A 5G and TSN fusion network flow scheduling framework and resource allocation method

Abstract

The present invention relates to a 5G (including 5G-U that uses both licensed and unlicensed bands) and TSN fusion network flow scheduling framework and resource allocation method, belonging to the field of communication technology, which includes three parts: one is a 5G and TSN fusion network architecture for industrial Internet, the second is a flow scheduling framework, and the third is a resource allocation method. The present invention uses a semi-continuous scheduling mechanism of hybrid static and dynamic resource allocation according to traffic type and priority, and combines a dynamic resource allocation algorithm based on maximum and minimum fair allocation to allocate resource blocks to each service, so as to achieve low latency guarantee and fair allocation of resources for high priority traffic in 5G and TSN fusion network flow scheduling. The present invention can reduce the average end-to-end latency of delay-sensitive traffic, improve the resource utilization of scheduling, and improve the fairness of flow scheduling.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology and relates to a 5G and TSN fusion network flow scheduling framework and resource allocation method. Background Art

[0002] Traditional industrial automation systems achieve low-latency and high-reliability communication requirements through wired networks. Time-sensitive networking (TSN) is a new solution to the cross-domain flexibility of existing industrial Ethernet technologies. It further enhances time synchronization, deterministic flow scheduling, and highly reliable seamless redundancy based on standard Ethernet, providing high reliability, deterministic bounded low latency, and low jitter service guarantees for end-to-end data transmission. In addition, TSN supports highly flexible configuration according to topology and communication requirements, and can provide a unified industrial network for all Ethernet-based communications. However, with the rapid development of wireless networks, especially 5G networks, in order to meet the highly flexible network system requirements envisioned in the industrial Internet, using 5G networks as an extension of TSN in the wireless domain is a promising approach. Wireless communication achieves higher flexibility and modularity, and further realizes situations involving mobile devices such as mobile vehicles, automatic guided vehicles (AGVs), or mobile consoles. At present, the main fusion solution of 5G and TSN is to use 5G networks as logical bridges for TSN, that is, 5G networks as extensions of TSN in wireless networks. This extended converged network can bring many advantages to industrial scenarios. The high flexibility brought by wireless communication enables industrial networks to support mobile devices and use cases. In addition, due to the unified management and control of network convergence, configuration and maintenance can be simplified, bringing better vertical integration and modularity.

[0003] The pre-allocation scheme of the Semi-Persistent Scheduling (SPS) technology is a promising method to reduce access latency. In semi-persistent scheduling (SPS), the gNB configures periodic time and frequency resources for the UE through a packet containing an initial control command. The number of commands required for resource allocation is greatly reduced, so the control channel overhead, alignment time, transmission time and processing time can be significantly reduced. In the industrial Internet, the high density and mobility of large-scale machine equipment require frequent resource allocation requests, which will become a heavy load on the central base station and cause network congestion and large signaling and processing delays, seriously offsetting the benefits of ultra-low latency optimization in the 5G standard and endangering the effectiveness of security applications. Compared with dynamic scheduling, SPS has higher computational efficiency and can significantly reduce the load on the control channel. SPS was originally designed to improve the QoS of VoIP (Voice over Internet Protocol) services, because their transmission rates are usually fixed and known, so they can be perfectly scheduled. At present, there are several main problems in the scheduling of 5G and TSN converged flows in industrial scenarios:

[0004] Guarantee of service QoS requirements: Traffic from TSN networks may contain multiple service types, which can generally be divided into delay-sensitive traffic and non-delay-sensitive traffic, all of which have different QoS requirements. Packets of various services arriving from TSN transmission enter the 5G system through the TSN converter. Among them, control, synchronization, and cyclic service flows are most sensitive to delay and jitter requirements and have the highest priority and criticality. Configuration diagnosis, audio and video, and best effort (BE) service types have lower priority and criticality, and have lower requirements for delay and jitter. How to ensure that traffic of different service types is forwarded and processed in a timely manner and avoid conflicts with other service traffic is an important issue.

[0005] Latency guarantee for time-sensitive traffic: TSN networks provide deterministic low-latency services at the millisecond level or even lower for industrial applications, providing data transmission guarantees for latency-sensitive traffic. Although the ultra-reliable low-latency communication (URLLC) framework of the 5G system supports real-time applications with latency requirements as low as 1ms, scheduling in wireless systems is usually more complex than in wired systems due to the time-varying behavior of wireless channels, and traditional dynamic access schemes in cellular networks cannot guarantee strict timeliness due to their complex "scheduling request-scheduling grant (SR-SG)" signaling process. Therefore, how to ensure the low latency requirements of latency-sensitive traffic in 5G wireless systems is an important issue.

[0006] Resource allocation fairness guarantee: Although SPS can greatly reduce the control signaling overhead and effectively reduce latency, unlike the fixed and known traffic transmission rate in VoIP, the traffic in the industrial Internet may contain multiple service types. Traffic with multiple different QoS requirements is usually scattered and variable in industry. If the traditional SPS is directly applied to the uplink scheduling in the 5G and TSN converged network for the industrial Internet without improvement, the access delay will be reduced at the expense of spectrum utilization. In addition, for variable bit rate (VBR) traffic, insufficient RBs allocation will lead to longer latency, and excessive RBs will lead to waste of wireless resources. The traditional static resource allocation algorithm cannot make full use of the 5GAN resources with less load, which makes the transmission delay of other QoS flows continue to increase, resulting in a gradual decrease in network throughput. In the dynamic resource allocation algorithm, the UE dynamically adjusts the weight and allocates wireless resources for data transmission by requesting the base station. If the actual resource allocation requirements of each user are not considered, the fairness of resource allocation cannot be achieved, resulting in a waste of bandwidth. Therefore, a reasonable resource allocation mechanism is needed to ensure the fair allocation of bandwidth resources. Summary of the invention

[0007] In view of this, the purpose of the present invention is to provide a 5G and TSN converged network flow scheduling framework and resource allocation method, which can ensure the end-to-end low latency of high-priority traffic during 5G and TSN converged network flow scheduling, as well as the fairness of bandwidth resource allocation and other performance.

[0008] In order to achieve the above object, the present invention provides the following technical solutions:

[0009] On the one hand, the present invention provides a 5G (including 5G-U that uses both licensed and unlicensed bands) and TSN (Time-Sensitive Networking) converged network flow scheduling framework, including a TSN network centralized network configuration (CNC) module, a 5G network application function (AF) module, a policy control function (PCF) module, a session management function (SMF) module and a TSN converter; the 5G network, as a logical bridge in the TSN network, obtains QoS information from TSN service traffic through AF, sends it to the PCF module and is mapped to the 5G QoS framework by the PCF, generates a 5QI mapping table and corresponding forwarding control rules, and then the SMF sends the configuration information to the RAN (Radio Access Network) according to the received forwarding control rules to realize resource allocation; the TSN converter includes a network side converter (NW-TT) and a device side converter (DS-TT), which serve as logical ports of 5G and TSN networks to realize the adaptation of the 5G system and the TSN network.

[0010] Furthermore, the feature mapping process of QoS from TSN to 5G service flow is as follows: in the control plane, the 5G system communicates with the TSN CNC through the TSN AF function to complete the mapping and intercommunication of network configuration and management information, obtains the QoS parameters of the traffic from the TSN network, including three QoS characteristics of delay requirement, data packet size and priority, and sends them to the PCF; the PCF maps the QoS flow from TSN to the corresponding 5QI value that meets its traffic characteristics based on the obtained QoS feature parameters and refers to the 3GPP 5QI standardized QoS feature mapping table, and then generates a 5QI mapping table and sends it to the SMF to configure the scheduling of QoS flow in the 5G system; in the data plane, the QoS flow from TSN is transmitted to the 5G system through the TSN converter (NW-TT / DS-TT), and the forwarding control rules are configured according to the mapping provided by the PCF to establish the corresponding QoS flow in the 5G system; for different types of QoS flows, the corresponding 5QI is configured for them, so that they are forwarded and processed in the 5GS.

[0011] Furthermore, for the eight types of TSN QoS flows, the data transmission guarantees of the five most critical QoS flows (priority values ​​7 to 3) have higher latency requirements. For these TSN flows, they are mapped to the delay-critical GBR type 5QI; since configuration and diagnostic TSN flows require bandwidth guarantees and the GBR type guarantees the bandwidth of the PDU session, this type of flow should be mapped to the GBR type 5QI; for audio and video and best-effort TSN flows, their criticality is low and bandwidth guarantees are not required, so they are mapped to the Non-GBR type 5QI; and considering the latency requirements, packet size and priority traffic characteristics of the QoS flow, a statically mapped 5QI mapping table is obtained.

[0012] On the other hand, the present invention provides a 5G and TSN converged network flow scheduling resource allocation method, the 5G PCF generates corresponding forwarding control rules according to the 5QI mapping table under the mapped 5G QoS framework and sends it to the SMF; the 5G SMF function provides the scheduling forwarding control rules to the RAN, and then the RAN allocates resource blocks according to the requirements and characteristics of each QoS flow; after completing the resource allocation, the QoS flow uses the allocated resource blocks to complete the data transmission.

[0013] Furthermore, a semi-persistent scheduling mechanism that mixes static and dynamic resource allocation is used according to traffic type and priority, and a dynamic resource allocation algorithm based on Max-min Fairness (MMF) is combined to allocate radio resource blocks to each QoS flow.

[0014] Furthermore, the method specifically comprises the following steps:

[0015] Step 1: The QoS flow from TSN is transmitted to the 5G system through the TSN converter. The relevant QoS information from the TSN traffic is obtained according to the TSN AF function of 5G, and converted to the QoS framework of the 5G system by PCF, and a 5QI mapping table is generated to obtain the traffic QoS information under the 5G QoS framework, and generate the forwarding control rules of the QoS flow;

[0016] Step 2: The 5G base station identifies the traffic type and criticality of the QoS flow according to the mapped 5QI mapping table and forwarding control rules, and uses a semi-continuous scheduling mechanism that mixes static and dynamic resource allocation to allocate resources; for time-sensitive deterministic periodic flow (TS-DPF), a certain number of resource blocks are statically allocated in advance for data transmission to ensure low latency in its scheduling, and other types of traffic obtain wireless resource blocks through dynamic resource allocation;

[0017] Step 3: Calculate whether the continuously pre-allocated resource blocks meet the resource requirements of the continuously statically allocated resource traffic. If so, put this part of the QoS flow set that does not request dynamic resource allocation into If not, the QoS flow needs to request dynamic resources from the base station. The set of QoS flows requesting dynamic resource allocation is v, and the required dynamic resource blocks are calculated;

[0018] Step 4: For the QoS flows in set v that need to be dynamically allocated resources, the resource allocation weight of each QoS flow is obtained based on the maximum-minimum fair allocation algorithm and the traffic priority function, and then the remaining resource blocks are dynamically allocated to each QoS flow according to the weight; if the allocated resource blocks meet the resource requirements of the traffic, the resource blocks that exceed the traffic resource requirements are recycled and removed from set v; if the allocated resource blocks do not meet the resource requirements, proceed to step 5;

[0019] Step 5: Calculate the number of resource blocks remaining for dynamic allocation after the previous allocation, and update the set v that needs to request dynamic resources, and return to step 4; until all QoS flows are allocated resource blocks that meet the resource requirements or the remaining allocatable resource blocks are exhausted, the dynamic resource allocation ends.

[0020] The beneficial effects of the present invention are:

[0021] 1) The present invention can reduce the average end-to-end delay of delay-sensitive traffic. The average end-to-end delay of the converged network is simulated and analyzed using this solution. Compared with the existing resource allocation solution based on dynamic scheduling, the average end-to-end delay of the delay-sensitive traffic scheduled by the present invention is smaller. At the same time, compared with the existing resource allocation solution based on static scheduling, the scheduling success rate of the present invention is higher.

[0022] 2) The present invention can improve the resource utilization of scheduling. The present invention uses the scheme to simulate and analyze the resource utilization of the network. Compared with the existing static resource allocation scheme, it improves the resource utilization of the entire network and the throughput of the entire system at the expense of a small amount of end-to-end delay performance. At the same time, compared with the existing dynamic scheduling, the present invention reduces the control signaling overhead and significantly reduces the computing cost.

[0023] 3) The present invention can improve the fairness of flow scheduling. The semi-persistent scheduling mechanism that uses a hybrid static and dynamic resource allocation and combines it with a dynamic resource allocation algorithm based on the maximum-minimum fair allocation algorithm not only ensures the deterministic transmission of delay-sensitive traffic, but also fully considers the data transmission opportunities and QoS requirements of other traffic. Compared with existing solutions, it can effectively prevent bandwidth abuse and starvation problems and improve the fairness of network bandwidth allocation.

[0024] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0026] Figure 1 The 5G and TSN fusion network architecture adopted by the present invention;

[0027] Figure 2 A schematic diagram of the resource allocation process for 5G and TSN converged network flow scheduling based on semi-persistent scheduling;

[0028] Figure 3 Schematic diagram of the working state of this embodiment;

[0029] Figure 4 The network resource utilization result diagram of this scheme, the static scheduling scheme and the dynamic scheduling scheme;

[0030] Figure 5 The average end-to-end delay results of this scheme, the static scheduling scheme, and the dynamic scheduling scheme. DETAILED DESCRIPTION

[0031] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0032] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0033] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "front", "rear", etc. indicate the orientation or position relationship, they are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0034] A 5G and TSN (Time-Sensitive Networking) fusion network flow scheduling framework and resource allocation method, such as Figure 1 As shown, it includes a 5G and TSN converged network architecture for the industrial Internet, including a TSN network CNC module; a 5G network AF module, a PCF module, an SMF module, and a TSN converter. The 5G network serves as a logical bridge in the TSN network. The AF obtains and analyzes the QoS information from the TSN service traffic, sends it to the PCF module, and maps it to the 5G QoS framework by the PCF, generates a 5QI mapping table and a corresponding forwarding control mechanism, and then the SMF sends the configuration information to the RAN according to the received forwarding control mechanism to realize resource allocation; the TSN converter includes NW-TT and DS-TT as logical ports of the 5G and TSN networks to realize the adaptation of the 5G system and the TSN network.

[0035] The process of mapping the characteristics of QoS flows from TSN to 5G is as follows: in the control plane, the 5G system communicates with the TSN CNC through the TSN AF function to complete the mapping and intercommunication of network configuration and management information, obtain the QoS parameters of the traffic from the TSN network, mainly including three QoS characteristics of delay requirements, packet size and priority, and send them to the PCF; based on the obtained QoS characteristic parameters, the PCF refers to the 3GPP 5QI standardized QoS characteristic mapping table, maps the QoS flow from TSN to the corresponding 5QI value that meets its traffic characteristics, and then generates a 5QI mapping table, which is sent to the SMF to configure the scheduling of QoS flows in the 5G system. In the data plane, the QoS flow from TSN is transmitted to the 5G system through the TSN converter (NW-TT / DS-TT), and the forwarding control mechanism is configured according to the mapping provided by the PCF to establish the corresponding QoS flow in the 5G system. For different types of QoS flows, the corresponding 5QI is configured for them, so that they can be forwarded and processed in the 5GS.

[0036] For the eight types of TSN QoS flows, the data transmission guarantees of the five most critical QoS flows (priority values ​​7 to 3) have higher latency requirements. Therefore, for these TSN flows, they are mapped to the latency-critical GBR type 5QI; since the configuration and diagnostic types of TSN flows require bandwidth guarantees and the GBR type guarantees the bandwidth of the PDU session, this type of flow should be mapped to the GBR type 5QI; for audio and video and best-effort types of TSN flows, they are less critical and do not require bandwidth guarantees, so they are mapped to the Non-GBR type 5QI. Then, considering the latency requirements, packet size, and priority flow characteristics of the QoS flow, the statically mapped 5QI mapping table is shown in Table 1:

[0037] Table 1 QoS mapping from TSN to 5G

[0038]

[0039] The 5G system considers the QoS requirements of different traffic flows, adopts a semi-continuous scheduling mechanism that mixes static and dynamic resource allocation, and uses a dynamic resource allocation algorithm based on Max-min Fairness (MMF) to achieve deterministic low-latency scheduling of delay-sensitive traffic. The goal of this solution (SPS+MMF) is to ensure end-to-end low-latency scheduling of high-priority delay-sensitive traffic, fairly allocate network resources for each QoS flow, thereby avoiding network collisions, improving network throughput, and achieving a reasonable trade-off between low latency and spectrum efficiency in resource scheduling.

[0040] The present invention also provides a 5G and TSN converged network flow scheduling resource allocation method, specifically: 5G PCF generates a corresponding forwarding control mechanism according to the 5QI mapping table under the mapped 5G QoS framework and sends it to SMF; the 5G SMF function provides the scheduling forwarding control mechanism to RAN, and then RAN allocates resource blocks according to the needs and characteristics of each QoS flow; after completing the resource allocation, the QoS flow uses the allocated resource blocks to complete data transmission.

[0041] The 5G and TSN converged network flow scheduling resource allocation method uses a semi-persistent scheduling mechanism that mixes static and dynamic resource allocation according to traffic type and priority, and combines a dynamic resource allocation algorithm based on maximum-min fairness (MMF) to allocate wireless resource blocks to each QoS flow.

[0042] The specific steps of this program are as follows Figure 2 :

[0043] Step 201, DS-TT transmits the QoS flow from the TSN network to the 5G system. The TSN AF function of 5G obtains the statistical feature information of the QoS flow to identify the type of traffic and related QoS information, converts it to the QoS framework of the 5G system, and obtains the traffic QoS information under the 5G QoS framework; then the QoS flow is cached in the queue of the TSN converter and waits for scheduling.

[0044] In step 202, the gNB obtains QoS information such as 5QI and traffic type according to the 5QI mapping table, identifies the traffic type and criticality of the QoS flow, determines whether it is periodic delay-sensitive traffic, and adopts a semi-continuous scheduling mechanism that mixes static and dynamic resource allocation. For TS-DPF, a certain number of resource blocks are pre-continuously and statically allocated for data transmission to ensure low latency in its scheduling. Other types of traffic obtain wireless resources through dynamic resource allocation.

[0045] Step 203: Calculate whether the continuously pre-allocated resource blocks meet the resource requirements of the continuously statically allocated resource traffic. If so, the QoS flow can directly use the pre-allocated resource blocks for data transmission; if not, the QoS flow needs to calculate the required dynamic resource blocks and request dynamic resources from the gNB.

[0046] Step 204: For all QoS flows that require dynamic resource allocation, the resource allocation weight of each QoS flow is obtained based on the traffic priority function based on the maximum-minimum fair allocation algorithm, and then the remaining resource blocks are dynamically allocated to each QoS flow based on the weight. It is determined whether the allocated resource blocks meet the resource requirements of the QoS flow. If so, the resource blocks with excess traffic resource requirements are recycled, and the QoS flow can use the allocated resources for data transmission; if the resource requirements are not met, the remaining RBs are used to continue to allocate wireless resources until the available RBs are exhausted.

[0047] Step 205: After the resource allocation algorithm is completed, the gNB sends an "authorization message" to the UE, which contains the resource allocation results within this period.

[0048] Step 206: The gNB updates each UE QoS flow resource request message and updates the resource allocation result.

[0049] The dynamic resource allocation of non-TS-DPF adopts a dynamic resource allocation algorithm based on the maximum-minimum fair allocation algorithm, through which network resources are fairly allocated to each QoS flow that needs to communicate. Network resources are allocated according to the priority of traffic flow, and the resources obtained by each QoS flow are limited to be less than or equal to the resources requested. QoS flows that are not satisfied continue to share the remaining wireless resources. Therefore, for the QoS flows of the entire network, fairness in network bandwidth allocation can be achieved to prevent bandwidth abuse and starvation.

[0050] The following will be combined with the attached Figure 3The implementation of uplink resource allocation in multiple cycles of the 5G and TSN converged network flow scheduling framework and resource allocation method in the present invention is described in detail. In this embodiment, the 5G and TSN networks are interconnected through the TSN converter and the TSN AF function. When the network starts running, the TSN AF function first analyzes the traffic characteristics from the TSN, including the byte size, priority, delay requirements and other characteristic information of the traffic, and then sends it to the 5GPCF function to map the QoS information to the 5G QoS framework to generate a 5QI mapping table and a QoS flow forwarding control policy, identify whether it is a delay-sensitive deterministic periodic flow or a non-delay-sensitive QoS flow, and send it to the 5G SMF function to establish the configuration and management of the QoS flow, and then cache the traffic in the UE. The gNB implements a semi-persistent scheduling mechanism that mixes static and dynamic resource allocation. After receiving the scheduling resource request information from the UE, it determines whether it is a delay-sensitive deterministic periodic flow. If it is a TS-DPF, it continuously pre-allocates fixed time-frequency resource blocks for it, and then runs a dynamic allocation algorithm based on maximum-minimum fair allocation to dynamically allocate RAN resources to TS-DPF and non-TS-DPF flows that do not meet resource requirements; after completing the resource allocation within each transmission time interval (TTI), it sends an authorization message to the UE to inform the resource allocation result within the period, and the UE sends the current cache information of its own node. After receiving the scheduling request of the UE node, the gNB updates the cache queue length of all UEs and repeats the above resource allocation process.

[0051] This example demonstrates the uplink resource allocation implementation process of the 5G and TSN converged network flow scheduling framework and resource allocation scheme over multiple cycles.

[0052] The resource allocation in the present invention "A 5G and TSN converged network flow scheduling framework and resource allocation method" is different from the traditional SPS-based 5G network resource allocation scheme. The traditional SPS-based network resource allocation mechanism mostly dynamically allocates bandwidth through the weight factors of each business flow, without considering the differences in QoS requirements of different flows, and cannot guarantee the deterministic low-latency transmission of high-priority delay-sensitive traffic in industrial scenarios and prevent bandwidth abuse and starvation problems. A 5G and TSN converged network flow scheduling framework and resource allocation method are adopted, and the scheme is used to achieve end-to-end low latency guarantee and fairness of bandwidth resource allocation for high-priority traffic in 5G and TSN converged network flow scheduling. Compared with the traditional 5G network resource allocation scheduling scheme, the present invention has the following advantages:

[0053] 1) The present invention can improve the resource utilization of traffic scheduling, and simulate the resource allocation mechanism scheme in the present invention and perform numerical analysis on the network resource utilization. Figure 4As shown, with the increase of network load (i.e., the number of QoS flows), the network resource utilization in the static scheduling (SS) scheme, the dynamic scheduling (DS) scheme, and the SPS+MMF scheme is constantly increasing, and the resource utilization of the SPS+MMF scheme is almost as high as that of the DS scheme, and compared with the dynamic scheduling, this scheme can ensure that each QoS flow is fairly allocated to resources, effectively preventing bandwidth abuse and starvation problems. At the same time, compared with the existing static scheduling (SS) scheme, the network resource utilization of the present invention is higher, and when the network load is above 200 (i.e., the network load is large), the resource utilization advantages of the SPS+MMF and DS schemes are more obvious. Therefore, the present invention has more advantages in improving the performance of network resource utilization and improves the overall throughput of the system.

[0054] The present invention can reduce the average end-to-end delay of QoS flow scheduling. The resource allocation scheme in the present invention is simulated and the average end-to-end delay is numerically analyzed. Compared with the existing dynamic scheduling (DS) scheme, the average end-to-end delay of the QoS flow scheduling of the present invention is smaller, especially the end-to-end delay of delay-sensitive traffic is guaranteed. Figure 5 As shown, as the network load increases, the average end-to-end delay in the SS scheme, DS scheme and SPS+MMF scheme continues to increase. Since the SPS+MMF scheme saves the scheduling request-scheduling grant (SR-SG) process of delay-sensitive traffic through SPS scheduling, its average end-to-end delay is smaller than that of the DS scheme. Compared with the SS scheme, due to the sacrifice of a small amount of end-to-end delay to improve resource utilization, its end-to-end delay is slightly more than that of the SS scheme, but the SPS+MMF scheme can still achieve an uplink delay of less than 4ms under various traffic loads, which can meet the extremely low delay requirements of TSN networks and URLLC applications. Therefore, the present invention has good average end-to-end delay performance and can achieve deterministic low-latency scheduling.

[0055] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A 5G and TSN fusion network flow scheduling resource allocation method, characterized in that: 5G PCF generates corresponding forwarding control rules based on the 5QI mapping table under the mapped 5GQoS framework and sends them to SMF; 5G SMF function provides scheduling forwarding control rules to RAN, and then RAN allocates resource blocks according to the requirements and characteristics of each QoS flow; after completing resource allocation, QoS flow uses the allocated resource blocks to complete data transmission; A semi-persistent scheduling mechanism that uses a hybrid static and dynamic resource allocation mechanism is used according to traffic type and priority, and a dynamic resource allocation algorithm based on the maximum-minimum-fairness (MMF) algorithm is used to allocate radio resource blocks to each QoS flow. This method specifically comprises the following steps: Step 1: The QoS flow from TSN is transmitted to the 5G system through the TSN converter. The relevant QoS information from the TSN traffic is obtained according to the TSN AF function of 5G, and converted to the QoS framework of the 5G system by PCF, and a 5QI mapping table is generated to obtain the traffic QoS information under the 5G QoS framework, and generate the forwarding control rules of the QoS flow; Step 2: The 5G base station identifies the traffic type and criticality of the QoS flow according to the mapped 5QI mapping table and forwarding control rules, and uses a semi-continuous scheduling mechanism that mixes static and dynamic resource allocation to allocate resources. For delay-sensitive deterministic periodic traffic, a certain number of resource blocks are statically allocated in advance for data transmission to ensure low latency in its scheduling. Other types of traffic obtain wireless resource blocks through dynamic resource allocation. Step 3: Calculate whether the continuously pre-allocated resource blocks meet the resource requirements of the continuously statically allocated resource traffic. If so, put this part of the QoS flow set that does not request dynamic resource allocation into ; If not, the QoS flow needs to request dynamic resources from the base station. The set of QoS flows requesting dynamic resource allocation is , and calculate the required dynamic resource blocks; Step 4: For the collection For the QoS flows that need to dynamically allocate resources, the resource allocation weight of each QoS flow is obtained according to the traffic priority function based on the maximum-minimum fair allocation algorithm, and then the remaining resource blocks are dynamically allocated to each QoS flow according to the weight; if the allocated resource blocks meet the resource requirements of the traffic, the resource blocks that exceed the traffic resource requirements are recycled and removed from the set If the allocated resource block does not meet the resource requirement, proceed to step five; Step 5: Calculate the number of resource blocks remaining for dynamic allocation after the previous allocation, and update the set of dynamic resources that need to be requested , return to step 4; until all QoS flows are allocated resource blocks that meet resource requirements or the remaining allocatable resource blocks are exhausted, the dynamic resource allocation ends.

Images