Scheduling method and system for minimizing average information age

Abstract

The invention relates to a scheduling method and system for minimizing average information age, which can realize information age minimization through an active packet loss strategy and a low-complexity scheduling algorithm under deterministic transmission time and calculation time, and simultaneously considers a packet loss scene and a resource utilization rate. The method comprises the following steps: acquiring transmission time, calculation time and a sending period of a data packet generated by a source node; according to the transmission time, the calculation time and the sending period of the data packet, an optimization problem is constructed, and the optimization problem aims at minimizing the average information age of the system; solving the optimization problem according to an active packet loss scheduling strategy and a low-complexity scheduling algorithm to obtain an optimal data packet transmission and calculation sequence; and scheduling the data packets according to the optimal data packet transmission and calculation sequence.

Description

Technical Field

[0001] This application relates to the field of mobile edge computing technology, and in particular to a scheduling method and system that minimizes the average information age. Background Technology

[0002] With the rapid development of Internet of Things (IoT) and Mobile Edge Computing (MEC) technologies, the demand for information freshness in real-time applications is increasing. Information Age (AoI), a key indicator for measuring information timeliness, is defined as the time interval from the generation of a data packet to its receipt by the destination node. It holds significant value in fields such as remote monitoring, autonomous driving, and smart grids.

[0003] Existing technologies for optimizing information age suffer from several drawbacks, including insufficient assumptions about transmission and computation times, difficulty in balancing queue latency and resource waste, insufficient research on information age optimization in proactive packet loss scenarios, and the challenge of applying highly complex algorithms in real-time. To address these issues, recent research has proposed the following improvements: joint transmission and computation scheduling to reduce overall processing time by coordinating the allocation of transmission and computation resources; proactive packet loss strategies to allow the discarding of some data packets to reduce the transmission queue cycle; and a maximum parallel computation-transmission overlap scheduling mechanism to achieve maximum parallelism of transmission and computation operations under the resource constraints of single-channel transmission and single computation unit by optimizing the scheduling strategy. However, existing methods still suffer from insufficient optimization in deterministic time scenarios, a disconnect between packet loss handling and scheduling strategies, and high algorithm complexity.

[0004] A search of existing technical literature revealed a patent application (application number 202411529357.9) titled "Scheduling Method for Vehicle-Mounted Edge Computing Networks Based on Information Age." This patent optimizes the scheduling strategy in a computing network system model where fixed servers and vehicle-mounted mobile servers coexist. It utilizes a deep reinforcement learning scheduling algorithm improved with a forking structure and action masking to optimize the scheduling strategy across both types of servers providing computing services. Specifically, it schedules the computing tasks of device nodes to idle server nodes for processing in each time slot. This minimizes the long-term average information age of the system model while satisfying resource constraints of edge computing servers, link conflict constraints, and the location constraints of vehicle-mounted mobile servers. However, this patent suffers from insufficient assumptions regarding transmission and computation times, high algorithm complexity, and low resource utilization. Summary of the Invention

[0005] Therefore, it is necessary to provide a scheduling method and system that minimizes the average information age to address the above-mentioned technical problems. This method and system can achieve data packet scheduling with low computational complexity, effective handling of packet loss, high resource utilization, and minimized information age under deterministic transmission and computation time conditions.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, the present invention provides a scheduling method for minimizing the average information age, the method comprising: Obtain the transmission time, calculation time, and sending period of the data packets generated by the source node; Based on the data packet transmission time, computation time, and sending cycle, an optimization problem is constructed, the optimization problem of which aims to minimize the average information age of the system model; The optimization problem is solved using an active packet loss scheduling strategy and a low-complexity scheduling algorithm to obtain the optimal data packet transmission and computation sequence. Based on the optimal data packet transmission and calculation sequence, data packet scheduling is achieved.

[0007] Preferably, the average information age The calculation is expressed as follows:

[0008] in, Indicates the sending period. This indicates the information age of each data packet.

[0009] Preferably, the optimization problem is calculated using the following expression:

[0010]

[0011]

[0012]

[0013]

[0014] in, Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The transmission time of a data packet from the source node to the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The computation time of each data packet on the edge server Indicates the first The transmission time of a data packet from the source node to the edge server. Indicates the first The computation time of each data packet on the edge server.

[0015] Preferably, solving the optimization problem includes: S301: An active packet loss scheduling strategy is adopted for the first set of multiple data packets, and the second set of data packets is obtained by retaining them. S302: A low-complexity scheduling algorithm is used for the second set of data packets to obtain the optimal data packet transmission and computation sequence.

[0016] Preferably, the proactive packet loss scheduling strategy includes: S311: Determine whether the sum of the total transmission time and total calculation time of the data packet set is greater than the sending period. If it is not greater, do not execute the active packet loss scheduling strategy. If it is greater, execute step S312. S312: Discard the data packets with the largest transmission time and calculation time in the data packet set in sequence until the sum of the total transmission time and total calculation time of the data packet set is not greater than the sending period. When all the data packets with the largest transmission time and calculation time in the data packet set are discarded, if the sum of the total transmission time and total calculation time of the data packet set is still greater than the sending period, proceed to step S313. S313: Discard the data packets with the largest sum of transmission time and computation time in the data packet set in turn, until the sum of the total transmission time and total computation time of the data packet set is not greater than the sending period.

[0017] Preferably, the low-complexity scheduling algorithm includes: S321: Obtain the transmission time and computation time of each data packet in the second data packet set, and combine the transmission time and computation time of each data packet into a two-dimensional data set. The second data packet set contains... Two-dimensional data; S322: Sort each two-dimensional data according to its transmission time and computation time, resulting in two queues; S323: Based on the system model's ability to simultaneously process the transmission of one data packet and the computation of another, the transmission time and computation time of the two-dimensional data with the longest transmission time are matched to obtain data packet transmission and computation columns. This allows the system model to maximize the parallel transmission of the data packet with the longest transmission time and the computation of the data packet with the longest computation time. The computation time and transmission time of the two-dimensional data with the longest transmission time are combined to obtain virtual two-dimensional data. The virtual two-dimensional data and the remaining two-dimensional data are then re-sorted according to transmission time and computation time, respectively. S324: Repeat execution In step S323, we obtain Each data packet transmission and computation column constitutes an optimal data packet transmission and computation sequence.

[0018] Preferably, after step S323, the method further includes recording the time saved by matching the two-dimensional data with the longest transmission time and the two-dimensional data with the longest calculation time.

[0019] Preferably, after step S302, the method further includes calculating the average information age of the system model based on the optimal data packet transmission and calculation sequence.

[0020] Preferably, the average information age of the system model includes: S331: Calculate the transmission and calculation points for each data packet based on the optimal data packet transmission and calculation sequence; S332: Calculate the information age of each data packet based on the transmission and calculation points of each data packet; S333: Calculate the average information age of the system model based on the information age of each data packet.

[0021] On the other hand, the present invention provides a scheduling system that minimizes the average information age, the system comprising: a model building module, a problem building module, a problem solving module, and a practical application module; The model building module is used to obtain the transmission time, calculation time, and sending period of the data packets generated by the source node; The problem construction module is used to construct an optimization problem based on the data packet transmission time, calculation time, and sending cycle. The optimization problem aims to minimize the average information age of the system model. The problem-solving module is used to solve the optimization problem and obtain the optimal data packet transmission and calculation sequence; The practical application module is used to schedule data packets based on the optimal data packet transmission and calculation sequence.

[0022] Compared with the prior art, the beneficial effects of the present invention are: This invention provides a scheduling method and system for minimizing the average information age. Under the condition of satisfying deterministic transmission time and computation time constraints, the optimal data packet transmission and computation sequence is determined by an active packet loss strategy and a low-complexity scheduling algorithm based on the maximum parallel computation-transmission overlap scheduling mechanism, thereby achieving data packet scheduling while minimizing the average information age of the system model. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating a scheduling method that minimizes the average information age in one embodiment. Figure 2 This is a schematic diagram of a scheduling method system model that minimizes the average information age in one embodiment; Figure 3 This is a schematic diagram of a scheduling system that minimizes the average information age in one embodiment. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0025] Example 1 like Figure 1 As shown in the figure, this embodiment proposes a scheduling method that minimizes the average information age, the method comprising: S1: Obtain the transmission time, calculation time, and sending period of the data packets generated by the source node; The specific implementation method for this step is as follows: like Figure 2 As shown, a system model is defined with a source node, an edge server, and a destination node. Data packets are sent from the source node to the edge server during the transmission time, and are processed by the edge server and received by the destination node during the computation time. In real-world scenarios (such as scenarios with preset tasks or fixed resource allocation), the transmission time and computation time are often deterministic or predictable. The transmission time, computation time, and packet loss probability of multiple data packets to be sent in a single task are determined. Based on the time constraints of the task and the limitations of system resources, the sending cycle is dynamically adjusted to ensure that data packets are transmitted and processed within the specified time.

[0026] S2: Based on the data packet transmission time, calculation time, and sending cycle, construct an optimization problem, the optimization problem of which aims to minimize the average information age of the system model; The specific implementation method for this step is as follows: The proactive packet loss information age minimization scheduling problem based on mobile edge computing is defined as: given a set of... A set of data packets to be sent Each data packet has a size that requires both transmission and computation, and the transmission time and computation time may differ; therefore, a data packet also has the time required for transmission. and calculation time At the same time, the sending period was limited. Within this transmission period, we need to minimize the average information age of the system model as much as possible. It's worth noting that we don't necessarily need to send all m data packets; we can send only a portion of them. ( ≤m) numbers that minimize the average information age of the system model.

[0027] Information age (AoI) is defined as the time interval from when a data packet is generated to when the destination node receives it. Average information age. It can be defined as:

[0028] First, assume that the kth group is in time. Send, and at time Arrive at the server. Then, at time Calculate it, and in time Delivered to its destination. Start time of transmission. and start calculation time The following constraints must be satisfied. First, only one packet can be transmitted at a time. Therefore, we have:

[0029] Second, the edge server must process the data packets sequentially, one by one:

[0030] Third, the k-th message can only be calculated after it is received by the edge server. Therefore, we have ≥0 and

[0031] Finally, all packets will be sent and calculated before the total time T, that is,

[0032] Therefore, the problem can be formulated as: under the constraints of the formula above, by... Select from data packets The goal is to obtain valuable data packets and ensure that all packets are sent and computation is completed within the sending cycle, thereby minimizing the average AoI of the system model.

[0033]

[0034] S3: Solve the optimization problem based on the active packet loss scheduling strategy and low-complexity scheduling algorithm to obtain the optimal data packet transmission and calculation sequence; The specific implementation method for this step is as follows: Active packet loss scheduling strategy: Based on the specific attributes of the problem, quantitatively discard excess data packets and retain as many remaining data packets as possible. By quantifying the AoI gain of data packets, high-value data packets are prioritized for retention, while low-value data packets are discarded. For example, if the time required for the set of sendable data packets is much greater than the transmission period, then selecting the data packets with shorter transmission times can meet the transmission period constraint. If the total transmission and computation time of the data packet set is no greater than the transmission period, then we determine that there is no need to actively discard data packets. The specific implementation method is as follows: (1) Determine the time required for all transmissions in the data packet set. and calculation time If the sum is greater than the transmission period T, then this strategy can be skipped. That is,

[0035] If the above conditions are met, skip the proactive packet loss strategy. If not, proceed to the next step.

[0036] (2) If the condition in (1) is not met, then the transmission time needs to be sequentially included in the data packet. and calculation time The largest value is discarded first, until all transmissions take time. and calculation time Is the sum less than the transmission period? .

[0037] (3) If all data packets that meet condition (2) are discarded, what is the total transmission time of all data packets? and calculation time If the sum is still greater than the transmission period T, then the data packet with the largest sum of transmission time and computation time will be discarded first.

[0038] Low-complexity scheduling algorithm: While we have secured sufficient data packets through an active packet loss scheduling strategy, not every arbitrary sending sequence will yield optimal results. We employ a low-complexity algorithm to minimize the time required for sending and scheduling computation given a set number of data packets. The specific implementation method is as follows: (1) In In each data packet, the transmission time and computation time of each data packet are combined into a two-dimensional data, that is, Given a set of two-dimensional data, sort the computation time and transmission time of all the two-dimensional data to obtain two queues. Prioritize matching the transmission time of the two-dimensional data with the longest transmission time with the computation time of the two-dimensional data with the longest computation time to form a pair, resulting in a data packet transmission and computation column. For example, the two two-dimensional data are (…). ), ( The first two-dimensional data represents the longest transmission time, and the second two-dimensional data represents the longest computation time. Therefore, the transmission and computation columns for this data packet are (…). After matching these two two-dimensional data, virtual two-dimensional data is returned. This means that the computation time when the first two-dimensional data is not matched and the transmission time when the second two-dimensional data is not matched are combined into a virtual two-dimensional data, and we also need to record the overlap between them to save time. This time is

[0039] (2) After implementation of (1), n ​​two-dimensional data can be regarded as n-1 two-dimensional data, where we give the order of the two largest two-dimensional data before and after, that is, the data packet transmission and calculation column ( However, the transmission and calculation relationship between the virtual two-dimensional data and the other two-dimensional data relative to this first data packet is still unknown. Therefore, subsequent processing of the remaining data packets is performed... -2 operations (1) are performed to finally obtain Each data packet transmission and computation sequence constitutes an optimal data packet sending and computation sequence. Under this sequence, we can obtain the shortest transmission and sending time period.

[0040] Suppose we need to process 4 data packets using a low-complexity scheduling algorithm. Based on these 4 packets, we obtain 4 two-dimensional data sets: (2, 7), (1, 6), (5, 3), and (4, 8). We sort the computation and transmission times of all the two-dimensional data sets, matching the transmission time of the longest transmission time in the two-dimensional data set (5, 3) with the computation time of the longest computation time in the two-dimensional data set (4, 8), resulting in a data packet transmission and computation column (5, 8). We then return the virtual two-dimensional data set (4, 3). For (4, 3), (2, 7), and (1, 6), we repeat the above operation twice more. The computation and transmission times of all two-dimensional data (4,3), (2,7) and (1,6) are sorted once. The transmission time of the two-dimensional data (4,3) with the longest transmission time and the computation time of the two-dimensional data (2,7) with the longest computation time are matched as a pair to obtain a data packet transmission and computation column (4,7), and the virtual two-dimensional data (2,3) is returned. The computation and transmission times of all two-dimensional data (2,3) and (1,6) are sorted twice. The transmission time of the two-dimensional data (2,3) with the longest transmission time and the computation time of the two-dimensional data (1,6) with the longest computation time are matched as a pair, resulting in a data packet transmission and computation column (2,6). The virtual two-dimensional data (1,3) is returned, and finally four data packet transmission and computation columns (5,8), (4,7), (2,6) and (1,3) are obtained.

[0041] S4: Based on the optimal data packet transmission and calculation sequence, data packet scheduling is implemented.

[0042] The specific implementation method for this step is as follows: The AoI of the system model is obtained from the optimal packet sending and computation sequence: After implementing the above-mentioned proactive packet loss scheduling strategy and the low-complexity scheduling algorithm based on the maximum parallel computation-transmission overlap scheduling mechanism, packet scheduling can be achieved, and the average information age of the system model can be calculated. We have determined the optimal packet sending and computation sequence. Next, we need to calculate the average information age (AoI) of the system model based on this sequence. The specific implementation method is as follows: (1) Determine the transmission and computation time points for each data packet: Based on the optimal data packet transmission and computation sequence, calculate the start transmission time and completion computation time for each data packet. The determination of these time points needs to meet the constraints mentioned above, such as only one data packet can be transmitted at a time, only one data packet can be computed at a time, and the data packet must be completed before it can enter the computation queue.

[0043] (2) Calculate the information age at each time point: For each time point t, the information age Δ(t) is defined as the time interval from the generation of the data packet to that time point. If the last updated data packet before time point t was at time t... If generated, then the information age Δ(t) = t - .

[0044] (3) Calculate the average information age: Based on the definition of average information age, calculate the average information age over the entire transmission period T:

[0045] By following the steps above, we can obtain the average information age of the system model under a given scheduling strategy.

[0046] Example 2 This embodiment further supplements the description of the scheduling method for minimizing the average information age proposed in the above embodiment.

[0047] The average information age The calculation is expressed as follows:

[0048] in, Indicates the sending period. This indicates the information age of each data packet.

[0049] The optimization problem is calculated using the following expression:

[0050]

[0051]

[0052]

[0053]

[0054] in, Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The transmission time of a data packet from the source node to the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The computation time of each data packet on the edge server Indicates the first The transmission time of a data packet from the source node to the edge server. Indicates the first The computation time of each data packet on the edge server.

[0055] Solving the optimization problem includes: S301: An active packet loss scheduling strategy is adopted for the first set of multiple data packets, and the second set of data packets is obtained by retaining them. S302: A low-complexity scheduling algorithm is used for the second set of data packets to obtain the optimal data packet transmission and computation sequence.

[0056] The proactive packet loss scheduling strategy includes: S311: Determine whether the sum of the total transmission time and total calculation time of the data packet set is greater than the sending period. If it is not greater, do not execute the active packet loss scheduling strategy. If it is greater, execute step S312. S312: Discard the data packets with the largest transmission time and calculation time in the data packet set in sequence until the sum of the total transmission time and total calculation time of the data packet set is not greater than the sending period. When all the data packets with the largest transmission time and calculation time in the data packet set are discarded, if the sum of the total transmission time and total calculation time of the data packet set is still greater than the sending period, proceed to step S313. S313: Discard the data packets with the largest sum of transmission time and computation time in the data packet set in turn, until the sum of the total transmission time and total computation time of the data packet set is not greater than the sending period.

[0057] The low-complexity scheduling algorithm includes: S321: Obtain the transmission time and computation time of each data packet in the second data packet set, and combine the transmission time and computation time of each data packet into a two-dimensional data set. The second data packet set contains... Two-dimensional data; S322: Sort each two-dimensional data according to its transmission time and computation time, resulting in two queues; S323: Based on the system model's ability to simultaneously process the transmission of one data packet and the computation of another, the transmission time and computation time of the two-dimensional data with the longest transmission time are matched to obtain data packet transmission and computation columns. This allows the system model to maximize the parallel transmission of the data packet with the longest transmission time and the computation of the data packet with the longest computation time. The computation time and transmission time of the two-dimensional data with the longest transmission time are combined to obtain virtual two-dimensional data. The virtual two-dimensional data and the remaining two-dimensional data are then re-sorted according to transmission time and computation time, respectively. S324: Repeat execution In step S323, we obtain Each data packet transmission and computation column constitutes an optimal data packet transmission and computation sequence.

[0058] After step S323, the process also includes recording the time saved by matching the two-dimensional data with the longest transmission time and the two-dimensional data with the longest computation time.

[0059] Following step S302, the method further includes calculating the average information age of the system model based on the optimal data packet transmission and calculation sequence.

[0060] The average information age of the computational system model includes: S331: Calculate the transmission and calculation points for each data packet based on the optimal data packet transmission and calculation sequence; S332: Calculate the information age of each data packet based on the transmission and calculation points of each data packet; S333: Calculate the average information age of the system model based on the information age of each data packet.

[0061] Example 3 like Figure 3 As shown, this embodiment proposes a scheduling system that minimizes the average information age. The system includes: a model building module, a problem building module, a problem solving module, and a practical application module. The model building module is used to obtain the transmission time, calculation time, and sending period of the data packets generated by the source node; The problem construction module is used to construct an optimization problem based on the data packet transmission time, calculation time, and sending cycle. The optimization problem aims to minimize the average information age of the system model. The problem-solving module is used to solve the optimization problem and obtain the optimal data packet transmission and calculation sequence; The practical application module is used to schedule data packets based on the optimal data packet transmission and calculation sequence.

Claims

1. A scheduling method that minimizes the average information age, characterized in that, include: S1: Obtain the transmission time, calculation time, and sending period of the data packets generated by the source node; S2: Based on the data packet transmission time, calculation time, and sending cycle, construct an optimization problem, the optimization problem of which aims to minimize the average information age of the system model; S3: Solve the optimization problem based on the active packet loss scheduling strategy and low-complexity scheduling algorithm to obtain the optimal data packet transmission and calculation sequence; S4: Based on the optimal data packet transmission and calculation sequence, data packet scheduling is implemented.

2. The scheduling method for minimizing the average information age according to claim 1, characterized in that, The average information age The calculation is expressed as follows: in, Indicates the sending period. This indicates the information age of each data packet.

3. The scheduling method for minimizing the average information age according to claim 2, characterized in that, The optimization problem is calculated using the following expression: in, Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The start time of transmission of each data packet from the source node to the edge server. Indicates the first The transmission time of a data packet from the source node to the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The time when each data packet begins to be calculated on the edge server. Indicates the first The computation time of each data packet on the edge server. Indicates the first The transmission time of a data packet from the source node to the edge server. Indicates the first The computation time of each data packet on the edge server.

4. The scheduling method for minimizing the average information age according to claim 1, characterized in that, Solving the optimization problem includes: S301: An active packet loss scheduling strategy is adopted for the first set of multiple data packets, and the second set of data packets is obtained by retaining them. S302: A low-complexity scheduling algorithm is used for the second set of data packets to obtain the optimal data packet transmission and computation sequence.

5. The scheduling method for minimizing the average information age according to claim 4, characterized in that, The proactive packet loss scheduling strategy includes: S311: Determine whether the sum of the total transmission time and total calculation time of the data packet set is greater than the sending period. If it is not greater, do not execute the active packet loss scheduling strategy. If it is greater, execute step S312. S312: Discard the data packets with the largest transmission time and calculation time in the data packet set in sequence until the sum of the total transmission time and total calculation time of the data packet set is not greater than the sending period. When all the data packets with the largest transmission time and calculation time in the data packet set are discarded, if the sum of the total transmission time and total calculation time of the data packet set is still greater than the sending period, proceed to step S313. S313: Discard the data packets with the largest sum of transmission time and computation time in the data packet set in turn, until the sum of the total transmission time and total computation time of the data packet set is not greater than the sending period.

6. The scheduling method for minimizing the average information age according to claim 4, characterized in that, The low-complexity scheduling algorithm includes: S321: Obtain the transmission time and computation time of each data packet in the second data packet set, and combine the transmission time and computation time of each data packet into a two-dimensional data set. The second data packet set contains... Two-dimensional data; S322: Sort each two-dimensional data according to its transmission time and computation time, resulting in two queues; S323: Based on the system model's ability to simultaneously process the transmission of one data packet and the computation of another, the transmission time and computation time of the two-dimensional data with the longest transmission time are matched to obtain data packet transmission and computation columns. This allows the system model to maximize the parallel transmission of the data packet with the longest transmission time and the computation of the data packet with the longest computation time. The computation time and transmission time of the two-dimensional data with the longest transmission time are combined to obtain virtual two-dimensional data. The virtual two-dimensional data and the remaining two-dimensional data are then re-sorted according to transmission time and computation time, respectively. S324: Repeat execution In step S323, we obtain Each data packet transmission and computation column constitutes an optimal data packet transmission and computation sequence.

7. The scheduling method for minimizing the average information age according to claim 6, characterized in that, After step S323, the process also includes recording the time saved by matching the two-dimensional data with the longest transmission time and the two-dimensional data with the longest computation time.

8. The scheduling method for minimizing the average information age according to claim 4, characterized in that, Following step S302, the method further includes calculating the average information age of the system model based on the optimal data packet transmission and calculation sequence.

9. The scheduling method for minimizing the average information age according to claim 8, characterized in that, The average information age of the computational system model includes: S331: Calculate the transmission and calculation points for each data packet based on the optimal data packet transmission and calculation sequence; S332: Calculate the information age of each data packet based on the transmission and calculation points of each data packet; S333: Calculate the average information age of the system model based on the information age of each data packet.

10. A scheduling system that minimizes the average information age, characterized in that, include: The module includes a model building module, a problem building module, a problem solving module, and a practical application module. The model building module is used to obtain the transmission time, calculation time, and sending period of the data packets generated by the source node; The problem construction module is used to construct an optimization problem based on the data packet transmission time, calculation time, and sending cycle. The optimization problem aims to minimize the average information age of the system model. The problem-solving module is used to solve the optimization problem and obtain the optimal data packet transmission and calculation sequence; The practical application module is used to schedule data packets based on the optimal data packet transmission and calculation sequence.

Images