CN117255354A

CN117255354A - Service delay optimization method, device, equipment and storage medium

Info

Publication number: CN117255354A
Application number: CN202310894545.0A
Authority: CN
Inventors: 郑英; 朱英; 仇勇; 魏芹; 高源�; 孙正辉; 程超
Original assignee: China Mobile Communications Group Co Ltd; China Mobile Group Jiangsu Co Ltd
Current assignee: China Mobile Communications Group Co Ltd; China Mobile Group Jiangsu Co Ltd
Priority date: 2023-07-19
Filing date: 2023-07-19
Publication date: 2023-12-19

Abstract

The invention discloses a service delay optimization method, a device, equipment and a storage medium, wherein the method comprises the following steps: measuring and monitoring the uplink time delay and the downlink time delay of the wireless network side, and optimizing the uplink time delay and the downlink time delay according to the measurement result and the monitoring result; splitting the optimized uplink time delay and the optimized downlink time delay, and dividing the split downlink time delay and the split uplink time delay parameter set according to service requirements; and optimizing the divided uplink delay parameter set and the divided downlink delay parameter set according to a preset delay optimization model. The invention optimizes the uplink time delay and the downlink time delay, then splits and divides the optimized uplink time delay and downlink time delay, optimizes the divided uplink time delay parameter group and the divided downlink parameter group according to the preset time delay optimizing model, thereby expanding analysis and optimization on the processing process of the service mechanism, meeting the service quality requirement and improving the perception of the service time delay.

Description

Service delay optimization method, device, equipment and storage medium

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a method, an apparatus, a device, and a storage medium for service delay optimization.

Background

At present, differentiation is realized mainly by identifying 5QI (5G QoS Identifier,5G service quality identifier) or QCI (QoS Class Identifier, scale value) of different users or service bearers, then by 5QI and ARP (Address Resolution Protocol ) resource scheduling weights, GBR (Guaranteed Bit Rate ) uplink and downlink minimum guarantee rates, speed limit, pre-scheduling and other basic QoS (Quality of Service, service quality) means, the differentiation QoS is realized only by adopting a small number of functions or parameters of an MAC layer (data link layer), analysis and optimization are not carried out in the processing process of QoS flows, service perception cannot be effectively improved, and 5QI and QCI pipeline carrying users with great current network resource weight and high flow proportion cannot be effectively realized, so that the service quality requirement is met.

Disclosure of Invention

The invention mainly aims to provide a service delay optimization method, device, equipment and storage medium, which aim at solving the technical problem of how to analyze and optimize the QoS processing process and improve the perception of service delay.

In order to achieve the above object, the present invention provides a service delay optimization method, which includes the following steps:

measuring and monitoring uplink time delay and downlink time delay of a wireless network side, and optimizing the uplink time delay and the downlink time delay according to a measurement result and a monitoring result;

splitting the optimized uplink time delay and the optimized downlink time delay, and dividing the split downlink time delay and the split uplink time delay parameter set according to service requirements;

and optimizing the divided uplink delay parameter set and the divided downlink delay parameter set according to a preset delay optimization model.

Optionally, the step of optimizing the divided uplink delay parameter set and the divided downlink delay parameter set according to a preset delay optimization model includes:

screening a target parameter set from the divided uplink delay parameter set and the divided downlink delay parameter set according to a preset delay optimization model;

and optimizing the service delay according to the service requirement and the target parameter set.

Optionally, the step of screening the target parameter set from the divided uplink delay parameter set and the divided downlink delay parameter set according to the preset delay optimization model includes:

inputting the divided uplink delay parameter set and the divided downlink delay parameter set into a preset delay optimization model;

reserving the effective parameter sets of the divided uplink delay parameter set and the divided downlink delay parameter set according to an update gate of a preset delay optimization model;

and deleting the invalid parameter sets of the divided uplink time delay parameter set and the divided downlink time delay parameter set according to a reset gate of a preset time delay optimization model.

Optionally, the step of optimizing the service delay according to the service requirement and the target parameter set includes:

and combining all parameters in the target parameter set and all parameter settings according to the service demand to obtain a delay optimization strategy to optimize service delay.

Optionally, the step of splitting the optimized uplink delay and the optimized downlink delay, and dividing the split downlink delay and the split uplink delay parameter set according to the service requirement includes:

splitting the optimized uplink time delay and the optimized downlink time delay into an uplink and downlink media access layer time delay, a wireless link layer time delay and a packet data convergence layer time delay;

and integrating and dividing the parameter sets of the media access layer delay, the wireless link layer delay and the packet data convergence layer delay according to service requirements.

Optionally, the step of optimizing the uplink delay and the downlink delay according to the measurement result and the monitoring result includes:

determining the time delay requirements of the uplink time delay and the downlink time delay according to the measurement result and the monitoring result;

and when the uplink time delay and the downlink time delay meet the time delay requirement, optimizing the uplink time delay and the downlink time delay according to a target optimization strategy.

Optionally, the step of determining the delay requirements of the uplink delay and the downlink delay according to the measurement result and the monitoring result includes:

determining the time delay from the wireless network side to the service server according to the measurement result and the monitoring result;

and determining the time delay requirement of the uplink time delay and the time delay requirement of the downlink time delay according to the service requirement time delay and the time delay.

In addition, in order to achieve the above object, the present invention further provides a service delay optimization device, where the service delay optimization device includes: an optimizing module and a dividing module;

the optimizing module is used for measuring and monitoring the uplink time delay and the downlink time delay of the wireless network side, and optimizing the uplink time delay and the downlink time delay according to a measurement result and a monitoring result;

the division module is used for dividing the optimized uplink time delay and the optimized downlink time delay, and dividing the divided downlink time delay and the divided uplink time delay parameter groups according to service requirements;

the optimizing module is further used for optimizing the divided uplink time delay parameter set and the divided downlink time delay parameter set according to a preset time delay optimizing model.

In addition, in order to achieve the above object, the present invention also proposes a service delay optimization device, which includes a memory, a processor, and a service delay optimization program stored on the memory and operable on the processor, the service delay optimization program being configured to implement the service delay optimization method as described above.

In addition, in order to achieve the above object, the present invention further proposes a storage medium having a service delay optimization program stored thereon, which when executed by a processor implements the service delay optimization method as described above.

Drawings

FIG. 1 is a schematic structural diagram of a service delay optimization device of a hardware operating environment according to an embodiment of the present invention;

fig. 2 is a schematic flow chart of a first embodiment of the service delay optimization method of the present invention;

fig. 3 is a schematic flow chart of a second embodiment of the service delay optimization method of the present invention;

fig. 4 is a schematic flow chart of a third embodiment of the service delay optimization method of the present invention;

fig. 5 is a flow chart of RAN-side delay optimization according to an embodiment of the service delay optimization method of the present invention;

fig. 6 is a block diagram of a first embodiment of a service delay optimizing apparatus according to the present invention.

The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

Referring to fig. 1, fig. 1 is a schematic structural diagram of a service delay optimization device of a hardware running environment according to an embodiment of the present invention.

As shown in fig. 1, the service delay optimizing apparatus may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display (Display), and the optional user interface 1003 may also include a standard wired interface, a wireless interface, and the wired interface for the user interface 1003 may be a USB interface in the present invention. The network interface 1004 may optionally include a standard wired interface, a Wireless interface (e.g., a Wireless-Fidelity (Wi-Fi) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) or a stable Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.

Those skilled in the art will appreciate that the architecture shown in fig. 1 does not constitute a limitation of the traffic delay optimization device, and may include more or fewer components than shown, or may combine certain components, or may be a different arrangement of components.

As shown in fig. 1, an operating system, a network communication module, a user interface module, and a traffic delay optimization program may be included in a memory 1005, which is considered a type of computer storage medium.

In the service delay optimization device shown in fig. 1, the network interface 1004 is mainly used for connecting to a background server, and performing data communication with the background server; the user interface 1003 is mainly used for connecting user equipment; the service delay optimizing device invokes a service delay optimizing program stored in the memory 1005 through the processor 1001, and executes the service delay optimizing method provided by the embodiment of the present invention.

Based on the hardware structure, the embodiment of the service delay optimization method is provided.

Referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of a service delay optimization method according to the present invention, and the first embodiment of the service delay optimization method according to the present invention is provided.

Step S10: and measuring and monitoring the uplink time delay and the downlink time delay of the wireless network side, and optimizing the uplink time delay and the downlink time delay according to the measurement result and the monitoring result.

It should be noted that, the execution body of the present embodiment may be a computer software service device with functions of data processing, network communication and program running, for example, a service delay optimization device, etc., or other electronic devices capable of implementing the same or similar functions, which is not limited in this embodiment.

It should be appreciated that the 5G network serves as a new infrastructure, which is the infrastructure of the industrial internet floor, and is responsible for the importance of intelligent manufacturing and intelligent transformation. Most 5G wireless sites carry both personal and industry applications, except for individual industry specific sites. Business applications vary widely and network demands vary. The 5G network wants to implement and meet the Service-level Agreement (SLA) Service requirements using slicing and QoS policies. At present, aiming at some characteristic users or services, signing on different 5QI or QCI, and then using different MAC layers to differentially schedule based on the 5QI or QCI, mainly starting pre-scheduling, adjusting different resource scheduling weight factors according to quality of service grades, differentially ensuring uplink and downlink minimum guaranteeing rate, or carrying out basic Qos capacities such as uplink and downlink rate guaranteeing through GBR bearing. The resource guarantee capability of the 5G newly added slice ensures the differentiation of the service. However, to meet the latency requirement of a small number of services, opening pre-scheduling based on cell level and unified 5QI/QCI results in large uplink resource overhead; cell-level DRX for saving electricity and starting causes unstable time delay of industry application; the measures of resource weight proportion difference, slice resource reservation, pre-scheduling, speed limiting and the like are more problems of how to allocate the resource allocation. The prior art means mainly adopts a means and a measure for differentiation through the resource allocation proportion, and other optimization means and measures based on the service processing process are not considered to improve the service delay perception.

In order to overcome the above-mentioned drawbacks, in this embodiment, based on the whole flow processing procedure of the service user plane data flow at the wireless network side, the RAN side (wireless network side) is configured to parse the delay layer by layer, find out the specific level of the data link layer associated with the segmentation delay, train and learn the policy parameter sets of different protocol processing units, and formulate different combination policies, so as to implement hierarchical differential scheduling to meet the service quality requirement.

It should be noted that, according to the service requirement and attribute, carrying out service perception analysis; different services have respective service perception evaluation basis and means, such as a katon, an uplink and downlink delay quality difference cell and the like, but the delay is usually one of the most important evaluation basis. The business types comprise personal business requirements, industry businesses and different applications; the specific requirements and characteristics of related services are known, including the data service characteristics including the requirements and characteristics of data packet size, uplink and downlink delay and rate requirements, packet sending frequency, delay requirements, delay stability and the like. The game and video of the individual user are different, and the service demands of video under tremble audio and communication video of the video service are obviously different; the business requirements of PLC, AGV, video feedback, video definition and the like applied in the industry are different and the requirements of the industry on uplink time delay, speed and time delay stability are higher. Collecting, determining or optimizing and adjusting the bearing and slicing of the service subscription; according to the service type and QoS requirement, knowing or optimizing and adjusting the subscription conditions of the corresponding 5QI and slice; if the service is very important and the time delay is sensitive, but the default load or the general slice is signed, the service requirement cannot be effectively met; the default or general slice has a plurality of service types, and the differentiated policy can have larger influence on other services or users or consume more resources; for important services or services with high service demands, subscription dedicated carriers or dedicated slices are required.

It should be noted that, judging whether the service quality delay meets the requirements can be generally implemented by time delay indexes according to the service requirements, such as time delay, time delay stability and rate requirements, and the service perception is generally implemented by the conditions of blocking, jitter and the like. The time delay meets the requirements of packet loss rate, speed and the like; the embodiment utilizes the time delay meeting condition to evaluate the service perception or quality meeting condition; the service delay is usually the delay optimization of the wireless network side, and the delay from the base station to the server is usually stable; judging the time delay of the wireless network side, and if the service perception or quality is satisfied, performing according to the existing scheduling strategy of the service, and keeping the service still; and if the service awareness is not satisfied, analyzing and optimizing the wireless network side segmentation delay.

It can be understood that the periodic measurement and monitoring of the uplink and downlink segment delay can be mainly performed on the cells, 5QI and slices according to the QoS monitoring policy. The RAN side is mainly composed of CU and DU UP delays, the above behavior examples: RAN side delay; uldelay_nr_ SNw =uldelay_gbcuup_ SNw +uldelay_gbdu_ SNw; CU user plane uplink delay ul.delay_gbcuup=drb.pdcpreorddelayul+drb.pdcpf 1delay; DU user plane uplink delay uldelay_gnbdu=drb.rlcdelayui+drb.air ifdelayui. The time delay of CU and DU is based on the starting point and key meaning of time delay statistics, and the content of each time period is known and mastered; each sub-layer of the data link layer is influenced by the upper layer and the lower layer respectively, and the related processing routes are different; for example, the starting point is that the base station issues DCI authorization, the end point is that the uplink data is sent by the UE, correctly decoded by the base station L1, and sent to the RLC layer (Radio Link Control, radio link layer control protocol) to stop after being processed by the MAC, and a plurality of links including air interface transmission are experienced in the middle. When the uplink PDCP layer (Packet Data Convergence Protocol, packet data convergence protocol layer) is delayed and sent up, data is transmitted "wired" through the F1 interface, and bandwidth is basically not limited. The uplink can be sent to the PDCP layer as soon as the RLC layer completes the grouping, and only the processing power of PDCP is limited.

It can be appreciated that in a 5G network, CU and DU are separated, for example, delay insensitive user plane functions are placed in a Centralized processing Unit (CU), delay sensitive user plane functions are placed in a Distributed processing Unit (DU), and transmission is performed between CU and DU through a frontaul; based on the above description, the delay insensitive user plane functions mainly include PDCP layer, including header compression, ciphering integrity protection, retransmission, sender sequence number maintenance and receiver ordering for higher layer service data, etc.; the time delay sensitive user plane functions are mainly RLC, MAC and physical layer, including data segmentation, concatenation, re-segmentation, reorganization, multi-logic channel multiplexing, HARQ hybrid automatic repeat request, etc. The functions of the MAC medium access control layer include scheduling, logic channel priority and the like which are realized in DUs, so that the data transmission of CUs and DUs is more close to the air interface capability, the service quality QoS is met, and the transmission characteristics between CUs and DUs are met. The CU and the DU are connected through an F1 interface.

It should be noted that, if the uplink delay and the downlink delay do not meet the uplink delay requirement and the downlink delay requirement, the parameter policy of the service granularity of the existing network is directly invoked, and analysis and policy optimization are performed on the service granularity.

Further, in order to improve the service delay awareness, step S10 of this embodiment may include:

The wireless network side delay mainly comprises CU and DU UP delay, and is specifically decomposed into three parts of wireless network side uplink and downlink PDCP delay, MAC layer delay and RLC layer delay; the time delay statistic granularity can be a cell level, a 5QI level and a slice level; and according to the service subscription condition, time delays with different granularities can be selected for analysis and optimization.

It can be understood that the target optimization strategy can be a parameter strategy of the current network service granularity, and when the uplink and downlink delays of a specific slice corresponding to the service under the cell do not meet the service requirement, the uplink and downlink delays of the slice granularity and the segmentation delays are required to be combined for analysis and optimization; the present embodiment eliminates the case of poor wireless delay caused by the coverage and quality reasons of the wireless network side, and the coverage and quality problems of the wireless network side are usually accompanied by various problems such as wireless connection rate, disconnection rate, packet loss rate, etc., and can be solved by daily optimization means. In combination with uplink and downlink delay and segmentation delay conditions of service granularity and with combination of service characteristics, differential optimization is performed, and for convenience of understanding, table 1 is a policy optimization table of a wireless network side, where the table is an optimization result of uplink delay of the wireless network side, downlink delay of the wireless network side, uplink PDCP layer delay, uplink MAC layer delay, uplink RLC layer delay, downlink PDCP layer delay, downlink MAC layer delay and downlink RLC layer delay according to different policies.

Table 1_wireless network side policy optimization table

Further, in order to improve the service delay perceptibility, the step of determining the delay requirements of the uplink delay and the downlink delay according to the measurement result and the monitoring result includes:

In a specific implementation, according to the service demand time delay minus the time delay from the RAN side to the service server, the time delay demand of the uplink time delay and the time delay demand of the downlink time delay of the wireless network side can be calculated, and the comparison is carried out according to the same granularity of the service; when the uplink time delay and the downlink time delay meet the time delay requirement of the uplink time delay and the time delay requirement of the downlink time delay, directly calling a parameter strategy of the current network service granularity; and if not, analyzing and optimizing the strategy aiming at the granularity of the service.

Step S20: splitting the optimized uplink time delay and the optimized downlink time delay, and dividing the split downlink time delay and the split uplink time delay parameter sets according to service requirements.

It should be noted that splitting the optimized uplink delay and the optimized downlink delay may be divided into an uplink PDCP delay, an uplink RLC delay, an uplink MAC layer delay, a downlink PDCP delay, a downlink RLC delay, and a downlink MAC layer delay, and dividing parameter sets of the uplink PDCP delay, the uplink RLC delay, the uplink MAC layer delay, the downlink PDCP delay, the downlink RLC delay, and the downlink MAC layer delay.

The PDCP packet data convergence protocol layer is used for transmitting user plane and control plane data, reordering, supporting disorder delivery, repeated discarding, ROHC and the like; the PDCP parameter sets include related parameters such as a reordering timer, an out-of-order commit, a discard timer, PDCP duplication, and the like. The RLC layer is mainly responsible for cutting RLC data, removing repeated data and retransmitting RLC, and relates to parameters including RLC segmentation, uplink and downlink POLLING timers, uplink and downlink retransmission timers, reassembly timers, uplink and downlink forbidden state report timers, RLC AM/UM/TM different mode selection, RLC SN size and the like; the MAC layer medium access layer, mapping, scheduling, multiplexing and demultiplexing between logical channels and transport channels, HARQ (Hybrid Automatic Repeat reQuest ), logical channel priority setting, etc., and parameters related to pre-scheduling related parameters, BLER (block error rate), MCS (Modulation and Coding Scheme, modulation and coding strategy), 5QI bearer related QoS parameters, DRX, scheduling-free, etc.; the uplink and downlink scheduling function is realized by the MAC layer, and proper resource scheduling, MCS coding and the like are selected according to the reported CSI information (Channel State Information ), service requirements and the like. The PDCP layer, the RLC layer and the MAC layer can carry out differential strategies and carry out targeted optimization according to different services.

Step S30: and optimizing the divided uplink delay parameter set and the divided downlink delay parameter set according to a preset delay optimization model.

It should be noted that the preset time delay optimization model may be a GRU algorithm, and GRU (Gate Recurrent Unit) is one of cyclic neural networks (Recurrent Neural Network, RNN); compared with LSTM, the GRU can achieve a considerable effect, and training is easier to perform in comparison, so that training efficiency can be improved to a great extent. The GRU model has only two gates, namely an update gate and a reset gate, namely a zt update gate and a rt reset gate in the figure, and the two gates are in the range of 0-1. The front memory information of the gate control is updated, the data quantity at the current moment can be kept, and the past information of the gate control is reset; the update gate is used to control the extent to which the state information at the previous time is brought into the current state, a larger value of the update gate indicates that the state information at the previous time is brought more. The reset gate is used to control the degree to which state information at a previous time is ignored, a smaller value of the reset gate indicating more is ignored. The gating z can simultaneously perform forgetting and selection memory; the range of the gating signal z is 0-1, and the closer the gating signal is to 1, the more data are stored; the closer to 0 then the more "forgotten" is represented; reset gates are a vector of 0 to 1 that measure the magnitude of the gate opening. For example, if the gating value corresponding to an element is 0, then the information representing the element is completely forgotten.

In a specific implementation, according to service related bearing and scheduling granularity, a GRU algorithm is utilized to perform data training calculation, PDCP/RLC/MAC layer scheduling strategy parameters meeting different service requirements are found out, a new parameter set strategy is constructed, and different PDCP/RLC/MAC parameter set strategies are differentially scheduled according to the minimum granularity of the service.

It should be noted that the calculation may be performed separately or simultaneously as needed; and the influence of each layer of parameter setting on the upper layer and the lower layer of time delay is considered in separate calculation. The time delay input parameter vectors are as follows: { PDCP group ID different parameter setting, RLC group ID different parameter setting, MAC group ID different parameter setting, service type, service packet size, per GB flow uplink PRB and CCE resource ratio, segmentation delay or RAN side delay of service association granularity (5 QI, slice, etc.), etc.; the comparison before and after the time delay can be performed according to the same granularity, for example, 5QI is adopted, and 5QI is adopted before and after the time delay; if slicing is adopted, slicing is adopted before and after. The per-uplink traffic unit uplink PRB resource overhead (Physical Resource Block ) and uplink CCE (Control Channel Element, aggregation level) overhead are increased while being calculated from the delay improvement dimension. Different service data packets have different sizes and different influences on resources; when compared with the service, the ratio of each GB flow resource is not much different; if the resource cost ratio changes greatly, the relation with unreasonable parameter setting is larger; unnecessary resource expenditure caused by low-delay requirements is avoided as much as possible, and network interference is increased. When the delay is improved, a value between 0 and 1 is set according to the improvement condition, and finally different parameter combination strategies meeting different service demands are found out, different PDCP strategy IDs, RLC strategy IDs and MAC strategy IDs are constructed, and service perception and resource utilization efficiency are improved.

In order to facilitate understanding, referring to fig. 5, fig. 5 is a RAN side delay optimization flow chart, in which attributes such as service type and service QoS guarantee requirement, service subscription bearer and slice are determined, whether service delay is satisfied is determined, when service delay is satisfied, processing is not performed, an existing policy is maintained, when service delay is not satisfied, uplink and downlink delays of different granularity segments of the RAN side are split into uplink and downlink CU-UP delays, uplink and downlink DU-UP delays are split into uplink and downlink PDCP delays, uplink and downlink RLC layer delays and uplink and downlink MAC layer delays, and finally, a combination policy suitable for the service is found according to a GRU algorithm, that is, a suitable target parameter set is found, including a PCDC differentiation policy and an RLC differentiation policy, that is, an MAC differentiation policy, and a secondary combination policy is performed.

The embodiment measures and monitors the uplink time delay and the downlink time delay of the wireless network side, and optimizes the uplink time delay and the downlink time delay according to the measurement result and the monitoring result; splitting the optimized uplink time delay and the optimized downlink time delay, and dividing the split downlink time delay and the split uplink time delay parameter set according to service requirements; and optimizing the divided uplink delay parameter set and the divided downlink delay parameter set according to a preset delay optimization model. In the embodiment, the uplink time delay and the downlink time delay are optimized, then the optimized uplink time delay and downlink time delay are split and divided, and the divided uplink time delay parameter set and the divided downlink parameter set are optimized according to the preset time delay optimizing model, so that analysis and optimization are developed in the processing process of a service mechanism, the service quality requirement is met, and the perceptibility of the service time delay is improved.

Referring to fig. 3, fig. 3 is a flow chart of a second embodiment of the service delay optimization method according to the present invention, and based on the first embodiment shown in fig. 2, the second embodiment of the service delay optimization method according to the present invention is proposed.

In a second embodiment, the step S30 includes:

step S301: screening a target parameter set from the divided uplink delay parameter set and the divided downlink delay parameter set according to a preset delay optimization model;

step S302: and optimizing the service delay according to the service requirement and the target parameter set.

It should be noted that, the transmission of the data packet at the wireless network side is mainly performed between the UE and the data link layer and the air interface at the RAN side; the data link layer comprises a PDCP layer, an RLC layer, a MAC layer and an SDAP layer (Service Data Adaptation Protocol) and a service data adaptation protocol layer; the SDAP mainly performs QoS Flow to DRB mapping on uplink and downlink data, and the differentiation strategy is mainly implemented in a PDCP layer, an RLC layer and a MAC layer. Each parameter set comprises different parameters and different parameter settings, and different target parameter sets can be formed by specific setting combination aiming at different parameters in the same parameter set.

In a specific implementation, examples are illustrated with a set of scheduling parameters: the scheduling parameter group comprises a scheduling parameter group identifier ID, an uplink and downlink IBLER target value, an uplink pre-scheduling interval, an uplink pre-scheduling data volume, an uplink and downlink MCS maximum value, an uplink and downlink MCS minimum value, an uplink intelligent pre-scheduling duration time, an SR scheduling head packet TB size, an SR period and the like; the uplink and downlink IBLER target values may be set to different values of 0.001%, 0.01%, 0.1%, 1%, 3%, 5%, 10%, etc., where different IBER values are suitable for different scenes, and may be combined with other parameters, such as SR scheduling periods (5 slots, 10 slots, 20 slots, etc.), to form different scheduling target parameter sets.

Further, in order to improve accuracy of optimizing the service delay, step S301 of this embodiment may further include:

In a specific implementation, by means of a GRU algorithm, service characteristic data comprise data packet size, packet sending period, delay jitter, resident 5G or specific frequency band and the like, service wireless environment parameters such as RSRP, SINR and the like, and a wireless network side parameter strategy such as a PDCP parameter group ID, an RLC parameter group ID, an MAC layer parameter group ID, a scheduling parameter group ID and a DRX parameter group ID is used as an input item, and parameters related to reservation and forgetting are selected according to the condition of meeting the uplink delay or the downlink delay or the segmentation delay of a wireless network side at a certain moment; if the parameters with invalid wireless uplink and downlink time delay or segmented time delay are forgotten through a reset gate, the parameters with valid time delay are reserved through an update gate; training in combination with specific services and parameter settings.

Further, in order to improve the accuracy of optimizing the service delay, the step of optimizing the service delay according to the service requirement and the target parameter set includes:

It can be understood that specific parameters and actual parameter settings suitable for actual services and service granularity can be found, target parameter sets meeting service requirements can be used, and parameter IDs of different target parameter sets can be combined according to the actual parameter settings meeting the requirements, so that differentiation of specific services and refined service requirements are realized.

According to the embodiment, a target parameter set is screened from the divided uplink time delay parameter set and the divided downlink time delay parameter set according to a preset time delay optimization model; and optimizing the service delay according to the service requirement and the target parameter set. In the embodiment, the target parameter set is screened from the divided uplink delay parameter set and the divided downlink delay parameter set, and the target parameter set optimizes the service delay, so that the optimization accuracy of the cassia service delay is improved.

Referring to fig. 4, fig. 4 is a schematic flow chart of a third embodiment of the service delay optimization method according to the present invention, and based on the first embodiment shown in fig. 2, the third embodiment of the service delay optimization method according to the present invention is proposed.

In a third embodiment, the step S20 further includes:

step S201: splitting the optimized uplink time delay and the optimized downlink time delay into an uplink and downlink media access layer time delay, a wireless link layer time delay and a packet data convergence layer time delay;

step S202: and integrating and dividing the parameter sets of the media access layer delay, the wireless link layer delay and the packet data convergence layer delay according to service requirements.

It can be appreciated that the medium access layer delay is the MAC layer delay, the radio link layer delay is the RLC layer delay, and the packet data convergence layer delay is the PDCP layer.

Note that, the PDCP, RLC, MAC layer parameter set is a main parameter; the scheduling parameter set, the DRX parameter set and the srs period also belong to the MAC layer parameters, and are divided independently according to the unused functions; the parameter group division and integration can be performed according to actual needs.

It should be noted that the same-frequency switching parameter set, the different-frequency switching parameter set, and the different-system parameter set belong to the RRC layer, and a differentiation policy may be performed on parameters that need to be differentially moved, where the switching class differentiation policy is more specific to the cell-level or 5 QI-level ToC service and a part of the ToB service that needs to be moved.

The embodiment divides the optimized uplink time delay and the optimized downlink time delay into an uplink and downlink media access layer time delay, a wireless link layer time delay and a packet data convergence layer time delay; and integrating and dividing the parameter sets of the media access layer delay, the wireless link layer delay and the packet data convergence layer delay according to service requirements. In the embodiment, when the uplink and downlink time delay is split into the uplink and downlink media access layer time delay, the wireless link layer time delay and the packet data convergence layer, the parameter sets are integrated and divided, so that the wireless network side time delay is optimized in a batch of pertinence, and the accuracy of optimizing the service time delay is improved.

In addition, the embodiment of the invention also provides a storage medium, wherein the storage medium is stored with a service delay optimization program, and the service delay optimization program realizes the service delay optimization method when being executed by a processor.

In addition, referring to fig. 6, an embodiment of the present invention further provides a service delay optimization device, where the service delay optimization device includes: an optimization module 10 and a division module 20;

the optimizing module 10 is configured to measure and monitor an uplink delay and a downlink delay of a wireless network side, and optimize the uplink delay and the downlink delay according to a measurement result and a monitoring result;

the dividing module 20 is configured to split the optimized uplink delay and the optimized downlink delay, and divide the split downlink delay and the split uplink delay into parameter sets according to service requirements;

the optimizing module 10 is further configured to optimize the post-division uplink delay parameter set and the post-division downlink delay parameter set according to a preset delay optimizing model.

Based on the first embodiment of the service delay optimizing device of the present invention, a second embodiment of the service delay optimizing device of the present invention is provided.

In this embodiment, the optimizing module 10 is configured to screen the target parameter set from the divided up-delay parameter set and the divided down-delay parameter set according to a preset delay optimizing model.

Further, the optimizing module 10 is further configured to optimize service delay according to the service requirement and the target parameter set.

Further, the optimizing module 10 is further configured to input the divided uplink delay parameter set and the divided downlink delay parameter set to a preset delay optimizing model.

Further, the optimizing module 10 is further configured to reserve, according to an update gate of a preset delay optimizing model, the valid parameter sets of the partitioned uplink delay parameter set and the partitioned downlink delay parameter set.

Further, the optimizing module 10 is further configured to delete the invalid parameter sets of the partitioned uplink delay parameter set and the partitioned downlink delay parameter set according to a reset gate of a preset delay optimizing model.

Further, the optimizing module 10 is further configured to combine each parameter in the target parameter set and each parameter setting according to the service requirement, and obtain a delay optimizing policy to optimize the service delay.

Further, the dividing module 20 is further configured to split the optimized uplink delay and the optimized downlink delay into an uplink and downlink media access layer delay, a radio link layer delay, and a packet data convergence layer delay.

Further, the dividing module 20 is further configured to integrate and divide the parameter sets of the medium access layer delay, the radio link layer delay and the packet data convergence layer delay according to service requirements.

Further, the optimizing module 10 is further configured to determine the delay requirement of the uplink delay and the downlink delay according to the measurement result and the monitoring result.

Further, the optimizing module 10 is further configured to optimize the uplink delay and the downlink delay according to a target optimizing policy when the uplink delay and the downlink delay meet the delay requirement.

Further, the optimizing module 10 is further configured to determine a delay from the wireless network side to the service server according to the measurement result and the monitoring result.

Further, the optimizing module 10 is further configured to determine a delay requirement of the uplink delay and a delay requirement of the downlink delay according to the service requirement delay and the delay.

Other embodiments or specific implementation manners of the service delay optimization device of the present invention may refer to the above method embodiments, and are not described herein again.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.

The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. read only memory mirror (Read Only Memory image, ROM)/random access memory (Random Access Memory, RAM), magnetic disk, optical disk), comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present invention.

The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.

Claims

1. The service delay optimization method is characterized by comprising the following steps of:

2. The service delay optimization method according to claim 1, wherein the step of optimizing the divided uplink delay parameter set and the divided downlink delay parameter set according to a preset delay optimization model comprises:

3. The service delay optimization method of claim 2, wherein the step of screening the target parameter set from the divided up-delay parameter set and the divided down-delay parameter set according to the preset delay optimization model comprises:

4. The service delay optimization method of claim 2, wherein the optimizing the service delay according to the service requirement and the target parameter set comprises:

5. The service delay optimization method of claim 1, wherein the splitting the optimized uplink delay and the optimized downlink delay and dividing the parameter sets of the split downlink delay and the split uplink delay according to service requirements comprises:

6. The service delay optimization method of claim 1, wherein the step of optimizing the uplink delay and the downlink delay according to the measurement result and the monitoring result comprises:

7. The service delay optimization method of claim 6, wherein the step of determining the delay requirements of the uplink delay and the downlink delay according to the measurement result and the monitoring result comprises:

8. A traffic delay optimizing apparatus, the traffic delay optimizing apparatus comprising: an optimizing module and a dividing module;

9. A traffic delay optimization device, the traffic delay optimization device comprising: memory, a processor and a traffic delay optimization program stored on the memory and executable on the processor, which when executed by the processor, implements the steps of the traffic delay optimization method according to any one of claims 1 to 7.

10. A storage medium having stored thereon a traffic delay optimization program which, when executed by a processor, implements the steps of the traffic delay optimization method according to any of claims 1 to 7.