Methods, apparatus and equipment for scheduling and transmitting business data

Abstract

This application provides a method, apparatus, and device for scheduling and transmitting service data. The scheme includes: obtaining the service priority of a first service; selecting a target frequency band from a frequency band pool corresponding to the frequency band priority of the service priority, based on the mapping relationship between the service priority and frequency band priority, wherein the service priority and / or frequency band priority can be dynamically adjusted according to a preset strategy; and allocating service fragments of the first service to the target frequency band for transmission. This application not only allocates high-frequency band resources to services with high service priority but also dynamically adjusts the mapping relationship between service priority and frequency band priority, further improving matching accuracy and providing end-to-end latency guarantees for high-priority services while improving frequency band resource utilization.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method, apparatus and device for scheduling and transmitting service data. Background Technology

[0002] In the current field of wireless communication, there is a contradiction between the scarcity of frequency band resources and the surge in bandwidth demand. To alleviate this contradiction and improve transmission capacity, existing technologies often employ a multi-band parallel transmission mechanism. As an example, parallel transmission is carried out on a shared transmission channel with a fixed total physical bandwidth. This channel can be divided into several sub-channels, each allocated to a different frequency band (e.g., band A, band B, and band C) to simultaneously carry the data transmission tasks of multiple wireless radio frequency signals.

[0003] Existing technologies do not take into account the characteristics of frequency bands A, B, and C, such as the unit data delay of the frequency bands. For example, if the unit data delay of frequency band A is less than that of frequency band B, frequency band A may be used to transmit services that are not sensitive to unit data delay. However, when a service that is sensitive to unit data delay arrives, it is necessary to wait for the service data transmission on frequency band A to be completed before it can be used to transmit this service that is sensitive to unit data delay, thus resulting in a technical problem of low resource utilization. Summary of the Invention

[0004] This application provides a method, apparatus, and device for scheduling and transmitting business data to solve the technical problem of low resource utilization in the prior art.

[0005] In a first aspect, embodiments of this application provide a data scheduling and transmission method, comprising: obtaining the service priority of a first service; selecting a target frequency band from a frequency band pool corresponding to the frequency band priority of the service priority, based on the mapping relationship between the service priority and frequency band priority, wherein the service priority and / or frequency band priority can be dynamically adjusted according to a preset strategy; and allocating service fragments of the first service to the target frequency band for transmission. The technical solution provided by embodiments of this application not only allocates high-frequency band resources to services with high service priority but also dynamically adjusts the mapping relationship between service priority and frequency band priority, further improving the accuracy of matching. This not only improves the utilization rate of frequency band resources but also provides end-to-end latency guarantees for high-priority services.

[0006] In one possible implementation, the preset strategy includes: adjusting the service priority of the first service based on its comprehensive evaluation value and a preset comprehensive evaluation value, wherein the comprehensive evaluation value is determined at least by the service priority of the first service, the remaining data volume of the first service, the unit data delay of the initial frequency band, and the remaining service time of the remaining service fragments in the initial frequency band; and / or adjusting the frequency band priority of the initial frequency band based on the unit data delay and a delay threshold, wherein the delay threshold is determined at least by the historical delay of the initial frequency band, the current load, and the volume of high-priority services. In this embodiment, during the service fragmentation transmission process, the service priority can be dynamically adjusted based on the comprehensive evaluation value of the service, thereby reallocating frequency band resources according to the adjusted service priority, thus improving the accuracy of frequency band resource allocation and increasing resource utilization. Similarly, the frequency band priority can also be dynamically adjusted based on the frequency band delay threshold, thereby reducing the load on high-priority frequency bands when the volume of high-priority services increases, and improving resource utilization when the volume of high-priority services decreases.

[0007] In one possible implementation, the frequency band priority of the initial frequency band is adjusted based on the unit data latency of the initial frequency band and the latency threshold. This includes: if the unit data latency of the initial frequency band is greater than a first preset multiple of the latency threshold, the frequency band priority of the initial frequency band is reduced, wherein the first preset multiple is greater than 1; or if the unit data latency of the initial frequency band is less than a second preset multiple of the latency threshold, and the current load is less than or equal to a preset percentage, the frequency band priority of the initial frequency band is increased, wherein the second preset multiple is greater than zero and less than 1. Through this dynamic adjustment mechanism, the allocation of frequency band resources can closely align with actual business needs, thereby ensuring the service quality of high-priority services while also improving the overall utilization rate of system resources.

[0008] In one possible implementation, the method further includes: determining whether the estimated waiting time of the remaining service fragments of the first service in the target frequency band exceeds a preset threshold; if so, determining a second frequency band for transmitting the remaining service fragments from a frequency band pool corresponding to the frequency band priority of the first service's own service priority, wherein the estimated waiting time of the remaining service fragments in the second frequency band is less than the estimated waiting time of the remaining service fragments in the target frequency band, and the difference between the unit service time of the second frequency band and the unit service time of the target frequency band is within a preset time range, wherein the estimated waiting time is the time for the remaining service fragments to wait for transmission to be completed in each frequency band. In this embodiment, when the estimated waiting time of a remaining service fragment in a frequency band is too long, the remaining service fragment can be switched to a frequency band with slightly higher unit data latency but shorter estimated waiting time, thereby reducing the overall waiting time of the service and ensuring minimal end-to-end latency.

[0009] In one possible implementation, determining a second frequency band for transmitting the remaining service fragment from a frequency band pool corresponding to the frequency band priority of the first service's own service priority includes: determining whether the difference between the time cost value of the remaining service fragment in the target frequency band and the time cost value of the remaining service fragment in the second frequency band is greater than a preset time cost value, wherein the time cost value of the remaining service fragment in the target frequency band is determined by the estimated waiting time of the remaining service fragment in the target frequency band and the unit data delay of the target frequency band; if it is greater than the preset time cost value, determining a second frequency band for transmitting the remaining service fragment from a frequency band pool corresponding to the frequency band priority of the first service's own service priority. This technical solution avoids invalid frequency band switching.

[0010] In one possible implementation, if the time factor exceeds the preset time value, the remaining service fragments are switched from the target frequency band to the second frequency band. This includes: pausing the transmission of the first service fragment on the target frequency band; switching the remaining service fragments from the target frequency band to the second frequency band; and sending switching information to the receiving device. The switching information includes at least the frequency band identifier of the second frequency band and the sequence number of the first service fragment transmitted on the second frequency band. In this embodiment, during frequency band switching, switching information is also sent to the receiving device, thereby enabling lossless frequency band switching.

[0011] In one possible implementation, upon receiving a second service, the service priority of the second service is the same as the priority of the first service. The method further includes: determining whether the volume of high-priority services is greater than or equal to a security threshold, wherein the security threshold is obtained by weighted calculation based on a basic capacity threshold, a latency sensitivity threshold, and a system stability threshold; if it is greater than the security threshold, then a third frequency band is selected from the frequency band pool corresponding to the frequency band priority of the second service, the frequency band priority of the third frequency band is increased, and the third frequency band is allocated to the frequency band pool corresponding to the adjusted frequency band priority, wherein the priority level of the first service is higher than the priority level of the second service. This technical solution ensures sufficient frequency band resources for services with a priority level of first service, guaranteeing system stability.

[0012] Secondly, embodiments of this application also provide a scheduling and transmission apparatus for service data, comprising:

[0013] The scheduling module is used to obtain the service priority of the first service; according to the mapping relationship between the service priority and the frequency band priority, it selects the target frequency band that can minimize the end-to-end delay from the frequency band pool corresponding to the frequency band priority of the service priority. The service priority and / or frequency band priority can be dynamically adjusted according to a preset strategy.

[0014] The transmission module is used to allocate service fragments of the first service to the target frequency band for transmission.

[0015] In one possible implementation, the preset strategy includes:

[0016] Based on the comprehensive evaluation value of the first service and a preset comprehensive evaluation value, the service priority of the first service is adjusted. The comprehensive evaluation value is determined at least by the service priority of the first service and the remaining data volume of the first service. The remaining data volume is the data volume of the remaining service fragments of the first service, and the remaining service fragments are the service fragments of the first service that have not been transmitted; and / or

[0017] The frequency band priority of the target frequency band is adjusted based on the unit data delay and the delay threshold of the target frequency band, wherein the delay threshold is determined at least by the historical delay and current load of the target frequency band, and the unit data delay of the target frequency band is determined at least by the propagation delay and the transmission delay.

[0018] In one possible implementation, the comprehensive evaluation value is further determined by the unit data latency of the target frequency band and the remaining service time of the remaining service slice in the target frequency band, wherein the remaining service time is calculated based on the unit data latency of the target frequency band and the remaining data volume.

[0019] In one possible implementation, the latency threshold is also determined by the volume of high-priority traffic.

[0020] In one possible implementation, the scheduling module is further configured to:

[0021] Determine whether the estimated waiting time of the remaining service segments of the first service in the target frequency band exceeds a preset threshold;

[0022] If so, a second frequency band for transmitting the remaining service fragment is determined from the frequency band pool corresponding to the frequency band priority of the first service's own service priority. The estimated waiting time of the remaining service fragment on the second frequency band is less than the estimated waiting time of the remaining service fragment on the target frequency band, and the difference between the unit service time of the second frequency band and the unit service time of the target frequency band is within a preset time range. The estimated waiting time is the time that the remaining service fragment waits to be transmitted on each frequency band.

[0023] In one possible implementation, when the scheduling module determines the second frequency band for transmitting the remaining service fragments from the frequency band pool corresponding to the frequency band priority of the first service, it specifically performs the following:

[0024] Determine whether the difference between the time cost value of the remaining service slice in the target frequency band and the time cost value of the remaining service slice in the second frequency band is greater than a preset time cost value. The time cost value of the remaining service slice in the target frequency band is determined by the estimated waiting time of the remaining service slice in the target frequency band and the unit data delay of the target frequency band.

[0025] If the time value is greater than the preset time value, a second frequency band for transmitting the remaining service fragments is determined from the frequency band pool corresponding to the frequency band priority of the first service.

[0026] In one possible implementation, when the scheduling module switches the remaining service fragment from the target frequency band to the second frequency band, it is specifically used for:

[0027] Pause the transmission of the first service segment on the target frequency band;

[0028] The remaining service fragments are switched from the target frequency band to the second frequency band, and switching information is sent to the receiving device. The switching information includes at least the frequency band identifier of the second frequency band and the sequence number of the first service fragment transmitted on the second frequency band.

[0029] Thirdly, embodiments of this application also provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes a scheduling and transmission method for service data as described in the first aspect or any of the first aspects.

[0030] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this specification. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of a system architecture shown in one embodiment of this application;

[0032] Figure 2a This is a schematic diagram illustrating, according to an embodiment of this application, that node A and node B are located inside the same device;

[0033] Figure 2b This is a schematic diagram of the structure of node A shown in one embodiment of this application;

[0034] Figure 2c This is a schematic diagram illustrating, according to an embodiment of this application, that node A and node B are located inside different devices;

[0035] Figure 3 This application provides a schematic flowchart illustrating a service data transmission scheduling method according to an embodiment;

[0036] Figure 4 This application provides a flowchart illustrating a method for determining a target frequency band from a frequency band pool, as shown in an embodiment.

[0037] Figure 5 This is a schematic diagram illustrating a process for adjusting the business priority of a service according to an embodiment of this application;

[0038] Figure 6 This is a schematic diagram of the structure of a computer device according to an embodiment of this application. Detailed Implementation

[0039] The exemplary embodiments will now be described in detail. When the description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this specification; they are merely exemplary embodiments of apparatuses and methods consistent with some aspects of this specification.

[0040] The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of this specification. The singular forms “a,” “described,” and “the” as used herein are also intended to include the plural forms unless the context clearly indicates otherwise.

[0041] It should be understood that the terms "first," "second," "third," etc., may be used in this specification to describe various information or structural modules for the purpose of more clearly describing the solution. These terms should not be construed as indicating or implying relative importance or implicitly specifying the number, order, or position of the indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this specification, unless otherwise stated, "a plurality of" means two or more; "if" can be interpreted as "when," "when," or "in response to determination."

[0042] In this specification, "and / or" describes the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following associated objects are in an "or" relationship.

[0043] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0044] Figure 1 This application provides a system architecture 100, which includes nodes 110 and 120, and a transmission medium 130 connecting the two. In this embodiment, the transmission medium 130 can be a physical channel with a fixed total bandwidth, and nodes 110 and 120 can be located within the same device or in different devices.

[0045] In one possible implementation, please refer to Figure 2a Nodes 110 and 120 are located within the same terminal device, achieving chip-level interconnection. For example, Figure 2a The terminal devices include nodes 110 and 120, which are stacked in three dimensions through through-silicon via (TSV) technology. In this case, the transmission medium is a TSV channel 130.

[0046] For details on the implementation process, please refer to [link / reference]. Figure 2bNode 110 includes at least a scheduling module 1101, a transmission module 1102, and a processing module 1103. The scheduling module 1101 can be used to match the frequency band corresponding to the service priority of the service data. The processing module 1103 can be a processing chip, such as a Central Processing Unit (CPU), a Data Processing Unit (DPU), or other types of chips. Node 120 includes at least a scheduling module 1201, a transmission module 1202, and a storage module 1203. The storage module 1203 can be a storage chip, such as a Double Data Rate (DDR) memory, or other types of storage chips.

[0047] In another possible implementation, please see Figure 2c Nodes 110 and 120 are located in different devices, such as Node 110 being a terminal device and Node 120 being a base station. In this case, the transmission medium 130 is an air interface wireless channel between the terminal device and the base station.

[0048] The technical solution provided in this application can be adapted to any scenario with differentiated requirements for the reliability, latency, or priority of service transmission, such as: factory automation (e.g., robotic arm control), telemedicine (e.g., medical image data transmission), autonomous driving and satellite communication, or other application scenarios, without limitation.

[0049] In this embodiment of the application, taking the transmission medium 130 as a TSV channel as an example, since the total physical bandwidth provided by TSV is fixed, it can be provided to wireless radio frequency signals of different frequency bands for data transmission. As an example, the TSV channel can be divided into 3 channels, which are used for wireless radio frequency signal transmission of frequency band A, frequency band B and frequency band C respectively.

[0050] Existing technologies do not take into account the characteristics of frequency bands A, B, and C, such as the unit data delay of the frequency bands. For example, if the unit data delay of frequency band A is less than that of frequency band B, frequency band A may be used to transmit services that are not sensitive to unit data delay. However, when a service that is sensitive to unit data delay arrives, it is necessary to wait for the service data transmission on frequency band A to be completed before it can be used to transmit this service that is sensitive to unit data delay, thus resulting in a technical problem of low resource utilization.

[0051] To address the aforementioned issues, this application provides a method for scheduling and transmitting service data, comprising: obtaining the service priority of a first service; selecting a target frequency band from a frequency band pool corresponding to the frequency band priority of the service priority, based on the mapping relationship between the service priority and frequency band priority, wherein the service priority and / or frequency band priority can be dynamically adjusted according to a preset strategy; and allocating service fragments of the first service to the target frequency band for transmission. The technical solution provided by this application selects a target frequency band from a frequency band pool matching the service priority based on the mapping relationship between the service priority and frequency band priority. The mapping relationship between the service priority and frequency band priority dynamically changes due to adjustments in the service priority and / or frequency band priority, minimizing the end-to-end latency of the service. In other words, the technical solution of this application not only allocates high-frequency band resources to services with high service priority but also dynamically adjusts the mapping relationship between the service priority and frequency band priority, further improving the matching accuracy. This not only improves the utilization rate of frequency band resources but also provides end-to-end latency guarantees for high-priority services.

[0052] Firstly, this application provides a detailed description of a method for scheduling and transmitting service data; please refer to [link to relevant documentation]. Figure 3 The method includes the following steps:

[0053] S301: The scheduling module 1101 obtains the service priority of the first service;

[0054] Before introducing the scheduling and transmission method for this business data, it's important to note that the specific business can be file transfer, video streaming, web page requests, etc. Each business generates data packets according to its own logic. For example, sending a 10MB file will be split into ten 1MB data packets. When the size of a data packet exceeds the maximum transmission unit (MTU), it will be further split into business fragments of the same size. Business data scheduling is then performed according to these business fragments.

[0055] In this embodiment, the scheduling module 1101 acquires service 1, which is the first service. In the specific implementation, the scheduling module 1101 first acquires the service priority information of service 1. The service priority of service 1 is one of at least two service priorities, including a first service priority, a second service priority, and a third service priority. The service priority can be understood as the importance and / or urgency of the service. For example, the first service priority is higher than the second service priority, and the second service priority is higher than the third service priority.

[0056] The scheduling module 1101 obtains the service priority information of service 1. This information can be a direct priority value, such as 0, 1, or 2, or an identifier, index, or tag used to query the priority strategy. The scheduling module 1101 can parse this identifier and map the priority strategy required to process the data stream according to local configuration or predefined rules.

[0057] S302: The scheduling module 1101 selects the target frequency band that is estimated to minimize the end-to-end delay from the frequency band pool that corresponds to the frequency band priority of its own service priority, based on the mapping relationship between its own service priority and frequency band priority. The own service priority and / or frequency band priority can be dynamically adjusted according to a preset strategy.

[0058] Frequency band refers to the range of electromagnetic waves, measured in Hertz (Hz), such as the sub-6GHz band, the Long Term Evolution (LTE) band, and the millimeter-wave band. In this application, in... Figure 1 When the system architecture shown is in the initialization state, the frequency bands that the scheduling module 1101 can use are divided into three priority frequency bands: a first priority frequency band, a second priority frequency band, and a third priority frequency band, based on a fixed latency threshold. The unit data latency of the first priority frequency band is less than that of the second priority frequency band, and the unit data latency of the second priority frequency band is less than that of the third priority frequency band. Specifically, taking a fixed latency threshold of 5 milliseconds (ms) as an example, frequency bands with a unit data latency less than or equal to 5 ms are assigned to the first priority frequency band pool. This means that the unit data latency of each candidate frequency band in the first priority frequency band pool is less than or equal to 5 ms. Correspondingly, frequency bands with a unit data latency greater than 5 ms and less than or equal to 20 ms are assigned to the second priority frequency band pool, and frequency bands with a unit data latency greater than 20 ms are assigned to the third priority frequency band pool. The fixed latency threshold here can be set according to actual conditions and is not limited to 5 ms. It should be noted here that the unit data delay of a frequency band is determined by at least the transmission delay and the propagation delay. The transmission delay is the data packet size of Service 1 divided by the bandwidth of the frequency band it is allocated to, and the propagation delay is the time required for an electromagnetic wave to travel from the transmitting end to the receiving end in the medium.

[0059] In this embodiment of the application, there is an initial mapping relationship between the service priority and the frequency band priority, as detailed in Table 1 below.

[0060] Table 1

[0061]

[0062] In this embodiment of the application, the determination of the business priority of service 1 can include the following two stages, which will be described in detail below.

[0063] Phase 1: The business segment of Business 1 is arriving for the first time.

[0064] In scenario 1, the service priority of service 1 is the initial service priority that is sent along with the service data. In this case, the initial frequency band is determined directly based on the initial mapping relationship in Table 1.

[0065] In the specific implementation process, for information on how to determine the initial frequency band from the frequency band priority pool corresponding to the service priority, please refer to [link to relevant documentation]. Figure 4 Specifically, it includes the following steps:

[0066] S401: The scheduling module 1101 calculates the estimated waiting time of the service fragment of the first service on each candidate frequency band, wherein the estimated waiting time is the time when the first service fragment is transmitted.

[0067] In this embodiment of the application, taking the first service as service 1, the service segment of service 1 as service segment 1, and the service priority of service 1 as the first service priority as an example, it is necessary to calculate the expected waiting time of service segment 1 on each candidate frequency band in the first frequency band priority frequency band pool. The expected waiting time can be understood as the time that service segment 1 waits to be served in the queue of each candidate frequency band, that is, the time that service segment 1 waits to be transmitted to completion in the queue of each candidate frequency band.

[0068] In this embodiment, the estimated waiting time for service segment 1 in each candidate frequency band is the sum of the service times of all service segments already queued in the queue of each candidate frequency band and the remaining service time of the service segment currently being processed. As an example, the estimated waiting time is used... Indicate, then = (Sum of service times for all queued service shards) + Remaining service time for the currently being processed service shard. Here, the currently being processed service shard is service shard 1 of service 1, and all queued services are the service shards preceding service 1. Since service shard 1 is the first service shard of service 1, all queued services can be service shards of other services. Because service time is proportional to data volume, if the remaining service time for the currently being processed service shard 1 is used... Indicate, then =s* Where s is the amount of remaining data currently being processed in business shard 1. This represents the unit data delay for frequency band i. Here, unit data delay can usually refer to the total time that a service segment takes from the sending end to the receiving end, which is the end-to-end delay, and is determined by at least the transmission delay and the propagation delay.

[0069] After executing step S401, step S402 is executed: After calculating the expected waiting time of service segment 1 on each candidate frequency band, the scheduling module 1101 calculates the time cost of service segment 1 on each candidate frequency band based on the expected waiting time of service segment 1 on each candidate frequency band and the unit data delay of each candidate frequency band.

[0070] As an example, we can first construct the time cost function. , =a* +b* Here, a and b are weighting coefficients, and a >> b. This means that the importance of unit data latency is greater than the importance of expected waiting time; in other words, unit data latency is the primary consideration. Therefore, when business segment 1 of business 1 arrives, the time cost function is calculated. The frequency band with the smallest time cost function value is selected from multiple candidate frequency bands as the initial frequency band for transmitting service fragment 1.

[0071] In the specific calculation process, it is assumed that the first frequency band priority frequency band pool of Service 1 includes three frequency bands, namely frequency band 1, frequency band 2, and frequency band 3, where the unit data latency of frequency band 1 is... =1ms, unit data delay in band 2 =1.1ms, unit data delay in band 3 =1ms; a=10, b=0.001; the corresponding estimated latency for service segment 1 on frequency band 1 is =1000ms, the estimated latency of service segment 1 on frequency band 2. =500ms, the estimated latency of service segment 1 on frequency band 3. =2000ms, then calculate the time cost function of service slice 1 in each candidate frequency band to obtain the corresponding time cost value, that is:

[0072] =10* +0.001* =11

[0073] =10* +0.001* =11.5

[0074] =10* +0.001* =12

[0075] After executing step S402, step S403 is executed: the scheduling module 1101 determines the initial frequency band for transmitting the first service segment from the multiple candidate frequency bands based on the time value of the first service segment on each candidate frequency band. That is, frequency band 1 in the above candidate frequency bands is used as the initial frequency band for transmitting service segment 1 of service 1.

[0076] It should be noted here that, in Figure 1 The system architecture shown is in the initialization or reset state. At this time, all frequency bands are empty and not used for transmission services. If service 1 arrives at this time, the service priority of service 1 is the first service priority. In this case, the frequency band with the smallest unit data delay can be selected from the first frequency band priority frequency band pool as the initial frequency band.

[0077] It should be noted that service scheduling is performed through service fragmentation. In addition to service data, each service fragment includes a service identifier and a sequence number, enabling the receiving end to sort the received data based on these identifiers and sequence numbers. In this embodiment, the size of the service fragment can be preset or determined based on factors such as the channel state information of a specific frequency band or the service type of service 1.

[0078] Phase 2: Business segmentation of Business 1 not reaching the target for the first time

[0079] When the business shard of business 1 is not being reached for the first time, for example, business 1 includes 10 business shards, and 4 business shards have already been reached, then the 5th business shard that is being reached for the first time is considered to be not being reached for the first time.

[0080] In the specific implementation process, when the service fragment of Service 1 is not reached for the first time, it is also necessary to consider whether the service priority of Service 1 itself and / or the frequency band priority of the frequency band have changed. The following will explain the situation in different cases.

[0081] The first scenario: Adjust the service priority of the first service based on the comprehensive evaluation value of the first service and the preset comprehensive evaluation value. The comprehensive evaluation value is determined by the service priority of the first service and the remaining data volume of the first service. The remaining data volume is the data volume of the remaining service fragments of the first service, and the remaining service fragments are the service fragments that have not been transmitted in the first service.

[0082] To avoid ignoring significant differences within different service categories, for example, medical telemetry services (such as ECG monitoring data) are extremely sensitive to transmission reliability and packet loss rate, while occasional delays may be tolerable; while video call services, although also high-priority, have more stringent requirements for real-time performance and latency, but are more tolerant of packet loss. However, currently classifying these two services with significantly different characteristics as high-priority services prevents network resources from being accurately allocated according to the inherent needs of each service, resulting in low resource utilization.

[0083] Alternatively, current scheduling often prioritizes service needs while neglecting the physical characteristics of data packets, such as data volume, which significantly impact actual transmission latency. For example, when a small, high-real-time data packet arrives after a large packet, even if the former has a higher priority, it must wait for the latter to complete before being transmitted, thus compromising the quality of service for low-latency services.

[0084] To address the aforementioned technical issues, the service priority of Service 1 must consider not only its own service priority but also at least the remaining data volume of Service 1. The remaining data volume refers to the remaining service fragments of Service 1, where the remaining service fragments are those that have not yet been transmitted. For example, if 3 out of 10 service fragments have been transmitted, then 7 service fragments remain. Knowing the remaining service fragments and their data volumes allows the calculation of the remaining data volume. Then, a comprehensive evaluation value is calculated based on the service priority and the remaining data volume. The service priority of Service 1 is adjusted according to this comprehensive evaluation value and a preset comprehensive evaluation value. Finally, an appropriate target frequency band is matched to Service 1 based on the adjusted service priority.

[0085] Furthermore, in this embodiment of the application, in order to improve the accuracy of the classification of the service priority of the service itself, the comprehensive evaluation value is also determined by the unit data delay of the initial frequency band and the remaining service time of the remaining service fragments in the initial frequency band. The remaining service time is calculated based on the unit data delay of the initial frequency band and the remaining data volume.

[0086] Specifically, the comprehensive evaluation value of the first service is calculated based on its own service priority, the unit data latency of the initial frequency band, the remaining service time of the remaining service segments in the first service in the target frequency band, and the remaining data volume of the first service; the service priority of the first service is adjusted based on the comprehensive evaluation value of the first service and the preset comprehensive evaluation value.

[0087] In this embodiment, the comprehensive evaluation value of service 1 is calculated periodically. This comprehensive evaluation value is related to the service priority of service 1, the unit data delay of frequency band 1, the remaining data volume of service 1, and the remaining service time. The remaining service fragments of service 1 are the service fragments of service 1 that have not been transmitted. Continuing with the above example, the remaining service fragments are the other service fragments of service 1 besides service fragment 1. The remaining data volume is the data volume of the remaining service fragments. The remaining service time is determined by the remaining data volume and the unit data delay of frequency band 1. As an example, the comprehensive evaluation value of service 1 = *p+ *(1 / )+ * (1 / remaining data volume) + * (1 / remaining service time), where p is related to the service's own priority, p = 0, 1, or 2. p = 0 indicates that service 1's own priority is the third priority, p = 1 indicates that the service's priority is the second priority, and p = 2 indicates that the service's priority is the third priority. - This represents the weight, which can be pre-set or adjusted adaptively according to different application scenarios.

[0088] In the specific implementation process, the obtained comprehensive evaluation value can be compared with the preset comprehensive evaluation value to determine whether to adjust the business priority of the business itself. For example, if the calculated comprehensive evaluation value is greater than the preset comprehensive evaluation value, the business priority of business 1 will be upgraded. If the calculated comprehensive evaluation value is less than the preset comprehensive evaluation value, the business priority of the business will be downgraded.

[0089] In this embodiment, the preset comprehensive evaluation value can be set according to different application scenarios. For example, in a robotic arm control scenario, there are high requirements for end-to-end latency; even a small delay can cause control command misalignment and lead to accidents. For example, in the production process, a robotic arm is used to install vehicle components, and the installation path is calculated based on the vehicle's position milliseconds prior. If the command is delayed, the robotic arm will install the door in the wrong position, causing scratches on the door or body paint, damage to parts, and requiring rework of the entire vehicle, resulting in extremely high costs. Therefore, a lower preset comprehensive evaluation value is set in this scenario. In a surveillance video stream scenario, bandwidth requirements are high, but latency sensitivity is low; therefore, a higher preset comprehensive evaluation value is also set in this scenario. In the following description, a preset comprehensive evaluation value of 1 will be used as an example.

[0090] Assume that business 1 is a business within a robotic arm control scenario. =0.4, =0.3, =0.2, =0.1, Service 1 has the second priority, p=1, the initial unit data delay d=1ms, the remaining data volume is 128 bytes, and the remaining service time is 2ms. Therefore, the comprehensive evaluation value of this service is 0.4×1+0.3×1+0.2×7.8+0.1×0.5=2.31. This value is greater than the preset comprehensive evaluation value, so the service priority of Service 1 needs to be upgraded. As an example, the service priority of Service 1 will be adjusted from the second priority to the first priority.

[0091] In a specific implementation, if the priority of the first service is adjusted from the second service priority to the first service priority, the method further includes: determining whether there is an idle frequency band in the frequency band pool corresponding to the frequency band priority of the first service priority; if the idle frequency band exists, using the idle frequency band as the target frequency band. In a specific implementation, before switching the remaining service fragments of service 1 from the initial frequency band to the idle frequency band, the method further includes: pausing the transmission of service 1 on frequency band A; switching the remaining service to the idle frequency band and sending switching information to the receiving end, the switching information including the frequency band identifier of the idle frequency band and the sequence number of the first service fragment transmitted on the idle frequency band.

[0092] If it does not exist, select a first frequency band from the second frequency band priority frequency band pool, increase the frequency band priority of the first frequency band, allocate the first frequency band to the first frequency band priority frequency band pool, and use the first frequency band as the target frequency band.

[0093] Continuing with the example of the robotic arm scenario above, if the priority of business 1 is adjusted from the second priority to the first priority, it is necessary to determine whether there is an idle frequency band in the first priority frequency band pool. If there is an idle frequency band, then the idle frequency band is used as the target frequency band.

[0094] If no free frequency bands exist in the first frequency band priority candidate pool, to prevent performance degradation or service interruption of services with first service priority due to insufficient resources, a frequency band with second frequency band priority will be determined. This frequency band can be a suitable frequency band in the second frequency band priority candidate pool. Suitable can be understood as the frequency band with the lowest unit data latency in the second frequency band priority candidate pool, such as frequency band 4 in Table 1 above. The frequency band priority of frequency band 4 will be raised to first frequency band priority. In this case, other services with second service priority (taking service 3 as an example) that were originally transmitted on frequency band 4 are no longer suitable for transmission on frequency band 4. If a frequency band for transmitting service 3 can be determined from the second frequency band priority frequency band pool, taking frequency band 5 as an example, service 3 can be switched to frequency band 5; otherwise, the service priority of service 3 will be reduced from second service priority to third service priority, and then a frequency band suitable for transmitting service fragments of service 3 will be determined from the third frequency band priority frequency band pool. The specific execution process can be found in section 5.

[0095] As another example, suppose service 1 is a service in a video streaming monitoring scenario. =0.4, =0.3, =0.2, =0.1, Service 1's own service priority is second service priority, p=0, the initial unit data delay of the frequency band is 5ms, the remaining data volume is 1500B, and the remaining service time is 100ms. Therefore, the overall priority of Service 1 is 0.4×1+0.3×0.2+0.2×0.67+0.1×0.01=0.595, which is less than the overall priority threshold. Therefore, Service 1's own service priority needs to be downgraded, that is, adjusted from second service priority to third service priority. In this case, frequency band 1 needs to be released, and then a frequency band suitable for transmitting Service 1 needs to be determined from the third priority frequency band pool, such as frequency band 7 in Table 1 above, thus ensuring sufficient high-priority frequency band resources.

[0096] In the second scenario, the frequency band priority of the initial frequency band is adjusted based on the unit data delay and the delay threshold of the initial frequency band. The delay threshold is determined at least by the historical delay and current load of the initial frequency band, and the unit data delay of the initial frequency band is determined at least by the propagation delay and the transmission delay. The initial frequency band is the frequency band allocated to the first service when the service fragment of the first service is first reached.

[0097] Prioritizing frequency bands based on fixed thresholds can lead to a disconnect between resource allocation and real-time demand. For example, during peak business hours, user access requests surge, resulting in a massive volume of services requiring low latency. At this time, the total amount of frequency band resources meeting the first priority is fixed and cannot be expanded. This causes numerous high-priority services to compete for high-priority resources, leading to severe overload on that band. This results in a sharp increase in queuing latency, packet loss rate, and service quality degradation. Conversely, during off-peak hours, fewer requests from high-priority services leave a large amount of low-latency resources idle, unusable by medium- or low-priority services, resulting in the idleness and waste of high-value resources.

[0098] On the other hand, different business types, and even different application scenarios of the same business, have vastly different sensitivities to latency. For example, cloud virtual reality and industrial automation control both require low latency, but their acceptable specific thresholds and jitter ranges may be completely different, thus lacking a certain degree of flexibility and adaptability.

[0099] To address the aforementioned technical problems, in this embodiment of the application, the frequency band priority of the initial frequency band is adjusted based on the unit data delay and delay threshold of the initial frequency band, wherein the delay threshold is determined at least by the historical delay and current load of the initial frequency band.

[0100] In the specific implementation process, to further improve the accuracy of frequency band priority allocation, in this embodiment, the latency threshold is also determined by the high-priority traffic volume. Specifically: the latency threshold of the initial frequency band is calculated based on the historical latency, current load, and high-priority traffic volume of the initial frequency band; the frequency band priority of the initial frequency band is adjusted based on the unit data latency of the initial frequency band and the latency threshold.

[0101] In the specific implementation process, if the delay threshold is used ,but = ·d̄ᵢ+ ·load(t) + · Where d̄ᵢ represents the historical delay, reflecting the historical delay level of the frequency band to prevent misjudgment due to brief jitter. During periods of stable load, With the weight increased to 0.5, the system relies more on stable historical trends to avoid unnecessary fluctuations in frequency band priority; during periods of sudden traffic surges, In this scenario, the system reduces the impact of historical values ​​and focuses more on real-time changes to quickly respond to sudden traffic spikes. `load(t)` represents the current load, used to sense resource pressure. When the load increases, the upgrade probability is increased; when the load decreases, upgrade space is released. Here, `load(t) = Σ(current queue length / maximum queue capacity)`. When the load is greater than 80%, ... Increase to 0.4 when the load is less than 40%. A drop to 0.2 makes the system more inclined to maintain or lower the priority, thus freeing up resources; This is used to indicate the demand pressure of high-priority services. When the demand for high-priority services surges, the probability of upgrading the frequency band's priority will increase. =Number of high-priority packets × Average packet size, when When it exceeds the set upper limit threshold, When Qhigh rises to 0.35 and is less than the set lower threshold, γ drops to 0.25, which means that there is less demand for high-priority services, reducing the urgency of upgrading this frequency band.

[0102] After calculating the delay threshold of the initial frequency band, the current priority of the initial frequency band is adjusted according to the delay threshold and the unit data delay of the initial frequency band. The following is a detailed description of each case.

[0103] Scenario 1

[0104] If the unit data delay of the initial frequency band is greater than a first preset multiple of the delay threshold, the frequency band priority of the initial frequency band is reduced, wherein the first preset multiple is greater than 1.

[0105] In the specific implementation process, when the unit data delay of the initial frequency band exceeds a first preset multiple of the delay threshold, it indicates that the initial frequency band can no longer meet the QoS requirements of the service. Here, the first preset multiple can be 1.20 or 1.50, etc. In this embodiment, the first preset multiple is 1.2 as an example. If the frequency band priority of the initial frequency band is the first frequency band priority, then a frequency band priority downgrade action is performed. The initial frequency band will immediately stop accepting services whose service priority is the first service priority. Services that have not been completed on the initial frequency band will undergo frequency band switching. The initial frequency band enters the resource recycling center and is reallocated to the second frequency band priority frequency band pool.

[0106] As an example, Service 1 is a video call, and its own service priority is the second service priority. When the load surges, the unit data latency allocated to it in frequency band 1 increases from 8ms to 12ms, while the latency threshold is 9ms. In this case, it is necessary to adjust frequency band 1 from the second frequency band priority to the third frequency band priority.

[0107] Scenario 2

[0108] If the unit data delay of the initial frequency band is less than a second preset multiple of the delay threshold, and the current load is less than or equal to a preset percentage, the frequency band priority of the initial frequency band is increased, for example, from the second frequency band priority to the first frequency band priority, or from the third frequency band priority to the second frequency band priority.

[0109] In the specific implementation process, when the unit service time of frequency band 1 is less than a second preset multiple of the delay threshold, and the current load is less than or equal to a preset percentage, frequency band 1 is released into the resource recycling center and reassigned to a higher-priority frequency band pool, for example, from the second-priority frequency band pool to the first-priority frequency band pool. Services transmitted on frequency band 1 are not affected. The second preset multiple can be 0.2 times, 0.5 times, or 0.8 times. In this embodiment, the second preset multiple is 0.8 times and the preset percentage is 60% as an example.

[0110] As an example, business 1 is a background download, and its own business priority is the third business priority. During the nighttime load trough, the unit data latency of frequency band 1 is 15ms, while the latency threshold is 20ms, and the current load is 45%. Therefore, frequency band 1 will be upgraded from the third frequency band priority frequency band pool to the second frequency band priority frequency band pool.

[0111] Through the aforementioned dynamic adjustment mechanism, the allocation of frequency band resources can closely align with actual business needs, thereby ensuring the service quality of high-priority services while also improving the overall utilization rate of system resources.

[0112] The third scenario: Based on the comprehensive evaluation value of the first service and a preset comprehensive evaluation value, adjust the service priority of the first service, wherein the comprehensive evaluation value is determined at least by the service priority of the first service and the remaining data volume of the first service, the remaining data volume is the data volume of the remaining service fragments of the first service, and the remaining service fragments are the service fragments of the first service that have not been transmitted; and based on the unit data delay and delay threshold of the target frequency band, adjust the frequency band priority of the target frequency band, wherein the delay threshold is determined at least by the historical delay and current load of the target frequency band, and the unit data delay of the target frequency band is determined at least by the propagation delay and the transmission delay.

[0113] After the scheduling module 1101 determines the target frequency band from the frequency band pool corresponding to the service priority based on the mapping relationship between its own service priority and frequency band priority, it executes step S303: the transmission module 1102 allocates the service fragments of the first service to the target frequency band for transmission.

[0114] In the process of service segmentation and transmission of Service 1, in order to improve the utilization of frequency band resources and reduce the end-to-end latency of services with the highest service priority, the frequency band carrying the service segmentation and transmission can be switched in the specific implementation process. The following will introduce the different situations.

[0115] Scenario 1: The remaining service fragments in Service 1 have been waiting for more than a preset threshold on Frequency Band 1. This mainly includes the following steps:

[0116] Determine whether the estimated waiting time of the remaining service segments of the first service in the first frequency band exceeds a preset threshold;

[0117] In the specific implementation, the frequency band allocated to service fragment 5 of service 1 is the target frequency band. Continuing with the example above, a data packet of service 1 is divided into 100 service fragments. Service fragment 5 of the 100 service fragments is transmitted on the target frequency band. The remaining service fragments are the remaining service fragments excluding service fragment 1. During the transmission of service fragment 1, the estimated waiting time of the remaining service fragments of service 1 on the target frequency band can be calculated in real time or at a certain time period, and it can be determined whether the estimated waiting time of the remaining service fragments on the target frequency band exceeds a preset threshold. The preset threshold can be 500ms or other thresholds, which are not limited here.

[0118] If so, a second frequency band for transmitting the remaining service fragment is determined from the frequency band pool corresponding to the frequency band priority of the first service. The expected waiting time of the remaining service fragment on the second frequency band is less than the expected waiting time of the remaining service fragment on the target frequency band, and the difference between the unit service time of the second frequency band and the unit service time of the target frequency band is within a preset time range, wherein the preset time range is (0-1ms), or other examples, which are not limited here.

[0119] In the specific implementation process, to avoid invalid handover, that is, when the improvement in latency is not significant when the remaining service fragments are switched from the target frequency band to the second frequency band, the method further includes the following steps before determining the second frequency band for transmitting the remaining service fragments from the frequency band pool corresponding to the frequency band priority of the first service's own service priority:

[0120] Determine whether the difference between the time cost value of the remaining service slice in the target and the time cost value of the remaining service slice in the second frequency band is greater than a preset time cost value;

[0121] If so, the remaining service segments will be switched from the target frequency band to the second frequency band.

[0122] If we continue with the above example, if service fragment 5 is transmitted on the target frequency band, and the estimated waiting time of the remaining service fragments of service 1 on the target frequency band is used... This indicates that the estimated latency of the remaining service segments of Service 1 on the second frequency band is... This indicates that before switching the remaining data from the target frequency band to the second frequency band, it is also necessary to determine... +a* Is it greater than +a* And judge +a* and +a* If the difference is greater than the preset time cost, the remaining service segments will be switched from the target frequency band to the second frequency band.

[0123] In the actual implementation process, to further prevent accidental handovers, the following can be considered:

[0124] like +a* Is it greater than + a* +C, and +a* and( +a* If the difference between +C) is greater than the preset time cost, the remaining service segments will be switched from the target frequency band to the second frequency band. Here, C is a time constant, which is used to characterize the switching overhead of switching the remaining service segments from the target frequency band to the second frequency band.

[0125] The above technical solutions can further reduce the end-to-end latency of services with the highest priority.

[0126] In this embodiment of the application, in order to prevent high-priority services from experiencing performance degradation or service interruption due to insufficient resources, and to avoid unnecessary resource contention leading to a decrease in system stability, the frequency band priority will also be adjusted in this case, which will be described separately below.

[0127] Example 1: Receive service 2, service 2 has the highest service priority; determine whether the volume of high-priority services is greater than or equal to a security threshold, wherein the security threshold is determined by a basic capacity threshold, a latency-sensitive threshold, and a system stability threshold;

[0128] If the priority is greater than the security threshold, a third frequency band is selected from the frequency band pool corresponding to the frequency band priority of the second service priority, the frequency band priority of the third frequency band is increased, and the third frequency band is assigned to the frequency band pool corresponding to the adjusted frequency band priority, wherein the level of the first service priority is higher than the level of the second service priority.

[0129] The base capacity threshold refers to the minimum amount of guaranteed resources reserved in the high-priority frequency band pool, which is used to ensure that high-priority services still have basic resources available even when the system is fully loaded. = γ, where, γ represents the total bandwidth capacity of the high-priority frequency band pool, and γ is the reservation coefficient, which typically ranges from 10% to 20%.

[0130] Latency Sensitivity: This reflects the stringent latency requirements of high-priority services. It is determined by both the service's tolerance for latency and the system's real-time load. The calculation method is as follows: =c⋅ +d⋅load(t). Wherein... For high-priority services, such as the real-time control commands for industrial robotic arms and the control information signals for remote surgery, the maximum tolerable end-to-end latency (e.g., 5ms) is defined as load(t), where load(t) is the system load, and c and d are weighting coefficients used to balance the weights of latency and load effects.

[0131] The system stability guardrail is used to prevent excessive preemption from causing resource fragmentation or low-priority service avalanche. It is achieved by limiting the overall system utilization to limit the overall cluster utilization.

[0132] = ⋅δ

[0133] in, It is the instantaneous comprehensive utilization rate of the system's total bandwidth; δ is the buffer coefficient, which usually takes the value range of (0.8~0.9), meaning that at least 10%-20% of system resources are reserved for low-priority services to ensure that their basic functions are not affected.

[0134] The safety threshold is a weighted sum of the above components:

[0135] = + +

[0136] in, , , As a dynamic weight, it can be adjusted according to different business scenarios. As an example, in latency-sensitive scenarios, it can be set to... It has a higher weight, making the safety threshold more sensitive to changes in latency.

[0137] When Service 2 is received, the system first checks whether the volume of high-priority services is greater than or equal to the security threshold calculated above. If it is greater, it means that the current high-priority resource pool can no longer meet the service quality requirements of the service and resource preemption needs to be triggered. The specific resource preemption process is the same as the resource preemption that occurs when the service priority is upgraded as described above, and will not be repeated here.

[0138] Example 2: Receive service 3, the service priority of service 3 is the first service priority; determine whether there is an idle frequency band in the first frequency band priority frequency band pool; if not, determine the fourth frequency band from the second frequency band priority frequency band pool and increase the frequency band priority of the fourth frequency band.

[0139] When a high-priority service arrives, the system first checks whether there is an idle frequency band in the first priority frequency band pool. If not, resource preemption is triggered. The specific implementation of this resource preemption is the same as the resource preemption process that occurs when the service priority is upgraded, as described above. To avoid redundancy, it will not be described in detail here.

[0140] Secondly, based on the same inventive concept as the service data scheduling and transmission method in the first aspect, embodiments of this application also provide a service data scheduling and transmission apparatus, including:

[0141] The scheduling module is used to obtain the service priority of the first service; according to the mapping relationship between the service priority and the frequency band priority, it selects the target frequency band that can minimize the end-to-end delay from the frequency band pool corresponding to the frequency band priority of the service priority. The service priority and / or frequency band priority can be dynamically adjusted according to a preset strategy.

[0142] The transmission module is used to allocate service fragments of the first service to the target frequency band for transmission.

[0143] In one possible implementation, the preset strategy includes:

[0144] Based on the comprehensive evaluation value of the first service and a preset comprehensive evaluation value, the service priority of the first service is adjusted. The comprehensive evaluation value is determined at least by the service priority of the first service and the remaining data volume of the first service. The remaining data volume is the data volume of the remaining service fragments of the first service, and the remaining service fragments are the service fragments of the first service that have not been transmitted; and / or

[0145] The frequency band priority of the target frequency band is adjusted based on the unit data delay and the delay threshold of the target frequency band, wherein the delay threshold is determined at least by the historical delay and current load of the target frequency band, and the unit data delay of the target frequency band is determined at least by the propagation delay and the transmission delay.

[0146] In one possible implementation, the comprehensive evaluation value is further determined by the unit data latency of the target frequency band and the remaining service time of the remaining service slice in the target frequency band, wherein the remaining service time is calculated based on the unit data latency of the target frequency band and the remaining data volume.

[0147] In one possible implementation, the latency threshold is also determined by the volume of high-priority traffic.

[0148] In one possible implementation, the scheduling module is further configured to: determine whether the estimated waiting time of the remaining service fragments of the first service in the target frequency band exceeds a preset threshold; if so, determine a second frequency band for transmitting the remaining service fragments from a frequency band pool corresponding to the frequency band priority of the first service's own service priority, wherein the estimated waiting time of the remaining service fragments in the second frequency band is less than the estimated waiting time of the remaining service fragments in the target frequency band, and the difference between the unit service time of the second frequency band and the unit service time of the target frequency band is within a preset time range, wherein the estimated waiting time is the time for the remaining service fragments to wait to be transmitted in each frequency band.

[0149] In one possible implementation, when the scheduling module determines the second frequency band for transmitting the remaining service fragment from the frequency band priority pool corresponding to the frequency band priority of the first service, it is specifically configured to: determine whether the difference between the time cost value of the remaining service fragment in the target frequency band and the time cost value of the remaining service fragment in the second frequency band is greater than a preset time cost value, wherein the time cost value of the remaining service fragment in the target frequency band is determined by the expected waiting time of the remaining service fragment in the target frequency band and the unit data delay of the target frequency band;

[0150] If the time value is greater than the preset time value, a second frequency band for transmitting the remaining service fragments is determined from the frequency band pool corresponding to the frequency band priority of the first service.

[0151] In one possible implementation, when the scheduling module switches the remaining service fragment from the target frequency band to the second frequency band, it is specifically used to: suspend the transmission of the first service fragment on the target frequency band; switch the remaining service fragment from the target frequency band to the second frequency band, and send switching information to the receiving device, the switching information including at least the frequency band identifier of the second frequency band and the sequence number of the first service fragment transmitted on the second frequency band.

[0152] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of the solution in this specification according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0153] Thirdly, please see Figure 6 Based on the same inventive concept as the service data scheduling and transmission method in the first aspect, this application also provides a computer device 600, which includes: a memory 601, a processor 602, a network interface 603, and a computer program stored in the memory and executable on the processor, wherein the processor 602 executes as described in the first aspect. Figures 3-5 The method for scheduling and transmitting business data described herein. The processor 602 of the computer device 600 provides computing and control capabilities. The memory 601 of the computer device 600 includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory 601 provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device 600 is used to store data. The network interface 603 of the computer device 600 is used for communication with external terminals via a network connection.

[0154] The above description is merely a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

Claims

1. A method for scheduled transmission of service data, characterized by, The method comprises: obtaining a self-service priority of a first service; selecting a target frequency band from a frequency band pool corresponding to a frequency band priority corresponding to the self-service priority, which is estimated to minimize an end-to-end delay, according to a mapping relationship between the self-service priority and the frequency band priority, wherein the self-service priority and / or the frequency band priority can be dynamically adjusted according to a preset strategy; allocating service fragments of the first service to the target frequency band for transmission; wherein the method of dynamically adjusting according to the preset strategy comprises: adjusting the self-service priority of the first service according to a comprehensive evaluation value of the first service and a preset comprehensive evaluation value, wherein the comprehensive evaluation value is determined by at least the self-service priority of the first service and a remaining data amount of the first service, the remaining data amount being a data amount of a remaining service fragment of the first service, and the remaining service fragment being a service fragment of the first service that has not been transmitted; and / or adjusting the frequency band priority of an initial frequency band according to a unit data delay of the initial frequency band and a delay threshold, wherein the delay threshold is determined by at least a historical delay of the initial frequency band and a current load, and the unit data delay of the initial frequency band is determined by at least a propagation delay and a transmission delay, the initial frequency band being a frequency band allocated to the first service when a service fragment of the first service is first transmitted.

2. The method of claim 1, wherein: the comprehensive evaluation value is further determined by a unit data delay of the initial frequency band and a remaining service time of the remaining service fragment in the initial frequency band, the remaining service time being calculated based on the unit data delay of the initial frequency band and the remaining data amount.

3. The method of claim 2, wherein, if the self-service priority of the first service is adjusted from a second service priority to a first service priority, the level of the first service priority being higher than that of the second service priority, the method further comprises: determining whether there is an idle frequency band in a frequency band pool corresponding to the frequency band priority corresponding to the first service priority; if there is the idle frequency band, determining the idle frequency band as the target frequency band; or if there is not, selecting a first frequency band from a frequency band pool corresponding to the frequency band priority corresponding to the second service priority, increasing the frequency band priority of the first frequency band, dividing the first frequency band into a frequency band pool corresponding to the adjusted frequency band priority, and taking the first frequency band as the target frequency band.

4. The method of claim 1, wherein, the delay threshold is further determined by a high-priority service amount.

5. The method of claim 4, wherein, adjusting the frequency band priority of the initial frequency band according to the unit data delay of the initial frequency band and the delay threshold comprises: if the unit data delay of the initial frequency band is greater than a first preset multiple of the delay threshold, decreasing the frequency band priority of the initial frequency band, wherein the first preset multiple is greater than 1; or if the unit data delay of the initial frequency band is less than a second preset multiple of the delay threshold, and the current load is less than or equal to a preset percentage, increasing the frequency band priority of the initial frequency band, wherein the second preset multiple is greater than zero and less than 1.

6. The method of claim 1, wherein, the method further comprises: determining whether a predicted waiting time of the remaining service fragments of the first service on the target frequency band exceeds a preset threshold value; if yes, determining a second frequency band for transmitting the remaining service fragments from a frequency band pool corresponding to a frequency band priority of a self-service priority of the first service, a predicted waiting time of the remaining service fragments on the second frequency band being less than the predicted waiting time of the remaining service fragments on the target frequency band, and a difference between a unit service time of the second frequency band and a unit service time of the target frequency band being within a preset time range, the predicted waiting time being a time for the remaining service fragments to wait to be transmitted on each frequency band.

7. The method of claim 6, wherein, determining a second frequency band for transmitting the remaining service fragments from a frequency band pool corresponding to a frequency band priority of a self-service priority of the first service, including: determining whether a difference between a time value of the remaining service fragments on the target frequency band and a time value of the remaining service fragments on the second frequency band is greater than a preset time value, the time value of the remaining service fragments on the target frequency band being determined by the predicted waiting time of the remaining service fragments on the target frequency band and a unit data delay of the target frequency band; if greater than the preset time value, determining a second frequency band for transmitting the remaining service fragments from a frequency band pool corresponding to a frequency band priority of a self-service priority of the first service.

8. The method of claim 7, wherein, if greater than the preset time value, switching the remaining service fragments from the target frequency band to the second frequency band, including: suspending transmission of service fragments of the first service on the target frequency band; switching the remaining service fragments from the target frequency band to the second frequency band, and sending switching information to a receiving device, the switching information at least including a frequency band identifier of the second frequency band and a sequence number of a first transmitted service fragment on the second frequency band.

9. The method of claim 1, wherein, when receiving a second service, the service priority of the second service being a first service priority, the method further comprising: determining whether a high-priority service volume is greater than or equal to a safety threshold value, wherein the safety threshold value is obtained by weighting based on a basic capacity threshold value, a time delay sensitivity threshold value, and a system stability threshold value; if greater than the safety threshold value, selecting a third frequency band from a frequency band pool corresponding to a frequency band priority of the second service priority, increasing the frequency band priority of the third frequency band, and dividing the third frequency band into a frequency band pool corresponding to the adjusted frequency band priority, the first service priority being higher than the second service priority.

10. A device for scheduled transmission of service data, characterized by including: a scheduling module, configured to obtain a self-service priority of a first service, and select a target frequency band that is estimated to minimize an end-to-end delay from a frequency band pool corresponding to a frequency band priority of the self-service priority according to a mapping relationship between the self-service priority and the frequency band priority, wherein the self-service priority and / or the frequency band priority can be dynamically adjusted according to a preset strategy; a transmission module, configured to allocate service fragments of the first service to the target frequency band for transmission; wherein the scheduling module dynamically adjusts according to the preset strategy, including: adjusting a service priority of the first service according to a comprehensive evaluation value of the first service and a preset comprehensive evaluation value, wherein the comprehensive evaluation value is determined at least by the service priority of the first service and a remaining data amount of the first service, the remaining data amount being a data amount of a remaining service fragment of the first service, the remaining service fragment being a service fragment not transmitted in the first service; and / or adjusting a frequency band priority of the target frequency band according to a unit data delay of the target frequency band and a delay threshold, wherein the delay threshold is determined at least by a historical delay of the target frequency band and a current load, and the unit data delay of the target frequency band is determined at least by a propagation delay and a transmission delay.

11. The apparatus of claim 10, wherein the comprehensive evaluation value is further determined by a unit data delay of the target frequency band and a remaining service time of the remaining service fragment on the target frequency band, the remaining service time being calculated based on the unit data delay of the target frequency band and the remaining data amount.

12. The apparatus of claim 10, wherein, the delay threshold is further determined by a high-priority service amount.

13. The apparatus of claim 10, wherein, The scheduling module is further configured to: determine whether a predicted waiting time of the remaining service fragment of the first service on the target frequency band exceeds a preset threshold value; if yes, determine a second frequency band for transmitting the remaining service fragment from a frequency band pool corresponding to a frequency band priority of the service priority of the first service, the predicted waiting time of the remaining service fragment on the second frequency band being less than the predicted waiting time of the remaining service fragment on the target frequency band, and a difference between a unit service time of the second frequency band and a unit service time of the target frequency band being within a preset time range, the predicted waiting time being a time for the remaining service fragment to wait to be transmitted on each frequency band.

14. The apparatus of claim 13, wherein, When the scheduling module determines the second frequency band for transmitting the remaining service fragment from the frequency band pool corresponding to the frequency band priority of the service priority of the first service, the scheduling module is specifically configured to: determine whether a difference between a time value of the remaining service fragment on the target frequency band and a time value of the remaining service fragment on the second frequency band is greater than a preset time value, the time value of the remaining service fragment on the target frequency band being determined by the predicted waiting time of the remaining service fragment on the target frequency band and the unit data delay of the target frequency band; if greater than the preset time value, determine the second frequency band for transmitting the remaining service fragment from the frequency band pool corresponding to the frequency band priority of the service priority of the first service.

15. The apparatus of claim 14, wherein, When the scheduling module switches the remaining service fragment from the target frequency band to the second frequency band, the scheduling module is specifically configured to: suspend transmission of the service fragment of the first service on the target frequency band; switch the remaining service fragment from the target frequency band to the second frequency band, and send switching information to a receiving device, the switching information at least including a frequency band identifier of the second frequency band and a sequence number of a first transmitted service fragment on the second frequency band.

16. A computer device, comprising: The device comprises a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor performs the method for scheduling transmission of service data according to claims 1-9.

Images