Service scheduling parameter determination methods, devices, equipment and storage media
By acquiring business scenario information and latency requirements, determining the granularity of business scheduling, and using a long short-term memory neural network model to train a scheduling algorithm with preset parameters, the problem of not being able to meet the latency requirements of business in 5G networks was solved, and dynamic adjustment and resource optimization were achieved.
Patent Information
- Application Number
- CN202310781075.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-28
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2043-06-28
AI Technical Summary
In existing technologies, 5G networks, without considering network congestion scenarios, cannot effectively meet service latency requirements, resulting in wasted network resources and an inability to respond promptly to latency issues.
By acquiring the service scenario information and latency requirements of the communication services to be scheduled, the granularity of service scheduling is determined, service quality monitoring is performed, and a long short-term memory neural network model is used to train a scheduling algorithm with preset parameters to dynamically adjust the service scheduling parameters to meet the service latency requirements.
It enables dynamic adjustment of scheduling parameters based on business needs, meeting actual latency requirements, reducing network load and resource waste, and improving user experience.
Smart Images

Figure CN116744464B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of communication technology, and in particular to a method, apparatus, device, and storage medium for determining service scheduling parameters. Background Technology
[0002] Based on the 5G QoS Identifier (5QI, an identifier used in 5G networks to distinguish different service quality requirements. 5QI is associated with specific QoS characteristics, such as transmission latency, data rate, and packet loss rate, to meet the needs of different business scenarios), the Quality of Service (QoS) provides the Packet Delay Budget (PDB) attribute. For major users, under uncongested network conditions, 98% of data packet delays cannot exceed the 5QI PDB attribute. Reliability cannot meet business requirements, and without considering network congestion scenarios, it cannot effectively reflect or meet the latency requirements of business needs. Network technology has mainly focused on latency requirements and low-latency guarantee capabilities through technical improvements; lower BLE, lower MCS, and PDCP repetition ensure network latency stability, while pre-scheduling and mini-slot technologies reduce latency. However, it lacks association with specific business latency requirements, making it impossible to promptly and effectively detect moments where latency is not met, and there is no timely and effective response.
[0003] In current network environments, a relatively static combination of policies and service scheduling parameters is typically used, lacking response strategies based on service requirements. While the application of various low-latency technologies across the network effectively reduces network-side latency, it leads to wasted network resources and fails to meet the needs of different services within the same cell or under the same 5QI. Existing technologies are all network-side services, lacking agreed-upon latency requirements based on actual service needs, as well as policy combinations for different latency requirements, and lacking the ability to dynamically optimize and adjust according to environmental changes.
[0004] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention
[0005] The main objective of this invention is to provide a method, apparatus, device, and storage medium for determining service scheduling parameters, aiming to solve the technical problem of network resource waste caused by the use of a relatively static combination of service scheduling parameters to reduce network-side latency in the prior art.
[0006] To achieve the above objectives, the present invention provides a method for determining service scheduling parameters, the method comprising the following steps:
[0007] Obtain the service scenario information and service latency requirements of the communication services to be scheduled;
[0008] The granularity of service scheduling is determined based on the aforementioned business scenario information;
[0009] Based on the service scheduling granularity, the quality of service of the communication service to be scheduled is monitored to obtain latency data;
[0010] The service scheduling parameters are determined based on the service latency requirements and the latency data.
[0011] Optionally, the step of determining the service scheduling parameters based on the service latency requirements and the latency data includes:
[0012] Determine the end-to-end latency of the service based on the aforementioned service latency requirements;
[0013] The target latency from the wireless access network to the user plane function is determined based on the latency data.
[0014] The service scheduling parameters are determined based on the service end-to-end latency and the target latency.
[0015] Optionally, the step of determining the service scheduling parameters based on the service end-to-end latency and the target latency includes:
[0016] The service wireless side latency is determined based on the service end-to-end latency and the target latency.
[0017] Determine whether the latency data meets the preset conditions based on the wireless side latency of the service.
[0018] If the latency data does not meet the preset conditions, the service scheduling parameters are determined based on the service wireless side latency using a preset parameter scheduling algorithm.
[0019] Optionally, after the step of determining whether the latency data meets the preset conditions based on the service wireless side latency, the method further includes:
[0020] When the latency data meets the preset conditions, the target service scheduling parameters corresponding to the service scheduling granularity are obtained;
[0021] The target service scheduling parameters are used as the service scheduling parameters.
[0022] Optionally, before the step of determining the service scheduling parameters based on the service latency requirements and the latency data, the method further includes:
[0023] Obtain relevant parameters for wireless side scheduling;
[0024] Different data combinations are performed on the wireless side scheduling-related parameters to obtain the initial scheduling parameter combination;
[0025] Determine the wireless side latency data corresponding to the initial scheduling parameter combination;
[0026] The long short-term memory neural network model is trained based on the initial scheduling parameter combination and the wireless side latency data to obtain a preset parameter scheduling algorithm.
[0027] Optionally, the step of training the long short-term memory neural network model based on the initial scheduling parameter combination and the wireless side latency data to obtain a preset parameter scheduling algorithm includes:
[0028] An initial sequence dataset is generated based on the initial scheduling parameter combination and the wireless side latency data;
[0029] The initial sequence dataset is cleaned and normalized to obtain a time-series data pool;
[0030] The long short-term memory neural network model is trained based on the time-series data pool to obtain a scheduling algorithm with preset parameters.
[0031] Optionally, after the step of determining the service scheduling parameters based on the service latency requirements and the latency data, the method further includes:
[0032] Obtain the pending business and determine the business requirements corresponding to the pending business;
[0033] Based on the aforementioned business requirements, determine the target business that is consistent with the business requirements of the business to be processed;
[0034] The service scheduling parameters of the target service are used as the service scheduling parameters of the service to be processed.
[0035] Furthermore, to achieve the above objectives, the present invention also provides a service scheduling parameter determination device, the device comprising:
[0036] The acquisition module is used to acquire the service scenario information and service latency requirements of the communication service to be scheduled.
[0037] The service scheduling granularity determination module is used to determine the service scheduling granularity based on the service scenario information;
[0038] The service quality monitoring module is used to monitor the service quality of the communication service to be scheduled based on the service scheduling granularity and obtain latency data.
[0039] The service scheduling parameter determination module is used to determine service scheduling parameters based on the service latency requirements and the latency data.
[0040] Furthermore, to achieve the above objectives, the present invention also proposes a service scheduling parameter determination device, the device comprising: a memory, a processor, and a service scheduling parameter determination program stored in the memory and executable on the processor, the service scheduling parameter determination program being configured to implement the steps of the service scheduling parameter determination method described above.
[0041] In addition, to achieve the above objectives, the present invention also proposes a storage medium storing a service scheduling parameter determination program, wherein when the service scheduling parameter determination program is executed by a processor, it implements the steps of the service scheduling parameter determination method described above.
[0042] This invention acquires service scenario information and service latency requirements of communication services to be scheduled; determines the service scheduling granularity based on the service scenario information; performs quality of service monitoring on the communication services to be scheduled based on the service scheduling granularity to obtain latency data; and determines service scheduling parameters based on the service latency requirements and the latency data. Because this invention performs quality of service monitoring on the communication services to be scheduled based on the service scheduling granularity to obtain latency data, and determines service scheduling parameters based on the service latency requirements and latency data, compared to existing methods that use a common parameter scheduling strategy for data communication, the above method of this invention can dynamically adjust service scheduling parameters to meet actual service latency and stability requirements, achieving network-driven service, and reducing the impact on network load and noise floor. Attached Figure Description
[0043] Figure 1 This is a schematic diagram of the structure of the service scheduling parameter determination device for the hardware operating environment involved in the embodiments of the present invention;
[0044] Figure 2 This is a flowchart illustrating the first embodiment of the service scheduling parameter determination method of the present invention;
[0045] Figure 3 This is a flowchart illustrating the second embodiment of the service scheduling parameter determination method of the present invention;
[0046] Figure 4 This is a flowchart illustrating the third embodiment of the service scheduling parameter determination method of the present invention;
[0047] Figure 5 This is a schematic diagram of model training for the third embodiment of the service scheduling parameter determination method of the present invention;
[0048] Figure 6 This is a structural block diagram of the first embodiment of the service scheduling parameter determination device of the present invention.
[0049] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation
[0050] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.
[0051] Reference Figure 1 , Figure 1 This is a schematic diagram of the device structure for determining the service scheduling parameters of the hardware operating environment involved in the embodiments of the present invention.
[0052] like Figure 1 As shown, the service scheduling parameter determining device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to establish communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or stable non-volatile memory (NVM), such as a disk drive. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.
[0053] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the device for determining service scheduling parameters and may include more or fewer components than shown, or combine certain components, or have different component arrangements.
[0054] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and a service scheduling parameter determination program.
[0055] exist Figure 1In the service scheduling parameter determination device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the service scheduling parameter determination device of the present invention can be set in the service scheduling parameter determination device, and the service scheduling parameter determination device calls the service scheduling parameter determination program stored in the memory 1005 through the processor 1001 and executes the service scheduling parameter determination method provided in the embodiment of the present invention.
[0056] Based on the aforementioned service scheduling parameter determination device, this embodiment of the invention provides a service scheduling parameter determination method, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the service scheduling parameter determination method of the present invention.
[0057] In this embodiment, the method for determining service scheduling parameters includes the following steps:
[0058] Step S10: Obtain the service scenario information and service latency requirements of the communication service to be scheduled.
[0059] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a mobile phone, tablet computer, or personal computer, or an electronic device or service scheduling parameter determination device capable of performing the above functions. The following description uses the service scheduling parameter determination device as an example to illustrate this embodiment and the subsequent embodiments.
[0060] It should be noted that the communication service to be scheduled can be a data interaction service using a 5G network, such as voice service, gaming service, and personal user communication service. The service scenario information can include network demand information corresponding to the communication service to be scheduled, such as personal service, enterprise service, and gaming scenario service. The service latency requirement can include the user's network performance requirements for the end-to-end latency, service rate, etc., of the communication service to be scheduled.
[0061] In practical implementation, both personal and industry-specific services can obtain information about the network requirements of the communication services to be scheduled through various means, including end-to-end latency and service speed requirements. Industry users can obtain this information through communication with the enterprise, while personal users can obtain it through various channels such as service providers, complaints, satisfaction levels, live network testing, service platform lag rates, and poor uplink and downlink latency. There are many metrics for service perception, but most can be used to average out the perceived speed of service delivery based on latency compliance.
[0062] Step S20: Determine the service scheduling granularity based on the service scenario information.
[0063] It should be noted that the granularity of service scheduling refers to the fineness of resource allocation and task arrangement when scheduling service traffic. Smaller granularity indicates greater control over service traffic scheduling, allowing for more precise resource allocation based on the specific needs of each service, application, or user. However, excessively small granularity may increase scheduling complexity and system overhead. The service scheduling granularity in this embodiment may include 5QI and slice granularity.
[0064] In practical implementation, the service scheduling parameter determination device first determines the 5QI, slice subscription, and PCF policy status of the communication service bearer to be scheduled. The base station (gNodeB, gNB) in the 5G network maps the Quality of Service (QoS) Flow (the basic unit for transmitting IP packets in a 5G network. It associates packets with specific QoS attributes to ensure that the QoS requirements of different service types and application scenarios are met. QoS Flow is jointly managed by the core network 5GC and the radio access network RAN, and is identified through 5QI. A user equipment (UE) can have multiple QoS Flows to support concurrent QoS requirements.) onto the bearer. The mapping relationship between QoS Flow and radio bearer channel can be many-to-one or one-to-one. Without PCC subscription rules, the default 5QI bearer is typically subscribed to according to UDM rules. For important or special business needs, the 5QI bearer can be modified through PCF rules. Specific 5QI bearers can be subscribed to at the user level as the default bearer, or differentiated based on individual user groups versus industry user groups. For important businesses, GBR or other Non-GBR bearers can be subscribed to through PCF rules. There are default slices for individual users and default slices for industries, as well as their own dedicated slices. According to the URSP rules defined by PCF, different services can use different slices. Currently, the default bearer for individual users is typically 5QI9, while the default bearers for industry users are 5QI6 and 8. VONR, as a voice service, typically uses 5QI1 and 2; gaming services can subscribe to GBR bearers such as 5QI4; for industry services, dedicated slices are usually configured. Different slices can be carried on the same 5QI, and the same default slice can be carried on different 5QI bearers. Furthermore, the granularity of service scheduling depends on the number of users involved in the service and their needs. For example, if the gaming service is carried on 5QI4, the gaming user group is large and the service requirements are consistent. Based on the latency requirements of the gaming service, it can be carried on 5QI4. Industry users are defaulted to 5QI6. Port PLC services require low latency and high reliability. The service requirements of users on 5QI6 are inconsistent. A combination of different strategies for slicing granular scheduling can better meet the service requirements without having a greater impact on other services and network load.
[0065] Step S30: Perform quality of service monitoring on the communication service to be scheduled based on the service scheduling granularity to obtain latency data.
[0066] It should be noted that the service quality monitoring of the communication service to be scheduled based on the service scheduling granularity, and the resulting latency data, can be various indicators of the communication service to be scheduled, which communicates based on the bearer and service scheduling parameter strategies corresponding to the service scheduling granularity. These indicators include key performance metrics such as latency, packet loss rate, jitter, and throughput. These can be measured through QoS monitoring.
[0067] It should be understood that wireless latency is closely related to changes in the network's wireless environment, which is related to factors such as network coverage, interference, load, visibility, frequency bands, network standards, and user distribution. QoS monitoring is an end-to-end characteristic of the network.
[0068] In practical implementation, a QoS Monitoring request can be issued through the PCF or SMF to initiate latency measurement; the SMF triggers QoS monitoring based on the PCF policy or local policy, issues a QoS Monitoring request, and initiates latency measurement at the Per UE Per QoS Flow granularity. The terminal sends periodic measurement reports, which include uplink and downlink latency to the base station and UPF, including uplink and downlink radio side latency, average and maximum application layer-application layer RTT latency, etc.; the SMF can activate QoS monitoring for all GTP-U paths between UPF and AN according to locally configured policy requests.
[0069] Step S40: Determine the service scheduling parameters based on the service latency requirements and the latency data.
[0070] It should be noted that determining the service scheduling parameters based on the service latency requirements and the latency data can involve judging whether the latency data meets the service latency requirements. If not, the service scheduling parameters are redefined to meet the service latency requirements. If they do meet the requirements, the target service scheduling parameters corresponding to the service scheduling granularity are used as the service scheduling parameters.
[0071] Furthermore, in order to improve the speed of service scheduling and enhance the user experience, after step S40, the method further includes: acquiring the service to be processed and determining the service requirements corresponding to the service to be processed.
[0072] Based on the aforementioned business requirements, determine the target business that is consistent with the business requirements of the business to be processed;
[0073] The service scheduling parameters of the target service are used as the service scheduling parameters of the service to be processed.
[0074] It should be noted that the service requirements may include high-bandwidth services and low-latency services.
[0075] In practical implementation, when the service requirements of the communication service to be scheduled are consistent with those of the communication service with determined service scheduling parameters (e.g., both are low-latency services), the service scheduling parameters of the communication service with determined service scheduling parameters can be obtained and used as the service scheduling parameters of the communication service to be scheduled. For example, when the commonalities of service requirements for scheduling tasks based on slice granularity are very obvious, parameter group scheduling can be performed using slice granularity, such as for PLC services; when the services carried by 5QI are relatively clear or have basically the same commonalities, differentiated scheduling can be performed using 5QI, such as for gaming services; when the services carried based on default slices or 5QI are not obvious, the default scheduling strategy of the existing cell can be adopted; this achieves the goal of minimizing network demand satisfaction and minimizing network resource overhead.
[0076] This embodiment acquires the service scenario information and service latency requirements of the communication service to be scheduled; determines the service scheduling granularity based on the service scenario information; performs quality of service monitoring on the communication service to be scheduled based on the service scheduling granularity to obtain latency data; and determines service scheduling parameters based on the service latency requirements and the latency data. Since this embodiment performs quality of service monitoring on the communication service to be scheduled based on the service scheduling granularity to obtain latency data, and determines service scheduling parameters based on the service latency requirements and latency data, compared to existing methods that use a common parameter scheduling strategy for data communication, the above method in this embodiment can dynamically adjust the service scheduling parameters to meet actual service latency and stability requirements, achieving network-driven service, and reducing the impact on network load and noise floor.
[0077] refer to Figure 3 , Figure 3 This is a flowchart illustrating the second embodiment of the service scheduling parameter determination method of the present invention.
[0078] Based on the first embodiment described above, in this embodiment, step S40 includes:
[0079] Step S401: Determine the end-to-end latency of the service based on the service latency requirements.
[0080] It should be noted that determining the end-to-end latency of a service based on the service latency requirements can be done by extracting the maximum end-to-end latency of a service that meets the user's network requirements from the service latency requirements.
[0081] Step S402: Determine the target latency from the wireless access network to the user plane function based on the latency data.
[0082] It should be noted that the Radio Access Network (RAN) and User Plane Function (UPF) are involved. The target latency from the RAN to the UPF can be the time required for a data packet to travel from the RAN to the core network (typically the 5G core network in the 5G era). This latency measures the data transmission speed within the network and is particularly critical for applications requiring high real-time performance (such as online gaming, video calls, and autonomous driving). To ensure the accuracy of the various latencies measured, in this embodiment, average latency data over a certain period can be used for all latencies.
[0083] It should be understood that RAN-to-UPF latency is affected by multiple factors, including radio signal quality, network congestion, and equipment performance. In 5G networks, the goal is to reduce this latency to the millisecond level to meet higher performance requirements and support real-time applications. To achieve this goal, 5G networks employ advanced technologies such as MIMO (Multiple-Input Multiple-Output), small cells, network slicing, and Quality of Service (QoS) policies.
[0084] Step S403: Determine the service scheduling parameters based on the service end-to-end latency and the target latency.
[0085] It should be noted that determining the service scheduling parameters based on the service end-to-end latency and the target latency can involve determining the user's service wireless side latency requirement based on the service end-to-end latency and the target latency, and then determining whether the service wireless side latency requirement is currently met based on the service wireless side latency requirement and latency data. If not, the target service scheduling parameters corresponding to a more granular service scheduling level can be used as the service scheduling parameters. If the requirement is met, the target service scheduling parameters corresponding to the current service scheduling level can be used as the service scheduling parameters.
[0086] Furthermore, in order to reduce network resources and meet the user's latency requirements, step S403 may include: determining the service radio-side latency based on the service end-to-end latency and the target latency;
[0087] Determine whether the latency data meets the preset conditions based on the wireless side latency of the service.
[0088] If the latency data does not meet the preset conditions, the service scheduling parameters are determined based on the service wireless side latency using a preset parameter scheduling algorithm.
[0089] It should be noted that determining the service wireless side latency based on the service end-to-end latency and the target latency can be achieved by subtracting the target latency from the service end-to-end latency. The service wireless side latency can be the maximum wireless side latency while meeting the user's latency requirements. Determining whether the latency data meets a preset condition based on the service wireless side latency can be done by determining whether the service wireless side latency is less than or equal to the actual wireless side latency measured in the latency data. If it is less than or equal to, the latency data is deemed to meet the preset condition, and the target service scheduling parameter corresponding to the service scheduling granularity can be directly used as the service scheduling parameter. If it does not meet the condition, it indicates that the target service scheduling parameter corresponding to the service scheduling granularity cannot meet the user's latency requirements. In this case, the service scheduling parameter is determined based on the service wireless side latency using a preset parameter scheduling algorithm. The preset parameter scheduling algorithm can be a pre-trained algorithm capable of deriving combinations of scheduling parameters at different granularities. The service scheduling parameters can be re-determined using the preset parameter scheduling algorithm until the user's latency requirements are met. To meet user needs, the service wireless side latency can be periodically compared with the actual service wireless side latency in the latency data. If the actual service wireless side latency is lower than the service wireless side latency, the service scheduling parameters continue to be used. Otherwise, the service scheduling parameters are re-determined based on the service wireless side latency using a preset parameter scheduling algorithm.
[0090] This embodiment determines the end-to-end latency of the service based on the service latency requirements; determines the target latency from the wireless access network to the user plane function based on the latency data; and determines the service scheduling parameters based on the end-to-end latency and the target latency. This embodiment compares the latency data at the service scheduling granularity with the service latency requirements to determine whether the current latency requirements are met. If not, the service scheduling parameters are redefined, thus satisfying the user's network latency requirements without increasing network load and improving the user experience.
[0091] refer to Figure 4 , Figure 4 This is a flowchart illustrating the third embodiment of the service scheduling parameter determination method of the present invention.
[0092] Based on the above embodiments, in this embodiment, before step S40, the method further includes:
[0093] Step S001: Obtain relevant parameters for wireless side scheduling.
[0094] It should be noted that the wireless side scheduling related parameters can be parameters that can affect the wireless side data transmission latency, including but not limited to: uplink pre-scheduling switch, pre-scheduling data volume, pre-scheduling duration, pre-scheduling period, DRX switch, uplink and downlink BLER target values, uplink and downlink rate guarantee, uplink and downlink MCS initial values, uplink and downlink MCS maximum values, MCS adjustment step size, SR period, etc.
[0095] Step S002: Combine the wireless side scheduling-related parameters with different data to obtain the initial scheduling parameter combination.
[0096] It should be noted that the initial scheduling parameter combination can be a data combination containing different parameters and / or different values of the same parameter, as shown in the following formula:
[0097]
[0098] Where X is used to characterize the initial scheduling parameter combination, and x11-xmn are used to characterize various different parameters and / or parameter values. For example, x11 means the uplink pre-scheduling switch is on, x1n means the pre-scheduling data volume is 10, xm1 means the pre-scheduling duration is 10, and xmn means the DRX switch is off, etc.
[0099] Step S003: Determine the wireless side latency data corresponding to the initial scheduling parameter combination.
[0100] It should be noted that the determination of the radio-side latency data corresponding to the initial scheduling parameter combination can be achieved by measuring the latency data at different network element granularities during network data interaction based on the initial scheduling parameter combination through QoS monitoring, and determining the radio-side latency data corresponding to the initial scheduling parameter combination based on the measured latency data.
[0101] Step S004: Train the long short-term memory neural network model according to the initial scheduling parameter combination and the wireless side latency data to obtain the preset parameter scheduling algorithm.
[0102] It should be noted that training the long short-term memory neural network model based on the initial scheduling parameter combination and the wireless side latency data can be achieved by training the long short-term memory neural network model based on the initial scheduling parameter combination and the wireless side latency data corresponding to the initial scheduling parameter combination, thereby obtaining a preset parameter scheduling algorithm.
[0103] Furthermore, in order to enable the preset parameter scheduling algorithm to accurately calculate the service scheduling parameters, before step S004, the method further includes: generating an initial sequence dataset based on the initial scheduling parameter combination and the wireless side latency data;
[0104] The initial sequence dataset is cleaned and normalized to obtain a time-series data pool;
[0105] The long short-term memory neural network model is trained based on the time-series data pool to obtain a scheduling algorithm with preset parameters.
[0106] It should be noted that generating the initial sequence dataset based on the initial scheduling parameter combination and the radio-side latency data can be achieved by constructing the initial sequence dataset based on multiple initial scheduling parameter combinations and the corresponding radio-side latency data. Cleaning and normalizing the initial sequence dataset to obtain the time-series data pool can be performed by cleaning the initial sequence dataset, removing non-standard data, and normalizing the data to form the time-series data pool.
[0107] In specific implementation, it can be referred to Figure 5 , Figure 5 This is a schematic diagram of model training for the third embodiment of the service scheduling parameter determination method of the present invention; see reference. Figure 5 A forget gate value of 0 means that information from ct-1 will not be passed to the calculation of ct, while a value of 1 means that all information from ct-1 will be passed to the calculation of ct. In this example, if the wireless latency calculated using the preset parameter scheduling algorithm is lower than the wireless latency required by the service, the value is 1, and the relevant information is passed down; if the calculated wireless latency is greater than the wireless latency required by the service, the value is 0, and no information is passed. The preset parameter scheduling algorithm calculates and finds differentiated combination parameters based on meeting the service latency requirements as the service scheduling parameters.
[0108] In this embodiment, based on different input scheduling parameters and their varying value ranges, a preset parameter scheduling algorithm can calculate service scheduling parameter combinations suitable for different service latency requirements. Different service scheduling parameter combinations can then be assigned different ID numbers, allowing for the identification of the corresponding service scheduling parameter combination based on the ID number. The X input item can be at the 5QI level or at the slice level. The algorithm minimizes the granularity of service requirements, reducing the impact on other services. For example, port PLCs and AGVs use slice granularity (few users, personalized needs), while game services use 5QI granularity (many users, 5QI independent carrying). Service scheduling parameter combinations suitable for game service latency requirements can be merged into a single ID or divided into DRXID groups, pre-scheduled ID groups, etc., as needed. When using slice-based scheduling of service scheduling parameter combinations, if the 5QI-level or cell-level identical function strategy fails, the service scheduling parameter combination corresponding to the slice granularity takes precedence; that is, the service scheduling parameter combination corresponding to the smaller granularity has the highest priority.
[0109] In practical implementation, based on the latency requirements of the wireless side of the service, the preset parameter scheduling algorithm can derive different combinations of scheduling parameters at different granularities. Considering the limited network wireless resources, in principle, scheduling based on slices is implemented according to slice granularity for users with small groups. For groups with commonalities and minimal impact, different granularities are differentiated according to 5QI. Others are set independently according to cell granularity. Example: Vertical industry PLC requirements vary across industries. The latency requirements of remote control PLCs and AGVs in ports also differ. Port PLCs and AGVs each have their own dedicated slice S-NSSAI, shared with other industry users on 5QI6. For port PLC and AGV control, differentiated scheduling based on slice granularity is adopted. If 5QI-level differentiated scheduling is used to meet the latency of PLCs and AGVs, it is not targeted enough and will cause a lot of resource waste, a large increase in network noise floor, and high resource consumption. Moreover, the latency requirements of other services are not entirely the same. Calculations are performed according to 5QI and slice granularity.
[0110] This embodiment acquires wireless-side scheduling-related parameters; combines these parameters with different data to obtain initial scheduling parameter combinations; determines the wireless-side latency data corresponding to the initial scheduling parameter combinations; and trains a Long Short-Term Memory (LSTM) neural network model based on the initial scheduling parameter combinations and the wireless-side latency data to obtain a preset parameter scheduling algorithm. This embodiment can train an LSM model based on the initial scheduling parameter combinations and the wireless-side latency data to obtain a preset parameter scheduling algorithm, enabling the determination of service scheduling parameters that meet user latency requirements, minimizing resource overhead, and thus satisfying the differentiated needs of different services.
[0111] Reference Figure 6 , Figure 6 This is a structural block diagram of the first embodiment of the service scheduling parameter determination device of the present invention.
[0112] like Figure 6 As shown, the service scheduling parameter determination device proposed in this embodiment of the invention includes:
[0113] The acquisition module 10 is used to acquire the service scenario information and service latency requirements of the communication service to be scheduled;
[0114] The service scheduling granularity determination module 20 is used to determine the service scheduling granularity based on the service scenario information.
[0115] Service quality monitoring module 30 is used to monitor the service quality of the communication service to be scheduled based on the service scheduling granularity and obtain latency data;
[0116] The service scheduling parameter determination module 40 is used to determine service scheduling parameters based on the service latency requirements and the latency data.
[0117] This embodiment acquires the service scenario information and service latency requirements of the communication service to be scheduled; determines the service scheduling granularity based on the service scenario information; performs quality of service monitoring on the communication service to be scheduled based on the service scheduling granularity to obtain latency data; and determines service scheduling parameters based on the service latency requirements and the latency data. Since this embodiment performs quality of service monitoring on the communication service to be scheduled based on the service scheduling granularity to obtain latency data, and determines service scheduling parameters based on the service latency requirements and latency data, compared to existing methods that use a common parameter scheduling strategy for data communication, the above method in this embodiment can dynamically adjust the service scheduling parameters to meet actual service latency and stability requirements, achieving network-driven service, and reducing the impact on network load and noise floor.
[0118] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this invention. In practical applications, those skilled in the art can select some or all of the workflow to achieve the purpose of this embodiment according to actual needs, and no restrictions are imposed here.
[0119] In addition, for technical details not described in detail in this embodiment, please refer to the service scheduling parameter determination method provided in any embodiment of the present invention, which will not be repeated here.
[0120] Based on the first embodiment of the service scheduling parameter determination device of the present invention, a second embodiment of the service scheduling parameter determination device of the present invention is proposed.
[0121] In this embodiment, the service scheduling parameter determination module 40 is further configured to determine the end-to-end service latency based on the service latency requirements;
[0122] The target latency from the wireless access network to the user plane function is determined based on the latency data.
[0123] The service scheduling parameters are determined based on the service end-to-end latency and the target latency.
[0124] Furthermore, the service scheduling parameter determination module 40 is also used to determine the service radio side latency based on the service end-to-end latency and the target latency;
[0125] Determine whether the latency data meets the preset conditions based on the wireless side latency of the service.
[0126] If the latency data does not meet the preset conditions, the service scheduling parameters are determined based on the service wireless side latency using a preset parameter scheduling algorithm.
[0127] Furthermore, the service scheduling parameter determination module 40 is also used to obtain the target service scheduling parameters corresponding to the service scheduling granularity when the latency data meets the preset conditions;
[0128] The target service scheduling parameters are used as the service scheduling parameters.
[0129] Furthermore, the service scheduling parameter determination module 40 is also used to obtain wireless side scheduling-related parameters;
[0130] Different data combinations are performed on the wireless side scheduling-related parameters to obtain the initial scheduling parameter combination;
[0131] Determine the wireless side latency data corresponding to the initial scheduling parameter combination;
[0132] The long short-term memory neural network model is trained based on the initial scheduling parameter combination and the wireless side latency data to obtain a preset parameter scheduling algorithm.
[0133] Furthermore, the service scheduling parameter determination module 40 is also used to generate an initial sequence dataset based on the initial scheduling parameter combination and the wireless side latency data;
[0134] The initial sequence dataset is cleaned and normalized to obtain a time-series data pool;
[0135] The long short-term memory neural network model is trained based on the time-series data pool to obtain a scheduling algorithm with preset parameters.
[0136] Furthermore, the service scheduling parameter determination module 40 is also used to acquire the service to be processed and determine the service requirements corresponding to the service to be processed.
[0137] Based on the aforementioned business requirements, determine the target business that is consistent with the business requirements of the business to be processed;
[0138] The service scheduling parameters of the target service are used as the service scheduling parameters of the service to be processed.
[0139] Other embodiments or specific implementations of the service scheduling parameter determination device of the present invention can be referred to the above-described method embodiments, and will not be repeated here.
[0140] Furthermore, this embodiment of the invention also proposes a storage medium storing a service scheduling parameter determination program, which, when executed by a processor, implements the steps of the service scheduling parameter determination method described above.
[0141] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.
[0142] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.
[0143] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.
[0144] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.
Claims
1. A method of determining service scheduling parameters, characterized by, The method for determining service scheduling parameters includes the following steps: Obtain the service scenario information and service latency requirements of the communication services to be scheduled; The granularity of service scheduling is determined based on the aforementioned business scenario information; Based on the service scheduling granularity, the quality of service of the communication service to be scheduled is monitored to obtain latency data; Determine the end-to-end latency of the service based on the aforementioned service latency requirements; The target latency from the wireless access network to the user plane function is determined based on the latency data. The service scheduling parameters are determined based on the service end-to-end latency, the target latency, and a preset parameter scheduling algorithm. The service scheduling granularity includes 5QI bearer granularity and / or network slice granularity; Before determining the end-to-end latency of the service based on the service latency requirements, the method further includes: Obtain relevant parameters for wireless side scheduling; Different data combinations are performed on the wireless side scheduling-related parameters to obtain the initial scheduling parameter combination; Determine the wireless side latency data corresponding to the initial scheduling parameter combination; The long short-term memory neural network model is trained based on the initial scheduling parameter combination and the wireless side latency data to obtain a preset parameter scheduling algorithm.
2. The traffic scheduling parameter determination method of claim 1, wherein, The step of determining the service scheduling parameters based on the service end-to-end latency and the target latency includes: The service wireless side latency is determined based on the service end-to-end latency and the target latency. Determine whether the latency data meets the preset conditions based on the wireless side latency of the service. If the latency data does not meet the preset conditions, the service scheduling parameters are determined based on the service wireless side latency using a preset parameter scheduling algorithm.
3. The traffic scheduling parameter determination method of claim 2, wherein, After the step of determining whether the latency data meets the preset conditions based on the service wireless side latency, the method further includes: When the latency data meets the preset conditions, the target service scheduling parameters corresponding to the service scheduling granularity are obtained; The target service scheduling parameters are used as the service scheduling parameters.
4. The traffic scheduling parameter determination method of claim 3, wherein, The step of training the long short-term memory neural network model based on the initial scheduling parameter combination and the wireless side latency data to obtain a preset parameter scheduling algorithm includes: An initial sequence dataset is generated based on the initial scheduling parameter combination and the wireless side latency data; The initial sequence dataset is cleaned and normalized to obtain a time-series data pool; The long short-term memory neural network model is trained based on the time-series data pool to obtain a scheduling algorithm with preset parameters.
5. The method for determining service scheduling parameters as described in any one of claims 1-3, characterized in that, After the step of determining the service scheduling parameters based on the service end-to-end latency, the target latency, and the preset parameter scheduling algorithm, the method further includes: Obtain the pending business and determine the business requirements corresponding to the pending business; Based on the aforementioned business requirements, determine the target business that is consistent with the business requirements of the business to be processed; The service scheduling parameters of the target service are used as the service scheduling parameters of the service to be processed.
6. A service scheduling parameter determination apparatus characterized by comprising: The service scheduling parameter determination device includes: The acquisition module is used to acquire the service scenario information and service latency requirements of the communication service to be scheduled. The service scheduling granularity determination module is used to determine the service scheduling granularity based on the service scenario information; The service quality monitoring module is used to monitor the service quality of the communication service to be scheduled based on the service scheduling granularity and obtain latency data. The service scheduling parameter determination module is used to determine the end-to-end service latency based on the service latency requirements. The target latency from the wireless access network to the user plane function is determined based on the latency data. The service scheduling parameters are determined based on the service end-to-end latency and the target latency, and a preset parameter scheduling algorithm. The preset parameter scheduling algorithm is obtained by training a long short-term memory neural network model based on the wireless side scheduling related parameters and the wireless side latency data corresponding to the wireless side scheduling related parameters. The service scheduling granularity includes 5QI bearer granularity and / or network slice granularity; Before determining the end-to-end latency of the service based on the service latency requirements, the method further includes: Obtain relevant parameters for wireless side scheduling; Different data combinations are performed on the wireless side scheduling-related parameters to obtain the initial scheduling parameter combination; Determine the wireless side latency data corresponding to the initial scheduling parameter combination; The long short-term memory neural network model is trained based on the initial scheduling parameter combination and the wireless side latency data to obtain a preset parameter scheduling algorithm.
7. A service scheduling parameter determination device characterized by comprising: The device includes: a memory, a processor, and a service scheduling parameter determination program stored in the memory and executable on the processor, the service scheduling parameter determination program being configured to implement the steps of the service scheduling parameter determination method as described in any one of claims 1 to 5.
8. A storage medium, characterized by The storage medium stores a service scheduling parameter determination program, which, when executed by a processor, implements the steps of the service scheduling parameter determination method as described in any one of claims 1 to 5.
Citation Information
Patent Citations
Method for scheduling terminal resource, method for determining resource scheduling granularity and device
CN107734661A
Wireless service quality determination method and device, computer equipment and storage medium
CN112584417A
Service demand sensing and scheduling method and system, electronic equipment and storage medium
CN116017375A
