Time-sensitive network flow scheduling method based on 5g network

Abstract

The invention provides a time-sensitive network flow scheduling method based on a 5G network. The method comprises the steps of data flow classification and priority ranking, time channel configuration, data flow scheduling, node forwarding rule configuration and exception handling. Data stream classification and priority ranking are performed by performing detailed analysis and classification on incoming data streams and performing priority ranking on the data streams according to the time sensitivity of the data streams. Configuration time channels allocate time channels with corresponding bandwidth and delay characteristics based on the category and priority of the data streams. And formulating a scheduling strategy for data stream scheduling, and performing data scheduling according to the time channels and priorities of various data streams. And configuring a node forwarding rule, so that the data can be transmitted according to a preset time channel and priority in a transmission process. When the network is abnormal, the system carries out processing based on a preset processing strategy.

Description

Technical Field

[0001] The present invention belongs to the field of 5G communications, and more specifically relates to a time-sensitive network traffic scheduling method based on a 5G network. Background Art

[0002] With the development and application of Internet technologies, network data traffic continues to grow, and a wide variety of network services are emerging. These different network services have varying requirements for the time sensitivity of data transmission. For example, applications such as video streaming and online gaming have high time sensitivity requirements, while applications such as email and web browsing have lower time sensitivity requirements. Therefore, how to effectively manage the various data flows in the network to optimize the use of network resources and ensure the quality of user experience has become a critical research topic.

[0003] Disadvantages of existing technology:

[0004] Difficulty meeting diverse needs: Existing technologies are often optimized for a single application and struggle to meet the needs of diverse application services. For example, a network scheduling strategy optimized for file downloads may not be suitable for video uploads.

[0005] Unreasonable resource allocation: Existing network routing algorithms often focus on network load balancing while ignoring the time sensitivity of data flows, which may result in services with higher real-time requirements not being given priority.

[0006] Real-time scheduling is difficult to achieve: For real-time traffic in the network, the existing packet scheduling mechanism often only schedules based on the arrival time of the packet, making it difficult to achieve real-time scheduling for different priorities.

[0007] The disadvantage of CN2024118185858 in the existing technology is that its centralized control architecture and static strategy are difficult to adapt to the high dynamics of 5G networks: on the one hand, the mode that relies on CNC to globally calculate the optimal path and time slot allocation will generate high control signaling overhead and delay in scenarios where 5G terminals have strong mobility and frequent wireless channel fluctuations, and the priority division mechanism with a fixed threshold (Dth / BWth) cannot dynamically identify sudden high-sensitive traffic (such as VR real-time interactive data), resulting in insufficient adaptability to complex business characteristics; on the other hand, its static path planning and time slot offset adjustment based on the PSO algorithm are prone to cross-slice conflicts due to resource fragmentation when facing the dynamic isolation requirements of 5G network slice resources, and fault recovery relies on the CNC to recalculate the backup path mechanism (calculation delay is about hundreds of milliseconds), which is difficult to meet the real-time requirements of 5G ultra-low latency services (such as industrial automation requiring 1ms recovery). Summary of the Invention

[0008] The main purpose of the present invention is to solve the problems existing in the existing network data flow management technology and to achieve more effective and reasonable management of online data flow.

[0009] In order to achieve the above object, the present invention is implemented by adopting the following technical solutions: the traffic scheduling method includes:

[0010] Data flow classification and prioritization: detailed analysis and classification of incoming data flows, prioritizing them based on their time sensitivity;

[0011] Configure time channels and allocate time channels with corresponding bandwidth and delay characteristics based on the data flow category and its priority, ensuring that highly time-sensitive data flows have sufficient resources for transmission;

[0012] Data flow scheduling: formulate scheduling strategies, schedule data according to the time channels and priorities of various data flows, and ensure that the time requirements of various data flows are met;

[0013] Configure node forwarding rules and load balancing strategies, and configure forwarding rules for each node in the network so that data can be transmitted according to the predetermined time channel and priority during transmission;

[0014] Exception handling: When an exception occurs in the network, the system will handle it based on the preset processing strategy.

[0015] In one embodiment, the data flow classification and prioritization includes:

[0016] S101, determining a data stream to be classified;

[0017] S102. Define the characteristics of the data flow, including protocol type, service type, and QoS requirements;

[0018] S103. Define classification rules and use a decision tree model based on the characteristics and requirements of the data flow;

[0019] S104, implementing data flow classification, classifying the data flow according to the characteristics of the data flow and the defined classification rules;

[0020] S105. Prioritize the classification results and define different priorities based on the importance or sensitivity of the results and each type of data;

[0021] S106 , allocating a channel to each classification of the data stream, and allocating different time channels to the classified data streams based on their priorities.

[0022] In one embodiment, configuring the time channel includes:

[0023] S201. Create time channels: For each type of data stream, create a time channel i according to its priority, where i ∈ {1, 2,..., n}, and n is the number of time channels;

[0024] S202. Allocate bandwidth: Based on the priority and time sensitivity requirements of each data stream, allocate the bandwidth of each time channel; The bandwidth allocation policy is represented as a function f, where f(i) = BW_i, and BW_i is the bandwidth allocated to time channel i;

[0025] S203. Set the delay characteristics of the channels: For data streams with higher priorities, set lower network delays; The setting of delay characteristics is represented as a function g, where g(i) = D_i, and D_i is the delay of time channel i;

[0026] S204. Dynamically adjust the attributes of time channels: According to the real-time load status of the network, dynamically adjust the bandwidth and delay of time channels. During peak traffic periods, reduce the bandwidth of low-priority time channels to ensure that critical services are not affected.

[0027] In one solution, the scheduling policy includes:

[0028] S301. Sort all data streams according to the arrival time a[i] of each data stream, and obtain a new sequence E', such that a[i1] <= a[i2] <=... <= a[iN];

[0029] S302. At any time point t, calculate the execution time t[i] of all data streams that have arrived but have not started to be executed, and select the one with the shortest execution time for processing, and find the data stream i that satisfies t[i] = min(t[j] | a[j] <= t < f(j));

[0030] S303. Start processing the selected data stream. For data stream i, define a function f(i) = max(a[i], f[i - 1]) + t[i] to calculate and update the completion time of each data stream;

[0031] S304. When all data streams have been processed, that is, when f(i) <= t holds for all i, the scheduling process ends.

[0032] In one solution, the configuration node forwarding rules include:

[0033] S401. Calculate the response time of each node: Send a small request to each node, and then estimate the actual response time based on the response time;

[0034] S402 finds the node with the shortest response time among all nodes: This is accomplished by comparing the response time of each node; the node with the shortest response time is preferentially selected for sending and processing data packets;

[0035] S403 sends the data stream to the node with the shortest response time: when there are multiple data streams to transmit, the node with the shortest response time is selected according to their respective priorities.

[0036] S404 updates the node's response time: after a node processes a request, it is necessary to recalculate the node's response time and redistribute the data flow based on the new response time.

[0037] In one embodiment, the load balancing strategy includes:

[0038] Assume that m is the total number of requests, n is the number of servers, and k is the number of samples, where k = n; then, for the i-th server, the probability of being selected is;

[0039] P = cumulative probability that the random number generated from the 1st request to the mth request falls within the range of the reservoir = ∑(the probability that the random number generated from the 1st to the mth request falls within the range of the reservoir);

[0040] Right now:

[0041] Therefore, the probability of each server being selected should be 1 / n, which means that each request is equally shared;

[0042] Here are the steps:

[0043] (1) Initialize a sampling pool of size N; each bit in this sampling pool represents a server;

[0044] (2) When each client request arrives, a random number i is generated in the range [0, the current total number of requests]. Assuming that the current request is the jth request in total, this random number i is generated.

[0045] (3) If the random number i is within the sampling pool range (0 to N-1), then the request is processed by the server number i;

[0046] (4) According to the reservoir sampling rule, the probability of each server being selected should be the same; therefore, after n requests are completed, it is expected that each server will process an equal number of requests; that is, for the i-th server, the probability of it being selected is 1 / n; this achieves overall load balancing.

[0047] Beneficial effects of the present invention:

[0048] 1. Time-sensitive judgment and prioritization: By accurately classifying and prioritizing data streams, the method can ensure that time-sensitive data streams are processed first, thereby improving user experience and service quality.

[0049] 2. Dynamic time channel configuration: The method dynamically configures the time channel, which can not only meet the specific needs of different types of data flows, but also make timely adjustments when the network environment changes.

[0050] 3. Efficient data flow scheduling: Through the shortest job first (SJF) algorithm, the method can achieve efficient data scheduling and improve the overall network efficiency.

[0051] 4. Flexible node forwarding and load balancing: This method combines the minimum response time method and the reservoir sampling method to perform node configuration and load balancing, which helps to avoid network congestion and ensure smooth transmission of data streams.

[0052] This method provides a comprehensive and adaptive solution for effectively scheduling time-sensitive network traffic and can greatly improve network utilization efficiency and service quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flow chart of the present invention;

[0054] Figure 2 Flowchart for classifying and prioritizing data flows;

[0055] Figure 3 To configure the time channel flow chart;

[0056] Figure 4 Schedule a flow chart for the data flow. DETAILED DESCRIPTION

[0057] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0058] like Figure 1 As shown, a time-sensitive network traffic scheduling method based on 5G network includes:

[0059] Step 1: Data flow classification and prioritization

[0060] Perform detailed analysis and classification of incoming data flows, which involves examining packet metadata (e.g., source, destination, protocol type, etc.). Prioritize data flows based on their time sensitivity. For example, time-sensitive services such as live video streaming and voice communications should be given high priority, while less time-sensitive services such as email transmission and web page loading should be given low priority.

[0061] S101: Determine the data stream to be classified and the data stream to be allocated channels.

[0062] S102: Define the characteristics of the data flow, which include the above factors, such as protocol type, service type, QoS requirements, etc.

[0063] S103. Define classification rules. Use a decision tree model based on the characteristics of the data stream and our needs.

[0064] It uses information gain to select the best partitioning attribute. Given a data set D, the information gain G(D,a) of attribute a is defined as the difference between the information entropy E(D) of data set D and the conditional information entropy E(D / a) after a given attribute a value, that is:

[0065] G(D,a)=E(D)-E(D|a)

[0066] Among them, information entropy E(D) represents the purity of the data set D, which can be calculated using the following formula:

[0067] E(D)=-SUM((|Di| / |D|)*log2(|Di| / |D|))

[0068] Here, D represents the data set, |D| represents the number of samples in the data set D, Di represents the set of class i samples in the data set D, and |Di| represents the number of class i samples.

[0069] Conditional information entropy E(D|a) represents the expected purity of the dataset D under the condition that the value of attribute a is known, and can be calculated using the following formula:

[0070] E(D|a)=SUM((|Dv| / |D|)*E(Dv))

[0071] Here, Dv represents the sample subset whose value is v on attribute a, |Dv| represents the number of samples in the subset, and E(Dv) represents the information entropy of the subset.

[0072] Then, the information gain of each possible attribute is calculated, and the attribute with the largest information gain is selected as the partitioning attribute. Through repeated partitioning, a decision tree can be constructed.

[0073] S104: Implement data flow classification: Use the selected classification method to classify the data flow according to the characteristics of the data flow and the defined classification rules.

[0074] S105. Prioritize the classification results. Define different priorities based on the results and the importance or sensitivity of each type of data.

[0075] S106: Allocate a channel for each classification of the data stream. According to the priority, different time channels are allocated to the classified data streams.

[0076] Step 2: Configure the time channel

[0077] Based on the data flow category and its priority, a time channel with corresponding bandwidth and delay characteristics is allocated, thus ensuring that high-priority (i.e., highly time-sensitive) data flows have sufficient resources for transmission.

[0078] Step 2, configuring time channels, ensures sufficient resources for high-priority, time-sensitive data flows within the solution, allowing for dynamic adjustments based on actual conditions. We use an improved least-connection load balancing algorithm to configure time channels. The specific method is as follows.

[0079] S201. Create a time channel: For each type of data stream, create a time channel i according to its priority, i∈{1,2,...,n}, where n is the number of time channels.

[0080] S202. Allocate bandwidth: Based on the priority and time sensitivity of each data stream, allocate bandwidth to each time channel. The bandwidth allocation strategy can be formulated as a function f, f(i) = BWi, where BWi is the bandwidth allocated to time channel i.

[0081] S203, setting the delay characteristics of the channel: for data flows with higher priority, set a lower network delay. Setting the delay characteristics can be formulated as a function g, g(i) = Di, where Di is the delay of time channel i.

[0082] S204. Dynamically adjust time channel attributes: Dynamically adjust the bandwidth and delay of the time channel based on the real-time load status of the network. For example, during peak traffic periods, the bandwidth of low-priority time channels can be reduced to ensure that critical services are not affected.

[0083] In the improved least-connection load balancing algorithm, the dynamic adjustment of bandwidth BW and delay D can be expressed by the following formula:

[0084]

[0085]

[0086] Here, t represents time, ΔBW and ΔD are the changes in bandwidth and delay depending on the network status, LCi(t) is the number of connections on time channel i at time t, and LCmax is the maximum number of connections on all time channels.

[0087] The basic idea of this algorithm is that if a time channel has a large number of links (LCs), it indicates that the channel is heavily loaded, and its bandwidth should be appropriately reduced, increasing latency. Conversely, if the number of links is small, it indicates that the channel is lightly loaded, and its bandwidth should be increased and latency reduced to process transmission tasks as quickly as possible, increase network throughput, and improve user experience. This prevents overloading of highly loaded channels and avoids wasting resources on less loaded channels, achieving efficient utilization of network resources.

[0088] The above steps implement the allocation of time channels, bandwidth division, delay setting and dynamic adjustment, forming an improved least connection load balancing strategy.

[0089] Step 3: Data flow scheduling

[0090] Develop a scheduling strategy to schedule data based on the time channel and priority of each data flow to ensure that the time requirements of each data flow are met. For example, the transmission of real-time video streaming and voice communication may be met first, and then the transmission of other non-time-sensitive data flows.

[0091] Suppose there are N data flows, each data flow i is defined by a tuple (a[i], t[i]), where a[i] represents the arrival time of data flow i, and t[i] represents the execution time (job length) of data flow i.

[0092] Define an expression E to describe the arrival time and execution time of each data flow: E = {(a[i], t[i])|i = 1, 2, ..., N}.

[0093] Define function f(i) to represent the completion time of data stream i and initialize it to 0: f(i) = 0, i = 1, 2, ..., N.

[0094] Next, we analyze the process of data flow scheduling according to the SJF algorithm. At this time, the current execution time is t.

[0095] S301. Sort all data streams according to the arrival time a[i] of each data stream, and obtain a new sequence E' such that a[i1]<=a[i2]<=...<=a[iN].

[0096] S302. At any time point t, calculate the execution time t[i] of all the data streams that have arrived but have not started to be executed, and select the one with the shortest execution time for processing, that is, find the data stream i that satisfies t[i]=min(t[j]|a[j]<=t<f(j)).

[0097] S303. Start processing the selected data stream. For the data stream i, define the function f(i)=max(a[i],f[i - 1])+t[i] to calculate and update the completion time of each data stream.

[0098] S304. When all the data streams have been processed (that is, f(i)<=t holds for all i), the scheduling process ends.

[0099] By minimizing the execution time of each data stream, SJF can effectively improve the overall efficiency of the system.

[0100] Step 4: Configure node forwarding rules and load - balancing strategies

[0101] Configure forwarding rules for each node in the network so that data can be transmitted according to the predetermined time channels and priorities during the transmission process. This involves the load - balancing strategies of each node in the network and the packet - selection strategies when the data traffic is large and the network is congested.

[0102] Configuring node forwarding rules is mainly to effectively transfer data between nodes. High - priority data streams can be transmitted first, and network congestion can be avoided through load - balancing strategies.

[0103] Configure forwarding rules: During the process of node configuration using the Least Response Time Method (LRTM), we need to monitor each node, understand its response time, and then send new requests to the node with the shortest response time. The response time refers to the time from when a node receives a request to when it finishes processing, considering the time required to process the request and the current network latency. In the network, the calculation formula for the node response time is:

[0104] rj = pj + dj

[0105] where rj is the response time of node j, pj is the time for node j to process a request, and dj is the current network latency. We need to find the node with the shortest response time among all nodes, that is, find min{r1, r2, r3…rn}. Here, n is the number of nodes in the network.

[0106] Detailed implementation process of configuring node forwarding rules using the least response time method:

[0107] S401 calculates the response time of each node: a small request is sent to each node, and then the actual response time is estimated based on the response time.

[0108] S402: Find the node with the shortest response time among all nodes: This is accomplished by comparing the response time of each node. The node with the shortest response time is preferentially selected for sending and processing data packets.

[0109] S403 sends the data stream to the node with the shortest response time: when there are multiple data streams to be transmitted, they need to be selected according to their respective priorities and the node with the shortest response time.

[0110] S404 Update Node Response Time: After a node processes a request, it is necessary to recalculate the node's response time and redistribute the data flow based on the new response time. It should be noted that because the network state is dynamic, the node's response time also changes in real time and needs to be updated regularly.

[0111] The main advantage of the minimum response time method is that it can dynamically distribute data flows based on the dynamic network environment and the processing capabilities of the nodes themselves, thereby ensuring the efficiency and stability of the network. This method can maximize the efficiency of network utilization.

[0112] Load balancing strategy: Each node in the network dynamically adjusts forwarding rules based on the current network load and data flow priority. For heavily loaded nodes, data flows are forwarded to less loaded nodes whenever possible, creating a load-balanced network. The load balancing algorithm uses a reservoir sampling method, meaning each node takes turns processing data flows.

[0113] Assume that m is the total number of requests, n is the number of servers, and k is the number of samples (here k=n). Then, for the i-th server, the probability of it being selected is.

[0114] P (selecting the i-th server) = the probability that the random number generated from the 1st request to the m-th request falls within the range of the reservoir = ∑ (the probability that the random number from the 1st to the m-th request falls within the range of the reservoir).

[0115] Right now:

[0116] Therefore, the probability of each server being selected should be 1 / n, which means that each request is shared equally.

[0117] Here are the steps:

[0118] (1) Initialize a sampling pool (reservoir) of size N. Each bit in the sampling pool represents a server.

[0119] (2) When each client request arrives, a random number i is generated in the range [0, the current total number of requests]. Assuming that the current request is the jth request in total, then this random number i is generated.

[0120] (3) If the random number i is within the sampling pool range (0 to N-1), then the request is processed by the i-th server.

[0121] (4) According to the reservoir sampling rule, each server should have the same probability of being selected. Therefore, after n requests have been processed, we expect each server to handle an equal number of requests. In other words, the probability of the i-th server being selected is 1 / n. This way, we can achieve overall load balancing.

[0122] In real-world applications, the number of servers n is often much smaller than the typical number of requests m. In this case, selecting the first n requests in the reservoir sampling essentially ensures that the request load across servers is balanced when m>n. This method balances the sampling results of a complete sampling process. It cannot guarantee complete balance for each individual sampling. Looking at a single moment, some servers may have higher request volumes, but this does not affect the overall load balance.

[0123] Step 5: Exception handling

[0124] When network anomalies such as packet loss and delay occur, the system will handle them based on the preset processing strategy. For example, when an increase in network delay is detected, the time channel may need to be reconfigured to improve the efficiency of data transmission.

[0125] Through the above steps, the time-sensitive network traffic scheduling method can not only improve the utilization of network resources and ensure that time-sensitive data streams can be transmitted first, but also adapt to changes in network traffic, dynamically adjust and optimize, flexibly handle exceptions, and facilitate management, thereby effectively ensuring and improving the quality and efficiency of network services.

[0126] The core difference between this application and the existing technology lies in the differentiated design of technical architecture, scheduling mechanism and applicable scenarios. The existing technology adopts a centralized control architecture (CNC / CUC), divides priorities through global topology perception and fixed thresholds, combines the PSO algorithm to generate static optimal paths and the SAC algorithm to optimize time slot allocation. Its core goal is to achieve global resource optimization and fault recovery in traditional TSN scenarios. This application focuses on the fusion scenario of 5G network and TSN, and introduces a dynamic time channel mechanism and distributed load balancing strategy: in data flow classification, this application adopts a decision tree model to replace the fixed bandwidth / delay threshold, which can handle more complex traffic characteristics (such as protocol type, QoS dynamic changes) and enhance classification flexibility; in terms of resource allocation, through the dynamic bandwidth and delay characteristic configuration of the time channel (rather than fixed time slot offset adjustment), combined with the load balancing achieved by reservoir sampling, it better adapts to the high mobility and slice resource fluctuation characteristics of the 5G network. In addition, the exception handling mechanism of this application responds quickly through preset strategies, rather than relying on CNC to recalculate the backup path in the existing technology, which is more suitable for the low latency requirement scenario of 5G network.

[0127] The key problem solved by this application is that the traditional centralized TSN scheduling mechanism is difficult to adapt to the dynamic characteristics of 5G networks: first, the high dynamics of 5G network topology (such as sudden changes in link quality caused by terminal movement) requires a more flexible priority division mechanism, and the existing technology based on fixed threshold (Dth / BWth) priority division cannot effectively identify sudden high-sensitivity traffic; second, the resource isolation requirements brought about by 5G network slicing require a time channel mechanism to realize the mapping of physical resources and logical channels, and the global time slot allocation of the existing technology may cause resource conflicts between slices; third, the distributed load balancing requirements in 5G edge computing scenarios. Compared with the centralized path calculation of the existing technology, the reservoir sampling strategy of this application can reduce the control plane signaling overhead. The technical effects brought about by these innovations include: in the 5G-TSN hybrid networking scenario, the end-to-end transmission delay of the data stream is reduced (dynamic bandwidth allocation of the time channel reduces queuing delay), the network throughput is improved (load balancing strategy optimizes link utilization), and the abnormal recovery time is shortened to 50ms (preset strategy fast switching), which is significantly better than the hundreds of milliseconds fault recovery performance of the existing technology. In addition, the decision tree classification model improves the accuracy of high-priority traffic identification and is more adaptable to complex business scenarios than the threshold method in existing technologies.

[0128] Example 1:

[0129] Assume that there are three types of data flows in our network: real-time video streaming, file transfer, and email.

[0130] Step 1: Classify and sort

[0131] Let's first categorize these three data streams: live video streaming is Category 1, file transfers are Category 2, and emails are Category 3. In terms of time sensitivity, live video streaming has the highest priority, email has the lowest, and file transfers are in between. Therefore, we prioritize these three data streams as follows: live video streaming > file transfers > emails.

[0132] Step 2: Configure the time channel

[0133] Based on classification and priority, we configured three time channels, with bandwidths designated BW1, BW2, and BW3, corresponding to real-time video streaming, file transfer, and email, respectively. Assuming a total bandwidth of 100 Gbps, we set BW1 to 60 Gbps, BW2 to 30 Gbps, and BW3 to 10 Gbps, giving real-time video streaming the highest bandwidth. We also added a latency feature, setting the latency for real-time video streaming to 1 ms, file transfer to 10 ms, and email to 50 ms.

[0134] Step 3: Data Scheduling

[0135] We use the SJF algorithm for scheduling. We process live video streams first, followed by file transfers, and finally emails. For example, if there are 10 live video streams, 5 file transfers, and 20 emails, we will process the 10 live video streams first, then the 5 file transfers, and finally the 20 emails.

[0136] Step 4: Configure node forwarding rules and load balancing strategies

[0137] Suppose there are four nodes, configured using the minimum response time method. Assume that the response time of node 1 is 0.5ms, that of node 2 is 1ms, that of node 3 is 2ms, and that of node 4 is 3ms. Node 1 will be prioritized for data packet transmission and processing. After processing is complete, the node's response time is updated and the configuration is reconfigured. Furthermore, a reservoir sampling method is used for load balancing. When more data streams need to be transmitted, each node is polled to process the same data stream.

[0138] Step 5: Exception handling

[0139] Assume that during data flow transmission, packet loss occurs at node 3. After the system detects this anomaly, it adjusts the bandwidth of this node to 0 and reallocates time channels among other normal nodes to ensure that high-priority data flows can be transmitted in a timely manner.

[0140] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-readable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0141] It should be understood that the detailed description of the technical solutions of the present invention using the preferred embodiments above is illustrative and not restrictive. A person skilled in the art, after reading the present specification, may modify the technical solutions described in the embodiments or replace some of the technical features therein with equivalents; such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A time-sensitive network traffic scheduling method based on 5G network, characterized by: The described traffic scheduling method includes: Data stream classification and priority ranking, which involves detailed analysis and classification of incoming data streams, and priority ranking according to the time sensitivity of the data streams; Configuring time channels, which involves allocating time channels with corresponding bandwidth and delay characteristics according to the category and priority of the data streams, ensuring that data streams with high time sensitivity have sufficient resources for transmission; Data stream scheduling, which involves formulating a scheduling strategy and performing data scheduling according to the time channels and priorities of various data streams, ensuring that the time requirements of various data streams are met; Configuring node forwarding rules and load balancing strategies, which involves configuring forwarding rules for each node in the network so that data can be transmitted according to the predetermined time channels and priorities during the transmission process; Exception handling, which involves the system processing based on a preset processing strategy when an exception occurs in the network.

2. The time-sensitive network traffic scheduling method based on 5G network according to claim 1 is characterized in that: The described data stream classification and priority ranking includes: S101. Determine the data stream to be classified; S102. Define the characteristics of the data stream, including protocol type, service type, and QoS requirements; S103. Define classification rules, which involves using a decision tree model according to the characteristics and requirements of the data stream; S104. Implement data stream classification, which involves classifying the data stream according to the characteristics of the data stream and the defined classification rules; S105. Perform priority ranking on the classification results, which involves defining different priorities according to the results and the importance or sensitivity of each type of data; S106. Allocate channels for each classification of the data stream, which involves allocating different time channels for the classified data streams according to the priority.

3. The time-sensitive network traffic scheduling method based on 5G network according to claim 1 is characterized in that: The described configuration of time channels includes: S201. Create time channels: For each type of data stream, create a time channel i according to its priority, where i ∈ {1, 2,..., n}, and n is the number of time channels; S202. Allocate bandwidth: Based on the priority and time sensitivity requirements of each data stream, allocate the bandwidth of each time channel; the bandwidth allocation strategy is represented as a function f, where f(i) = BW_i, and BW_i is the bandwidth allocated to time channel i; S203. Set the delay characteristics of the channel: For data streams with higher priorities, set lower network delays; the setting of delay characteristics is represented as a function g, where g(i) = D_i, and D_i is the delay of time channel i; S204. Dynamically adjust the time channel attributes: According to the real-time load status of the network, dynamically adjust the bandwidth and delay of the time channels. During the traffic peak period, reduce the bandwidth of low-priority time channels to ensure that key services are not affected.

4. The time-sensitive network traffic scheduling method based on 5G network according to claim 1 is characterized in that: The described scheduling strategy includes: S301. Sort all data streams according to the arrival time a[i] of each data stream and obtain a new sequence E', such that a[i1] <= a[i2] <=... <= a[iN]; S302. At any time point t, calculate the execution time t[i] of all data streams that have arrived but have not started to execute, and select the one with the shortest execution time for processing, and find the data stream i that satisfies t[i] = min(t[j] | a[j] <= t < f(j)); S303, start processing the selected data stream, define the function f(i)=max(a[i],f[i-1])+t[i] for data stream i, and calculate and update the completion time of each data stream; S304: When all data streams are processed, that is, f(i)<=t holds for all i, the scheduling process ends.

5. The time-sensitive network traffic scheduling method based on 5G network according to claim 1 is characterized in that: The configuration node forwarding rules include: S401 calculates the response time of each node: a small request is sent to each node, and the actual response time is estimated based on the response time; S402 finds the node with the shortest response time among all nodes: This is accomplished by comparing the response time of each node; the node with the shortest response time is preferentially selected for sending and processing data packets; S403 sends the data stream to the node with the shortest response time: when there are multiple data streams to transmit, the node with the shortest response time is selected according to their respective priorities; S404 updates the node's response time: after a node processes a request, it is necessary to recalculate the node's response time and redistribute the data flow based on the new response time.

6. The time-sensitive network traffic scheduling method based on 5G network according to claim 1 is characterized in that: The load balancing strategy includes: Assume that m is the total number of requests, n is the number of servers, and k is the number of samples, where k = n; then, for the i-th server, the probability of being selected is; P = cumulative probability that the random number generated from the 1st request to the mth request falls within the range of the reservoir = ∑(the probability that the random number generated from the 1st to the mth request falls within the range of the reservoir); Right now: Therefore, the probability of each server being selected should be 1 / n, which means that each request is equally shared; Here are the steps: (1) Initialize a sampling pool of size N; each bit in this sampling pool represents a server; (2) When each client request arrives, a random number i is generated in the range [0, the current total number of requests]. Assuming that the current request is the jth request in total, this random number i is generated. (3) If the random number i is within the sampling pool range (0 to N-1), then the request is processed by the server number i; (4) According to the reservoir sampling rule, the probability of each server being selected should be the same; therefore, after n requests are completed, it is expected that each server will process an equal number of requests; that is, for the i-th server, the probability of it being selected is 1 / n; this achieves overall load balancing.

Images