Traffic scheduling method for vehicle-mounted time-sensitive network

Abstract

The embodiment of the invention provides a traffic scheduling method for a vehicle-mounted time sensitive network. The method is applied to the technical field of industrial internet. The method comprises the following steps: constructing a network topology structure of a vehicle-mounted time-sensitive network; modeling the traffic of the vehicle-mounted time-sensitive network into a six-tuple, and constructing a traffic set by using the modeled traffic; constructing a correlation matrix, wherein the element value of the correlation matrix is used for indicating the dependency relationship between the traffic; determining a transmission path of each flow based on a network topology structure of the vehicle-mounted time-sensitive network; performing chain flow division on the traffic set based on the correlation matrix to obtain a chain flow set; determining a weighted priority of each chain flow in the chain flow set, and sorting all the chain flows in the chain flow set according to the weighted priorities from high to low to obtain a chain flow scheduling sequence; and each chain flow is scheduled in sequence according to the chain flow scheduling sequence, so that the calculation complexity is reduced, and the scheduling success rate and the operation efficiency are improved.

Description

Technical Field

[0001] This application relates to the field of industrial internet technology, and in particular to a traffic scheduling method for vehicular time-sensitive networks. Background Technology

[0002] As in-vehicle electronic and electrical architectures evolve towards centralized and regional control units, in-vehicle networks need to simultaneously carry multiple service flows, including large-scale data streams from sensors such as cameras and LiDAR, periodic control flows from body and chassis control, and multimedia streams from in-vehicle entertainment and information interaction. These flows not only have high bandwidth and low latency requirements, but also exhibit complex timing dependencies and coordination constraints between different functional domains. Improper scheduling can lead to delays in control commands or failure to transmit sensor information in a timely manner, thereby affecting the safety and stability of autonomous driving and advanced driver assistance systems. Time-Sensitive Networking (TSN), as a next-generation deterministic Ethernet technology, provides fundamental support for deterministic transmission in in-vehicle networks through time synchronization and time-slot-based scheduling mechanisms.

[0003] However, in vehicular TSN environments, the volume of service traffic is enormous and highly correlated. Traditional distributed or heuristic scheduling methods are prone to conflicts under high load conditions, leading to a decrease in scheduling success rate. Furthermore, directly scheduling the entire global flow set presents a massive problem with high computational complexity and insufficient efficiency, failing to meet the real-time requirements of vehicular networks. Moreover, most existing scheduling methods treat service flows as independent entities, lacking system modeling of the dependencies between flows and failing to fully utilize the potential for scheduling optimization.

[0004] Therefore, there is an urgent need for a new scheduling method that can effectively characterize the dependencies between business flows, reduce the complexity of the overall problem, and improve the scheduling success rate and operational efficiency while ensuring real-time performance. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a traffic scheduling method for vehicular time-sensitive networks, the method comprising:

[0006] Constructing the network topology of an in-vehicle time-sensitive network;

[0007] The traffic of the vehicle-mounted time-sensitive network is modeled as a six-tuple, and the traffic set is constructed using the modeled traffic.

[0008] Construct a correlation matrix, the element values ​​of which are used to indicate the dependencies between traffic flows;

[0009] The transmission path of each traffic item is determined based on the network topology of the vehicular time-sensitive network.

[0010] The traffic set is divided into chain flows based on the correlation matrix to obtain the chain flow set;

[0011] Determine the weighted priority of each chain in the chain flow set, and sort all the chain flows in the chain flow set from high to low according to the weighted priority to obtain the chain flow scheduling sequence;

[0012] Each link is scheduled sequentially according to the link scheduling sequence.

[0013] Optionally, the traffic of the vehicular time-sensitive network is modeled as a six-tuple, as follows:

[0014]

[0015] in, Indicates the source node, Indicates the target node. Indicates the size of the traffic data packets. Indicates the transmission period of the traffic. Indicates the priority of traffic. Indicates the deadline for data transmission.

[0016] Optionally, constructing the correlation matrix includes:

[0017] Build Create a matrix and initialize all elements in the matrix to 0.

[0018] The correlation matrix is ​​obtained by updating all elements in the matrix according to the preset dependencies between traffic flows.

[0019] The correlation matrix is ​​represented as follows:

[0020]

[0021]

[0022] in, For the first Traffic, For the first Traffic volume.

[0023] Optionally, the network topology based on the vehicular time-sensitive network determines the transmission path of each traffic flow, including:

[0024] Dijkstra's algorithm is used to search for the shortest path from each traffic source node to the target node in the network topology of a vehicle-mounted time-sensitive network.

[0025] The shortest path from the source node to the target node is used as the transmission path for traffic.

[0026] Optionally, the step of partitioning the traffic set based on the correlation matrix to obtain the chain flow set includes:

[0027] Determine whether there is a dependency relationship between two traffic streams based on the element values ​​in the correlation matrix;

[0028] Traffic with dependencies is grouped into the same chain flow, and a chain flow set is constructed based on the obtained chain flows.

[0029] Optionally, the weighted priority of each chain stream in the chain stream set is determined according to the following formula:

[0030]

[0031] in, For weighted priority, This indicates the number of flows contained in the chain. This indicates the normalized traffic priority. This indicates the normalized size of the traffic data packets. This represents the reciprocal of the normalized traffic transmission deadline. The weighting parameter for traffic priority. The weighting parameter is the size of the traffic data packets. The weighted parameter for the traffic transmission deadline.

[0032] Optionally, scheduling each link stream sequentially according to the link stream scheduling sequence includes:

[0033] Establish constraints, including: periodic constraints, deadline constraints, and flow-related constraints;

[0034] For the current chain flow to be scheduled, a solution set is constructed in the candidate solution space using the remaining available time slots, and an improved particle swarm optimization algorithm is used to search for and solve the solution to generate a scheduling solution.

[0035] Optionally, the improved particle swarm optimization algorithm includes:

[0036] The initial scheduling solution is generated by combining random distribution with boundary constraints;

[0037] The generated scheduling solution is evaluated using a preset objective function, which is expressed as follows:

[0038]

[0039] in, This indicates the penalty for violating the constraints. Indicates the scheduling failure rate. Indicates the delay ratio. The weight parameter represents the penalty. The weighted parameter representing the scheduling failure rate, Weighting parameters for latency ratio;

[0040] The generated scheduling solution is iteratively updated based on dynamic weights and a comprehensive learning mechanism until the maximum number of iterations is reached or the objective function converges.

[0041] This application provides a traffic scheduling method for vehicular time-sensitive networks (VSNs). The method involves: constructing the network topology of the VSN; modeling the VSN traffic as six-tuples and using the modeled traffic to construct a traffic set; constructing a correlation matrix, where the elements of the correlation matrix indicate the dependencies between traffic flows; determining the transmission path of each traffic flow based on the VSN network topology; partitioning the traffic set into chain flows based on the correlation matrix; determining the weighted priority of each chain flow in the chain flow set and sorting all chain flows in the chain flow set from high to low weighted priority to obtain a chain flow scheduling sequence; and scheduling each chain flow sequentially according to the chain flow scheduling sequence. This method reduces computational complexity and improves scheduling success rate and operational efficiency. Attached Figure Description

[0042] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0043] Figure 1 A flowchart illustrating the traffic scheduling method for vehicular time-sensitive networks provided in this application embodiment;

[0044] Figure 2 A schematic diagram of the network topology of the vehicle-mounted time-sensitive network provided in the embodiments of this application;

[0045] Figure 3 A schematic diagram of the chain flow division provided in the embodiments of this application;

[0046] Figure 4 This is a schematic diagram of the chain flow scheduling process provided in the embodiments of this application.

[0047] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0049] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein.

[0050] In this application, the terms "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0051] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0052] Figure 1 This is a flowchart illustrating a traffic scheduling method for vehicular time-sensitive networks provided in an embodiment of this application. Figure 1 As shown, the traffic scheduling method for vehicular time-sensitive networks provided in this embodiment includes:

[0053] S1: Construct the network topology of the vehicle-mounted time-sensitive network.

[0054] In this embodiment, the topology of the in-vehicle time-sensitive network is as follows: Figure 2As shown, the network topology includes a vehicle central control unit (VCC), multiple zone control units (ZCUs), and corresponding switches (SWs). The vehicle is divided into six zones based on spatial location, managed by six zone control units (ZCUs): front left (FL), front right (FR), center left (ML), center right (MR), rear left (RL), and rear right (RR). Each zone's ZCU is connected to a switch to aggregate data streams generated by sensors and actuators within the zone. Each zone's ZCU is further connected to the VCC, enabling cross-zone data exchange and centralized scheduling. The switches maintain a ring interconnection to provide multi-path selection, improving system reliability and fault tolerance.

[0055] Furthermore, the vehicular time-sensitive network topology can be represented as follows: ,in For a set of nodes, Let be the set of edges. Nodes represent communication nodes in the vehicular time-sensitive network (VSN), including all switches and terminal devices in the VSN, and edges represent communication links in the VSN.

[0056] S2: Model the traffic of the vehicle time-sensitive network as a six-tuple, and use the modeled traffic to construct a traffic set.

[0057] Understandably, traffic in a vehicle time-sensitive network can originate from different regions, such as sensor streams in the environmental perception domain, control command streams in the execution control domain, and multimedia streams in the entertainment information domain. In order to uniformly schedule different service traffic, VCC obtains the basic attributes of each traffic by parsing the system configuration file during the network initialization phase.

[0058] Each traffic entry represents a periodic data transfer from one node to another in a directed graph. Modeled as a six-tuple, it is represented as follows:

[0059]

[0060] in, Indicates the source node, Indicates the target node. Indicates the size of the traffic data packets. The transmission cycle indicates the frequency of data transmission. The priority of a flow indicates its relative importance. This indicates the transmission deadline for the traffic, which is the maximum time required for transmission. Successful transmission within the deadline is considered a successful scheduling, while failure to do so indicates a scheduling failure.

[0061] Aggregate all traffic to form a traffic collection There are a total of n traffic flows. This flow set describes all data transmission requirements in the entire vehicular time-sensitive network, providing a global perspective for subsequent path allocation, link flow partitioning, and scheduling optimization.

[0062] S3: Construct a correlation matrix, the element values ​​of which are used to indicate the dependencies between traffic flows.

[0063] Understandably, there are often logical relationships between different service traffic in vehicular time-sensitive networks. For example, the traffic from the front camera must be completed before the braking control command, which is a time-series dependency; while data from multiple sensors need to arrive at the same destination within a certain time error range, i.e., the arrival time difference is less than or equal to the error threshold to meet the system's cooperation requirements, which is a cooperation dependency.

[0064] Specifically, any two streams Heliu Relationship , means as follows:

[0065]

[0066] Constructing a correlation matrix This allows for a systematic representation of the correlation between all traffic flows. Each element in the matrix clearly expresses the relationship between each pair of traffic flows, thus transforming complex business constraints into a clear and computable data structure. The correlation matrix is ​​represented as follows:

[0067]

[0068] Wherein, the matrix size is Initially, all elements in the matrix are set to zero, indicating that there are no dependencies between different traffic flows. Dependencies between different service flows are then preset. During the system initialization phase, the traffic is traversed and the correlation matrix is ​​updated item by item to obtain a complete matrix expression. When new traffic is added, the matrix can be maintained by incremental update.

[0069] S4: Determine the transmission path for each traffic flow based on the network topology of the vehicle-mounted time-sensitive network.

[0070] Specifically, Dijkstra's algorithm is used to search for the shortest path from the source node to the target node for each traffic flow in the network topology of the vehicular time-sensitive network; the shortest path from the source node to the target node is used as the transmission path of the traffic flow.

[0071] Dijkstra's shortest path algorithm, using hop count as a weight, searches for the shortest path from the source node to the target node in a directed graph, thereby reducing the number of intermediate nodes and forwarding latency. For any traffic to be scheduled, a feasible transmission path can be found within the given network topology. ,in Represents a stream This method effectively shortens the transmission path length and reduces link occupancy latency.

[0072] After the transmission path calculation is completed, the transmission path for each flow is fixed and serves as a constraint in the scheduling process. Specifically, when generating the scheduling scheme, it is necessary to ensure that the time slots allocated to the flow are arranged on the link resources traversed by its path, avoiding link conflicts or port resource overlaps. If the path sets of two flows intersect, it must be ensured that their time slots within the same supercycle do not overlap. In this way, the scheduling process can guarantee the reachability and transmission determinism of all flows under a fixed topology.

[0073] S5: Based on the correlation matrix, the traffic set is divided into chain flows to obtain the chain flow set.

[0074] Specifically, the dependency relationship between two flows is determined based on the element values ​​in the correlation matrix; flows with dependency relationships are merged into the same chain flow, and a chain flow set is constructed based on the obtained chain flows.

[0075] Figure 3 This is a schematic diagram illustrating the flow segmentation process provided in an embodiment of this application. In this embodiment, the flow segmentation is performed from the global traffic set. Select one flow And identify other traffic that depends on this traffic, if traffic and corresponding elements in the matrix This indicates that the two have a temporal or collaborative dependency and need to be merged into the same flow; this is achieved by traversing the global flow set sequentially. , and all of them Directly or indirectly related traffic aggregation forms a complete chain. After the chain flow is formed, the flow within it is transferred from the global set. Remove it from the list and select a new remaining traffic stream as the baseline. Repeat the above identification and aggregation process until a set is formed. If it is empty, the final result will be a set of chain flows. ,in This represents the m-th chain flow. Each chain flow consists of a set of interdependent flows, for example... That is, to represent the chain flow Depend on These three traffic streams, through this step, decompose the complex global traffic scheduling problem into multiple smaller chained subproblems, reducing computational complexity and ensuring that related traffic streams can be coordinated in subsequent scheduling.

[0076] S6: Determine the weighted priority of each chain in the chain flow set, and sort all the chain flows in the chain flow set from high to low according to the weighted priority to obtain the chain flow scheduling sequence.

[0077] Understandably, after obtaining the set of link flows, to ensure that critical traffic or traffic highly sensitive to latency receives priority in time slots and resources during subsequent scheduling phases, it is necessary to determine the importance of each link flow in the overall scheduling. Each link flow is prioritized and assigned a weighted priority to reflect its relative importance in global scheduling. The weighted priority is calculated for each link flow. Satisfy the following formula:

[0078]

[0079] in, This indicates the number of flows contained in the chain. This indicates the normalized traffic priority. This indicates the normalized size of the traffic data packets. This represents the reciprocal of the normalized traffic transmission deadline. The weighting parameter for traffic priority. The weighting parameter is the size of the traffic data packets. The weighted parameter for the traffic transmission deadline.

[0080] Compute the set of chain flows Weighted priority of all links Then sort them from high to low to obtain an ordered sequence. ,satisfy: .

[0081] Through the above steps, the link flow partitioning and priority ranking ensures that related traffic is processed in groups during scheduling, and that critical links receive priority scheduling resources to guarantee the timely completion of important tasks. Meanwhile, lower-priority links are scheduled sequentially, provided system resources permit. This link flow priority ranking and scheduling mechanism effectively avoids resource conflicts.

[0082] S7: Schedule each link sequentially according to the link scheduling sequence.

[0083] Specifically, constraints are established, including periodic constraints, deadline constraints, and flow-related constraints. For the current chain flow to be scheduled, a solution set is constructed in the candidate solution space using the remaining available time slots, and an improved particle swarm optimization algorithm is used to search for and solve the solution to generate a scheduling solution.

[0084] Figure 4 This is a schematic diagram of the link flow scheduling process provided in this embodiment. In this embodiment, the scheduling time window of the vehicular time-sensitive network is first determined, which is the super-period formed by the least common multiple of the cycles of each service traffic. This time interval can fully reflect the periodic behavior of the system. To ensure the feasibility of the scheduling solution, various constraint rules are introduced on the available time slot set, including periodic constraints, deadline constraints, and traffic correlation constraints. Periodic constraints ensure that each traffic is scheduled within its cycle, and that transmission within the same time slot does not conflict with other traffic. Deadline constraints mean that each traffic must complete transmission before its specified deadline; otherwise, scheduling is considered a failure. Traffic correlation constraints ensure that each traffic is scheduled according to a predetermined order and coordination requirements, thereby guaranteeing the correctness of the business logic. These constraints effectively avoid resource overlap and timing conflicts, providing a foundation for subsequent optimization.

[0085] After the constraints are established, the chain flows are selected one by one for scheduling based on the chain flow ranking results obtained in the above steps. For the chain flow to be scheduled, a solution set is constructed in the candidate solution space using the remaining available time slots, and an improved particle swarm optimization algorithm is used to search for and solve it.

[0086] After scheduling a single link stream, the scheduling result is recorded and the remaining available time slots are updated. Then, the next link stream is scheduled. This process is repeated until all links streams have been scheduled, resulting in a global scheduling scheme.

[0087] Specifically, the improved particle swarm optimization algorithm includes:

[0088] (1) The initial scheduling solution is generated by combining random distribution with boundary constraints.

[0089] In the initialization phase, the initial particle positions are generated by combining random distribution with boundary constraints and mapped to the legal search space to ensure that the initial solution has basic feasibility.

[0090] Specifically, each particle, when representing the scheduling start slot of a certain flow, is limited to the period of that flow, as shown below:

[0091]

[0092] in, Indicates the first The particle in the first The starting transmission time slot position on each traffic flow Indicates the interval Uniformly distributed sampling within. For the first Each traffic cycle, Let be the number of sub-flows in the chain. This boundary constraint effectively avoids infeasible solutions caused by large-scale random initialization, thereby significantly improving the effectiveness of the initial solution and search efficiency.

[0093] (2) The generated scheduling solution is evaluated using a preset objective function, which is expressed as follows:

[0094]

[0095] in, This indicates the penalty for violating the constraints. Indicates the scheduling failure rate. Indicates the delay ratio. The weight parameter represents the penalty. The weighted parameter representing the scheduling failure rate, Weighting parameters for latency ratio;

[0096] Evaluating the generated scheduling solution using an objective function allows for the simultaneous consideration of key factors such as the degree of constraint satisfaction, scheduling success rate, and end-to-end latency, thereby ensuring a reasonable balance between reliability and real-time performance. The magnitude of the objective function value directly reflects the quality of the solution; a smaller objective function value indicates that the solution satisfies multiple constraints while possessing better scheduling performance, and is therefore considered a superior solution.

[0097] (3) The generated scheduling solution is iteratively updated according to the dynamic weight and the integrated learning mechanism until the maximum number of iterations is reached or the objective function converges.

[0098] During the iterative update phase, this embodiment introduces a dynamic inertia weight adjustment mechanism during the updating of particle velocity and position. Specifically, a larger inertia weight is set in the early stage of the search to enhance the global exploration capability of the particles, enabling them to search for potential solutions over a wider range; as the iteration progresses, the inertia weight is gradually reduced, thereby improving the local convergence accuracy and allowing the particle swarm to more closely approximate the optimal solution;

[0099]

[0100] in, This represents the inertia weight at the t-th iteration. This represents the initial maximum value of the inertia weight. The initial minimum value of the inertia weight is represented by T, and the maximum number of iterations is represented by T.

[0101] This embodiment employs a comprehensive learning mechanism, where the learning object of a particle is no longer limited to a single global optimal solution, but rather dynamically selects the learning object by combining the global optimal solution with the historical optimal solutions of other individuals. Specifically, a random number within the interval [0,1] is introduced. When the random number is less than 0.5, the global optimal solution is selected as the learning object; when the random number is greater than or equal to 0.5, the historical optimal solution of an individual is selected as the learning object. This achieves adaptive switching between global and individual information, effectively maintaining the diversity of the population and preventing particles from becoming overly reliant on a single experience and falling into local optima, thereby ensuring that the entire population converges stably towards the global optimum.

[0102] When the iteration meets the termination condition, such as reaching the upper limit of the number of iterations or the convergence threshold of the objective function, the current optimal solution is output as the scheduling scheme for the current chain flow. After all chains have been scheduled, the final complete scheduling scheme is output.

[0103] The traffic scheduling method for vehicular time-sensitive networks (VSNs) provided in this application reduces the solution scale by constructing a correlation matrix and dividing the traffic into chains based on dependencies. This transforms the complex global scheduling problem into several sub-problems. By combining weighted priority criteria with priority-order scheduling, it ensures that critical task traffic receives resources first, improving system determinism. Furthermore, an improved particle swarm optimization algorithm is employed during scheduling, significantly increasing the scheduling success rate and shortening the runtime. This method effectively meets the deterministic communication and high reliability requirements of VSNs and has strong potential for widespread application.

[0104] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0105] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A traffic scheduling method for vehicular time-sensitive networks, characterized in that, The method includes: Constructing the network topology of an in-vehicle time-sensitive network; The traffic of the vehicle-mounted time-sensitive network is modeled as a six-tuple, and the traffic set is constructed using the modeled traffic. Construct a correlation matrix, the element values ​​of which are used to indicate the dependencies between traffic flows; The transmission path of each traffic item is determined based on the network topology of the vehicular time-sensitive network. The traffic set is divided into chain flows based on the correlation matrix to obtain the chain flow set; Determine the weighted priority of each chain in the chain flow set, and sort all the chain flows in the chain flow set from high to low according to the weighted priority to obtain the chain flow scheduling sequence; Each link is scheduled sequentially according to the link scheduling sequence.

2. The method according to claim 1, characterized in that, The traffic of the vehicular time-sensitive network is modeled as a six-tuple, as follows: ； in, Indicates the source node, Indicates the target node. Indicates the size of the traffic data packets. Indicates the transmission period of the traffic. Indicates the priority of traffic. Indicates the deadline for data transmission.

3. The method according to claim 1, characterized in that, The construction of the correlation matrix includes: Build Create a matrix and initialize all elements in the matrix to 0. The correlation matrix is ​​obtained by updating all elements in the matrix according to the preset dependencies between traffic flows. The correlation matrix is ​​represented as follows: ； ； in, For the first Traffic, For the first Traffic volume.

4. The method according to claim 1, characterized in that, The network topology based on vehicular time-sensitive networking determines the transmission path of each traffic flow, including: Dijkstra's algorithm is used to search for the shortest path from each traffic source node to the target node in the network topology of a vehicle-mounted time-sensitive network. The shortest path from the source node to the target node is used as the transmission path for traffic.

5. The method according to claim 1, characterized in that, The process of partitioning the traffic set based on the correlation matrix to obtain the chain flow set includes: Determine whether there is a dependency relationship between two traffic streams based on the element values ​​in the correlation matrix; Traffic with dependencies is grouped into the same chain flow, and a chain flow set is constructed based on the obtained chain flows.

6. The method according to claim 1, characterized in that, The weighted priority of each chain stream in the chain stream set is determined by the following formula: ； in, For weighted priority, This indicates the number of flows contained in the chain. This indicates the normalized traffic priority. This indicates the normalized size of the traffic data packets. This represents the reciprocal of the normalized traffic transmission deadline. The weighting parameter for traffic priority. The weighting parameter is the size of the traffic data packets. The weighted parameter for the traffic transmission deadline.

7. The method according to claim 1, characterized in that, The step of scheduling each link stream sequentially according to the link stream scheduling sequence includes: Establish constraints, including: periodic constraints, deadline constraints, and flow-related constraints; For the current chain flow to be scheduled, a solution set is constructed in the candidate solution space using the remaining available time slots, and an improved particle swarm optimization algorithm is used to search for and solve the solution to generate a scheduling solution.

8. The method according to claim 7, characterized in that, The improved particle swarm optimization algorithm includes: The initial scheduling solution is generated by combining random distribution with boundary constraints; The generated scheduling solution is evaluated using a preset objective function, which is expressed as follows: ； in, This indicates the penalty for violating the constraints. Indicates the scheduling failure rate. Indicates the delay ratio. The weight parameter represents the penalty. The weighted parameter representing the scheduling failure rate, Weighting parameters for latency ratio; The generated scheduling solution is iteratively updated based on dynamic weights and a comprehensive learning mechanism until the maximum number of iterations is reached or the objective function converges.

Images