WO2025001747A1 - Communication network-based service scheduling method, apparatus, device and storage medium - Google Patents

Abstract

Disclosed in the present invention are a communication network-based service scheduling method, an apparatus, a device and a storage medium. A communication network comprises a service scheduling apparatus and at least one network device, the service scheduling apparatus executing the service scheduling method to schedule service data going to or coming from the network device via the communication network. The method comprises: in response to receiving service data, allocating the service data to a corresponding service queue according to a predetermined traffic allocation policy, and, according to a network device to which the service data belongs or the service type of the service data, allocating the service data to corresponding scheduling queues among a plurality of scheduling queues; and, according to the scheduling priorities of the plurality of scheduling queues, sending the service data in the plurality of scheduling queues.

Description

Communication network-based service scheduling method, device, equipment and storage medium

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to Chinese Patent Application No. 202310804968.9 filed on June 30, 2023, the contents of which are incorporated herein by reference in their entirety.

Technical Field

The present application relates to the field of communications, and in particular to a communication network-based service scheduling method, device, equipment, and computer-readable storage medium.

Background Art

In recent years, the rapid development of computers and the Internet has led to the emergence and rapid spread of real-time services. Applications such as virtual reality, cloud gaming, and online medical care have put forward very strict requirements on communication latency (1-10 milliseconds) and reliability. In addition, the access of a large number of multimedia devices in today's home and office environments makes it very important to support real-time services in communication networks. However, achieving the level of latency and reliability required by real-time services is a challenging problem. For example, there are two mainstream directions for solving this problem in WiFi networks: 1) Use a restricted timed wake-up mechanism (Restricted Target Wake Time, r-TWT) or similar channel reservation technology. This technology achieves the demand indicators of real-time services by monopolizing the use of channels for a period of time, but due to the relatively complex information interaction and scheduling between nodes, the implementation effect of this technology is not satisfactory. 2) Use enhanced distributed channel access (EDCA) multi-access category (AC) priority technology. This technology adjusts the EDCA parameters of different ACs to achieve low-latency channel access for real-time services. However, since both the access point (AP) and the station (STA) use multiple access technology based on carrier sensing, even if the current node has a high priority, the real-time service must wait for the current transmission to end when it needs to be transmitted. Although the priority division based on different ACs can alleviate this problem to a certain extent, the AC priority with a larger granularity is still insufficient for the high standards of real-time services. In addition, traditional WLAN scheduling technology does not distinguish between real-time services and ordinary services. Real-time services and ordinary services may be mixed in the same AC queue. Real-time services need to wait for the transmission of ordinary services with higher priority (dynamic priority) before they can be transmitted. This is almost intolerable for real-time services.

Large-granular AC priority division not only deteriorates the performance of real-time services, but also affects traditional audio and video services. The Quality of Service (QoS) requirements of different video, audio, or conference traffic are different. Audio and video services with different QoS requirements may be roughly placed in the same AC queue. Services in the same AC queue are provided with the same QoS, so the QoS of some services may not be met, resulting in a poor user experience.

Under the framework of EDCA-based Media Access Control (MAC), one direction to solve the current problem is to consider the priorities of different services in the scheduling algorithm. Real-time services and audio and video services in the same AC will be given different scheduling priorities, and higher-priority services will receive more transmission parties than lower-priority services, thereby ensuring the QoS of such services. However, frequent scheduling of high-priority services will also lead to fairness considerations, and lower-priority services may experience "starvation".

Therefore, the communication network in the related art has the problems of poor QoS support for real-time services, inability to distinguish QoS requirements for services in the same queue, and inability to ensure scheduling fairness.

Summary of the invention

An embodiment of the present application provides a service scheduling method based on a communication network, wherein the communication network includes a service scheduling device and at least one network device, wherein the service scheduling device executes the service scheduling method to schedule service data going to or coming from the network device via the communication network, and the method includes: in response to receiving service data, allocating the service data to corresponding service queues according to a predetermined traffic distribution strategy, wherein each service queue includes multiple scheduling queues with different scheduling priorities, each service queue has a corresponding device priority table, and the device priority table records the dynamic correspondence between the network device and the multiple scheduling queues; allocating the service data to corresponding scheduling queues in the multiple scheduling queues according to the network device to which the service data belongs or the service type of the service data; and sending the service data in the multiple scheduling queues according to the scheduling priorities of the multiple scheduling queues, wherein the service data in the scheduling queues with higher scheduling priorities are sent first.

An embodiment of the present application also provides a service scheduling device based on a communication network, wherein the communication network includes a service scheduling device and at least one network device, wherein the service scheduling device executes the service scheduling method to schedule service data going to or coming from the network device via the communication network, and the device includes: an allocation module, which, in response to receiving service data, allocates the service data to corresponding service queues according to a predetermined traffic allocation strategy, wherein each service queue includes multiple scheduling queues with different scheduling priorities, and each service queue has a corresponding device priority table, and the device priority table records the dynamic correspondence between the network device and the multiple scheduling queues; the allocation module is also used to allocate the service data to corresponding scheduling queues in the multiple scheduling queues according to the network device to which the service data belongs or the service type of the service data; a sending module is used to send the service data in the multiple scheduling queues according to the scheduling priorities of the multiple scheduling queues, wherein the service data in the scheduling queues with higher scheduling priorities are sent first.

An embodiment of the present application also proposes a service scheduling device based on a communication network, the device comprising a memory, a processor, a program stored on the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein the program implements the aforementioned method when executed by the processor.

An embodiment of the present application also provides a computer-readable storage medium, which stores one or more programs. The one or more programs can be executed by one or more processors to implement the aforementioned service scheduling method.

BRIEF DESCRIPTION OF THE DRAWINGS

By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easy to understand. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

FIG1 is a schematic diagram of a deployment of a communication network according to an embodiment of the present application;

FIG2 is a structural block diagram of an AP according to an embodiment of the present application;

FIG3 is a flow chart of a service scheduling method based on a communication network according to an embodiment of the present application;

4 is a schematic diagram of forcibly reducing service data of a network device with the highest priority to other scheduling queues according to an embodiment of the present application;

FIG5 is a schematic diagram of a service scheduling method based on a communication network according to an embodiment of the present application;

FIG6 is a schematic diagram of data transmission of multi-user OFDMA scheduling according to an embodiment of the present application;

FIG7 is a schematic diagram of data transmission of multi-user MUMIMO scheduling according to an embodiment of the present application;

FIG8 is a schematic diagram of a service scheduling device based on a communication network according to an embodiment of the present application.

DETAILED DESCRIPTION

The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

It should be understood that the terms "first", "second", "third", and "fourth" in the claims, specification, and drawings of the present application are used to distinguish different objects rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements, and/or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and/or their collections.

In the subsequent description, the suffixes such as "module", "component" or "unit" used to represent elements are only used to facilitate the description of the present invention, and have no special meanings. Therefore, "module", "component" or "unit" can be used in a mixed manner.

It should also be understood that the terms used in this application specification are only for the purpose of describing specific embodiments. It is not intended to limit the present application. As used in the present specification and claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should be further understood that the term "and/or" used in the present specification and claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations.

As used in this specification and claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context.

The specific implementation of the present application is described in detail below with reference to the accompanying drawings.

In this document, the terms "service", "traffic", "data", and "service data" may all be used to refer to services transmitted via a communication network.

In view of the problems in the related art that the communication network has poor QoS support for real-time services, the services in the same queue cannot distinguish QoS requirements, and the scheduling fairness cannot be guaranteed, the embodiment of the present application proposes a service scheduling method based on the communication network, which aims to ensure the needs of services of different priorities and can solve the scheduling needs of different types of services under the same service queue.

FIG. 1 is a schematic diagram of an example of an application environment applicable to the service scheduling method and service scheduling device according to an embodiment of the present application. The service scheduling method of the embodiment of the present application can be executed by a service scheduling device with service scheduling requirements in any communication network, and is used to schedule multiple services to or from a network device via the communication network. The service scheduling device can be included in an appropriate network device in the communication network. In FIG. 1 and the following embodiments, the application environment in the present application is described by taking a wireless local area network (WLAN) as an example. It can be understood that the service scheduling method, device, equipment and computer-readable storage medium in the present application are not limited to the application environment of this WLAN, and any communication network with scheduling requirements and multiple services is applicable to the present invention.

As shown in FIG1 , the wireless LAN includes multiple basic service sets (BBS), each of which includes an access point (AP), which can be associated with multiple stations (STA) and transmit traffic data to each other. For the sake of explanation, error! Reference source not found. Only two APs are deployed in the example, AP1 in basic service set 1 (BSS1) and AP2 in BSS2. AP1 has at least two associated STAs (STA1 and STA2) and works in the 5GHz and 6GHz frequency bands at the same time, and AP2 has at least three associated STAs (STA3, STA4 and STA5) and works alone in 5GHz. Due to the same frequency band range or communication range, the coverage area of AP1 overlaps with the coverage area of AP2, and STA1 is located in the overlapping area of AP1 and AP2. When traffic data is transmitted between AP and STA, AP or STA obtains the address of the traffic data by parsing the traffic data, and determines the receiver of the traffic data according to the address of the traffic data.

Generally speaking, the AP in Error! Reference source not found. is a fixed terminal that provides network services to STA. However, in some applications, the AP can be a mobile terminal or a non-fixed terminal; the STA can be a fixed, non-fixed or mobile terminal. STA Examples include, but are not limited to, smart phones, personal computers, multimedia devices, audio devices, devices for the Internet of Things (IOT), or other wireless terminal devices that require AP to provide network services. STA can also be called a mobile unit, a wireless unit, a mobile terminal, a user equipment (UE), or some other appropriate terms. STA can be implemented using a protocol stack. The protocol stack includes a physical layer for sending and receiving data according to the physical and electrical specifications of the wireless channel, a data link layer for wireless channel access, a network layer for processing data transmission from the source address to the destination address, a transport layer for data transmission management, and other layers necessary for establishing a network and connection.

Although described using WLAN deployment and IEEE 802.11 networks as examples, the techniques described in the present invention can be extended to other networks that adopt various standards or protocols. For example, including Bluetooth, Hiper LAN (a set of wireless standards that can be compared with the IEEE 802.11 standard and are mainly used in Europe), and in a wide area network (WAN), WLAN, personal area network (PAN) or other suitable networks now known or later developed. Therefore, the scheduling techniques presented in the present invention can be applied to any suitable wireless network or other network, regardless of the communication protocol used.

FIG2 is an example structural block diagram of an AP as an execution subject of a service scheduling device or service scheduling method according to an embodiment of the present application. It should be noted that the AP in the embodiment of the present application is only an example. For different communication networks, the structure of the AP may be different, and the present application does not impose any restrictions on this. As shown in FIG2, the AP 100 includes a memory 110, at least one processor 140, and a network interface module 130. The at least one processor 140 can communicate internally through a bus, and the bus is a communication system for transmitting data between multiple components or subcomponents of the AP 100. One or more processors 140 may include one or a combination of a baseband processor, a digital signal processor, or a transceiver processor. In the present application, the processor 140 includes a scheduling component 150 for executing the scheduling technology based on the communication network described in the present application, and the scheduling component 150 includes hardware, software, and can be configured to execute code or execution instructions stored in the memory 110 (e.g., a computer-readable storage medium). The scheduling component 150 can maximize the network capacity and ensure latency, throughput, and fairness for sending STA data. The memory 110 contains cached data of multiple STAs and various related characteristics of the data stored therein, such as certain network status or management information, real-time voice or video streaming data, etc. The data can be classified as low-latency, emergency, real-time or normal business according to these characteristics, which affects when and how to transmit the data. The network interface module 130 includes a WLAN 802.11 interface, and in other networks, it may include a Bluetooth interface, a WiMax interface or a wired Ethernet interface.

FIG3 is a flow chart of a service scheduling method based on a communication network according to an embodiment of the present application, and the scheduling method can be performed by a service scheduling device in the communication network. The service scheduling device can be, for example, the AP described in FIG1 and FIG2 or included therein, or can be a network device in other communication networks that has service scheduling requirements or included therein, and the service scheduling device is used to schedule services of each network device in the communication network. The scheduling method includes the following steps S301 to S303.

In step S301, in response to receiving service data, the service data is allocated to corresponding service queues according to a predetermined traffic distribution strategy, wherein each service queue includes multiple scheduling queues with different scheduling priorities, and each service queue has a corresponding device priority table, which records the dynamic correspondence between the network device and the multiple scheduling queues.

In step S302, the service data is allocated to a corresponding scheduling queue among the multiple scheduling queues according to the network device to which the service data belongs or the service type of the service data.

In step S303, the service data in the multiple scheduling queues are sent according to the scheduling priorities of the multiple scheduling queues, wherein the service data in the scheduling queues with higher scheduling priorities are sent preferentially.

The embodiment of the present application performs scheduling according to the priority of the service data of the network device. This scheduling method can guarantee the needs of different services and solve the scheduling needs of different types of services under the same service queue, so that queues with higher priority can obtain more scheduling opportunities.

In the embodiments of the present application, the service queue refers to a queue defined according to the communication protocol. For example, in the WLAN protocol, the number of general service queues can be determined according to the number of ACs or the number of TIDs. Assuming that there are 4 ACs in the WLAN protocol, 4 service queues can be set. In some embodiments, the service data is allocated to the corresponding service queue according to a predetermined traffic allocation strategy, and the traffic allocation strategy includes a priority division strategy or a load balancing allocation strategy.

Taking the wireless local area network as an example, during the scheduling process, the scheduling component in the AP first generates at least one service queue, and the traffic data transmitted from the upper layer (such as the network layer) is first stored in the corresponding STA, and then the STA (including the traffic data transmitted from the upper layer) will be assigned to the corresponding service queue according to the priority division strategy or the load balancing distribution strategy. In some embodiments, the priority division strategy may include dividing the priority according to the type of traffic data, for example, the traffic data type may include video data (VO), voice data (VI), best effort data (BE), background data (BK), etc., and may also be divided into priorities according to other methods, and the present application does not impose any restrictions on this.

In addition, it is understandable that the traffic data packets transmitted by the upper layer are mixed with the service data for each network device. Therefore, when the traffic data is allocated to multiple service queues according to the traffic data allocation strategy, each service queue may only include the service data of part of the network devices. It is also understandable that the traffic allocation strategy does not divide the service data absolutely mutually exclusive. Therefore, there is a network device data that belongs to at least two service queues.

The device priority table records the correspondence between network devices and scheduling queues. Each network device corresponds to a scheduling queue, and the service data belonging to the network device is in the corresponding scheduling queue. If any service of a network device has been assigned to this service queue, then the device priority table of this service queue will record the corresponding scheduling queue of this network device. The correspondence between network devices and scheduling queues changes dynamically. For example, the changes mentioned below are based on the queue transmission quota and the device transmission quota. The specific description and related processing steps are detailed below.

In some embodiments, the service data is sent to the network device to which the service data belongs or the service type of the service data. The business data is allocated to the corresponding scheduling queue among the multiple scheduling queues, including: determining the network device to which the business data belongs, and querying the device priority table of the business queue to determine the scheduling queue corresponding to the network device to which the business data belongs; and allocating the business data to the determined scheduling queue.

In some embodiments, if the network device to which the service data belongs is not found in the device priority table, the service type of the service data is determined; if the service type is the first service type, the service data is assigned to the scheduling queue with the highest scheduling priority among the multiple scheduling queues. In some embodiments, the first service type can be a real-time service type or other high-priority service type, which is not limited in this application.

In some embodiments, when the service type is a second service type, the network device to which the service data belongs is determined; and the service data is allocated to a corresponding scheduling queue among the multiple scheduling queues according to the static priority of the network device to which the service data belongs. Specifically, the second service type may be a normal service type. When the service type is a normal service type, the corresponding scheduling queue is determined according to the static priority of the network device corresponding to the service data.

In some embodiments, allocating the service data to a corresponding scheduling queue in the multiple scheduling queues according to the static priority of the network device to which the service data belongs includes: determining the corresponding scheduling priority according to the static priority of the network device to which the service data belongs; allocating the service data to the scheduling queue with the corresponding scheduling priority. Specifically, the service types of the network device include: a first service type such as a real-time service type, and a second service type such as a non-real-time service type. The static priority of the network device can be set according to a certain service requirement of a specific implementation. Service types may include game services, video services, voice services, and the like. Different service types need to guarantee different performance, and the static priority of the STA is determined according to the performance differentiation. For example, assuming that the STA is a game service, the game service requires a relatively low latency, so the latency will be prioritized. Determine the corresponding scheduling priority according to different service types.

In some embodiments, after the service data is assigned to the corresponding scheduling queue, the scheduling method further includes: updating the device priority table of the service queue to record the corresponding relationship between the network device to which the service data belongs and the corresponding scheduling queue in the device priority table. Specifically, the device priority table of the service queue first has an initialization value, and then is updated to the device priority table according to the corresponding relationship between the network device to which the service data belongs and the corresponding scheduling queue.

In some embodiments, each scheduling queue has a queue transmission quota, and each network device has a device transmission quota corresponding to each scheduling queue, wherein the queue transmission quota has a preset initial value and decreases accordingly with the transmission of service data in the scheduling queue, and the device transmission quota has a preset initial value and decreases accordingly with the transmission of service data of the network device, wherein after sending the service data, the scheduling method further includes: according to The sent business data updates the queue transmission quota of the scheduling queue; when the queue transmission quota is less than or equal to a first predetermined threshold, all business data in the scheduling queue are transferred to other scheduling queues in the multiple scheduling queues whose scheduling priority is lower than that of the scheduling queue.

In some embodiments, the queue transmission configuration of each scheduling queue is a common quota for all stations in the scheduling queue, while the device transmission quota of each network device in each scheduling queue belongs to its own transmission quota.

In some embodiments, the queue transmission quota of each scheduling queue is greater than the network transmission quota of a network device.

In some embodiments, the transmission quota of each scheduling queue can be determined according to the number of network devices in the scheduling queue and/or the performance requirements of the network devices in the scheduling queue corresponding to the business type requirements, and can also be determined according to empirical values. Similarly, the device transmission quota of each network device itself can also be determined according to the performance requirements of the network device corresponding to the business type requirements, or determined by experimental simulation or empirical values. The setting of the queue transmission quota of the scheduling queue and the device transmission quota of the network devices in the scheduling queue of the present application is not restricted. In the process of the network device transmitting business data, the queue transmission quota of the scheduling queue and the device transmission quota of each network device in the scheduling queue are dynamically changed. Each time the business data is sent, the queue transmission quota of the scheduling queue and the device transmission quota of the network device itself are consumed. The queue transmission quota of the scheduling queue is updated according to the sent business data. When the queue transmission quota is less than or equal to the first predetermined threshold, the business data in the scheduling queue will be forcibly transferred to a lower scheduling queue. The present application determines the priority of each network device in the scheduling queue according to the queue transmission quota of the scheduling queue and the device transmission quota of the network devices in the scheduling queue, and schedules the business data of the network devices in the scheduling queue according to the priority of each network device in the scheduling queue. It is understandable that, since the queue transmission quota of the scheduling queue and the device transmission quota of each network device in the scheduling queue change dynamically, the priority of each network device in the scheduling queue also changes dynamically.

Since the priority of a higher priority scheduling queue is higher than that of a lower priority scheduling queue, when there is a network device in the higher priority queue, the service scheduling device will always give priority to scheduling the network device in the higher priority scheduling queue; however, the network device in the higher priority scheduling queue cannot always exist, otherwise it will cause starvation to the network devices in the lower scheduling queue. Therefore, in some embodiments, when the device transmission capacity of the network device in the scheduling queue is less than or equal to the first predetermined threshold, the service data of the network device in the higher priority scheduling queue will be forcibly transferred to other scheduling queues in multiple scheduling queues with a scheduling priority lower than the scheduling queue, so as to ensure the fairness of scheduling.

In some embodiments, the service scheduling method further includes updating a queue transmission quota of the scheduling queue according to a transmission mode of the sent service data.

In some embodiments, the transmission mode includes single-user transmission and multi-user transmission, and multi-user transmission is further divided into orthogonal frequency division multiple access (OFDMA) transmission and multi-user MIMO (MUMIMO) transmission. Among them, single-user transmission refers to the one-to-one transmission between AP and STA. transmission, while multi-user transmission means that the communication network can transmit to multiple network devices at the same time. OFDMA means that a group of users can access the channel at the same time. OFDMA technology specifies that each user uses one (or a group) of all OFDM subcarriers. OFDMA divides the entire frequency band into smaller units, and multiple users can use the entire band at the same time, and its allocation mechanism is very flexible. It can dynamically allocate the number of subcarriers according to the size of the user's business volume, and the modulation mode and transmission power used on different subcarriers can also be different. MUMIMO means that in a wireless communication system, a base station serves multiple mobile terminals at the same time, and the base stations make full use of the antenna's airspace resources to communicate with multiple users at the same time.

In some embodiments, when the transmission is scheduled as a single-user transmission, the method for updating the queue transmission quota of the scheduling queue is: the transmission quota of the current scheduling queue minus the queue transmission quota used by the current scheduling queue for the current transmission.

In some embodiments, when the transmission mode of the transmission scheduling is OFDMA, the updated queue transmission quota is:

Slice _{STA i} =Slice _{STA i} −Slice _Use *BW ⁱ /BW _Total , wherein Slice _Use is the total transmission quota usage of this multi-user transmission, BW _i is the resource block size of the current scheduling queue, and BW _Total is the total resource block size of the multi-user transmission.

In some embodiments, when the transmission mode of the transmission scheduling is MUMIMO, the queue transmission quota of the updated scheduling queue is: Slice _{STA i} = Slice _{STA i} - Slice _Use * NSS _i / NSS _Total , where Slice _Use is the total transmission quota usage of this multi-user transmission, NSS _i is the number of spatial streams of the current scheduling queue, and NSS _Total is the total number of spatial streams of multi-user transmission.

In some embodiments, when the queue transmission quota is greater than a first predetermined threshold, the device transmission quota of the network device to which the service data belongs is updated according to the sent service data; when the device transmission quota is less than a second predetermined threshold, for the network device in the scheduling queue with the highest priority, all service data in the network device is transferred back to the scheduling queue with the highest priority; for other scheduling queues except the highest priority scheduling queue, all service data of the network device is transferred from the current scheduling queue to other scheduling queues among the multiple scheduling queues whose scheduling priority is lower than the current scheduling queue.

When the device transmission quota of a network device in a scheduling queue with the highest priority meets certain conditions, it will generally be added to the scheduling queue with the highest priority again. However, since the priority of the scheduling queue with the highest priority is higher than that of other scheduling queues, when there are network devices in the highest priority queue, the service scheduling device will always give priority to scheduling the network devices in the scheduling queue with the highest priority. Therefore, the network devices in the scheduling queue with the highest priority cannot always exist, otherwise it will cause starvation to the network devices in other scheduling queues. The embodiment of the present application determines whether the service data of a network device is transferred to a scheduling queue of other scheduling priorities based on the queue transmission quota of the scheduling queue. Specifically, when the queue transmission quota of the scheduling queue with the highest priority is less than or equal to the first threshold, All the service data in the scheduling queue will be transferred to other scheduling queues in the multiple scheduling queues whose scheduling priority is lower than that of the scheduling queue. When the queue transmission quota of the highest scheduling queue is greater than the first predetermined threshold, and the device priority of the network device in the scheduling queue with the highest priority is less than the second predetermined threshold, all the service data in the network device will be transferred back to the scheduling queue with the highest priority. For other scheduling queues that are not the highest priority, similarly, when the queue transmission quota is less than or equal to the first predetermined threshold, all the service data in the scheduling queue will be transferred to other scheduling queues in the multiple scheduling queues whose scheduling priority is lower than that of the scheduling queue. When the device transmission quota of the network device in other scheduling queues is less than the second predetermined threshold, all the service data of the network device will be transferred from the current scheduling queue to other scheduling queues in the multiple scheduling queues whose scheduling priority is lower than that of the current scheduling queue.

It should be noted that although the business data of the network device is moved into a scheduling queue that is one level lower than the business data of any network device, the static priority of the business data of any network device will not change. What changes is only the dynamic priority of the business data of the network device determined according to the device transmission quota.

In some embodiments, the second predetermined threshold mentioned above refers to the remaining transmission quota of the business data of any network device being less than the second predetermined threshold or the transmission quota used by the business data of any network device exceeding the transmission quota of the business data of any network device. That is, when the business data of any network device uses up its own transmission quota, but in most cases, the transmission quota of the business data of any network device is not exactly used up, and there may be a small amount of remaining quota of the business data of any network device, but it is not enough for another transmission of business data, or the actual transmission quota used by the business data of any network device exceeds the transmission quota of the business data of any network device itself.

In some embodiments, the second threshold is determined by a minimum transmission quota used for one transmission during a service data scheduling process of the network device.

In some embodiments, after all service data of the network device is transferred from the current scheduling queue to other scheduling queues among the multiple scheduling queues with a scheduling priority lower than the current scheduling queue, the device transmission quota of the network device corresponding to the device transferred to the other scheduling queue is:

Slice _cur = Slice+Factor*Slice _last , wherein Slice is the transmission quota corresponding to the other scheduling queue reallocated to the network device, Slice _last is the device transmission quota of the network device corresponding to the current scheduling queue, and Factor is a scaling factor.

Specifically, when the device transmission quota of the service data of the network device is negative, it means that the device transmission quota of the service data of the current network device exceeds its own transmission quota, and the Slice _last is negative. The service data of any network device is dropped from a high priority to a low priority queue. If the device transmission quota of the service data of any network device exceeds its own transmission quota, it is unfair to the service data of the low priority network device. Therefore, the excess will be penalized at the next level. Therefore, the scaling factor Factor is a value greater than 1, generally 1.1, or other values greater than 1.

In some embodiments, after the service data of the network device is transferred to other scheduling queues, the device priority table is updated, and the corresponding scheduling queue of the network device to which the transferred service data belongs is modified to the other scheduling queue to which the data is transferred. That is, the device priority table maintains the priority of the service data of a dynamic network device.

In some embodiments, the service queue also has an invalidation queue, and the service data in the invalidation queue is not sent. When the service data of all network devices in the service queue are transferred to the invalidation queue, the service data of the network devices in the invalidation queue are transferred back to the multiple scheduling queues.

Specifically, the invalidation queue is the lowest level queue in the service queue. When the service data of all network devices in the service queue drops to the invalidation queue, the service data of the network devices in the invalidation queue is re-added to the multiple scheduling queues. With the transmission of service data of multiple network devices, the device transmission quota of the service data of each network device is constantly changing, and then the priority of the service data of each network device is constantly changing. Specifically, the priority of the service data of each network device is constantly decreasing until it drops to the last level of priority, that is, it drops to the invalidation queue. When the service data of all network devices in the service queue drops to the invalidation queue, the service scheduling device will re-add the service data of all network devices in the invalidation queue to the scheduling queue in the service queue and re-schedule.

In some embodiments, the scheduling method also includes business data of newly added network devices and business data of rejoined network devices, wherein the business data of newly added network devices refers to a new network device connected to the communication network, such as a new mobile phone connected to a WIFI signal, etc. The rejoined network device refers to a network device in sleep mode being awakened, such as a mobile phone being awakened by reopening the software in the mobile phone after it is in sleep mode.

In some embodiments, when the business data of a network device in an invalid queue is re-added to multiple scheduling queues, the business data of a network device in an expired queue is re-added to the multiple scheduling queues based on the business type corresponding to the business data of the network device or the static priority of the business data of the network device.

Specifically, when the service data of the network device in the invalid queue corresponds to the service data of the scheduling queue with the highest priority, such as the service data of the real-time service type, the service data of the network device is re-added to the scheduling queue with the highest priority; when the service data of the network device in the invalid queue corresponds to the non-real-time service type (i.e., the ordinary service type), the service data of the network device is re-added to other scheduling queues according to the static priority of the service data of the network device.

In some embodiments, the scheduling method includes when the service data of a new network device is added to the plurality of scheduling When the service data of the queue or the sleeping network device is re-added to the multiple scheduling queues, the scheduling priority of the scheduling queue selected by the service data of the new network device or the service data of the sleeping network device _is Qin = min(Q _max , Q _out ), Q _out is the scheduling priority of the scheduling queue where the service data of the sleeping network device is located when the service data of the sleeping network device exits the scheduling queue, and Q _max is the highest scheduling priority of the scheduling queue in the device priority table of the service queue when the service data of the new network device is added.

In some embodiments, when the service data of a network device is newly added or the service data of a sleeping network device rejoins the scheduling queue, if it is directly added to the highest priority queue or the queue level when it was previously exited (for example, the level when the service data of the network device exits the scheduling queue after sleeping), the service data of the low-priority network device may experience a long period of starvation (the service data of the high-priority network device needs to reduce the priority level one level at a time), so in the embodiment of the present application, a highest priority level will be maintained, which is equal to the highest priority Q _max of the service data of all network devices in the scheduling queue when the service data of the network device is added. When the service data of the network device joins or rejoins the scheduling queue, the selected queue priority cannot be higher than Q _max , that is, _Qin = min (Q _max , Q _out ), and Q _out is the queue priority of the service data of the sleeping network device exiting the scheduling queue.

4 is a schematic diagram of forcibly reducing the service data of the network device in the highest priority to other scheduling queues according to an embodiment of the present application. The embodiment of the present application takes the service data of the service type in the highest priority as the first scheduling queue, the other scheduling queues as the second scheduling queue, and the network device as the site as an example for explanation. The first scheduling queue RealTime includes STA1 and STA2. After multiple traffic data transmissions, when the transmission quotas of STA1 and STA2 meet the preset conditions, STA1 and STA2 are forced to drop to the second scheduling queue (including Q4, Q3) Q4.

In some embodiments, the scheduling queue with the highest scheduling priority can be a real-time service queue, and the other scheduling queues are ordinary service queues. It is understandable that the scheduling queue with the highest scheduling priority and other scheduling queues can also be scheduling queues of other different priorities according to scheduling requirements, and this application does not limit this.

In some embodiments, other scheduling queues include multiple levels of second sub-queues, each second sub-queue has a different priority, and the business data of the network devices in the scheduling queue are gradient scheduled based on the transmission quota of the scheduling queue and the device transmission quota of the business data of each network device in the scheduling queue, including: when the transmission quota of the business data of any network device in the second sub-queue meets a preset condition, the business data of any network device is reduced to a scheduling queue that is one level lower than the business data of any network device, wherein the scheduling queue that is one level lower than the business data of any network device is a second sub-queue that is one level lower than the business data of any network device.

Specifically, the other scheduling queues include multiple second sub-queues of different priorities, and the priorities of the multiple second sub-queues decrease from top to bottom. The number of second sub-queues in the second scheduling queue can be set to 3, 4, 5 or more as required. Different from the scheduling queue with the highest priority, as long as the transmission quota of any station in the second sub-queue of the other scheduling queues meets the preset conditions, the service data of any network device will be reduced to a level lower than that of any station. It should be noted that although the service data of any network device is moved into a scheduling queue one level lower than the service data of any network device, the static priority of the service data of any network device will not change, and what changes is only the dynamic priority of the service data of any network device determined according to the transmission quota.

In some embodiments, after the service data of any network device is reduced to a scheduling queue one level lower than the service data of any network device, the scheduling method further includes: reallocating the transmission quota of the service data of any network device. Specifically, when the transmission quota of the service data of any network device is relatively small, that is, when the preset conditions mentioned above are met, the service data of any network device will be moved to a scheduling queue one level lower than the service data of the network device, and a certain transmission quota will be reallocated to the service data of the network device, so that the service data of the low-priority network device can be scheduled within a predicted time period in the future, thereby improving the fairness of scheduling.

In some embodiments, the scheduling method further includes: selecting a queue transmission quota and a device transmission quota based on a scheduling target, wherein the queue transmission quota and the device transmission quota are transmission time or transmission data volume. Specifically, based on the scheduling target, the queue transmission quota and the device transmission quota can be selected as transmission time or transmission data volume (bit). When the transmission time is selected as the quota, the quota can be configured as 10ms, which can achieve the space-time fairness target; when the transmission data volume is selected as the quota, the bandwidth fairness target can be achieved. The present application can also select other transmission quotas according to other scheduling targets.

In some embodiments, the service data priorities of network devices in the same scheduling queue in multiple scheduling queues of each service queue are the same, and the service data of network devices in the same scheduling queue are scheduled by polling or first-in-first-out. In an embodiment of the present application, the service data of multiple network devices are respectively mapped to different scheduling queues according to the priorities of the service data of the network devices, and the service data priorities of network devices in the same scheduling queue are the same.

In some embodiments, when there is service data STAi of a network device whose priority in at least two service queues is higher than a third threshold at the same time, the service data STAi of the network device is scheduled in one of the service queues TIDm, and the device transmission quota used is Slice _STAi,TIDm , then the device transmission quota of the service data STAi of the network device in the other service queue TIDn of the at least two service queues is: Slice _STAi,TIDn = Slice _STAi,TIDn -Factor _TiDn,m *Slice _STAi,TIDm , where Factor _TiDn,m is a scaling factor between at least two service queues.

Specifically, since different service queues may contain service data of the same destination network device, if the service data of a certain network device is at a higher priority in multiple service queues at the same time, the service data of the same destination network device may be scheduled multiple times in a short time interval, while the service data of other network devices with lower priority will be starved for a longer period of time. Therefore, a solution in the embodiment of the present application is to impose a quota penalty on the service data of network devices in other TID queues when the service data of a network device in a certain TID queue is scheduled. For example, when STA1 in queue TID5 is scheduled and Slice _{STA1 is used, the device in queue TID5} transmits When the transmission quota is increased, the transmission quota of the device with the same STA in other TID queues will be deducted as Slice _STA1,TIDn = Slice _{STA1, TIDn} - Factor _TIDn,5 * Slice STA1,TID5. Factor _TIDn,5 is the scaling factor between TID n and TID5. When the priority of TID n is greater than TID5, Factor _TIDn,5 is generally a value less than 1 and can be set to 0.2. When the priority of TID n is less than TID5, Factor _TIDn,5 is generally greater than or equal to the obtained value and can be set to 1.

The service scheduling method based on the communication network provided by the present application can avoid the complex dynamic priority calculation of the traditional scheduling technology. At the same time, in the process of traffic data transmission, the dynamic priority strategy of the service data of each network device is determined according to the dynamic transmission quota of the service data of each network device. Each priority corresponds to a scheduling queue. The priority of the service data of the network device in each queue will decrease with the decrease of the transmission quota usage, and the service data of the network device is redistributed through the service data transmission quota of the network device, the penalty of the device transmission quota, etc., and the service data priority of the desired network device is improved. For typical scenarios such as multi-user transmission in network transmission, the service data of the network device is newly added to the network or sleep, corresponding strategy optimization is also made, and the embodiment of the present application also takes into account the "hunger" phenomenon in certain scenarios, and further improves the fairness of scheduling.

Furthermore, the service scheduling method described in the embodiment of the present application can meet the performance requirements of real-time services. By setting a higher priority real-time service queue in the scheduling method, the real-time service can obtain a higher priority and more transmission opportunities than the ordinary service, and within a certain transmission time or data volume, the priority of the real-time service will not change.

Furthermore, after adopting the service scheduling method and device described in the embodiment of the present application, the transmission requirements of the customized priority services of different users can be met. Each service has a static priority label, and different users can have different static priorities for service devices. In the service scheduling method according to the embodiment of the present application, the static priority will be mapped to different dynamic service queues. The larger the static priority, the larger the dynamic priority corresponding to. The service with a larger dynamic priority will obtain more transmission opportunities to meet the throughput or latency requirements of the service.

Compared with related technologies, the optimization method involved in the present invention can meet the scheduling requirements of different business types, has better scheduling fairness and is less difficult to implement.

FIG5 is a schematic diagram of a communication network-based service scheduling method according to an embodiment of the present application.

As shown in Figure 5, the scheduling queue with the highest priority in a service queue is the real-time queue RealTime, and the other scheduling queues are ordinary queues, where other scheduling queues include Q4, Q3, Q2, Q1, Q0, and the invalid queue Expire, and the network device is STA as an example for explanation. At time T1, all STAs are assigned to the first scheduling queue RealTime or the Q queue of the second scheduling queue according to the size of the static priority. The static priority of the STA is allocated according to the service type. The STA of the real-time service type is placed in the RealTime queue, and the other STAs are placed in the highest service queue in the Q queue (the top queue Q4, the STA static priority is the same), such as STA1, STA2 and STA3. The STAs in the embodiments of the present application are assumed to be real-time services. After a period of traffic data transmission, at time T2, STA3, STA2, and STA1 are assigned to the real-time service queue according to their respective The remaining transmission quotas are eventually dropped to the ordinary queue Q4. After a period of traffic data transmission, at time T3, STA3 is in the Q2 queue. This is because in the time period [T2, T3], STA3 has used up the quotas in its Q4 and Q3 queues and has been moved down to the Q2 queue; STA1 has used up the quota of its Q4 queue and moved down to the Q3 queue, and STA2 has not used up the quota of the Q4 queue, so it remains in the Q4 queue. Since the service scheduling method in the embodiment of the present application will give priority to scheduling STAs in queues with higher priorities, if scheduling is performed at time T3, STA2 will obtain the highest priority. At time T4, STA3 and STA4 are moved from the scheduling queue to the Expire queue because they have used up the quota of each layer of Q queues in succession. The STA in the Expire queue will not be scheduled, so if scheduling is performed at time T4, STA1 will obtain the highest priority. The process of STA changing the service queue is like a staircase. Each time a STA uses up its quota, it goes down one staircase, and so on, until all STAs are dropped into the invalid queue. All STAs in the invalid queue are added back to the scheduling queue for a new round of scheduling.

The AP uses OFDMA and MUMIMO multi-user transmission modes respectively. Space-time is used as a quota indicator, and the quota for each reallocation can be set to 10ms. Figure 6 is a schematic diagram of data transmission of a multi-user OFDMA scheduling according to an embodiment of the present application. At time T1, the scheduling of the TID7 queue is as follows: STA1 and STA4 are in the Q4 queue, and STA2 is in the Q3 queue, and their quotas are 1ms, 10ms, and 5ms, respectively. Then, the AP obtains a multi-user group through the STA grouping component, which consists of STA1, STA2, and STA4, and uses the OFDMA mode to send. The resource allocation component obtains the RU allocation of each STA as 484Tone, 484Tone, and 242Tone, respectively; after this transmission is completed, the scheduling of the TID queue needs to be updated. At time T2, the scheduling of the TID7 queue is as follows: STA4 is in the Q4 queue, STA1 and STA2 are in the Q3 queue, and their quotas are 9ms, 8ms, and 4ms, respectively. Taking the quota calculation of STA1 as an example, the quota consumed by STA1 in this transmission is 5*484Tone/(484Tone+484Tone+242Tone)=2ms, and the remaining quota of STA1 before sending is 1ms, so the quota of STA1 is updated to: 10+1*(1-2)=9ms, and the scaling factor here is selected as 1. Since STA1 has consumed the quota of the Q4 queue, it is moved down to the Q3 queue, and the calculation of other STAs is similar. At T3, the scheduling of the TID6 queue is as follows: STA1 and STA5 are in the Q4 queue, and STA3 is in the Q3 queue. Their quotas are 7ms, 8ms, and 2ms, respectively. Different TID queues may have the same STA, because the same STA may have traffic from different TIDs at the same time.

Figure 7 is a schematic diagram of data transmission of a multi-user MUMIMO scheduling according to an embodiment of the present application. The multi-user group consists of STA1, STA3 and STA5, which are sent using the MUMIMO mode, and the NSS are 2, 1, and 2 respectively. When the transmission process is completed, the TID6 queue scheduling is as follows: STA1 is in the Q4 queue, STA3 and STA5 are in the Q3 queue, and their quotas are 5ms, 7ms, and 10ms respectively. Taking the quota calculation of STA5 as an example, the quota consumed by STA5 in this transmission is 5*2NSS/(2NSS+2NSS+1NSS)=2ms, and the remaining quota of STA1 before sending is 2ms, so the quota of STA1 is updated to: 10+1*(2-2)=10ms, and the scaling factor here is selected as 1. Since STA5 has consumed all The quota of the Q4 queue is therefore moved down to the Q3 queue. The calculation for other STAs is similar.

FIG8 is a schematic diagram of a service scheduling device 800 based on a communication network according to an embodiment of the present application. The service scheduling device 800 can be applied to the communication network deployment of the embodiment corresponding to FIG1 and the AP architecture of the embodiment corresponding to FIG2. It can also be applied to other communication network environments including multiple services. The present application does not limit this. The device 800 includes an allocation module 801 and a scheduling module 802, wherein:

The allocation module 801 is configured to allocate the service data to corresponding service queues according to a predetermined traffic allocation strategy in response to receiving the service data, wherein each service queue includes a plurality of scheduling queues with different scheduling priorities, and each service queue has a corresponding device priority table, and the device priority table records the dynamic correspondence between the network device and the plurality of scheduling queues;

The allocation module 801 is further configured to allocate the service data to a corresponding scheduling queue among the multiple scheduling queues according to the network device to which the service data belongs or the service type of the service data;

The sending module 802 is used to send the service data in the multiple scheduling queues according to the scheduling priorities of the multiple scheduling queues, wherein the service data in the scheduling queues with higher scheduling priorities are sent preferentially.

In some embodiments, the allocation module 801 is also used to determine the network device to which the business data belongs, and query the device priority table of the business queue to determine the scheduling queue corresponding to the network device to which the business data belongs; and allocate the business data to the determined scheduling queue.

In some embodiments, the allocation module 801 is also used to determine the service type of the service data if the network device to which the service data belongs is not found in the device priority table; and to allocate the service data to the scheduling queue with the highest scheduling priority among the multiple scheduling queues if the service type is the first service type.

In some embodiments, the allocation module 801 is also used to determine the network device to which the service data belongs when the service type is the second service type; and allocate the service data to a corresponding scheduling queue among the multiple scheduling queues according to the static priority of the network device to which the service data belongs.

In some embodiments, the allocation module 801 is further used to determine a corresponding scheduling priority according to a static priority of a network device to which the service data belongs; and allocate the service data to a scheduling queue having the corresponding scheduling priority.

In some embodiments, after the service data is assigned to the corresponding scheduling queue, the allocation module 801 is also used to update the device priority table of the service queue to record the correspondence between the network device to which the service data belongs and the corresponding scheduling queue in the device priority table.

In some embodiments, the sending module 802 is used to update the queue transmission quota of the scheduling queue according to the sent business data after sending the business data; when the queue transmission quota is less than or equal to the first predetermined threshold, transfer all business data in the scheduling queue to the scheduling queue with a scheduling priority lower than the scheduling queue. In other dispatch queues of the same scheduling queue.

In some embodiments, the sending module 802 is used to update the device transmission quota of the network device to which the business data belongs according to the business data sent when the queue transmission quota is greater than a first predetermined threshold; when the device transmission quota is less than a second predetermined threshold, for the network device in the scheduling queue with the highest priority, all business data in the network device is transferred back to the scheduling queue with the highest priority; for other scheduling queues except the scheduling queue with the highest priority, all business data of the network device are transferred from the current scheduling queue to other scheduling queues among the multiple scheduling queues whose scheduling priority is lower than the current scheduling queue.

In some embodiments, the sending module 802 is used to update the queue transmission quota of the scheduling queue according to the transmission mode of the sent business data.

In some embodiments, the sending module 802 is specifically configured to, when the transmission mode is orthogonal frequency division multiple access, update the queue transmission quota of the scheduling queue to:

Slice _{STA i} =Slice _{STA i} −Slice _Use *BW _i /BW _Total , wherein Slice _Use is the total transmission quota usage of this multi-user transmission, BW _i is the resource block size of the current scheduling queue, and BW _Total is the total resource block size of the multi-user transmission.

When the transmission mode is multi-user multiple-input multiple-output, the updated queue transmission quota of the scheduling queue is:

Slice _{STA i} =Slice _{STA i} -Slice _Use *NSS _i /NSS _Total , where Slice _Use is the total transmission quota usage of this multi-user transmission, NSS _i is the number of spatial streams in the current scheduling queue, and NSS _Total is the total number of spatial streams in the multi-user transmission.

In some embodiments, the sending module 802 is further configured to transfer all service data of the network device from the current scheduling queue to other scheduling queues in the multiple scheduling queues with a scheduling priority lower than that of the current scheduling queue, and the device transmission quota of the network device corresponding to the other scheduling queues to which the device is transferred is:

Slice _cur = Slice+Factor*Slice _last , wherein Slice is the transmission quota corresponding to other scheduling queues reallocated to the network device, Slice _last is the device transmission quota of the network device corresponding to the current scheduling queue, and Factor is the scaling factor.

In some embodiments, the allocation module 801 is further configured to select the queue transmission quota and the device transmission quota based on a scheduling target, wherein the device transmission quota and the device transmission quota are transmission time or transmission data volume.

In some embodiments, the sending module 802 is further configured to, when the service data of a new network device is added to the multiple scheduling queues or the service data of a sleeping network device is re-added to the multiple scheduling queues, the scheduling queue selected by the service data of the new network device or the service data of the sleeping network device _is Qin = min (Q _max , Q _out ), Q _out is the scheduling queue where the service data of the sleeping network device is located when the service data of the sleeping network device exits the scheduling queue, and Q _max is the highest scheduling queue among the service data of all network devices in the service queue when the service data of the new network device is added.

In some embodiments, the sending module 802 is also used for, when there is service data STAi of a network device whose priority in at least two service queues is higher than a third threshold at the same time, the service data STAi of the network device is scheduled in one of the service queues TIDm, and the device transmission quota used is Slice _STAi,TIDm , then the device transmission quota of the service data STAi of the network device in the other service queues TIDn of the at least two service queues is: Slice _STAi,TIDn = Slice _STAi,TIDn -Factor _TiDn,m *Slice _STAi,TIDm , wherein Factor _TiDn,m is the scaling factor between the at least two service queues.

An embodiment of the present application also provides a service scheduling device based on a communication network, which includes a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory. When the program is executed by the processor, any scheduling method in the above embodiments is realized.

An embodiment of the present application also provides a computer-readable storage medium, which stores one or more programs. When the one or more programs can be executed by one or more processors, a service scheduling method according to any of the above embodiments is implemented.

The service scheduling method, device, equipment, and computer-readable storage medium based on the communication network provided by the present application can avoid the complex dynamic priority calculation of traditional scheduling technology. At the same time, during the flow data transmission process, the dynamic priority strategy of each site is determined according to the dynamic transmission quota of each site. Each priority corresponds to a scheduling queue. The priority of the site in each queue will decrease with the decrease of the transmission quota usage, and the expected site priority is provided through the redistribution of the site transmission quota, the penalty of the transmission quota, etc., and corresponding strategy optimization is also made for typical scenarios such as multi-user transmission in network transmission, new nodes joining the network or sleeping, and the embodiment of the present application also takes into account the "hunger" phenomenon in certain scenarios, and further improves the fairness of scheduling.

Furthermore, the service scheduling method, device, equipment and computer-readable storage medium described in the embodiments of the present application can meet the performance requirements of real-time services. By setting a higher priority real-time service queue in the scheduling method, the real-time service can obtain a higher priority and more transmission opportunities than the ordinary service, and within a certain transmission time or data volume, the priority of the real-time service will not change.

Furthermore, the service scheduling method, device, equipment and computer-readable storage medium described in the embodiments of the present application can meet the transmission requirements of customized priority services of different users. Each service has a static priority label, and different users can have different static priorities for service devices. In the service scheduling method in the embodiments of the present application, the static priority will be mapped to different dynamic service queues. The larger the static priority, the larger the dynamic priority. Services with higher priority will get more transmission opportunities to meet the throughput or latency requirements of the services.

Compared with related technologies, the optimization method, device, equipment, and computer-readable storage medium involved in the present invention can meet the scheduling requirements of different business types, have better scheduling fairness, and are less difficult to implement.

Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules/units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

In hardware implementation, the division between the functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some physical components or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer-readable storage medium (or non-temporary medium) and a communication medium (or temporary medium). As known to those of ordinary skill in the art, the term computer-readable storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules or other data). Computer-readable storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the present invention is not limited thereby. Any modification, equivalent substitution and improvement made by those skilled in the art without departing from the scope and essence of the present invention shall be within the scope of the present invention.

Claims (23)

A service scheduling method based on a communication network, wherein the communication network includes a service scheduling device and at least one network device, the service scheduling device executes the service scheduling method to schedule service data to or from the network device via the communication network, the method comprising: In response to receiving the service data, allocating the service data to the corresponding service queue according to a predetermined traffic allocation strategy, wherein each service queue includes a plurality of scheduling queues with different scheduling priorities, each service queue has a corresponding device priority table, and the device priority table records the dynamic correspondence between the network device and the plurality of scheduling queues; Allocating the service data to a corresponding scheduling queue among the multiple scheduling queues according to the network device to which the service data belongs or the service type of the service data; The service data in the multiple scheduling queues are sent according to the scheduling priorities of the multiple scheduling queues, wherein the service data in the scheduling queues with higher scheduling priorities are sent preferentially. The service scheduling method according to claim 1, wherein the allocating the service data to a corresponding scheduling queue among the multiple scheduling queues according to the network device to which the service data belongs or the service type of the service data comprises: Determine the network device to which the service data belongs, and query the device priority table of the service queue to determine the scheduling queue corresponding to the network device to which the service data belongs; The service data is allocated to the determined scheduling queue. The service scheduling method according to claim 2, further comprising: if the network device to which the service data belongs is not found in the device priority table, determining the service type of the service data; When the service type is the first service type, the service data is allocated to a scheduling queue having the highest scheduling priority among the multiple scheduling queues. The service scheduling method according to claim 3, further comprising: determining the network device to which the service data belongs when the service type is the second service type; The service data is allocated to a corresponding scheduling queue among the multiple scheduling queues according to the static priority of the network device to which the service data belongs. The service scheduling method according to claim 4, wherein allocating the service data to a corresponding scheduling queue among the multiple scheduling queues according to the static priority of the network device to which the service data belongs comprises: Determine the corresponding scheduling priority according to the static priority of the network device to which the service data belongs; The service data is allocated to the scheduling queue having the corresponding scheduling priority. The service scheduling method according to claim 4 or 5, wherein, after allocating the service data to the corresponding scheduling queue, the method further comprises: updating the device priority table of the service queue to record the corresponding relationship between the network device to which the service data belongs and the corresponding scheduling queue in the device priority table middle. The service scheduling method according to claim 1, wherein each scheduling queue has a queue transmission quota, and each network device has a device transmission quota corresponding to each scheduling queue, wherein the queue transmission quota has a preset initial value and decreases accordingly with the transmission of service data in the scheduling queue, and the device transmission quota has a preset initial value and decreases accordingly with the transmission of service data of the network device, wherein after sending the service data, the method further comprises: Updating the queue transmission quota of the scheduling queue according to the sent service data; When the queue transmission quota is less than or equal to a first predetermined threshold, all service data in the scheduling queue are transferred to other scheduling queues in the multiple scheduling queues with a scheduling priority lower than that of the scheduling queue. The service scheduling method according to claim 7, further comprising: When the queue transmission quota is greater than a first predetermined threshold, updating a device transmission quota of a network device to which the service data belongs according to the sent service data; When the device transmission quota is less than a second predetermined threshold, for a network device in a scheduling queue with the highest priority, all service data of the network device is transferred back to the scheduling queue with the highest priority; for other scheduling queues except the scheduling queue with the highest priority, all service data of the network device is transferred from the current scheduling queue to other scheduling queues among the multiple scheduling queues whose scheduling priority is lower than that of the current scheduling queue. The service scheduling method according to claim 7 or 8 further includes: updating a device priority table, in which the corresponding scheduling queue of the network device to which the transferred service data belongs is modified to the other scheduling queue to which it is transferred. According to the service scheduling method according to claim 7, wherein the service queue also has an invalid queue, and the service data in the invalid queue is not sent, wherein when the service data of all network devices in the service queue are transferred to the invalid queue, the service data in the invalid queue is transferred back to the multiple scheduling queues. The service scheduling method according to claim 7, wherein the updating of the queue transmission quota of the scheduling queue according to the sent service data comprises: The queue transmission quota of the scheduling queue is updated according to the transmission mode of the sent service data. The service scheduling method according to claim 11, wherein, when the transmission mode is orthogonal frequency division multiple access, the updated queue transmission quota of the scheduling queue is: Slice _STAi = Slice _STAi - Slice _Use * ^BWi / BW _Total , where Slice _Use is the total transmission quota usage of this multi-user transmission, _BWi is the resource block size of the current scheduling queue, and BW _Total is the total resource block size of the multi-user transmission. The service scheduling method according to claim 11, wherein when the transmission mode is multi-user multiple-input multiple-output, the updated queue transmission quota of the scheduling queue is: Slice _STAi = Slice _STAi - Slice _Use * NSS ⁱ / NSS _Total , where Slice _Use is the total transmission quota usage of this multi-user transmission, NSS _i is the number of spatial streams in the current scheduling queue, and NSS _Total is the total number of spatial streams for multi-user transmission. The service scheduling method according to claim 8, wherein the second predetermined threshold is determined by a minimum device transmission quota used for one transmission during the service data scheduling process. The service scheduling method according to claim 8, wherein, after all service data of the network device is transferred from the current scheduling queue to other scheduling queues among the multiple scheduling queues whose scheduling priority is lower than that of the current scheduling queue, the device transmission quota of the network device corresponding to the other scheduling queue to which it is transferred is: Slice _cur = Slice+Factor*Slice _last , wherein Slice is the transmission quota corresponding to the other scheduling queues reallocated to the network device, Slice _last is the device transmission quota of the network device corresponding to the current scheduling queue, and Factor is a scaling factor. The service scheduling method according to claim 7, further comprising: The queue transmission quota and the device transmission quota are selected based on a scheduling target, wherein the device transmission quota and the device transmission quota are transmission time or transmission data volume. According to the service scheduling method of claim 1, wherein the service data of the network devices in the same scheduling queue in the multiple scheduling queues of each of the service queues have the same priority; the service data of the multiple network devices in the same scheduling queue are scheduled by polling or first-in-first-out. The service scheduling method according to claim 1 further includes: when the service data of a new network device is added to the multiple scheduling queues or the service data of a sleeping network device is re-added to the multiple scheduling queues, the scheduling priority of the scheduling queue selected by the service data of the new network device or the service data of the sleeping network device is _Qin = min(Q _max , Q _out ), Q _out is the scheduling priority of the scheduling queue where the service data of the sleeping network device is located when the service data of the sleeping network device exits the scheduling queue, and Q _max is the highest scheduling priority of the scheduling queue in the device priority table of the service queue when the service data of the new network device is added. The service scheduling method according to claim 1, wherein, when there is service data STAi of a network device whose priority in at least two service queues is higher than a third threshold at the same time, the service data STAi of the network device is scheduled in one of the service queues TIDm, and the device transmission quota used is Slice _STAi,TIDm , then the device transmission quota of the service data STAi of the network device in the other service queues TIDn of the at least two service queues is: Slice _STAi,TIDn = Slice _STAi,TIDn -Factor _TiDn,m *Slice _STAi,TIDm , wherein Factor _TiDn,m is a scaling factor between the at least two service queues. The service scheduling method according to claim 1, wherein the traffic distribution strategy includes a priority division strategy or a load balancing distribution strategy. A service scheduling device based on a communication network, wherein the communication network includes a service scheduling device and at least one network device, the service scheduling device executes the service scheduling method to schedule service data to or from the network device via the communication network, and the device includes: A distribution module is configured to distribute the service data to corresponding service queues according to a predetermined traffic distribution strategy in response to receiving the service data, wherein each service queue includes multiple scheduling queues with different scheduling priorities, each service queue has a corresponding device priority table, and the device priority table records the dynamic correspondence between the network device and the multiple scheduling queues; the distribution module is further configured to distribute the service data to a corresponding scheduling queue among the multiple scheduling queues according to the network device to which the service data belongs or the service type of the service data; The sending module is configured to send the service data in the multiple scheduling queues according to the scheduling priorities of the multiple scheduling queues, wherein the service data in the scheduling queues with higher scheduling priorities are sent preferentially. A service scheduling device based on a communication network, wherein the scheduling device includes a memory, a processor, a program stored in the memory and executable on the processor, and a data bus for realizing connection and communication between the processor and the memory, wherein the program, when executed by the processor, realizes the service scheduling method as described in any one of claims 1 to 20. A computer-readable storage medium having one or more programs stored thereon, wherein the one or more programs can be executed by one or more processors to implement the service scheduling method described in any one of claims 1 to 20.