A wireless resource scheduling method and related device

Abstract

The application provides a wireless resource scheduling method and related equipment. The application ensures that key control signaling related to system synchronization, stability and terminal access can be transmitted in time and reliably by preferentially judging and allocating resources of broadcast signaling and random access response in the initial stage of the scheduling process. The application effectively avoids the risk that control signaling is delayed or discarded due to resource competition when the system service load is too high, guarantees the stable operation of the entire network and the smooth terminal access, and significantly improves the robustness and reliability of the system. A multi-level and differentiated user service guarantee mechanism is established to achieve the optimal balance of resource allocation efficiency and fairness. The application organically integrates the priority scheduling of control signaling with the multi-level guarantee strategy of user data and intelligent link adaptation to form a complete solution. The application guarantees the QoS of key signaling and various services, and significantly improves the overall throughput, stability and resource utilization efficiency of the system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of point-to-multipoint wireless data backhaul control, and in particular to a wireless resource scheduling method and related equipment. BACKGROUND

[0002] In the field of modern wireless communication, a private protocol point-to-multipoint (PTMP) system based on a WiFi physical layer protocol transmits data from multiple dispersed remote stations (RTs) to a central station (base station), and has a wide range of applications, covering data backhaul of small base stations, data backhaul of cameras, control signaling and control parameter backhaul of industrial equipment, and many other types of scenarios. Under this architecture, multiple terminals are responsible for transmitting multiple types of services, and efficient, fair, and diversified service quality (QoS) requirement meeting data scheduling is the key to ensuring the overall performance of the system.

[0003] For a communication system of a standard protocol such as 4G / 5G, after the system completes the maintenance of each user and each QoS level queue, a scheduling result is generated according to a specific scheduling algorithm, and specific uplink and downlink time slot resources are allocated to each user accordingly. Generally, the minimum granularity of the allocation is one time slot / resource block. The object to be studied by the present system is a system based on a private MAC protocol of a WiFi physical layer, and how to implement a scheduling method for multiple terminals and multiple service types. Although the existing scheduling process can handle conventional unicast data services, it has problems of insufficient efficiency and real-time performance when dealing with system-level control signaling and initial access management. For example, in the traditional scheduling design, the transmission of a broadcast channel usually adopts a fixed and predefined period. However, when the broadcast transmission time arrives, if the scheduler is processing or has arranged a large amount of user data services, the transmission of the broadcast signaling may be delayed, or may compete for resources with high-priority data services. Such delay or conflict will directly affect the normal work of all terminals, and even cause the terminals to lose synchronization, thereby reducing the robustness of the system. Therefore, the prior art fails to organically integrate key system-level management signaling (such as broadcast) and access process signaling (such as random access response) into the dynamic resource scheduling process in an efficient, timely, and priority-protected manner. SUMMARY

[0004] The present application aims to at least solve one of the above technical defects, and therefore provides a wireless resource scheduling method and related equipment to solve the technical defect of low efficiency of wireless resource scheduling in the prior art.

[0005] The wireless resource scheduling method comprises: determining a first slave station and a second slave station of a target wireless system based on a preset first priority principle, wherein the bandwidth mode of the first slave station is a fixed bandwidth mode, and the bandwidth mode of the second slave station is a competitive bandwidth mode; obtaining a bandwidth request value of a quality of service level reported by each first slave station and each second slave station to a master station; allocating time slot resources to each first slave station according to a preset first allocation principle based on the bandwidth request value reported by each first slave station; calculating the current remaining time slot resources; determining the allocation priority of each service of each second slave station based on each service type of each second slave station; and allocating the current remaining time slot resources to each second slave station in turn according to a preset second allocation principle based on the bandwidth request value reported by each second slave station and the allocation priority of each service of each second slave station until all time slot resources are allocated.

[0006] Preferably, the second slave stations are allocated time slot resources in turn according to a preset second allocation principle based on the bandwidth request value reported by each second slave station and the allocation priority of each service of each second slave station, including: judging whether the current remaining time slot resources meet the bandwidth request of all second slave stations based on the bandwidth request value reported by each second slave station; if yes, allocating the current remaining time slot resources to each service of each second slave station in turn according to the allocation priority of each service of each second slave station; if not, judging whether the current remaining time slot resources meet the bandwidth request of the first service in all second slave stations; if the current remaining time slot resources meet the bandwidth request of the first service in all second slave stations, judging whether the time slot resources remaining after the bandwidth request of the first service in all second slave stations is removed meet the bandwidth request of the second service in all second slave stations; if the current remaining time slot resources can meet the bandwidth request of the second service in all second slave stations, judging whether the time slot resources remaining after the bandwidth request of the first and second services in all second slave stations is removed meet the bandwidth request of the third service in all second slave stations; if the time slot resources remaining after the bandwidth request of the first and second services in all second slave stations is removed can meet the bandwidth request of the third service in all second slave stations, allocating time slot resources to the third service in all second slave stations in turn according to the principle of ensuring the bandwidth rate requirement of the third service; after the time slot resources are allocated to the third service in all second slave stations in turn, judging whether the time slot resources remaining after the bandwidth request of the first, second and third services in all second slave stations is removed meet the bandwidth request of the fourth service in all second slave stations; if the time slot resources remaining after the bandwidth request of the first, second and third services in all second slave stations is removed can meet the bandwidth request of the fourth service in all second slave stations, allocating time slot resources to each fourth service of each second slave station in turn according to the bandwidth request of the fourth service in all second slave stations.

[0007] Preferably, the method further includes: if the current remaining time slot resources do not meet the bandwidth request of the first service in all second slave stations, sorting the priority of the first service in all second slave stations; and allocating time slot resources to the first service in all second slave stations in turn according to the principle of ensuring the bandwidth requirement of the first service according to the priority sorting result of all first services until all remaining time slot resources are allocated; wherein the priority determination formula of the first service in all second slave stations is as follows: ​

[0008] in, For the first business User number Priority of levels; To ensure the rate of service for the primary business; The current rate of the first service; Tolerable latency for the service quality level of the primary service; This represents the current latency of the first service.

[0009] Preferably, the method further includes: if the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first services in all the second slave stations, are insufficient to satisfy the bandwidth requests of the second services in all the second slave stations, then the priority of the second services in all the second slave stations is ranked; and according to the priority ranking result of all the second services, time slot resources are allocated to the second services in all the second slave stations sequentially according to the principle of ensuring the bandwidth requirements of the second services, until all remaining time slot resources are allocated; wherein, the priority determination formula for the second services in all the second slave stations is as follows:

[0010] in, For the second business User number Priority of levels; To ensure the rate of service for the second business; The current rate for the second service; Tolerable latency for the service quality level of the second service; This represents the current latency of the second service.

[0011] Preferably, the method further includes: if the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first and second services in all the second slave stations, are insufficient to satisfy the bandwidth requests of the third services in all the second slave stations, then the priorities of all the third services are ranked; and based on the priority ranking of all the third services, time slot resources are allocated sequentially to the third services in all the second slave stations according to the principle of ensuring the bandwidth requirements of the third services, until all remaining time slot resources are allocated; wherein, the priority determination formula for the third services in all the second slave stations is as follows:

[0012] ;

[0013] in, For the third business User number Priority of levels; To ensure the rate of third-party services; a current rate of the third service.

[0014] Preferably, the method further comprises: if the current remaining time slot resources cannot satisfy the bandwidth request of the fourth service in all second slave stations, allocating the current remaining time slot resources to each fourth service in all second slave stations in turn according to the proportional fair allocation principle.

[0015] Preferably, the method further comprises: if the current remaining time slot resources cannot satisfy the bandwidth request of the fourth service in all second slave stations, calculating the total time slot resources required by the fourth service in all second slave stations, and determining the mode of increasing modulation rate to satisfy the resource request of all fourth services according to the current remaining time slot resources.

[0016] Preferably, the method further comprises: after allocating the required time slot resources to each fourth service in all second slave stations, judging whether there are time slot resources remaining at the current time; if there are no time slot resources remaining at the current time, determining that the total time slot resources have been allocated; if there are time slot resources remaining at the current time, supplementing the remaining time slot resources to each second service in all second slave stations according to the priority of each second service in all second slave stations to satisfy the bandwidth request thereof; judging whether there are time slot resources remaining after supplementing the time slot resources remaining at the current time to each second service in all second slave stations; if there are time slot resources remaining after supplementing the time slot resources remaining at the current time to each second service in all second slave stations, supplementing the remaining time slot resources to each third service in all second slave stations according to the bandwidth request of each third service in all second slave stations until all time slot resources have been allocated.

[0017] Preferably, the method further comprises: setting the allocation priority of the control signaling data service as a first priority; setting the allocation priority of the voice data service as a second priority; setting the allocation priority of the video data service as a third priority; and setting the allocation priority of the basic data service as a fourth priority; wherein the priority levels of the first priority, the second priority, the third priority and the fourth priority decrease in turn.

[0018] The wireless resource scheduling device comprises: a first determining unit configured to determine first slaves and second slaves of a target wireless system based on a preset first priority principle when a preset wireless resource scheduling time arrives, wherein a bandwidth mode of the first slaves is a fixed bandwidth mode and a bandwidth mode of the second slaves is a competitive bandwidth mode; an obtaining unit configured to obtain a bandwidth request value of a quality of service grade reported by each of the first slaves and each of the second slaves to a master station; a first allocating unit configured to allocate time slot resources to each of the first slaves based on the bandwidth request value reported by each of the first slaves according to a preset first allocation principle; a calculating unit configured to calculate a current remaining time slot resource; a second determining unit configured to determine an allocation priority of each service of each of the second slaves based on each service type of each of the second slaves; and a second allocating unit configured to allocate the current remaining time slot resource to each of the second slaves according to a preset second allocation principle in sequence until all time slot resources are allocated, based on the bandwidth request value reported by each of the second slaves and the allocation priority of each service of each of the second slaves.

[0019] A wireless resource scheduling device comprises one or more processors and a memory; the memory stores computer readable instructions, and the computer readable instructions are executed by the one or more processors to implement the steps of the wireless resource scheduling method according to any one of the preceding descriptions.

[0020] A readable storage medium stores computer readable instructions, and the computer readable instructions are executed by one or more processors to make the one or more processors implement the steps of the wireless resource scheduling method according to any one of the preceding descriptions.

[0021] From the above introduction, it can be seen that the application can determine the first slave station and the second slave station of the target wireless system based on the preset first priority principle, wherein the bandwidth mode of the first slave station is a fixed bandwidth mode, and the bandwidth mode of the second slave station is a competitive bandwidth mode; so as to allocate time slot resources according to different types of slave stations, and then obtain the bandwidth request value of the quality of service level reported by each first slave station and each second slave station to the master station. Since the bandwidth mode of the first slave station is a fixed bandwidth mode, it is necessary to allocate time slot resources for it first in the allocation, so the total time slot resources can be allocated to each first slave station according to the bandwidth request value reported by each first slave station based on the preset first allocation principle. After allocating time slot resources for each first slave station, the current remaining time slot resources need to be calculated, and the allocation priority of each service type of each second slave station also needs to be determined, so that the current remaining time slot resources can be allocated to each second slave station in turn according to the bandwidth request value reported by each second slave station and the allocation priority of each service type of each second slave station based on the preset second allocation principle, until all time slot resources are allocated.

[0022] From the above introduction, in the point-to-multipoint system, the application determines the bandwidth mode of each slave station accessing the master station in the initial stage of the scheduling process, and preferentially allocates the time slot resources of the slave station with a fixed bandwidth mode, and then preferentially allocates the resources of broadcast signaling and random access response according to the service type of the slave station with a competitive bandwidth mode, so as to ensure that the key control signaling related to system synchronization, stability and terminal access can be transmitted in time and reliably, effectively avoiding the risk that the control signaling is delayed or discarded due to resource competition when the system service load is too high, thereby ensuring the stable operation of the entire network and the smooth terminal access, and significantly improving the robustness and reliability of the system. A multi-level and differentiated user service guarantee mechanism is established, and the optimal balance between resource allocation efficiency and fairness is achieved. By organically integrating the preferential scheduling of control signaling and the multi-level guarantee strategy of user data and intelligent link adaptation, a complete solution is formed. Not only the QoS of key signaling and various services is guaranteed in a micro sense, but also the overall throughput, stability and resource utilization efficiency of the system are significantly improved in a macro sense, which has high practical value and commercial prospect. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor. Figure 1 A flowchart of a wireless resource scheduling method; Figure 2It is a schematic diagram of a wireless resource scheduling device structure. Figure 3 It is a hardware structure block diagram of a wireless resource scheduling device. DETAILED DESCRIPTION

[0024] The technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0025] In actual application, the LTE\5G system using the standard protocol is not suitable for the fast deployment of the long-distance backhaul industrial scene because of the complex network topology, high investment construction cost and long cycle. However, the wireless interconnection system based on the MAC layer private protocol of the WiFi physical layer becomes an important complementary product in the industrial interconnection field except for 4G / 5G because of the high cost performance and fast deployment. The MAC layer private protocol system based on the WiFi physical layer is customized. The master station is deployed at the backhaul point of the wired broadband. The slave stations communicate with the master station through wireless. Each slave station hangs the user or small local area network. For example, a slave station A hangs the switch. The switch is connected to the voice gateway, device control data and device status alarm information. Another slave station B hangs the camera data. A slave station C hangs a fixed bandwidth private line network. Through the research of the applicant, it is found that the slave station C can be set as a fixed time slot slave station. The slave station C is allocated a fixed time slot. The slave stations A and B can be set as dynamic time slot mode. In this way, the fixed bandwidth of the device C can be guaranteed, and the high priority data of the devices A and A can be transmitted in real time.

[0026] Therefore, in view of the fact that most of the current wireless resource scheduling schemes are difficult to adapt to complex and changeable business demands, the applicant has researched a wireless resource scheduling scheme. The wireless resource scheduling method can organically integrate the control signaling priority scheduling, the user data multi-level guarantee strategy and the intelligent link adaptation to form a complete solution. Not only the QoS of the key signaling and various businesses is guaranteed in the micro aspect, but also the overall throughput, stability and resource utilization efficiency of the system are significantly improved in the macro aspect, which has extremely high practical value and commercial prospect. The method provided by the embodiments of the present application can be used in many general or special computing device environments or configurations. For example, personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor devices, distributed computing environments including any of the above devices or devices, etc. The present application provides a wireless resource scheduling method. The method can be applied to various unmanned aerial vehicle inspection management systems and can also be applied to various computer terminals or intelligent terminals. The execution subject can be the processor or server of the computer terminal or the intelligent terminal.

[0027] The wireless resource scheduling method provided by the embodiments of the present application is described below Figure 1 , and the flow of the wireless resource scheduling method provided by the embodiments of the present application is shown in Figure 1 , which can include the following steps:

[0028] Step S101, when the preset wireless resource scheduling time arrives, the first slave station and the second slave station of the target wireless system are determined based on the preset first priority principle.

[0029] Specifically, in practice, in a point-to-multipoint system, a master station (AP, access point) and a slave station (RT, Remote Terminal) are included, one master station can connect multiple slave stations, and one AP can connect multiple RTs; the AP generally connects the broadband through a wired connection to realize data backhaul; the AP connects multiple RTs through wireless signals; each RT can be connected to a camera; the networking of "wired broadband access point + AP + RT + camera / other equipment" realizes the data backhaul of the remote camera / other equipment to the wired broadband access point; that is, one AP and multiple RTs provide a backhaul link for the remote equipment, for example, the distance between the farthest RT and the AP in the industry can reach more than 15 km; the most important characteristics of such a long-distance point-to-multipoint wireless backhaul system are transmission distance and transmission data bandwidth; among them, the transmission distance mainly considers the transmission power, antenna gain and reception sensitivity; the transmission data bandwidth mainly considers the bandwidth of the wireless signal, the modulation mode and the channel utilization. In practice, the bandwidth of the wireless signal is fixed for a specific system, such as 20MHz per channel for the 2.4G wifi frequency band; the maximum bandwidth of the 5.8GHz 802.11ac is 80MHz. The modulation mode of the wireless signal refers to the debugging method of the physical layer. The modulation mode of the wireless signal is related to the SNR signal-to-noise ratio. The signal-to-noise ratio and the modulation mode have a corresponding fitting curve. The higher the signal-to-noise ratio, the higher the modulation mode, for example, when the signal-to-noise ratio exceeds 30dBm, the signal-to-noise ratio can reach 2048QAM; the SNR signal-to-noise ratio refers to the ratio of the power of the main signal to the noise signal, the stronger the power of the main signal, the higher the SNR, and the higher the corresponding modulation mode. However, for a specific device, a specific AP vs RT deployment distance, the modulation mode is basically fixed. When the bandwidth of the wireless signal and the debugging mode are fixed, how to improve the system to protect important data requires a reasonable data scheduling method to improve the overall performance of the system to meet the expected requirements, and the data scheduling of the RT is one of the methods to improve the performance of the system.

[0030] In point-to-multipoint systems, data is usually classified into periodic critical data (e.g. sensor periodic sampling data, controller's closed-loop control instructions. This kind of data requires extremely low latency and extremely high reliability, and must be delivered within a certain time window.) and aperiodic / event-driven data (e.g. alarm information, parameter configuration, diagnostic information, temporary query. This kind of data has relatively loose real-time requirements, but the system needs to be able to respond in a timely manner.) If all slaves use fixed bandwidth mode (e.g. fixed time slots in TDMA), each slave will monopolize a fixed amount of wireless resources (time slots / channels) regardless of whether it has data to send. This guarantees determinism, but when the number of slaves is large and the amount of aperiodic data is small, it will result in low resource utilization and limited system capacity. If all slaves use contention bandwidth mode (e.g. CSMA / CA), slaves contend for the channel when needed, which improves resource utilization, but brings uncertainty in access delay and collision risk, and cannot meet the real-time requirements of critical data.

[0031] In order to ensure the key communication certainty while improving the overall utilization of wireless resources. When allocating time slot resources, the hard real-time demand needs to be met first, and the remaining resources are used to improve the overall efficiency. Therefore, when the preset wireless resource scheduling time arrives, based on the preset first priority principle, the first slave station and the second slave station of the target wireless system are determined. Among them, the bandwidth mode of the first slave station is a fixed bandwidth mode, and the bandwidth mode of the second slave station is a competitive bandwidth mode. In practice, the slave station is set to two priorities, the first priority is a fixed time slot mode (i.e. fixed bandwidth mode); the second priority is a dynamic time slot mode (i.e. competitive bandwidth mode); for the slave station with higher priority, set to fixed time slot mode, these time slots are fixed for the slave station, and other slave stations cannot compete with the slave station; for the slave station with general priority, set to dynamic scheduling mode, these time slots are the remaining time slots after subtracting the fixed time slots from the total time slots of a single period, which can be dynamically scheduled. In practice, the master station first checks the communication requirements of all slave stations. Those slave stations that carry periodic key data, have strict time limit and reliability requirements are preferentially allocated to resources of fixed bandwidth mode (such as exclusive, periodically appearing time slots). These slave stations become the first slave station. Therefore, the communication of the first slave station is guaranteed to be deterministic, regardless of how busy the network is, their exclusive resources will not be infringed. After meeting all the first slave stations (fixed bandwidth requirements), there are still some scattered or unoccupied wireless resource blocks left in the system. The master station divides these remaining resources (time slots) into a "competition pool" or "shared window". Other slave stations that need to send non-periodic, non-critical data (such as alarms, diagnostics) are allocated as second slave stations, and use competitive bandwidth mode to use these shared resources. This part of the resource is fully utilized, and the system can flexibly handle burst data, and the overall throughput is improved. In order to ensure the determinism and real-time of the absolute guarantee of core control instructions and key sensing data (lifeline), the valuable wireless spectrum resources are maximally utilized to process other auxiliary and management data, so as to achieve the best balance between reliability and efficiency.

[0032] In step S102, the bandwidth request value of the quality of service level reported by each first slave station and second slave station to the master station is obtained.

[0033] Specifically, in a point-to-multipoint system, one central station (AP) serves multiple stations at the same time. Wireless resources are limited, mainly including time slots, frequency bands, power, and the demand of each station is dynamically changing, some stations are in video conference (need high bandwidth, low delay), some are browsing web pages (low bandwidth demand, but require fast response), some are downloading files in the background (can tolerate delay, but need large bandwidth). Users will not continuously send data at full load, and traffic is bursty. If the central station does not understand these dynamically changing demands, resource scheduling will become very inefficient. If all stations can send requests at any time, when the number of users is large, the request signals will collide with each other, causing the central station to be unable to receive any information, and the network efficiency will decrease sharply. Therefore, the central station sets a preset scheduling time for each station, and the preset time provides each station with an ordered "reporting window". Periodically collect requests, and then make a unified scheduling, which is much more efficient than processing scattered requests at any time. This reduces the overhead of control signaling. The central station can make the optimal scheduling decision based on a complete, current cycle demand snapshot, just like the time slice scheduling of CPU. The "quality of service level (QoS level)" of each RT is not a simple "I want bandwidth" request, but contains priority information of the service type. For example, high priority can include voice, online games, emergency instructions. These services need to be prioritized and resources allocated to them first, even if their total bandwidth demand may not be large. Low priority can include file downloads, software updates. These services can fully utilize the remaining resources not occupied by high-priority services. The QoS bandwidth request value is a quantitative demand information. The AP needs to know "who has demand, how much". Avoid waste or lack of resources. Therefore, in order to achieve accurate and dynamic allocation, maximize the utilization efficiency of wireless resources under the premise of ensuring fairness and service quality. When the preset wireless resource scheduling time arrives, the bandwidth request value of the quality of service level reported by each station can be obtained. In practice, the main body of wireless resource scheduling for each RT is the software of the AP, which obtains the QoS level reported by each RT through the control channel, and after calculation by the AP, the RT is notified of the type of data it should send through the control channel. The size of the bandwidth required by the corresponding RT device is specified in the QoS parameter, such as RT1 below which is a camera with a bandwidth requirement of 5 Mbps; such as RT2 below which is a small local area network with a bandwidth requirement of 100 Mbps.

[0034] Step S103, based on the bandwidth request value reported by each first station, the total time slot resource is allocated to each first station according to a preset first allocation principle.

[0035] Specifically, the total wireless time slot resource is a zero-sum game field. The time slots assigned to slave 1 cannot be used by slave 2. The primary goal of system design is to ensure that the key services are not interfered. The first slave (fixed bandwidth mode) carries the key control loop of the system, such as motor speed feedback, valve opening control, safety interlocking signal. The delay and jitter of these data streams must be strictly limited (such as less than 10ms). By first and independently allocating time slots for them, it is equivalent to drawing exclusive and protected "VIP channels" for these key data streams on the time axis. These channels will be opened on time and on schedule for the corresponding first slave in each scheduling period. During the system design phase, engineers need to calculate the end-to-end delay in the worst case. Only when the resources of the key data stream are predetermined and fixed, is this worst-case calculation possible and reliable. If the resources of the first slave need to be dynamically allocated together with the second slave, its delay will become a random variable, which cannot meet the predictability requirements of hard real-time systems. Therefore, based on the bandwidth request value reported by each first slave, the total time slot resource can be allocated to each first slave according to the preset first allocation principle.

[0036] The preset first allocation principle can be "earliest deadline first", or "fixed priority scheduling", or "periodic or rate-based scheduling (such as RM algorithm)". The operation logic of the preset first allocation principle is to determine the bandwidth request value (usually expressed as "how many time slots are needed per superframe") of all first slaves and their quality of service level (QoS, such as period, deadline, priority). The scheduler runs the algorithm to find and allocate fixed and conflict-free time slot positions that meet the QoS requirements of each first slave in the total time slot resource pool. Finally, a certain and periodic time slot allocation table is given. For example: "slave F occupies time slot 1 every frame; slave G occupies time slot 3 every frame". This allocation principle ensures that the time slots of any two first slaves do not overlap. During the allocation process, the system can immediately determine whether the current total resource is sufficient to meet the hard real-time needs of all first slaves. If not, the system will immediately trigger an alarm, rather than running in an unreliable state. Once allocated, this mode will usually remain unchanged in subsequent superframes, unless there is a major change in network topology or service demand, thereby providing time certainty.

[0037] The resource usage nature of the first slave and the second slave are fundamentally opposite. The first slave (fixed) needs long term, stable, predictable resource occupation, its resource is "reserved". The second slave (contending): needs short term, dynamic, on-demand resource occupation. Its resource is "borrowed" or "shared". Assume there are 10 slots in total. The first slave H needs 2 fixed slots, the second slaves I-K contend for the remaining resources. If the 10 slots are mixed together, trying to allocate for the first slave H (fixed) and the second slaves I-K (contending) at the same time with one algorithm, the scheduler cannot allocate "fixed" slots for the slaves I-K in advance, because their demand is bursty. But if the slots of the first slave H are not fixed, the slots of the first slave H can be "preempted" or affected by the requests of the second slaves I-K in dynamic scheduling, causing the determinacy of the first slave H A to be broken. The whole scheduling problem becomes extremely complex and cannot guarantee the key performance.

[0038] At step S104, the current remaining slot resources are calculated.

[0039] In particular, the remaining time slots are the only source of system "flexibility" and "scalability". After the fixed time slots are allocated to the first slave stations, the remaining time slot resources are calculated to establish a resource pool for the competitive scheduling of the second slave stations, and thus the current remaining time slot resources can be calculated. The remaining time slots constitute a dynamically shared resource area, and are all the resources available to the second slave stations (competitive mode). Without this clear boundary, the competitive mechanism cannot be started. By clearly defining the amount of remaining resources, the competitive rules (such as backoff algorithms, priority competition) can be designed to ensure that the competitive behavior of the second slave stations is strictly limited within this pool and will absolutely not interfere with the fixed time slots of the first slave stations. The number of remaining time slots is also a core indicator for measuring the load level and expansion capacity of the system. A high remaining rate (such as remaining > 50%) means that the system is lightly loaded, has sufficient resources to handle burst data, and can accommodate new nodes or improve the quality of service of non-critical services. A low remaining rate / close to zero means that the system is close to full load, and the competition of the second slave stations will be extremely fierce, and the delay of non-critical services will increase, which is an early warning signal that the system needs to be optimized or expanded. Zero remaining / negative remaining means that the fixed demand of the first slave stations has exhausted or even exceeded the total resources. This is a serious system design error or failure state, and must be immediately alarmed, as the determinacy of even the most basic critical services cannot be guaranteed. Based on the calculation results of the remaining time slots, the master station can make more intelligent decisions. If there are many remaining time slots, the master station can relax the competitive parameters of the second slave stations (such as shortening the competitive window), allowing them to access faster and further reducing the delay of non-critical services. If there are few remaining time slots, the master station can tighten the competitive parameters (such as increasing the competitive window) to reduce the probability of collision, although the average access delay increases, but it can guarantee the order of shared resources. The master station can also decide whether to accept new services or new nodes. When a new slave station requests to join the network (or an existing slave station has a new service flow), the master station first checks the remaining time slot resources. If the new request is hard real-time (requires fixed bandwidth), there must be enough and continuous remaining time slots to be accepted. If the new request is soft real-time / non-real-time (can compete for bandwidth), the master station can estimate its impact on the quality of service of the existing second slave stations according to the amount of remaining resources to decide whether to accept. Moreover, by recording and analyzing the number of remaining time slots in a plurality of consecutive scheduling periods, the growth trend of the fixed demand of the first slave stations or the competitive demand of the second slave stations can be identified. Optimize long-term planning to provide data support for network expansion, service migration, or protocol parameter optimization. In practice, the total bandwidth of the target wireless system and the total bandwidth required by each first slave station can be calculated first, and then the remaining bandwidth after subtracting the total bandwidth occupied by each first slave station from the total bandwidth of the target wireless system is the remaining time slot resource. The total bandwidth of the target wireless system is generally determined at the beginning of the design; the total bandwidth required by each first slave station is calculated according to the following formula: ; represents the A bandwidth rate level negotiated between a first slave station and a master station, the MCS of each first slave station is strongly related to the SNR between the first slave station and the master station; if 3 first slave stations are linked under a master station, the first first slave station has a high SNR due to a short distance, and the negotiated MCS with the master station is 13; the second first slave station negotiates a MCS of 5 with the master station; the third first slave station negotiates a rate level of 8 with the master station; the negotiation works as the physical layer of WiFi7; indicates the number of master stations allocated to the target wireless system, the cycle period of the target wireless system is generally set to 5ms, and one unit is 100us; indicates the first slave station in one cycle period.

[0040] Step S105, based on each service type of each second slave station, determine the allocation priority of each service of each second slave station.

[0041] Specifically, after determining the "competition pool" of the remaining time slot resources, if all the services of all the second slaves simply "compete equally" (like pure CSMA / CA), the system can run, but it will become rough and inefficient, and cannot meet the needs of complex industrial scenarios. Therefore, based on each service type of each second slave, the allocation priority of each service of each second slave can be determined, and a service type-based allocation priority mechanism is introduced, in order to establish a fine "micro-scheduling" order in a resource-limited competition environment. The remaining competition resources are scarce and uncertain, and "competition rules" must be established. The remaining time slots are "surplus resources" squeezed out by the system after meeting all hard real-time needs. This resource pool is usually scarce (most of which has been occupied by the first slave), scattered (the time slots may not be continuous in time), and dynamically changing (the remaining amount may be different in different scheduling periods). If all the services (such as emergency alarms, ordinary queries, and software upgrade packages) of all the second slaves are allowed to compete without discrimination and fairly, serious problems will occur, for example, a slave sending a large capacity, non-urgent diagnostic package may block a slave sending a critical alarm signal, causing important services to starve due to resource competition. It also cannot meet the diversified QoS requirements. Non-critical services in industrial networks also have different tolerances. Alarm delay needs to be less than 100 ms, while firmware upgrade can tolerate several seconds. Undiscriminating competition cannot reflect these differences. The priority mechanism can create a "deterministic gradient" in the competition environment. By determining the allocation priority of each service type of each second slave, the master station actually superimposes a layer of static and predictable preference rules on the competition mechanism. It can ensure timely response to critical aperiodic events, and the highest priority can be given to services such as "emergency shutdown alarm", "safety interlock trigger", "device fault alarm", etc. In competition, high-priority services can use shorter contention windows and more retransmission opportunities, thereby statistically ensuring extremely low access delay. Although it is still a competition, it is several orders of magnitude faster than low-priority services. It can achieve reasonable allocation of resources among services, and the priority divides the service flow into different "service levels". For example, when resources are extremely scarce, the system will naturally ensure that high-priority services are given priority, and low-priority services are delayed or buffered. This meets the logic of industrial operation. It avoids the worst case of "stuck together" of all services when the network is congested. High-priority services can still pass through, maintaining the basic monitorability and safety of the system. It provides a basis for more advanced scheduling of the master station (such as priority weighted round robin).

[0042] Furthermore, a single slave station may generate multiple services: a smart sensor may simultaneously generate periodic temperature data (first slave station service), occasional over-temperature alarms (second slave station, high priority), and periodic self-test reports (second slave station, low priority). If priorities are tied to devices, it becomes impossible to differentiate and process these services. The purpose of a network is to serve service flows, not simply devices. Alarm signals should be delivered quickly regardless of the device they originate from. Network administrators can flexibly adjust the behavior strategy of the entire network by configuring the priority of service types without modifying the underlying code of each device. The master station translates this priority information into parameters for a contention mechanism and sends them to the second slave station. This typically affects the following aspects:

[0043] 1. Contention window size: High-priority services use a smaller contention window, which means shorter contention backoff time and faster retries.

[0044] 2. Initial backoff count: High-priority services may receive a smaller initial backoff value.

[0045] 3. Retransmission limit: High-priority services are allowed more retransmissions to improve reliability.

[0046] 4. Independent competition queues: Within the slave station, services with different priorities enter different sending queues, and high-priority queues are always processed first.

[0047] Specifically, in practice, the priority of control signaling data services can be assigned as the first priority; the priority of voice data services can be assigned as the second priority; the priority of video data services can be assigned as the third priority; and the priority of basic data services can be assigned as the fourth priority. The priority levels of the first, second, third, and fourth priorities decrease in that order.

[0048] Step S106: Based on the bandwidth request value reported by each second slave station and the allocation priority of each service of each second slave station, the remaining time slot resources are allocated to each second slave station in sequence according to the preset second allocation principle until all time slot resources are allocated.

[0049] Specifically, in the case of only priority setting, the second slave's traffic still randomly competes in the remaining time slot pool. Priority is only a "probability advantage", not a "deterministic guarantee", high priority traffic has a smaller contention window, meaning statistically easier to win, but in extreme cases (such as a large number of low priority traffic sending at the same time), high priority traffic can still experience multiple backoff and collision, resulting in uncertain delay. The system also cannot handle "hard" non-periodic real-time traffic, some second slave's traffic may not be periodic, but once it occurs, it has a strict deadline (such as some safety warning signals). Pure competition mechanism cannot promise to complete transmission before the deadline. Resource utilization still has room for improvement, competition is inevitably accompanied by idle time slots (waiting for backoff) and collision time slots (data collision retransmission), which is a valuable waste in the remaining pool where resources are extremely scarce.

[0050] In practice, the preset second allocation principle is usually a hybrid scheduling algorithm, which can be set as, for example, priority weighted round robin, earliest deadline first, etc. By "active calculation and allocation" instead of "passive competition", the master station obtains the bandwidth request value (such as "alarm service needs 2 consecutive time slots") and priority of all second slave stations, and can actively reserve time slots in the remaining resources for them, like handling the first slave station. For example, for "emergency shutdown" service (highest priority), the master station will forcibly allocate the required resources in the nearest available remaining time slot, ensuring that it can definitely send in the next superframe, completely avoiding the uncertainty of competition. As a global controller, the master station can calculate a conflict-free and compact allocation scheme for all second slave station services with demand in the remaining discrete time slot resources, like playing chess. This eliminates the collision and backoff overhead caused by competition, and theoretically achieves 100% utilization of the remaining resources (allocation until used up). This allows the master station to support more complex service models and QoS requirements. The bandwidth request value contains the "amount" and "shape" information of the service (how many time slots are needed? Are consecutive time slots needed?). The allocation priority contains the "urgency" information of the service. The master station can combine the two to implement complex strategies. For example: Strategy A (guarantee key burst): first meet all the requests of high-priority services, and then use the remaining resources to meet medium and low-priority services in proportion. Strategy B (consider fairness): allocate resources to all services in proportion according to priority weight. Furthermore, "sequential allocation" embodies the strictness of the scheduling order. The allocation order is directly determined by the "second allocation principle" (such as from high to low priority). This ensures that the most important service gets the best remaining resources first (for example, time slots that are earlier in time and more continuous). "Allocation complete" avoids resource idling, and any remaining time slot that is not explicitly allocated will become "blank" and be wasted in this scheduling period. This creates a deterministic expectation. For low-priority services that are not allocated resources, the master station can explicitly inform them: "no resources in this round, please buffer data and wait for the next period". This is better than waiting and retrying indefinitely in competition, because the device obtains a deterministic feedback and can enter a low-power state or take other strategies. A completely allocated time slot mapping table makes the ownership of each time slot clear. All slave stations (including first and second slave stations) know exactly which precise time slot they can send or receive at the beginning of the superframe, and the network behavior becomes completely predictable and manageable.

[0051] As can be known from the above introduction, the application can, in the starting stage of the scheduling process, preferentially judge the bandwidth modes of each slave station accessing the master station, preferentially allocate the slave stations of the fixed bandwidth mode, and then preferentially allocate the resources of broadcast signaling and random access response according to the service types of the slave stations of the competitive bandwidth mode, so as to ensure that the key control signaling related to system synchronization, stability and terminal access can be transmitted in time and reliably, effectively avoid the risk that the control signaling is delayed or discarded due to resource competition when the system service load is too high, and thus ensure the stable operation of the entire network and the smooth terminal access, significantly improve the robustness and reliability of the system. A multi-level and differentiated user service guarantee mechanism is established, the optimal balance between resource allocation efficiency and fairness is achieved, a complete solution is formed by organically integrating the preferential scheduling of control signaling, the multi-level guarantee strategy of user data and intelligent link adaptation, and not only the QoS of key signaling and various services is guaranteed in microcosm, but also the overall throughput, stability and resource utilization efficiency of the system are significantly improved in macrocosm, which has extremely high practical value and commercial prospect.

[0052] As can be known from the above introduction, the application can, based on the bandwidth request value reported by each second slave station and the allocation priority of each service of each second slave station, allocate the current remaining time slot resources to each second slave station in turn according to the preset second allocation principle until all time slot resources are allocated, and the process is introduced as follows:

[0053] In step S201, whether the current remaining time slot resources meet the bandwidth requests of all second slave stations is judged based on the bandwidth request value reported by each second slave station.

[0054] Specifically, in order to achieve scientific and reasonable scheduling, the master station needs to know the macro relationship between the resources (current remaining assignable time slots) and the demand (total request bandwidth) it faces. If it is satisfied or even abundant, the system is in a relatively relaxed state. If it cannot be met, it means that the system has entered a state of resource scarcity, which means that simple "on-demand allocation" is no longer feasible, and more complex decision mechanisms must be started. Therefore, after obtaining the bandwidth request value of the service quality level reported by each slave station, the current remaining time slot resources can be judged based on the bandwidth request value reported by each second slave station whether they meet the bandwidth request of all second slave stations. Then the AP can get the total bandwidth of the system at this moment and this deployment position according to the air interface bandwidth of each second RT. For example, an AP accesses three RTs, which are RT1, RT2 and RT3. Since the communication between the AP and each RT is realized by using time division technology, the total communication period of the AP and all RTs is set to 5ms, and 5ms is divided into several time slots. In each time slot, the AP communicates with only one RT, and in the next time slot, it communicates with other APs. The throughput corresponding to each time slot can be calculated according to the time slot length and the payload proportion of the frame structure. That is, the time slot and the throughput are approximately proportional. RT1, RT2 and RT3 may have different bandwidth requests according to different time and different services. For example, RT1 is connected to a camera, and its bandwidth demand is fixed at 5Mbps. RT2 is connected to an office LAN, which may have 100Mbps during the day and 0Mbps at night. Even during the day, there may be a burst of high throughput, which causes the QoS reported by the RT at different time periods to be different values.Meanwhile, the same RT under the hanging data also have multiple types, such as IP phone service, this kind of service for delay requirement is very strict, the system according to the type of data packet will set this kind of service as the highest priority UGS service (Unsolicited Grant Service, unsolicited grant service), that is, the priority of UGS service can be set to the first priority, such as customer online UDP service, this kind of service is not sensitive to delay requirement and can be defined as BE service (Best Effort, best effort service, that is, the priority of BE service can be set to the fourth priority; Then AP needs to calculate in real time which time slot of a single communication cycle should be given to which RT, and what service of the RT should be given priority, which user's service should be satisfied first; If the time slot resource of AP is enough or the demand of RT for bandwidth is small, a 5ms cycle can meet the bandwidth demand of all RTs, which is the best; If the time slot resource of AP cannot meet the bandwidth demand of all data types of all RTs in a 5ms cycle, how to optimize the scheduling strategy to ensure that important data is transmitted first and non important data is transmitted later needs to be considered, thus triggering different scheduling strategies. Based on the bandwidth request value reported by each slave station (RT), it is judged whether the remaining time slot resource can meet the bandwidth request of all second slaves. This judgment result is equivalent to the "switch" of the working mode of the scheduling algorithm. When the time slot resource of the master station is sufficient (the current remaining time slot is greater than or equal to the total request bandwidth of all second slaves), the bandwidth request value of each second slave can be directly approved or basically approved, and the remaining time slot resource after meeting all demands can be used as "extra reward" for secondary allocation. The secondary allocation strategy can be equal distribution, or weighted distribution according to the historical performance, priority, etc. of the second slave, so as to further improve the overall throughput and user experience. When the time slot resource of the master station is insufficient (the remaining time slot resource is less than the total request bandwidth of all second slaves), the "QoS level" reported by each second slave becomes the most important basis for decision. For example, first of all, it is necessary to ensure that the services with the highest priority (such as voice and control signaling) are 100% satisfied. Even if the total time slot resource is tight, these services need to be allocated the required time slots first. Secondly, after meeting all high priority services, if the remaining time slot resource is still insufficient, the fairness algorithm needs to be started for services with the same or lower priority (such as video streaming and file download). This is usually not a simple "equal distribution", but a "proportional fair distribution according to weight". For example, based on QoS level, subscription bandwidth of slave station, historical average throughput, etc., the remaining available time slots can be allocated proportionally according to the weight of each request. For example, assuming that three slaves A, B and C of the same level request 4, 2 and 2 units of bandwidth respectively, but only 4 units are available, they may be allocated 2, 1 and 1 units respectively according to the request ratio. Secondly, for commercial network, this judgment is the key to realize SLA and QoS.The network promises high value users with service priority. When time slot resources are tight, through this judgment and subsequent priority scheduling, the experience of these users can be ensured not to be affected, thus fulfilling the SLA. Without judgment and priority-based scheduling, some low-priority services may not get resources at all in a fierce competition. A good scheduling will allow low-priority services to obtain "not enough but still available" resources through proportional fairness and other algorithms, even in a resource shortage, to maintain basic network functions. Therefore, if the current remaining time slot resources meet the bandwidth requests of all second slave stations, step S202 is executed. If the current remaining time slot resources cannot meet the bandwidth requests of all second slave stations, step S203 is executed.

[0055] In step S202, the current remaining time slot resources are allocated to each type of service of each second slave station in turn according to the allocation priority of each type of service of each second slave station.

[0056] Specifically, in practice, even if the time slot resources are sufficient, simply "average allocation by request value" or "first come first served" is a very low-level and inefficient strategy. According to the service allocation priority, it is allocated in turn to realize "value maximization", not just "throughput maximization". Therefore, if the current remaining time slot resources meet the bandwidth requests of all second slave stations, the current remaining time slot resources can be allocated to each type of service of each second slave station in turn according to the allocation priority of each type of service of each second slave station. The time slot resources are sufficient only as a "quantity" concept, but the requirements of different services for time sensitivity and importance are different "qualities". The scheduling strategy needs to reflect this qualitative difference. For example, even in the case of sufficient total resources, if the allocation process is concurrent or without priority, resource competition may still occur in a micro-instant. Suppose the system handles two requests at the same time: service A (voice call, high SLA) needs a small amount of time slot that can be used immediately. Service B (file download, low SLA) requests a large amount of time slot resources. If the system allocates a large block of continuous time slots to service B first, it may temporarily fragment the resources, causing the request of service A to be unable to be satisfied immediately, resulting in call lag and low-priority service blocking high-priority service. Therefore, in order to more reasonably use the time slot resources, all high-SLA services can be allocated the resources they need first. To ensure that the resources of all critical services are "definitely" reserved, and the time slots are continuous and low-latency. Then allocate the remaining resources to low-SLA services. This fundamentally eliminates the risk of blocking. Reasonable allocation of time slot resources also needs to meet the differentiated SLA requirements of different services. Service level agreement is not only about bandwidth, but also includes delay and jitter. For example, high-SLA services (such as industrial control, online games, VR / AR) are extremely sensitive to delay and jitter, and the resources they need are not only "quantity" enough, but also "quality" excellent, that is, the time slots allocated need to be continuous and advanced. Low-SLA services (such as file download, email reception) are not sensitive to delay and jitter, and only care about the final total throughput.

[0057] By allocating time slot resources to each type of service of each second slave station in turn according to the allocation priority of each type of service of each second slave station, the current remaining time slot resources are allocated to each type of service of each second slave station, and high-SLA services can be preferentially allocated continuous and front time slots to minimize their transmission delay. The remaining possibly discontinuous and rear time slots can be allocated to low-SLA services, which can make good use of these fragmented resources. The "optimal utility" of network resources can be achieved. The same time slot resources allocated to different services produce different overall network values and user experiences. For example, 1MBps of bandwidth used for voice calls creates a very high communication value. The same 1MBps used for background cloud synchronization creates a relatively low value. Therefore, even if the resources are sufficient, preferentially guaranteeing the resource requirements of high-value services can maximize the overall utility and user satisfaction of limited wireless resources. Assuming that the system has 100 units of time slots, two second slave stations request time slot resources: second slave station A (remote surgery, ultra-high SLA) requests 10 units, which must start transmitting at the first time slot, and the time slots must be continuous. Second slave station B (video monitoring backup, low SLA) requests 95 units. The total time slot resources requested by the two second slave stations is 105, but the current remaining time slot resources are only 100, and the time slot resources are insufficient. But if second slave station A (remote surgery, ultra-high SLA) requests 10 units. Second slave station B (video monitoring backup, low SLA) requests 85 units. The total time slot resources requested by the two second slave stations is 95, which is less than the current time slot resources 100, and the resources are sufficient. There can be two allocation strategies: Strategy 1: parallel allocation without priority, which can start from the front end of the physical resources, and the first 85 time slots are allocated to second slave station B, and the last 10 time slots are allocated to second slave station A. The data packet of second slave station A needs to wait for 85 time slots before it can start transmitting, with a very high delay, and the surgery control command cannot be transmitted in real time, and the service fails. Strategy 2: allocate in turn according to the allocation priority, and first round preferentially allocate 10 units required by second slave station A (for example, allocate the first 10 continuous time slots). The second round allocates the remaining 90 units to second slave station B. Then the command of second slave station A is almost zero-delayed, and the service is successful. Although second slave station B only gets 90 units instead of 85, it does not affect its backup function at all, and the overall utility is maximized. Therefore, even if the resources are sufficient, the time slot resources are allocated in turn according to the service allocation priority to ensure the low delay and low jitter requirements of high-SLA services. From the mechanism, low-priority services are prevented from affecting high-priority services. Let the limited wireless resources produce the maximum overall economic utility and user experience. It embodies the evolution of network scheduling from "connectivity" to "intelligence", from "best effort" to "quality assurance".

[0058] In step S203, it is judged whether the current remaining time slot resources meet the bandwidth request of the first service in all second slave stations.

[0059] Specifically, if the current remaining time slot resources cannot meet the bandwidth requests of all second slave stations, in order to ensure that the system can still maintain the most core services uninterrupted in the extreme case of serious resource shortage, and achieve graceful degradation instead of overall collapse, it is necessary to determine whether the current remaining time slot resources meet the bandwidth requests of the first service in all second slave stations, wherein the first service can be a control signaling data service, and the control signaling service has the highest priority, therefore, the allocation priority of the first service is set as the first priority, and general control signaling has strict requirements for time delay; the control signaling in the system can be reported through MQTT during transmission, and a priority identifier is added in the data packet frame header, so as to facilitate data scheduling. The first service can be an UGS (Unsolicited Grant Service, Unsolicited Grant Service) service. When it is found that the current time slot resources cannot meet all the demands, the primary problem is: "when a part of the service has to be sacrificed, what should be preserved?". The first service usually represents the lifeline service of the network, such as emergency communication (such as public safety, rescue instruction), key control signaling, management signaling for ensuring network basic operation and other key services, and once these services are interrupted, it may cause disastrous consequences or paralysis of the entire system. Therefore, the first service generally has "veto power" and must be prioritized in any case. If the current remaining time slot resources cannot meet the bandwidth requests of all second slave stations, the system will enter two completely different "crisis handling modes": if the remaining time slot resources can meet all the first services (the bottom line can be preserved), low-end services can be discarded to protect high-end services. All the remaining time slot resources are allocated to all the first services without any conditions. This is the first priority to ensure the core is not dead. After the first service is satisfied, the time slot resources are exhausted. For all non-first services, no resources will be allocated. These non-first services will experience interruption or waiting. Such allocation preserves the minimum core function, although the user experience is poor, but the system itself and the most critical application have not collapsed. If the remaining time slot resources cannot meet the first service, the time slot resources are extremely scarce, and even the lifeline is threatened. The resource allocation strategy needs to be adjusted to "overall degradation" or "emergency plan", at this time the regular scheduling algorithm has failed, and the unconventional emergency plan needs to be started, such as allocating resources to all first services in proportion. This means that even the first service cannot get its full request, but can only get a part of the resource according to the weight. This is better than randomly interrupting a key service. Some low-level but resource-intensive sessions may be forcibly interrupted, and resources are recycled and allocated to the first service. A serious alarm will be sent to the upper management system, indicating that the system is in an "overload" state and cannot guarantee core services, and external intervention (such as expansion and diversion) is needed.

[0060] If the current remaining time slot resources cannot satisfy the bandwidth request of the first level service in all second slave stations, it means that the system knows that it has already failed to meet the bottom line and must start the highest level crisis handling program and seek external help, then step S205 is executed; if the bandwidth request of the first service in all second slave stations can be satisfied, the system knows that its bottom line is stable and can execute the strategy of "keeping high and discarding low" with confidence, then step S204 is executed. This mechanism ensures that the system behavior is still predictable and manageable under extreme stress, rather than chaotic random interruption, which is crucial for building a highly reliable communication system.

[0061] In step S204, it is determined whether the current remaining time slot resources, excluding the time slot resources required by the bandwidth request of the first service in all second slave stations, can satisfy the bandwidth request of the second service in all second slave stations.

[0062] Specifically, when it is confirmed that the current remaining time slot resources, excluding the time slot resources required by the bandwidth request of the first service in all second slave stations, can satisfy the bandwidth request of all first services, after ensuring that the highest priority service is worry-free, it is necessary to immediately re-evaluate the current time slot resource situation to determine the allocation strategy of the next priority service. Therefore, if the current remaining time slot resources can satisfy the bandwidth request of the first level service in all second slave stations, it is determined whether the current remaining time slot resources, excluding the time slot resources required by the bandwidth request of the first service in all second slave stations, can satisfy the bandwidth request of the second service in all second slave stations. To confirm how much "freely disposable" resources are left. And the remaining resources are regarded as a "new, smaller total resource pool", and the above-mentioned resource allocation logic for the first service is repeated for the next priority. Among them, the second service can be a voice data service, such as IP-based voice data, which can be an rtPS service (i.e. rtPS (Real-Time Polling Service), the allocation priority of the second service can be set to the second priority, and the result of this judgment directly determines how to treat the second service, thereby switching again between the two modes of "completely meet" and "reduce as needed". Therefore, if the current remaining time slot resources, excluding the time slot resources required by the bandwidth request of the first service in all second slave stations, cannot satisfy the bandwidth request of the second service in all second slave stations, it means that the remaining time slot resources cannot completely satisfy the second service, and may need to be allocated as needed, then step S206 is executed. If it can satisfy the bandwidth request of the second service in all second slave stations, it means that the remaining time slot resources are sufficient to completely meet the demand of the second service, then step S207 can be executed.

[0063] Step S205, priority ranking of all first services in all second slaves; and according to the priority ranking result of all first services, time slot resources are allocated to all first services in all second slaves in turn according to the principle of ensuring the bandwidth requirement of the first service, until all remaining time slot resources are allocated.

[0064] Specifically, as can be known from the above description, when the time slot resources are so scarce that even the top-level first service cannot be fully satisfied, it means that the system has entered a "disaster recovery mode" or a "wartime state". At this time, the goal has changed from "optimizing efficiency" to "minimizing loss". The first service needs to be internally sorted and allocated in turn, and a "relative priority" order is established in the "absolute priority" group. The first service as a whole has absolute priority over other levels of services. However, when there is a time slot resource competition within this group, the original "first priority" label cannot solve the S-vs-S conflict. A more detailed decision basis is needed: at this time, a new and more detailed sorting dimension must be introduced within the first service to determine "who is more priority in the first service?". This sorting basis can be as follows: (1) service sub-type: such as the signal priority of a fire alarm sensor is higher than that of a normal security camera. (2) user identity / contract: such as the emergency call of a platinum VIP user is higher than that of a normal user. (3) system criticality: such as the management signaling of a core network element is higher than the data reporting of an edge device. (4) real-time requirement: control instructions with millisecond-level response are higher than status updates with second-level response. When the time slot resources are insufficient to save all services, the rational strategy is "save the most valuable as much as possible". If the first service is not sorted by priority, but is allocated evenly or randomly, it may result in all first services getting a little resource, but none of them can work normally. This is equivalent to wasting scarce time slot resources, but failing to protect any critical function. By sorting and allocating in turn, the highest priority S1 service can obtain all the resources it needs and run normally at 100%. If there is remaining resource, the second highest priority S2 service can also obtain all the time slot resources it needs and run normally at 100%. Since the resource is exhausted before being allocated to the lowest priority Sn service, it may not run at all, resulting in the sacrifice of some low-priority first services, but successfully protecting the most valuable and critical first services in the system. If the resources are allocated evenly, all first services will be in a semi-paralyzed state, resulting in a command failure and ultimately greater loss. When the remaining resources cannot meet all first-level services, internal sorting and allocation in turn is a strategy to pursue "optimal failure" in desperation. By establishing clear selection rules, scarce resources are used to protect the most valuable core functions in the system, thereby maximizing the survival ability of the system. The priority determination formula of all first services in all second slaves is as follows: wherein, a priority of the first service; a priority of the first service; a priority of the first service; a guaranteed rate of the first service; a current rate of the first service; a tolerance delay of the first service; a current delay of the first service.

[0065] Step S206, ranking the priorities of the second services in all second slave stations; and according to the ranking results of the priorities of all the second services, allocating time slot resources for the second services in all second slave stations in turn according to the principle of ensuring the bandwidth requirements of the second services, until all the remaining time slot resources are allocated.

[0066] Specifically, the second services as a whole are less important than the first services, but the values and real-time requirements of the individual second services are not completely homogeneous. Therefore, when all the first services are satisfied but the remaining resources are insufficient to satisfy all the second services, there is still a difference between the primary and the secondary in the "next highest priority" group. Therefore, if the current remaining time slot resources are insufficient to satisfy the bandwidth requirements of all the second services except for the time slot resources required by the bandwidth requests of the first services in all second slave stations, it means that the remaining time slot resources cannot completely satisfy the second services and may need to be allocated on demand, then the priorities of the second services in all second slave stations are ranked; and according to the ranking results of the priorities of all the second services, the second services in all second slave stations are allocated time slot resources in turn according to the principle of ensuring the bandwidth requirements of the second services, until all the remaining time slot resources are allocated. Wherein, the priority of the second service in all second slave stations is determined by the following formula: wherein, a priority of the first service; a priority of the first service; a priority of the first service; a guaranteed rate of the first service; a current rate of the first service; a tolerance delay of the first service; a current delay of the first service.

[0067] In practice, the second services are also heterogeneous. For example, in an enterprise network, the second services can include A1 (critical) for real-time production monitoring data (if the data is incomplete or delayed, it can cause production failure), A2 (important) for video conference of the management team (although important, short-term lag or reduced image quality is acceptable), and A3 (ordinary) for batch sales data synchronization (high throughput is required, but real-time is not sensitive). If the "proportional fairness" algorithm is used, A1, A2, and A3 will be proportionally reduced in bandwidth. This can cause the critical production monitoring data to be lost due to insufficient resources, causing substantial losses. While the video conference and data synchronization are reduced, the services themselves are not completely disabled. When the time slot resources are insufficient to allow all second services to run, the goal of the scheduler is the same as when processing the first service, that is, to ensure that the highest value subset can run 100% normally. If proportional fairness is allocated, the result is that "all second services are alive, but are in sub-health state". For A1, a critical service, sub-health (incomplete data) can be equivalent to unusable. If the second services are sorted internally and then allocated in turn, the result is that "some second services are completely healthy, and the other second services are temporarily disabled". This sacrifices low-priority second services, but in exchange for the absolute health and availability of high-priority second services. This is a more ruthless but also more rational "sacrifice pawns to protect the car" strategy, except that this strategy is now applied to the second internal. Secondly, according to the second service model, "hard real-time" and "soft real-time" services also need to be treated differently. Hard real-time services (such as A1) generally have a clear bandwidth threshold that must be met. Below this threshold, the service fails. It needs "assurance", not "fair sharing". The performance of soft real-time / elastic services (A2, A3) improves with increasing bandwidth, but there is no absolute failure point. They can make good use of any resources allocated by "proportional fairness". Through internal sorting and sequential allocation, the system is identifying and prioritizing these "hard real-time" second services, which have higher requirements for "assurance".

[0068] In step S207, it is determined whether the time slot resources remaining after the bandwidth requests of the first services and the second services in all second slave stations are removed satisfy the bandwidth requests of the third services in all second slave stations.

[0069] Specifically, if the time slot resources remaining after removing the time slot resources required by the bandwidth requests of the first services in all the second slave stations satisfy the bandwidth requests of the second services in all the second slave stations, it indicates that the time slot resources remaining after removing the time slot resources required by the bandwidth requests of the first services and the second services in all the second slave stations are sufficient to satisfy the bandwidth requests of the third services in all the second slave stations. The third services can be video data services, such as video data, to avoid picture jitter. The third services can be nrtPS services (i.e., Non-Real-Time Polling Service), and the allocation priority of the third services can be set as the third priority. When it is confirmed that there are still surplus time slot resources after satisfying the bandwidth requests of the first services and the second services, it is further determined whether the bandwidth requests of the third services are satisfied. After ensuring that all high-priority services (the first services and the second services) are fully guaranteed, the remaining time slot resources are pure "idle resources". If no determination is made, these resources can be wasted. If the time slot resources remaining after removing the time slot resources required by the bandwidth requests of the first services and the second services in all the second slave stations do not satisfy the bandwidth requests of the third services in all the second slave stations, step S208 is performed. If the bandwidth requests of the third services in all the second slave stations can be satisfied, step S209 is performed.

[0070] This can ensure that each time slot resource is used as much as possible to improve the overall network performance and user experience, from the first services with the highest priority to the third services with the lowest priority. This recursive determination mechanism makes the allocation of resources and the degradation of service quality present a clear and predictable hierarchy: the network planner can clearly know that when the total resources are greater than the demand of (the first services + the second services + the third services), all the first services, the second services, and the third services experience perfect. When the total resources are between the demand of (the first services + the second services) and the demand of (the first services + the second services + the third services), the first services and the second services experience perfect, and the third services experience degradation. When the total resources are between the demand of the first services and the demand of (the first services + the second services), the first services experience perfect, the second services experience degradation or partial interruption, and the third services cannot be used. This predictability is crucial for service level agreements, capacity planning, and troubleshooting. Determining whether the time slot resources remaining after satisfying the bandwidth requests of the first services and the second services satisfy the bandwidth requests of the third services ensures that all the remaining time slot resources are efficiently allocated to the low-priority services after the high-priority services are guaranteed. A clear and predictable quality of service ladder is built, making the system behave as expected under different loads.

[0071] Step S208, rank the priorities of all third traffics; and according to the ranking results of the priorities of all third traffics, allocate time slot resources for the third traffics of all second slaves in turn according to the principle of ensuring the bandwidth requirements of the third traffics, until all the remaining time slot resources are allocated.

[0072] Specifically, when the resource allocation proceeds to the third traffic, the same strict policy of "internal ranking, in turn allocation, ensuring requirement" as the first traffic and the second traffic is still adopted, instead of switching to the more relaxed "proportional fairness", in order to ensure that the "complete availability of traffic" is prioritized in any case, rather than the "average fairness of traffic". If the third traffic switches to "proportional fairness", it will cause inconsistency of rules, making the system behavior complex and unpredictable. Therefore, if the remaining time slot resources after removing the time slot resources required by the bandwidth requests of the first traffic and the second traffic in all second slaves cannot meet the bandwidth requests of the third traffic in all second slaves, rank the priorities of all third traffics; and according to the ranking results of the priorities of all third traffics, allocate time slot resources for the third traffics of all second slaves in turn according to the principle of ensuring the bandwidth requirements of the third traffics, until all the remaining time slot resources are allocated. Wherein, the priority determination formula of the third traffic in all second slaves is as follows:

[0073] , wherein, is the priority of the user's first class of the third traffic; is the guaranteed rate of the third traffic; is the current rate of the third traffic.

[0074] ​For any priority tier, when the slot resources are insufficient to meet all the traffic of that tier, the priority rules are applied again within that tier to ensure that the highest value subset can run at 100%. This is true for the first traffic, the second traffic, and the third traffic. Both the network planner and the user know that, among any two peers of the same tier, the one explicitly marked as higher priority will be guaranteed first in resource contention. Although the overall importance of the third traffic is lower than that of the first and second traffic, it also contains traffic that is of higher importance than others within it for the user experience or system functionality. This strategy makes the system's service degradation "cascade by value tier" rather than "uniformly degrade across the board". Uniform degradation is where everything slows down, but everything also kind of works. This sounds fair, but in reality it can cause the experience of all users to be below an acceptable threshold. If the system explicitly preserves the highest value traffic at the expense of the lowest value traffic. This guarantees that even under heavy pressure, the services that remain are of high quality and usable. For the end user, "some functionality is completely unusable" can be a better experience than "all functionality is painfully slow".

[0075] Step S209, allocate slot resources for the third traffic in all second slave stations in turn according to the principle of ensuring the bandwidth rate requirement of the third traffic.

[0076] Specifically, if the time slot resources remaining after removing the time slot resources required by the bandwidth requests of the first services and the second services in all the second slaves can satisfy the bandwidth requests of the third services in all the second slaves, it indicates that the current time slot resources are sufficient, and the time slot resources can be allocated to the third services in all the second slaves in sequence according to the principle of ensuring the bandwidth rate requirements of the third services. In practice, if the system simultaneously sends the “grant” signals to all the second slaves corresponding to the third services to allow them to immediately start using the allocated time slots, instantaneous conflicts or resource competitions may be caused at the physical layer or the MAC layer, even if the time slots are theoretically separated. Such competitions may cause data packet collisions and thus reduce the efficiency. Therefore, the third services can be allocated and activated in sequence according to a certain order (such as priority or slave ID number). This ensures that the channel is orderly even during the allocation process, and avoids any potential instantaneous overload or competition. Under the premise of “satisfaction”, all the third services can obtain full bandwidth. However, who gets the resources first is still crucial for the third services which are extremely sensitive to delay. “Allocation in sequence” is a fine measure of “further optimizing the delay of key services on the basis of ensuring bandwidth”. In practice, calculating and locking the specific time slot position for a service requires the consumption of the computing resources of the system. If parallel processing is used to calculate the time slot mapping for hundreds or thousands of third services at the same time, the CPU load of the controller may instantaneously surge, causing a processing bottleneck. If serial processing (allocation in sequence) is used, the system can smoothly distribute the computing tasks in a short period of time, avoid the instantaneous peak of computing resources, and make the system run more stably. Since the total resources are sufficient, such slight and orderly allocation delay has little effect on the services.

[0077] In step S210, after the time slot resources for the third services of all the second slaves are allocated in sequence, it is determined whether the time slot resources remaining after removing the time slot resources required by the bandwidth requests of the first services, the second services and the third services in all the second slaves can satisfy the bandwidth requests of the fourth services in all the second slaves.

[0078] In particular, in order to achieve global optimization and absolute fairness of resource allocation while strictly guaranteeing the SLA of high-priority services, after allocating time slot resources for the third services of all second slaves in turn, it is determined whether the current remaining time slot resources, excluding the time slot resources required by the bandwidth requests of the first services, the second services and the third services in all second slaves, satisfy the bandwidth requests of fourth services in all second slaves. The fourth services can be basic data services, and can include services such as online services, data download services, data alarm services, device parameter reporting services, etc. For example, the fourth services can be BE (Best Effort) services, and the allocation priority of the fourth services can be set as a fourth priority. In practice, when network resources are insufficient, the performance of high-priority services (the first services, the second services and the third services) must be absolutely guaranteed and cannot be affected by low-priority fourth services. If no remaining resource judgment is performed, and resource allocation for the fourth services is directly started, "resource fragmentation" or "over-allocation" can occur. That is, although the total remaining resources can satisfy the total requests of the fourth services in number, the distribution of these resources on the time axis can be scattered and cannot be effectively allocated to multiple fourth service streams. When the scheduler attempts to satisfy these scattered requests, it can inadvertently occupy "buffer resources" that should be reserved for high-priority services (new data of the first services, the second services or the third services that can come in the next scheduling period), thereby violating the commitment to the SLA thereof. If after allocation for the third services, the fourth services are immediately allocated as soon as scattered resources are available, the fourth services that are scheduled first can occupy resources, and the fourth services that are scheduled later can have no resources available. This is a kind of "first come, first served" local fairness, which is unfair to the fourth services. If the demands of all high-priority services are first aggregated, and the total remaining resources at the system level are calculated, then the total demands of all fourth services are uniformly reviewed. If the total resources are sufficient, the resources can be allocated to all fourth services in proportion or according to the weight in batches, to ensure that each fourth service can share the resources, and global fairness among all fourth services is achieved. If the total resources are insufficient, a unified degradation strategy can be adopted, such as proportionally reducing all fourth services, which is also a kind of global fairness. Calculating the total remaining resources at one time and then performing batch allocation is more efficient than the incremental allocation mode of "one allocation at a time", produces less resource fragmentation, and can better plan time slots to allocate continuous time slot blocks to the fourth services, which is very advantageous for transmitting large data blocks and can reduce encapsulation overhead and improve transmission efficiency. If the current remaining time slot resources, excluding the time slot resources required by the bandwidth requests of the first services, the second services and the third services in all second slaves, satisfy the bandwidth requests of the fourth services in all second slaves, step S211 can be performed.If the bandwidth request of the fourth service in all the second slave stations cannot be satisfied, step S212 or step S213 is performed.

[0079] In step S211, time slot resources are allocated to each fourth service of each second slave station in turn according to the bandwidth request of the fourth service in all the second slave stations.

[0080] Specifically, if the time slot resources remaining after removing the time slot resources required by the bandwidth requests of the first services, the second services and the third services in all the second slave stations satisfy the bandwidth requests of the fourth services in all the second slave stations, it indicates that the remaining time slot resources are sufficient, and the time slot resources can be allocated according to the resource requests of the fourth services in all the second slave stations in turn. In practice, when the system judges that the remaining resources satisfy the total request of all the fourth services, why still “allocate in turn according to the resource requests of the fourth services in all the slave stations”, instead of allocating in one batch, mainly because a general and stable process is needed to process all levels of services. This process is usually: “select a service flow -> calculate and allocate resources for it -> update the remaining resources -> select the next service flow”. The first services, the second services and the third services adopt “allocation in turn”, and then the third services also adopt the same way, which guarantees the simplicity and unity of the code and logic. This reduces the complexity and maintenance cost of the system. If more complex scheduling rules are introduced in the future, this “allocation in turn” framework can easily integrate these rules. If it is designed as a simple “batch average allocation”, it will be very difficult to add any differentiated processing later. The “fourth services” is a large category, but there may still be subtle priority differences or specific fairness principles to be followed between different service flows inside it. “Allocation in turn” provides operational space for implementing these fine-grained strategies. For example, although they are all fourth services, fourth service flows from different slave stations or different applications may have different “weights”. When the scheduler processes “in turn”, it can allocate according to this weight order, allowing fourth service flows with higher weights to prefer better time slot resources. Classic fair scheduling algorithms, such as “round robin” or “weighted fair queueing”, essentially serve in multiple queues in turn. Through “allocation in turn”, polling can be easily implemented among all fourth service flows, ensuring that each fourth service flow will not be starved, thereby maintaining fairness within the fourth services. Resource allocation is not just a matter of addition and subtraction, but also a mapping of service data to specific, physical time slot locations. “Allocation in turn” allows the actual layout of resources to be considered at each allocation. When scheduling, algorithms such as “first fit” or “best fit” can be used to find the most suitable, contiguous block of time slots for each fourth service flow, allowing more compact use of resources and reducing small fragments that cannot be used. Secondly, some services may be sensitive to latency (even if they are fourth priority), and want to be scheduled as soon as possible. In the process of “allocation in turn”, the fourth service flows that are scheduled first naturally get time slots earlier in time. During the allocation process, the state of the system may change. A long, one-time batch allocation calculation may be outdated by the time it is executed. If a high-priority first service or second service arrives suddenly (such as an urgent control signaling) during the allocation of the fourth services, it needs to be responded to immediately.The procedure of "sequential allocation" can be easily interrupted, and the resource allocation for the fourth service can be suspended to deal with the high priority service, and then the remaining resource allocation for the fourth service can be continued after the high priority service is dealt with. This is difficult to achieve in the batch allocation mode.

[0081] In step S212, all the remaining time slot resources are sequentially allocated to each fourth service in all the second slave stations according to the proportional fair allocation principle.

[0082] Specifically, if the remaining time slot resources after removing the time slot resources required by the bandwidth requests of the first service, the second service and the third service in all the second slave stations are not enough to meet the bandwidth requests of the fourth service in all the second slave stations, it is indicated that the remaining time slot resources are insufficient, and then all the remaining time slot resources are sequentially allocated to each fourth service in all the second slave stations according to the proportional fair allocation principle. The proportional fair principle embodies the fair concept of "all the same services share the shortage". The resource reduction proportion obtained by each fourth service is the same. If a service requests more resources, the absolute amount of reduction is also large; if the request is small, the absolute amount of reduction is also small. But the "pain degree" of everyone is the same. A key guarantee of proportional allocation is that as long as there is a little remaining resource in the system, each fourth service can be allocated a part. This ensures that no fourth service connection will be timed out or interrupted due to continuous lack of resources. Although the quality of service is reduced, the connection is maintained and the service can survive. Avoiding some users or services being completely "marginalized". The fourth service itself does not make any guarantee on the delay and bandwidth. The design goal is to serve as many people as possible and fairly when there is remaining resource in the network.

[0083] In step S213, the total time slot resources required by the fourth services of all the second slave stations are calculated, and the mode of increasing the modulation rate to meet the resource requests of all the fourth services is determined according to the current remaining time slot resources.

[0084] Specifically, in order to "create" more virtual resources by improving spectral efficiency, so as to solve the resource conflict without reducing the actual data amount. If the remaining time slot resources after removing the time slot resources required by the bandwidth requests of the first service, the second service and the third service in all second slave stations are not enough to meet the bandwidth request of the fourth service in all second slave stations, the total time slot resources required by the fourth service of all second slave stations can also be calculated, and the mode (MCS) of the modulation rate that needs to be increased is determined according to the current remaining time slot resources, so as to meet the resource request of all fourth services. The MCS level determines the number of data bits that can be carried by each time slot (symbol). The spectral efficiency of high-order MCS (such as 64QAM, 256QAM) is much higher than that of low-order MCS (such as QPSK). For example, when allocating resources for a service, there is a trade-off between "time slot number" and "MCS level". To transmit the same amount of data, "more time slots + lower MCS" can be used, or "fewer time slots + higher MCS" can be used. When it is clear how much the current resources are different, it is explored whether it is possible to fill the gap by improving the MCS. If the MCS of all C-level services is increased by one or more levels, how many time slots are needed to transmit the same amount of data? It means that by improving the MCS, the originally insufficient physical time slots are now enough! This is a "quality for quantity" strategy, which exchanges higher signal processing cost (which may cause the anti-noise ability to decrease) for more compact time slot occupation. This can guarantee the user data throughput and optimize the user experience. Compared with direct proportional allocation (which means directly discarding part of the data), the MCS improvement strategy completely transmits the data amount requested by all fourth services. Users will not feel a rate decrease or an increase in delay, and the experience is better. The valuable time slot resources can be used more efficiently. The system improves the overall throughput of the system by exploiting the potential of the physical layer to transmit more data in the same time and space resources. For the fourth service, simple proportional reduction is a "passive" best effort. While actively improving the MCS is a "proactive" best effort. Therefore, when the remaining time slot resources are insufficient, the total demand of all fourth services and the remaining resources are calculated, and it is judged whether the MCS can be improved to meet the demand, which is an efficient and intelligent scheduling optimization strategy. By balancing between the physical layer and the link layer, when the resources are insufficient, it is tried to compress the resource demand by improving the efficiency first, rather than directly reducing the data amount, so as to achieve a win-win in guaranteeing the service quality and improving the system efficiency. This is one of the core innovations and values in your patent scheme. In general, the bandwidth request of each service carries the size of each QoS queue buffer. After receiving, the AP needs to combine the current MCS of the scheduling, the MAC header overhead, the delimiter overhead, the pilot overhead, etc. to calculate the corresponding symbol number and time slot number, and then make a judgment on resource estimation.

[0085] In practice, after allocating the required time slot resources for each fourth service of all second slave stations, it can be determined whether there are remaining time slot resources at the current time; if there are no remaining time slot resources at the current time, it is determined that the total time slot resource allocation is complete; if there are remaining time slot resources at the current time, the remaining time slot resources are supplemented to each second service in all second slave stations according to the priority of each second service in all second slave stations to meet the bandwidth request thereof; it is determined whether there are remaining time slot resources after the remaining time slot resources at the current time are supplemented to each second service in all second slave stations; if there are remaining time slot resources after the remaining time slot resources at the current time are supplemented to each second service in all second slave stations, the remaining time slot resources are supplemented to each third service in all second slave stations according to the bandwidth request of each third service until all time slot resources are allocated.

[0086] Specifically, through the above scheme, the first stage resource allocation target is "to ensure survival". The system strictly follows the priority to ensure that the first service is 100% satisfied, then the second service, then the third service, and finally the fourth service if there is a surplus. The second stage resource reallocation target is "to pursue excellence". When the basic requests of all services are met, if there is still surplus resources, the system will not idle it, but will supplement it to high-priority services in reverse to exceed their basic needs and improve their performance. Allocating the remaining time slot resources to high-priority services can produce the greatest marginal benefit. Therefore, the remaining resources are preferentially returned to the second service that can produce the greatest value, and if there is still a surplus, it is returned to the third service. Therefore, after allocating the time slot resources for each fourth service of all second slave stations, it can be determined whether there are remaining time slot resources at the current time; if there are no remaining time slot resources at the current time, it is determined that the time slot resource allocation is complete; if there are remaining time slot resources at the current time, the remaining time slot resources are supplemented to each second service in all second slave stations according to the priority of each second service in all second slave stations to meet the bandwidth request thereof. This is to preferentially improve the experience of the highest value service. The system will allocate additional resources to each second service in order according to the priority within the second service until the performance of the service reaches an upper limit (or its request is completely met and cannot utilize more), or the remaining resources are exhausted. After supplementing the second service, it can be determined whether there are remaining time slot resources after the remaining time slot resources at the current time are supplemented to each second service in all second slave stations; if there are remaining time slot resources after the remaining time slot resources at the current time are supplemented to each second service in all second slave stations, it means that the resources are abundant, and the remaining time slot resources are supplemented to each third service in all second slave stations according to the bandwidth request of each third service to improve the performance thereof until all time slot resources are allocated, which can achieve complete utilization of resources.

[0087] For example, assume the total resource is 1000Mbps. The first service requests 50Mbps in total; the second service requests 200Mbps in total (including Al: 80Mbps, A2: 120Mbps); the third service requests 300Mbps in total; and the fourth service requests 150Mbps in total. In the first stage, the first service is allocated 50Mbps, leaving 950Mbps; then the second service is allocated 200Mbps, leaving 750Mbps; then the third service is allocated 300Mbps, leaving 450Mbps; then the fourth service is allocated 150Mbps, leaving 300Mbps; at this point, the basic requirements of all services have been met. In the second stage, the remaining time slot resources are allocated. According to the priority, the first Al service is supplemented. Al can have a bandwidth upper limit of 100Mbps, so it can still absorb 20Mbps at most, so 20Mbps is allocated to Al, leaving 280Mbps. Then the A2 service is supplemented, and A2 can have a bandwidth upper limit of 150Mbps, so it can still absorb 30Mbps at most, so 30Mbps is allocated to A2, leaving 250Mbps. At this point, the second service can no longer absorb more resources, but there are still 250Mbps left, so 250Mbps can be allocated to all third services in proportion or according to other fair policies. The third service originally requested a total of 300Mbps, but now actually obtains a total bandwidth of "300+250=550Mbps", and the performance is greatly improved until the resources are exhausted.

[0088] The wireless resource scheduling device provided in the present application is described below. The wireless resource scheduling device described below can be referred to in correspondence with the wireless resource scheduling method described above. Referring to Figure 2 , Figure 2 is a structural schematic diagram of a wireless resource scheduling device. As shown in Figure 2As shown, the wireless resource scheduling apparatus can comprise: a first determining unit 101, configured to determine a first slave station and a second slave station of a target wireless system based on a preset first priority principle when a preset wireless resource scheduling time arrives, wherein the bandwidth mode of the first slave station is a fixed bandwidth mode, and the bandwidth mode of the second slave station is a competitive bandwidth mode; an obtaining unit 102, configured to obtain a bandwidth request value of a quality of service level reported by each first slave station and each second slave station to a master station; a first allocating unit 103, configured to allocate time slot resources to each first slave station based on the bandwidth request value reported by each first slave station according to a preset first allocation principle; a calculating unit 104, configured to calculate a current remaining time slot resource; a second determining unit 105, configured to determine an allocation priority of each service type of each second slave station based on each service type of each second slave station; and a second allocating unit 106, configured to allocate the current remaining time slot resource to each second slave station in turn according to a preset second allocation principle until all time slot resources are allocated, based on the bandwidth request value reported by each second slave station and the allocation priority of each service type of each second slave station. The specific processing procedure of each unit comprised in the wireless resource scheduling apparatus can refer to the relevant description in the foregoing wireless resource scheduling method, and will not be described here again.

[0089] The wireless resource scheduling apparatus provided in the application can be applied to wireless resource scheduling equipment, such as terminals, mobile phones, computers, and the like. Optionally, Figure 3 A hardware structure block diagram of the wireless resource scheduling equipment is shown, which can refer to Figure 3The hardware structure of the wireless resource scheduling device can include at least one processor 1, at least one communication interface 2, at least one memory 3 and at least one communication bus 4. In the present application, the number of the processor 1, the communication interface 2, the memory 3 and the communication bus 4 is at least one, and the processor 1, the communication interface 2 and the memory 3 complete the communication with each other through the communication bus 4. The processor 1 can be a central processing unit CPU, or an application specific integrated circuit ASIC, or one or more integrated circuits configured to implement the embodiments of the present application, etc.; the memory 3 can include a high-speed RAM memory, and can also include a non-volatile memory such as at least one disk memory, etc.; wherein the memory stores a program, and the processor can call the program stored in the memory, and the program is used to implement the various processing procedures in the terminal wireless resource scheduling scheme. The present application also provides a readable storage medium, which can store a program suitable for the processor to execute, and the program is used to implement the various processing procedures of the terminal in the wireless resource scheduling scheme. Finally, it should be noted that in this paper, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the element defined by the statement "including a" does not exclude the existence of other identical elements in the process, method, article or device including the element. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between the various embodiments can be referred to each other. The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. The various embodiments can be combined with each other. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A wireless resource scheduling method, characterized in that, include: When the preset wireless resource scheduling time arrives, based on the preset first priority principle, the first slave station and the second slave station of the target wireless system are determined, wherein the bandwidth mode of the first slave station is fixed bandwidth mode and the bandwidth mode of the second slave station is contention bandwidth mode. Obtain the bandwidth request value of the quality of service level reported to the master station by each of the first slave station and each of the second slave stations; Based on the bandwidth request value reported by each of the first slave stations, the total time slot resources are allocated to each of the first slave stations according to the preset first allocation principle; Calculate the remaining time slot resources; Based on each service type of each second slave station, determine the allocation priority of various services for each second slave station; Based on the bandwidth request value reported by each second slave station and the allocation priority of each service of each second slave station, the remaining time slot resources are allocated to each second slave station in sequence according to the preset second allocation principle, until all time slot resources are allocated.

2. The method according to claim 1, characterized in that, Based on the bandwidth request value reported by each second slave station and the allocation priority of each service of each second slave station, the remaining time slot resources are allocated to each second slave station in sequence according to a preset second allocation principle, including: Based on the bandwidth request value reported by each second slave station, determine whether the remaining time slot resources can meet the bandwidth requests of all second slave stations. If satisfied, then based on the allocation priority of various services of each second slave station, the remaining time slot resources will be allocated to various services of each second slave station in sequence. If not, determine whether the remaining time slot resources can satisfy the bandwidth request of the first service in all second slave stations; If the remaining time slot resources satisfy the bandwidth request of the first service in all the second slave stations, then determine whether the remaining time slot resources, excluding the time slot resources required for the bandwidth request of the first service in all the second slave stations, satisfy the bandwidth request of the second service in all the second slave stations. If the remaining time slot resources can satisfy the bandwidth requests of the second services in all the second slave stations, then determine whether the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first and second services in all the second slave stations, can satisfy the bandwidth requests of the third services in all the second slave stations. If the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first and second services in all the second slave stations, can satisfy the bandwidth requests of the third services in all the second slave stations, then time slot resources will be allocated to the third services in all the second slave stations in sequence according to the principle of ensuring the bandwidth rate requirements of the third services. After allocating time slot resources to the third service of all second slave stations in sequence, determine whether the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first, second and third services in all second slave stations, can satisfy the bandwidth requests of the fourth service in all second slave stations. If the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first, second, and third services in all the second slave stations, can satisfy the bandwidth requests of the fourth services in all the second slave stations, then time slot resources will be allocated to each of the fourth services in each of the second slave stations in turn according to the bandwidth requests of the fourth services in all the second slave stations.

3. The method according to claim 2, characterized in that, The method also includes: If the remaining time slot resources are insufficient to meet the bandwidth requests of the first services in all second slave stations, then the priorities of the first services in all second slave stations are ranked. Based on this ranking, time slot resources are allocated sequentially to the first services in all second slave stations, ensuring their bandwidth requirements are met, until all remaining time slot resources are allocated. The formula for determining the priority of the first services in all second slave stations is as follows: ； in, For the first business User number Priority of levels; To ensure the rate of service for the primary business; The current rate of the first service; Tolerable latency for the service quality level of the primary service; This represents the current latency of the first service.

4. The method according to claim 2, characterized in that, The method also includes: If the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first services in all the second slave stations, are insufficient to satisfy the bandwidth requests of the second services in all the second slave stations, then the priority of the second services in all the second slave stations is ranked. Based on the priority ranking of all the second services, time slot resources are allocated sequentially to the second services in all the second slave stations according to the principle of ensuring the bandwidth requirements of the second services, until all remaining time slot resources are allocated. The formula for determining the priority of the second services in all the second slave stations is as follows: ； in, For the second business User number Priority of levels; To ensure the rate of service for the second business; The current rate for the second service; Tolerable latency for the service quality level of the second service; This represents the current latency of the second service.

5. The method according to claim 2, characterized in that, The method also includes: If the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first and second services in all second slave stations, are insufficient to satisfy the bandwidth requests of the third services in all second slave stations, then all third services are prioritized. Based on this priority ranking, time slot resources are allocated sequentially to the third services in all second slave stations according to the principle of ensuring their bandwidth requirements, until all remaining time slot resources are allocated. The formula for determining the priority of the third services in all second slave stations is as follows: ； in, For the third business User number Priority of levels; To ensure the rate of third-party services; This is the current rate for the third service.

6. The method according to claim 2, characterized in that, The method includes: If the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first, second, and third services in all the second slave stations, are insufficient to satisfy the bandwidth requests of the fourth services in all the second slave stations, then, according to the principle of proportional fair allocation, all the remaining time slot resources will be allocated sequentially to each of the fourth services in all the second slave stations.

7. The method according to claim 2, characterized in that, The method also includes: If the remaining time slot resources, after deducting the time slot resources required for the bandwidth requests of the first, second, and third services in all second slave stations, are insufficient to meet the bandwidth requests of the fourth service in all second slave stations, then the total time slot resources required for the fourth service in all second slave stations are calculated. Based on the remaining time slot resources, the mode that needs to increase the modulation rate is determined to meet the resource requests of all fourth services.

8. The method according to claim 2, characterized in that, The method also includes: After allocating the required time slot resources to each of the fourth services of all the second slave stations, determine whether there are still time slot resources available at the current moment; If there are no remaining time slots at the current moment, then the total time slot resources have been allocated. If there are remaining time slot resources at the current moment, the remaining time slot resources will be supplemented to each of the second services in all the second slave stations according to the priority of each second service in all the second slave stations to meet their bandwidth requests. Determine whether there are any remaining time slot resources after allocating the remaining time slot resources at the current moment to each of the second services in all the second slave stations; If, after allocating the remaining time slot resources at the current moment to each of the second services in all the second slave stations, there are still time slot resources remaining, then these remaining time slot resources will be allocated to each of the third services in all the second slave stations according to their bandwidth requests, until all time slot resources have been allocated.

9. The method according to claim 1, characterized in that, The step of determining the allocation priority of various services for each second slave station based on each service type of each second slave station includes: The priority for assigning control signaling data services is set to first priority. The priority for assigning voice and data services is the second priority. The priority for assigning video data services is the third priority. The priority of basic data services is assigned as the fourth priority; the priority levels of the first, second, third and fourth priorities decrease in that order.

10. A wireless resource scheduling device, characterized in that, include: The first determining unit is used to determine the first slave station and the second slave station of the target wireless system based on a preset first priority principle when the preset wireless resource scheduling time arrives, wherein the bandwidth mode of the first slave station is a fixed bandwidth mode and the bandwidth mode of the second slave station is a contention bandwidth mode. The acquisition unit is used to acquire the bandwidth request value of the quality of service level reported by each of the first slave stations and each of the second slave stations to the master station; The first allocation unit is used to allocate time slot resources to each of the first slave stations based on the bandwidth request value reported by each of the first slave stations, according to a preset first allocation principle. The computing unit is used to calculate the currently remaining time slot resources; The second determining unit is used to determine the allocation priority of various services of each second slave station based on each service type of each second slave station; The second allocation unit is used to allocate the remaining time slot resources to each of the second slave stations in sequence according to the bandwidth request value reported by each of the second slave stations and the allocation priority of each service of each of the second slave stations, until all time slot resources are allocated.

11. A wireless resource scheduling device, characterized in that, include: One or more processors, and a memory; the memory stores computer-readable instructions that, when executed by the one or more processors, implement the steps of the wireless resource scheduling method as described in any one of claims 1 to 9.

12. A readable storage medium, characterized in that: The readable storage medium stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the wireless resource scheduling method as described in any one of claims 1 to 9.

Images