (constructing a set of feasible paths).

[0072] Before initialization, for each stream Pre-compute the set of feasible paths :

[0073] The first K-Shortest Path algorithm is used to calculate the first K-Shortest Path. There are several alternative paths.

[0074] Filtering based on deadline constraints: If the transmission delay of a certain path... The deadline for this stream has been exceeded. If so, then that path will be removed. This represents the number of hops traversed. The time slot period.

[0075] This step ensures that any path the particle chooses during the search process physically meets the time delay requirements.

[0076] 3. The speed update mechanism guided by elite profiles.

[0077] To prevent premature convergence of the algorithm, this embodiment establishes an external Elite Archive. ), used to store the top 15% of feasible solutions in the current iteration in terms of fitness.

[0078] When updating particle velocities, a "multi-source guide" from the elite archives is introduced to break free from local optima and move particles toward the global optimum. The velocity update formula is as follows:

[0079]

[0080] in: c represents the particle's own historical optimal solution; c represents the population's global optimal solution. To extract from the elite archives A high-quality solution is randomly selected from the options. For acceleration coefficient, For inertial weights, It is a contraction factor; Let be the velocity of the particle in the t-th iteration; This represents the position of the particle during the t-th iteration. It is the optimal position for the entire group.

[0081] The corresponding position update formula is:

[0082]

[0083] in, For the first particle The position of the next iteration; For the first particle Next iteration position; For the first particle per-iteration speed.

[0084] 4. Two-stage flexible constraint repair mechanism.

[0085] Since the updated particle positions may lead to infeasible solutions (such as routing index out-of-bounds or slot conflicts), this embodiment introduces a repair mechanism:

[0086] Phase 1: Route Repair

[0087] If the updated route index If the solution is invalid, instead of a simple random reset, a random selection is made from the following options: (1) the shortest path; (2) the second shortest path; (3) the path with the lowest current network load. This ensures that the repaired solution still has high quality.

[0088] Phase Two: Time Slot Repair

[0089] If a time slot conflict is detected on a link (i.e., the same time slot is allocated to multiple flows), instead of using the greedy "earliest available time slot" strategy, the system searches for and allocates the most feasible time slot that is furthest from the existing conflicting time slot. This strategy can effectively distribute traffic and improve the diversity of solutions.

[0090] 5. Fitness assessment.

[0091] Calculate the particle swarm fitness and update the current particle swarm based on the calculated fitness. The optimization objective is to maximize the fitness function. To achieve this:

[0092]

[0093] This function takes into account three performance metrics: To achieve the proportion of successfully scheduled flows, all conflict and latency constraints must be satisfied; This represents the maximum link load rate in the network, used to optimize load balancing. The average end-to-end delay for all streams.

[0094] 6. Output and Execution.

[0095] When the algorithm reaches its maximum number of iterations When convergence conditions are met, output the global optimal solution. Based on the routing vector and initial timeslot vector in the solution, and combined with the timeslot propagation rules of the cyclic queuing and time-aware scheduling mechanism, the gate control list (GCL) of all switches in the entire network is calculated and the configuration is distributed to complete the scheduling.

[0096] Step S3: Based on the global optimal solution obtained from the optimization, determine the routing path of each traffic and the transmission time slots at each node port along the route, generate the gate control list (GCL) and send it to the TSN switch.

[0097] This embodiment achieves zero-jitter deterministic transmission and simplifies configuration complexity: by employing a cyclic queuing and time-aware scheduling mechanism, utilizing alternating dual cyclic queues and a "single time slot, single frame" gating mechanism, the uncertainty caused by queuing delay is eliminated from the physical mechanism, achieving microsecond-level zero-jitter transmission. Furthermore, compared to the traditional TAS mechanism, this embodiment does not require maintaining a complex global gating list, retaining the advantage of simple CQF queue management and reducing the difficulty of network configuration and maintenance.

[0098] The TSN-PSO algorithm proposed in this embodiment effectively solves the scheduling problem under large-scale traffic by introducing constraint-aware coding and elite profile guidance. Thanks to the multi-source guidance speed update mechanism and flexible conflict repair strategy, this embodiment can distribute traffic more evenly in network links, which effectively alleviates local link congestion and improves the overall network resource utilization.

[0099] Example 2

[0100] The purpose of this embodiment is to provide a routing and time-slot joint scheduling system for time-sensitive networks, including:

[0101] The directed graph building module is configured to model time-sensitive networks as directed graphs, using switches and terminals as nodes.

[0102] The optimization solution module is configured to: acquire a set of time-triggered traffic to be scheduled; and, with the goal of successful scheduling and load balancing, jointly optimize the routing paths and initial transmission time slots of all time-triggered traffic using an improved particle swarm optimization algorithm to determine the routing path of each time-triggered traffic and the transmission time slots at each node port along the route. The improved particle swarm optimization algorithm involves: introducing an external elite archive, which stores feasible solutions that meet the fitness requirements in the current iteration; and using these feasible solutions to guide particle velocity updates.

[0103] The scheduling module is configured to schedule time-triggered traffic based on the routing path of each time-triggered traffic and the transmission time slots at each node port along the route.

[0104] In further embodiments, the following is also provided:

[0105] An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When executed by the processor, the computer instructions perform the method described in Embodiment 1. For brevity, further details are omitted here.

[0106] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0107] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0108] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in Embodiment 1.

[0109] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, a detailed description is not provided here.

[0110] A computer program product includes a computer program that, when executed by a processor, implements the method described in Embodiment 1.

[0111] The present invention also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0112] The computer program code used to implement the methods of the present invention may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the computer or other programmable data processing device, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0113] In the context of this invention, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0114] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0115] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for joint routing and time-slot scheduling in time-sensitive networks, characterized in that, include: By treating switches and terminals as nodes, the time-sensitive network is modeled as a directed graph. Obtain the set of time-triggered traffic to be scheduled. With the goal of successful scheduling and load balancing, use an improved particle swarm optimization algorithm to jointly optimize and solve the routing path and initial transmission time slot of all time-triggered traffic to be scheduled, and determine the routing path of each time-triggered traffic and the transmission time slot at each node port along the way. The two-vector structure consisting of the routing decision vector and the time slot allocation vector of each time-triggered traffic path is used as a particle; wherein, each element in the routing decision vector is a specific path in the corresponding feasible path set; If the updated particle's routing index is invalid, the particle's route is determined based on the shortest path, the second shortest path, or the path with the least current network load; if a time slot conflict is detected on a link, a feasible time slot farthest from the existing conflicting time slot is searched for and allocated. The scheduling of time-triggered traffic is completed based on the routing path of each time-triggered traffic and the transmission time slots at each node port along the way; If time-triggered traffic is injected into the initial timeslot of the first link at the source node, it is... Then in the first Transmission slots on hop links Must meet: ; The improved particle swarm optimization algorithm is as follows: an external elite archive is introduced, which stores feasible solutions that meet the fitness requirements in the current iteration, and the particle velocity update is guided by the feasible solutions in the external elite archive.

2. The routing and time-slot joint scheduling method for time-sensitive networks as described in claim 1, characterized in that, With the goals of successful scheduling and load balancing, an improved particle swarm optimization algorithm is used to jointly optimize the routing paths and initial transmission time slots of all time-triggered traffic to be scheduled. Specifically: For each time-triggered traffic, a set of feasible paths is pre-calculated; An external elite archive is introduced, and the particle velocity update in the next iteration is guided by the feasible solutions in the external elite archive; wherein, the external elite archive stores feasible solutions that meet the fitness requirements in the current iteration; Calculate the particle swarm fitness and update the current particle swarm based on the calculated fitness; Until the number of iterations or the convergence condition is reached, the routing path of each time-triggered traffic and the transmission time slots at each node port along the way are obtained.

3. The routing and time-slot joint scheduling method for time-sensitive networks as described in claim 2, characterized in that, For each time-triggered traffic, a set of feasible paths is pre-calculated, specifically as follows: use - The shortest path algorithm calculates the first Alternative routes; If the transmission delay of a candidate path exceeds the deadline for time-triggered traffic, the corresponding candidate path will be eliminated.

4. The routing and time-slot joint scheduling method for time-sensitive networks as described in claim 1, characterized in that, A dual-circular queue structure is configured on each switch port. The dual-circular queue structure includes a highest priority queue and a second highest priority queue. Each priority queue is only allowed to receive or send one data frame in each time slot.

5. The routing and time-slot joint scheduling method for time-sensitive networks as described in claim 4, characterized in that, The specific workflow of the dual-circular queue structure configured on the switch port is as follows: In odd time slots, the highest priority queue is in the receive state and is used to buffer data frames from the previous hop; the second highest priority queue is in the send state and is used to send the data frames buffered in the second highest priority queue to the next hop; In even-numbered time slots, the second-highest priority queue is in the receiving state and is used to buffer data frames from the previous hop; The highest priority queue is in the sending state, used to send the data frames buffered in the highest priority queue to the next hop.

6. A routing and time-slot joint scheduling system for time-sensitive networks using the method described in claim 1, characterized in that, include: The directed graph building module is configured to model time-sensitive networks as directed graphs, using switches and terminals as nodes. The optimization solution module is configured to: acquire a set of time-triggered traffic to be scheduled; and, with the goal of successful scheduling and load balancing, jointly optimize the routing paths and initial transmission time slots of all time-triggered traffic using an improved particle swarm optimization algorithm to determine the routing path of each time-triggered traffic and the transmission time slots at each node port along the route. The improved particle swarm optimization algorithm involves: introducing an external elite archive, which stores feasible solutions that meet the fitness requirements in the current iteration; and using these feasible solutions to guide particle velocity updates. The scheduling module is configured to schedule time-triggered traffic based on the routing path of each time-triggered traffic and the transmission time slots at each node port along the route.

7. An electronic device, characterized in that, It includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor, which, when executed by the processor, perform the method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, Used to store computer instructions, which, when executed by a processor, perform the method described in any one of claims 1-5.

Images