WO2021259114A1 - 外环值确定方法、装置、设备及存储介质 - Google Patents

外环值确定方法、装置、设备及存储介质 Download PDF

Info

Publication number
WO2021259114A1
WO2021259114A1 PCT/CN2021/100443 CN2021100443W WO2021259114A1 WO 2021259114 A1 WO2021259114 A1 WO 2021259114A1 CN 2021100443 W CN2021100443 W CN 2021100443W WO 2021259114 A1 WO2021259114 A1 WO 2021259114A1
Authority
WO
WIPO (PCT)
Prior art keywords
outer loop
model
value
grid
air interface
Prior art date
Application number
PCT/CN2021/100443
Other languages
English (en)
French (fr)
Inventor
李建国
刘巧艳
史珂
马泽鹏
鲁大伟
庄梦溪
范淼
孙胜男
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Priority to US18/002,920 priority Critical patent/US20230246729A1/en
Priority to KR1020227045932A priority patent/KR20230016679A/ko
Priority to EP21829532.7A priority patent/EP4170941A4/en
Publication of WO2021259114A1 publication Critical patent/WO2021259114A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0009—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the channel coding
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B17/00—Monitoring; Testing
    • H04B17/30—Monitoring; Testing of propagation channels
    • H04B17/309—Measuring or estimating channel quality parameters
    • H04B17/336—Signal-to-interference ratio [SIR] or carrier-to-interference ratio [CIR]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0002—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the transmission rate
    • H04L1/0003—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the transmission rate by switching between different modulation schemes
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0015—Systems modifying transmission characteristics according to link quality, e.g. power backoff characterised by the adaptation strategy
    • H04L1/0019—Systems modifying transmission characteristics according to link quality, e.g. power backoff characterised by the adaptation strategy in which mode-switching is based on a statistical approach
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
    • H04L1/0033—Systems modifying transmission characteristics according to link quality, e.g. power backoff arrangements specific to the transmitter
    • H04L1/0034—Systems modifying transmission characteristics according to link quality, e.g. power backoff arrangements specific to the transmitter where the transmitter decides based on inferences, e.g. use of implicit signalling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/1607—Details of the supervisory signal
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/20—Arrangements for detecting or preventing errors in the information received using signal quality detector
    • H04L1/203—Details of error rate determination, e.g. BER, FER or WER

Definitions

  • This application relates to the field of wireless communication technology, and in particular to a method, device, device, and storage medium for determining an outer loop value.
  • AMC Adaptive Modulation and Coding
  • the traditional AMC needs to adjust the outer loop value (Acknowledge/Non-Acknowledge, ACK/NACK) to achieve the target reliability requirements of network settings.
  • ACK/NACK outer loop value
  • the above convergence process often requires dozens or even hundreds of transmission opportunities for user equipment; it will result in some user equipment that has not yet converged, and the information transmission has ended, or has converged, but the previous dozens or even hundreds of transmissions have not been transmitted.
  • the use of the best modulation and coding strategy restricts the improvement of spectrum efficiency.
  • This application provides methods, devices, equipment, and storage media for determining the outer loop value.
  • an embodiment of the present application provides a method for determining an outer loop value, including: determining a pre-trained outer loop initialization model based on the current feature data of the user equipment; and based on the current air interface measurement value of the user equipment and the outer loop initialization model Determine the initial outer loop value of the user equipment.
  • an embodiment of the present application provides a device for determining an outer loop value, including: a model determining module configured to determine a pre-trained outer loop initialization model based on current feature data of the user equipment; an outer loop value determining module configured To determine the initial outer loop value of the user equipment based on the current air interface measurement value of the user equipment and the outer loop initialization model.
  • an embodiment of the present application provides a device, including: one or more processors; a memory, configured to store one or more programs; when the one or more programs are used by the one or more processors Execution, so that the one or more processors implement any one of the methods in the embodiments of the present application.
  • an embodiment of the present application provides a storage medium that stores a computer program, and when the computer program is executed by a processor, any one of the methods in the embodiments of the present application is implemented.
  • FIG. 1 is a flowchart of a method for determining an outer loop value provided by an embodiment of the present application
  • FIG. 2 is a flowchart of learning an outer loop initialization model in an offline state provided by an embodiment of the present application
  • Fig. 3 is a schematic diagram of grid division provided by an embodiment of the present application.
  • Fig. 4 is a schematic diagram of grid division provided by an embodiment of the present application.
  • Fig. 5 is a flowchart of a model application method provided by an embodiment of the present application.
  • Fig. 6 is a flowchart of a triggering method from offline learning to online learning provided by an embodiment of the present application
  • FIG. 7 is a flowchart of an online learning method for an outer loop initialization model provided by an embodiment of the present application.
  • FIG. 8 is a flowchart of a method for correcting an outer loop initialization model provided by an embodiment of the present application.
  • FIG. 9 is an overall flow chart of determining the outer loop value provided by an embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of a device for determining an outer loop value provided by an embodiment of the present application.
  • Fig. 11 is a schematic structural diagram of a device provided by the present application.
  • the wireless channel Compared with the wired channel, the wireless channel has a narrower coherence bandwidth and shorter coherence time, which is a significant feature of the wireless channel. Due to the characteristics of rapid changes in wireless channels, it is difficult to make full use of spectrum resources if a fixed modulation and coding method is adopted. Therefore, adaptive coding and modulation technology is introduced in wireless communication technology to improve spectrum efficiency.
  • the basic idea of adaptive coding and modulation is: adapting to the channel conditions and adopting an appropriate channel modulation and coding strategy, thereby improving the spectrum efficiency and user perception.
  • the outer loop In order to achieve the adaptation of the modulation and coding strategy and the user channel conditions in the traditional adaptive modulation and coding technology, the outer loop needs to be adjusted according to the ACK/NACK information fed back by the user to achieve the target reliability requirements set by the system.
  • the above convergence process often requires dozens or even hundreds of transmission opportunities for user equipment; this will result in some user equipment that has not yet converged and the transmission has ended, or has converged, but the previous dozens or even hundreds of transmissions have not adopted the most advanced transmission.
  • Optimal modulation and coding strategy which in turn restricts the improvement of spectrum efficiency.
  • the optimal modulation and coding scheme is not adopted, it will also affect the user's perception of the service.
  • the present application provides a method for determining an outer loop value.
  • FIG. 1 is a flowchart of a method for determining an outer loop value provided in an embodiment of the present application. This embodiment is applicable to the case of adaptive coding and modulation technology. The method may be executed by the outer loop value determining device provided in the embodiment of the present application, and the device may be implemented in software and/or hardware.
  • the method for determining the outer loop value mainly includes steps S11 and S12.
  • S12 Determine an initializing outer loop value of the user equipment based on the current air interface measurement value of the user equipment and the outer loop initialization model.
  • the pre-trained outer loop initialization model is an outer loop initialization model trained in an offline state, and all offline models in this application may refer to a pretrained outer loop initialization model.
  • the outer loop initialization model includes a first outer loop model and a second outer loop model; the determination of the pre-trained outer loop initialization model based on the current feature data of the user equipment includes: The feature data and the grid division strategy determine the grid to which the user equipment belongs; obtain the first outer ring model and the second outer ring model corresponding to the grid to which the user equipment belongs.
  • the determining the initial outer loop value of the user equipment based on the current air interface measurement value of the user equipment and the outer loop initialization model includes: determining based on the current air interface measurement value and the first outer loop model A first outer loop value; determining a second outer loop value based on the current air interface measurement value and the second outer loop model; determining an initial outer loop value based on the first outer loop value and the second outer loop value.
  • the determining the first outer loop value based on the current air interface measurement value and the first outer loop model includes: determining the first outer loop value based on the current air interface measurement value and the first outer loop model The modulation and coding strategy MCS corresponding to the current air interface measurement value; wherein the first outer loop model is a corresponding relationship model between the air interface measurement value and the MCS under the set target BLER interval; the target BLER is obtained based on the determined MCS and the reference demodulation curve The corresponding system mapping SINR; the difference between the system mapping SINR and the current air interface measurement value is determined as the first outer loop value.
  • the determining the second outer loop value based on the current air interface measurement value and the second outer loop model includes: obtaining the second outer loop value based on the current air interface measurement value through the second outer loop model Two outer loop values; wherein, the second outer loop model is a corresponding relationship model between the air interface measurement value and the outer loop value under the set target BLER interval.
  • the method before determining the pre-trained outer loop initialization model based on the current feature data of the user equipment, the method further includes: training the outer loop initialization model based on the historical feature data of the user equipment.
  • the training of the outer loop initialization model based on the historical feature data of the user equipment includes: performing grid division and grid division based on the historical feature data; determining the confirmation response in each grid/ Deny response to the number of ACK/NACK information and outer loop value; calculate the block error rate of each grid based on the number of ACK/NACK information in each grid; determine based on the block error rate and outer loop value of each grid Initialize the model of the outer loop.
  • performing grid division and grid division based on the historical feature data includes: performing grid division on each cell in the network based on the historical feature data; for each grid grid according to the scheduled MCS and Corresponding air interface measurement values are divided into grids.
  • the determining the number of ACK/NACK information and the outer loop value in each grid includes: obtaining ACK/NACK (ACK/NACK) information corresponding to demodulation; The information and the corresponding outer loop information are delivered to the corresponding grid; the number of ACK/NACK messages in each grid and the outer loop value are counted.
  • ACK/NACK ACK/NACK
  • the determining the outer loop initialization model based on the BLER and outer loop value of each grid includes: in each grid, determining the set target BLER interval based on the BLER of each grid The MCS corresponding to the air interface measurement value and the outer loop value corresponding to the air interface measurement value; the first outer loop model is determined based on the correspondence between the air interface measurement value and the MCS; based on the correspondence between the air interface measurement value and the outer loop value The relationship determines the second outer ring model.
  • the determination of the MCS corresponding to the air interface measurement value and the outer loop value corresponding to the air interface measurement value under the set target block error rate BLER interval based on the block error rate of each grid includes: for one MCS, if The number of grids meeting the set block error rate BLER interval is greater than the preset value, and the weighting factor of each grid is calculated according to the number of ACK/NACK information in each grid; each grid is calculated according to the weighting factor of each grid Perform a weighted average calculation on the air interface measurement values corresponding to the grids to obtain the air interface measurement values corresponding to the MCS; perform a weighted average calculation on the outer loop values corresponding to each grid according to the weight factor of each grid to obtain the MCS Corresponding outer loop value; determining the outer loop value corresponding to the air interface measurement value according to the air interface measurement value corresponding to the MCS and the outer loop value corresponding to the MCS.
  • the determination of the MCS corresponding to the air interface measurement value under the target BLER interval and the outer loop value corresponding to the air interface measurement value based on the BLER of each grid includes: for an MCS, if the set error block is satisfied The number of grids in the rate interval is less than the preset value, find the air interface measurement value interval corresponding to the target BLER interval corresponding to the MCS on the reference demodulation curve; the median value in the air interface measurement value interval that is searched for is initialized with the system default outer loop The difference between the values is used as the air interface measurement value corresponding to the MCS; the system default initialization outer loop value is used as the air interface measurement value corresponding to the outer loop value.
  • the method further includes: obtaining the first network before the application model Performance index and the second network performance index after using the outer loop initialization model; when the first network performance index and the second network performance index meet preset conditions, start online learning to obtain the outer loop online model; Use the outer loop online model to determine the new initial outer loop value of the user equipment.
  • outer-loop online model is an outer-loop initialization model trained in an online state, and all online models in this application may refer to the outer-loop online model.
  • the method further includes: correcting the outer ring model of the corresponding grid based on the ACK/NACK information after the user in the grid uses the outer ring model for the first time; the outer ring model includes a pre-trained outer ring Initialize the model or the outer loop online model.
  • the correction of the outer ring model of the corresponding grid based on the ACK/NACK information after the user in the grid uses the outer ring model for the first time includes: for each grid, the calculation of the ACK/NACK information is denied. Respond to the proportion of NACK; when the proportion of NACK is greater than the target value, reduce the outer loop value of the outer loop model according to the corresponding strategy; when the proportion of NACK is less than the target value, press The corresponding strategy increases the outer loop value of the outer loop model.
  • a learning method of an outer loop initialization model in an offline state mainly includes the following steps:
  • the grid is divided according to the collected historical characteristic data of users in the network.
  • historical characteristic data includes: user transmission mode, scheduled air interface measurement value SINR, space division mark; space division number; supported maximum modulation order (64QAM, 256QAM); corresponding demodulation ACK/NACK; outer loop Information etc.
  • the cell is divided into four grids, corresponding to low interference low road loss grid 1, high interference low road Damage grid 2, low interference high road damage grid 3, high interference high road damage grid 4.
  • the MCS uses the 1st order as the granularity, as the horizontal axis, and the air interface measurement SINR value uses 0.5dB as the granularity, as the vertical axis, for two-dimensional division.
  • S25 Deliver the ACK/NACK information and outer loop information corresponding to the actual demodulation to the corresponding grid of the grid.
  • the BLER of the corresponding grid and the corresponding outer loop value are calculated according to the number of ACK/NACK counted by each grid in each grid. If the corresponding raster sample size is lower than a certain threshold, this raster is marked as invalid and recorded as NULL.
  • the specific method is, in the grid, for each MCS, search for a grid whose corresponding BLER meets the set BLER interval, thereby determining the air interface measurement value SINR corresponding to the MCS, and the outer loop value corresponding to the air interface measurement SINR.
  • the calculation method is: the ratio of the number of samples (total number of ACK/NACK) in the grid that meets the condition to the sum of the number of samples in all the grids that meet the condition.
  • the MCS does not meet the grid of the set BLER interval, find the interval of the SINR value corresponding to the target BLER interval of the MCS on the system reference demodulation curve (usually the simulated curve), and then take the SINR interval
  • the median value minus the system default initialized outer loop value is the air interface measurement value SINR corresponding to the MCS, and the default initialized outer loop value is its corresponding outer loop value.
  • each air interface measurement value SINR corresponds to the scheduling MCS that meets the target BLER interval
  • the outer loop value corresponding to each air interface measurement value SINR that is, the learned model is to set the target BLER interval
  • the relationship model between the measured value of the lower air interface SINR and the scheduling MCS is denoted as:
  • Set the target BLER interval air interface measurement value SINR and the outer loop value ⁇ SINR relationship model is recorded as:
  • a model application method is provided. As shown in Figure 5, the model application method mainly includes the following steps:
  • S31 Determine the attribution grid according to the current feature data corresponding to the user, and schedule a model of the corresponding grid.
  • the grid to which the user equipment belongs is determined, and the model of the corresponding grid is obtained.
  • a triggering method from offline learning to online learning is provided.
  • the triggering method from offline learning to online learning mainly includes the following steps:
  • an online learning method of the outer loop initialization model is provided. As shown in Figure 7, the online learning method of the outer loop model mainly includes the following steps:
  • S51 Determine feature segmentation points according to the distribution of the newly captured feature data, and then perform grid division.
  • the outer loop initialization model correction method mainly includes the following steps:
  • S61 Initialize the ACK/NACK information corresponding to the outer loop scheduling according to the adopted model of the grid cache, and calculate the NACK ratio.
  • S62 Calculate the model adjustment amount according to the relationship between the calculated NACK ratio and the target BLER.
  • an overall flow chart for determining the outer loop value is provided, as shown in FIG. 9, which mainly includes 7 steps: offline learning, model application, model evaluation, online learning, model application, effect evaluation, and model modification.
  • a method for determining the outer loop value mainly includes the following steps:
  • the grid is divided according to the collected historical characteristic data of users in the network.
  • historical characteristic data includes: user transmission mode, scheduled air interface measurement value SINR, space division mark; space division number; supported maximum modulation order (64QAM, 256QAM); corresponding demodulation ACK/NACK; outer loop Information etc.
  • space division According to the number of space divisions (2, 3, 4, 5, and greater than or equal to 6, divided into 5 types), two interference levels (high interference, low interference), and the maximum modulation order supported ( 64QAM, 256QAM) and the number of RBs (size RB) are divided into 40 grids.
  • grid-level division may also include user chip type information, demodulation algorithm information, user beam position information, and so on. It should be noted that the above-mentioned grid division method is only an exemplary description, not a limitation.
  • MCS uses the first order as the granularity, as the horizontal axis, and the SINR value uses 0.5dB as the granularity, as the vertical axis, for two-dimensional division.
  • the captured new characteristic data it is judged whether it is space division. If it is not space division, according to the transmission mode, interference level, supported maximum modulation order configuration information, and the number of scheduled RBs, find the user data attribution According to the scheduled MCS and air interface measurement value SINR, the corresponding actual demodulation corresponding ACK/NACK information and outer loop information are delivered to the corresponding grid of the grid, and recorded in each grid The number of ACK/NACK messages (two separately) and outer loop value.
  • space division according to the number of space division, interference level, the configuration information of the maximum modulation order supported, and the number of scheduled RBs, find the grid to which the user data belongs; and calculate it according to the scheduled MCS and air interface measurement value SINR
  • the corresponding ACK/NACK information corresponding to the actual demodulation, and the outer loop information are delivered to the corresponding grid of the grid where they are located, and the number of ACK/NACK (two separately recorded) and the outer loop value are recorded in each grid.
  • the BLER of the corresponding grid and the outer loop value of the corresponding grid are calculated according to the number of ACK/NACK counted by each grid in each grid. If the corresponding raster sample size is lower than a certain threshold, this raster is marked as invalid and recorded as NULL.
  • the specific method is, in the grid, for each MCS, search for a grid whose BLER meets the set BLER interval, so as to determine the air interface measurement value SINR corresponding to the MCS and the outer loop value corresponding to the air interface measurement value SINR.
  • the calculation method is: the ratio of the number of samples (total number of ACK/NACK) in the grid that meets the condition to the sum of the number of samples in all the grids that meet the condition.
  • the MCS does not meet the grid of the set BLER interval, find the interval of the demodulation value SINR corresponding to the target BLER interval of the MCS on the system reference demodulation curve (usually the simulated curve), and then take the interval
  • the median value of minus the system default initialization outer loop, the SINR value is measured for the air interface corresponding to the MCS, and the default initialized outer loop value is its corresponding outer loop value.
  • the relationship model between the air interface measurement value SINR of the target BLER interval and the scheduling MCS is recorded as:
  • D_SINR_Init min ⁇ max ⁇ (D_SINR0+D_SINR1)/2, -V_limt ⁇ , V_limt ⁇
  • V_limt is the protection limit value, its value is greater than 0, the default is 10.
  • the collection of the relationship between the SE of the corresponding network and the average air interface measurement SINR is: And count the proportion of user-level BLER falling within the target BLER interval as R1; where K 1 is the number of sample points, and Set_p 1 represents the set of network performance before the model is applied.
  • D_SE i represents the difference between the SE after the application model corresponding to the i-th element in the above intersection and the SE before the application model.
  • S8102 selecting an appropriate segmentation point according to the distribution of the feature quantity used for grid division of the current cell, for example, selecting 50% of the distribution as a threshold to perform grid division.
  • BLER( ⁇ ) represents the BLER under the corresponding combination
  • Represents the relationship model between the air interface measurement value SINR and the outer loop value corresponding to the j-th grid; j 1,...64;
  • N represents the number of MCS that the system can support.
  • D_SINR_Init min ⁇ max ⁇ (D_SINR0+D_SINR1)/2, -V_limt ⁇ , V_limt ⁇
  • V_limt is the protection limit value, its value is greater than 0, the default is 10.
  • S812 Determine whether
  • NACK_R NACK_R
  • the model adjustment is calculated according to the following relationship:
  • BLER_tart is the set target BLER
  • ⁇ is the model adjustment step size.
  • SINR i max ⁇ min(SINR i - ⁇ ,SINR i + ⁇ ),SINR i - ⁇ , get a new model:
  • an offline method for initializing the outer loop model to determine the outer loop value is provided.
  • the offline initialization outer loop model determines the outer loop value method mainly includes the following steps:
  • the maximum modulation order (64QAM, 256QAM) and the number of RBs (size RB) are divided into 24 grids; in the case of space division: according to the number of space divisions (2, 3, 4, 5, and greater than or equal to 6, points 5 types), two interference levels (high interference, low interference), the maximum modulation order supported (64QAM, 256QAM), and the number of RBs (size RB) are divided into 40 grids, and grid-level division is also possible Contains the user's chip type information, demodulation algorithm information, user's beam position information, and so on.
  • S92 Perform grid division for each grid according to the scheduled MCS and the corresponding air interface measurement channel quality value SINR.
  • MCS takes the first order as the granularity, as the horizontal axis
  • SINR value takes 0.5dB as the granularity, as the vertical axis, for two-dimensional division.
  • the new transmission characteristic data of the user determine whether it is space division. If it is not space division, find out which user data belongs to according to the transmission mode, interference level, maximum modulation order configuration information supported, and the number of scheduled RBs.
  • the weighting factor of each grid is: the number of samples contained in the grid that meets the condition (total ACK/NACK Number) and the ratio of the sum of the sample numbers of all grids that meet the conditions;
  • the MCS does not satisfy the grid of the set BLER interval, find the SINR interval corresponding to the target BLER interval of the MCS on the system reference demodulation curve (usually the simulated curve), and then take the middle of the SINR interval
  • the value minus the system default initialization outer loop is the air interface measurement value SINR corresponding to the MCS, and the default initialization outer loop is its corresponding outer loop.
  • each measured air interface measurement value SINR corresponds to the scheduling MCS that meets the target BLER interval
  • the outer loop value corresponding to the air interface measurement value SINR that is, the learned model
  • the relationship model between the air interface measurement value SINR of the target BLER interval and the scheduling MCS is recorded as:
  • BLER( ⁇ ) represents the BLER under the corresponding combination
  • j 1,...64
  • N represents the number of MCS that the system can support
  • D_SINR_Init min ⁇ max ⁇ (D_SINR0+D_SINR1)/2, -V_limt ⁇ , V_limt ⁇
  • V_limt is the protection limit value, its value is greater than 0, and the default is 10;
  • NACK_R calculates the model adjustment amount according to the following relational expression:
  • BLER_tart is the set target BLER
  • ⁇ is the model adjustment step size.
  • SINR i max ⁇ min(SINR i - ⁇ ,SINR i + ⁇ ),SINR i - ⁇ , get a new model:
  • a method of revising the online model is provided.
  • the grid is divided according to the collected historical characteristic data of users in the network.
  • historical characteristic data includes: user transmission mode, scheduled air interface measurement value SINR, space division mark; space division number; supported maximum modulation order (64QAM, 256QAM); corresponding demodulation ACK/NACK; outer loop Information etc.
  • space division According to the number of space divisions (2, 3, 4, 5, and greater than or equal to 6, divided into 5 types), two interference levels (high interference, low interference), and the maximum modulation order supported ( 64QAM, 256QAM) and the number of RBs (size RB) are divided into 40 grids.
  • grid-level division may also include user chip type information, demodulation algorithm information, user beam position information, and so on. It should be noted that the above-mentioned grid division method is only an exemplary description, not a limitation.
  • MCS uses the first order as the granularity, as the horizontal axis, and the SINR value uses 0.5dB as the granularity, as the vertical axis, for two-dimensional division.
  • the captured new characteristic data it is judged whether it is space division. If it is not space division, according to the transmission mode, interference level, supported maximum modulation order configuration information, and the number of scheduled RBs, find the user data attribution According to the scheduled MCS and air interface measurement value SINR, the corresponding actual demodulation corresponding ACK/NACK information and outer loop information are delivered to the corresponding grid of the grid, and recorded in each grid The number of ACK/NACK messages (two separately) and outer loop value.
  • space division according to the number of space division, interference level, the configuration information of the maximum modulation order supported, and the number of scheduled RBs, find the grid to which the user data belongs; and calculate it according to the scheduled MCS and air interface measurement value SINR
  • the corresponding ACK/NACK information corresponding to the actual demodulation, and the outer loop information are delivered to the corresponding grid of the grid where they are located, and the number of ACK/NACK (two separately recorded) and the outer loop value are recorded in each grid.
  • the BLER corresponding to each grid and the outer loop value corresponding to each grid are calculated according to the number of ACK/NACK counted by each grid in each grid. If the corresponding raster sample size is lower than a certain threshold, this raster is marked as invalid and recorded as NULL.
  • S105 Determine the correspondence model between the air interface measurement value SINR and the dispatched MCS and the correspondence relationship model between the air interface measurement value SINR and the outer loop value according to the BLER and the outer loop value under each grid in the grid.
  • the specific method is that in the grid, for each MCS, find the grid whose BLER meets the set BLER interval, so as to determine the air interface measurement value SINR corresponding to the grid and the outer loop value corresponding to the grid.
  • the calculation method is: the ratio of the number of samples (total number of ACK/NACK) in the grid that meets the condition to the sum of the number of samples in all the grids that meet the condition.
  • each air interface measurement value SINR corresponds to the scheduling MCS that meets the target BLER interval, and the outer loop value corresponding to the air interface measurement value SINR, that is, the learned model is:
  • the relationship model between the air interface measurement value SINR of the target BLER interval and the scheduling MCS is recorded as:
  • SINR0 the current air interface measurement value corresponding to the user, by the model
  • the corresponding MCS is determined
  • the median value of the corresponding SINR in the target BLER interval obtained by the MCS and the BLER curve referenced by the system is recorded as SINR1
  • D_SINR_Init min ⁇ max ⁇ (D_SINR0+D_SINR1)/2, -V_limt ⁇ , V_limt ⁇
  • V_limt is the protection limit value, its value is greater than 0, the default is 10.
  • D_SE i represents the difference between the SE after the application model corresponding to the i-th element in the above intersection and the SE before the application model.
  • S10111 Determine the grid to which the user belongs according to the characteristic data corresponding to the user and the strategy of dividing grids, and obtain the model of the corresponding grid And model
  • D_SINR_Init min ⁇ max ⁇ (D_SINR0+D_SINR1)/2, -V_limt ⁇ , V_limt ⁇
  • V_limt is the protection limit value, its value is greater than 0, and the default is 10;
  • D_SE i represents the difference between the SE after the application model and the SE before the application model corresponding to the i-th element in the aforementioned intersection.
  • D_SE_mean ⁇ Th 0 and R 2 ⁇ Th 1 and R 0 ⁇ R 1 ; or D_SE_mean ⁇ Th 0 use the cell default outer loop and stop the application model; otherwise, continue to use the current model for the outer loop of newly accessed users initialization
  • this application provides a method for determining an outer loop value.
  • FIG. 10 is a schematic structural diagram of a device for determining an outer loop value provided in an embodiment of this application. This embodiment is applicable to the case of adaptive coding and modulation technology. , The device can be implemented in software and/or hardware.
  • the method for determining the outer loop value mainly includes a step model determining module 101 and an outer loop value determining module 102.
  • the model determining module 101 is configured to determine a pre-trained outer loop initialization model based on the current feature data of the user equipment;
  • the outer loop value determining module 102 is configured to determine the initial outer loop value of the user equipment based on the current air interface measurement value of the user equipment and the outer loop initialization model.
  • the outer loop initialization model includes a first outer loop model and a second outer loop model. Ring model.
  • the model determination module 101 is configured to determine the grid to which the user equipment belongs based on the current feature data of the user equipment and the grid division strategy; obtain the first outer ring corresponding to the grid to which the user equipment belongs Model and the second outer ring model.
  • the outer loop value determining module 102 is configured to determine a first outer loop value based on the current air interface measurement value and the first outer loop model; based on the current air interface measurement value and the second outer loop value The outer loop model determines a second outer loop value; the initial outer loop value is determined based on the first outer loop value and the second outer loop value.
  • the determining the first outer loop value based on the current air interface measurement value and the first outer loop model includes: determining the current air interface based on the current air interface measurement value and the first outer loop model The modulation and coding strategy MCS corresponding to the measured value; wherein, the first outer loop model is the corresponding relationship model between the air interface measured value and the MCS under the set target block error rate BLER interval; the target is obtained based on the determined MCS and the reference demodulation curve The system mapped SINR corresponding to the BLER; the difference between the system mapped SINR and the current air interface measurement value is determined as the first outer loop value.
  • the determining the second outer loop value based on the current air interface measurement value and the second outer loop model includes: obtaining the second outer loop value based on the current air interface measurement value through the second outer loop model Two outer loop values; wherein, the second outer loop model is a corresponding relationship model between the air interface measurement value and the outer loop value under the set target BLER interval.
  • the device further includes: a model training module configured to train the outer loop based on historical feature data of the user equipment before determining the pre-trained outer loop initialization model based on the current feature data of the user equipment Ring initialization model.
  • a model training module configured to train the outer loop based on historical feature data of the user equipment before determining the pre-trained outer loop initialization model based on the current feature data of the user equipment Ring initialization model.
  • the model training module is configured to perform grid division based on the historical feature data; determine the number of acknowledgment/non-acknowledgement ACK/NACK messages and the outer loop value in each grid; The number of ACK/NACK information in the grid calculates the block error rate of each grid; and the outer loop initialization model is determined based on the block error rate and the outer loop value of each grid.
  • performing grid division based on the historical feature data includes: performing grid division on each cell in the network based on the historical feature data; for each grid grid according to the scheduled MCS and the corresponding air interface measurement The values are divided into grids.
  • the determining the number of ACK/NACK information in each grid includes: obtaining the ACK/NACK information corresponding to the demodulation acknowledgment response/denial response; combining the ACK/NACK information and the corresponding outer loop The information is delivered to the corresponding grid; the number of ACK/NACK messages in each grid and the outer loop value are counted.
  • the determining the outer loop initialization model based on the block error rate and the outer loop value of each grid includes: in each grid, determining the setting based on the block error rate of each grid The MCS corresponding to the air interface measurement value under the target block error rate BLER interval and the outer loop value corresponding to the air interface measurement value; the first outer loop model is determined based on the correspondence between the air interface measurement value and the MCS; based on the air interface measurement value and The corresponding relationship of the outer loop values determines the second outer loop model.
  • the determination of the MCS corresponding to the air interface measurement value under the set target block error rate BLER interval based on the block error rate of each grid and the air interface measurement value and the corresponding outer loop value includes: for one MCS, If the number of grids satisfying the set block error rate interval is greater than the preset value, the weighting factor of each grid is calculated; the air interface measurement value corresponding to each grid is weighted and averaged according to the weighting factor of each grid, Obtain the air interface measurement value corresponding to the MCS; perform a weighted average calculation on the outer loop value corresponding to each grid according to the weight factor of each grid to obtain the outer loop value corresponding to the MCS; The air interface measurement value and the outer loop value corresponding to the MCS determine the outer loop value corresponding to the air interface measurement value.
  • the determination of the MCS corresponding to the air interface measurement value under the set target block error rate BLER interval based on the block error rate of each grid and the air interface measurement value and the corresponding outer loop value includes: for one MCS, If the number of grids satisfying the set error rate interval is less than the preset value, find the air interface measurement value interval corresponding to the target block error rate interval corresponding to the MCS on the reference demodulation curve; change the median value in the air interface measurement value interval The difference between the system default initialized outer loop value and the system default initialization outer loop value is used as the air interface measurement value corresponding to the MCS; the system default initialization outer loop value is used as the outer loop value corresponding to the air interface measurement value.
  • the apparatus further includes: an online model learning module configured to determine the initial outer loop value of the user equipment based on the current air interface measurement value of the user equipment and the outer loop initialization model , Obtain the first network performance index before applying the model and the second network performance index after using the outer loop initialization model; when the first network performance index and the second network performance index meet the preset conditions, start the online Learn to obtain the outer loop online model; use the outer loop online model to determine the new initial outer loop value of the user equipment.
  • an online model learning module configured to determine the initial outer loop value of the user equipment based on the current air interface measurement value of the user equipment and the outer loop initialization model , Obtain the first network performance index before applying the model and the second network performance index after using the outer loop initialization model; when the first network performance index and the second network performance index meet the preset conditions, start the online Learn to obtain the outer loop online model; use the outer loop online model to determine the new initial outer loop value of the user equipment.
  • the device further includes: an online model adjustment module configured to correct the outer ring model of the corresponding grid based on the ACK/NACK information after the user in the grid uses the outer ring model for the first time;
  • the ring model includes an outer ring initialization model or an outer ring online model.
  • the adjusting the online model based on the ACK/NACK information after using the online model includes: calculating the proportion of NACK in the ACK/NACK information in the ACK/NACK information for each grid; in the When the proportion of NACK is greater than the target value, reduce the outer loop value of the outer loop model according to the corresponding strategy;
  • the outer loop value of the outer loop model is increased according to a corresponding strategy.
  • the outer loop value determination device provided in this embodiment can execute the outer loop value determination method provided in any embodiment of the present application, and has the corresponding functional modules and beneficial effects for executing the method.
  • the outer loop value determination method provided in any embodiment of this application please refer to the outer loop value determination method provided in any embodiment of this application.
  • FIG. 11 is a schematic structural diagram of a device provided by the present application.
  • the device includes a processor 111, a memory 112, an input device 113, and an output device 114;
  • the number of processors 111 may be one or more.
  • one processor 111 is taken as an example; the processor 111, the memory 112, the input device 113 and the output device 114 in the device may be connected by a bus or other means, FIG. 11 Take the bus connection as an example.
  • the memory 112 can be used to store software programs, computer-executable programs, and modules, such as program instructions/modules corresponding to the outer loop value determination method in the embodiment of the present application (for example, the outer loop value determination device The model determination module 101 and the outer loop value determination module 102 in ).
  • the processor 111 executes various functional applications and data processing of the device by running software programs, instructions, and modules stored in the memory 112, that is, implements any method provided in the embodiments of the present application.
  • the memory 112 may mainly include a program storage area and a data storage area.
  • the program storage area may store an operating system and an application program required by at least one function; the data storage area may store data created according to the use of the device, and the like.
  • the memory 112 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.
  • the memory 112 may further include a memory remotely provided with respect to the processor 111, and these remote memories may be connected to the device through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
  • the input device 113 can be used to receive inputted numeric or character information, and generate key signal input related to user settings and function control of the device.
  • the output device 114 may include a display device such as a display screen.
  • the embodiment of the present application also provides a storage medium containing computer-executable instructions, when the computer-executable instructions are executed by a computer processor, are used to perform an outer loop value determination method, including:
  • the initial outer loop value of the user equipment is determined based on the current air interface measurement value of the user equipment and the outer loop initialization model.
  • a storage medium containing computer-executable instructions provided by the embodiments of the present application and the computer-executable instructions are not limited to the above-mentioned operations, and can also execute the outer loop value determination method provided by any embodiment of the present application. Related operations.
  • the method, device, device, and storage medium for determining the outer loop value determine the pre-trained outer loop initialization model based on the current feature data of the user equipment; determine based on the current air interface measurement value of the user equipment and the outer loop initialization model
  • the technical solution for initializing the outer loop value of the user equipment solves the problems of resource waste and reduced user perception caused by non-convergence of small packet users, and realizes the improvement of the convergence speed of the adaptive modulation and coding technology.
  • user terminal encompasses any suitable type of wireless user equipment, such as a mobile phone, a portable data processing device, a portable web browser, or a vehicle-mounted mobile station.
  • the various embodiments of the present application can be implemented in hardware or dedicated circuits, software, logic or any combination thereof.
  • some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software that may be executed by a controller, microprocessor, or other computing device, although the application is not limited thereto.
  • Computer program instructions can be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code written in any combination of one or more programming languages or Object code.
  • ISA instruction set architecture
  • the block diagram of any logic flow in the drawings of the present application may represent program steps, or may represent interconnected logic circuits, modules, and functions, or may represent a combination of program steps and logic circuits, modules, and functions.
  • the computer program can be stored on the memory.
  • the memory can be of any type suitable for the local technical environment and can be implemented using any suitable data storage technology, such as but not limited to read-only memory (ROM), random access memory (RAM), optical storage devices and systems (digital multi-function discs) DVD or CD disc) etc.
  • Computer-readable media may include non-transitory storage media.
  • the data processor can be any type suitable for the local technical environment, such as but not limited to general-purpose computers, special-purpose computers, microprocessors, digital signal processors (DSP), application-specific integrated circuits (ASIC), programmable logic devices (FGPA) And processors based on multi-core processor architecture.
  • DSP digital signal processors
  • ASIC application-specific integrated circuits
  • FGPA programmable logic devices

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Probability & Statistics with Applications (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

一种外环值确定方法、装置、设备及存储介质，首先根据网络中的用户特征数据进行宫格划分和栅格划分，训练模型；基于用户设备的当前特征数据确定预训练的外环初始化模型（S11）；基于用户设备的当前空口测量值和所述外环初始化模型确定用户设备的初始化外环值（S12），应用模型；根据应用效果评估模型；根据评估效果来修正模型。

Description

外环值确定方法、装置、设备及存储介质
相关申请的交叉引用
本申请基于申请号为202010591941.2、申请日为2020年6月24日的中国专利申请提出，并要求该中国专利申请的优先权，该中国专利申请的全部内容在此引入本申请作为参考。
技术领域
本申请涉及无线通信技术领域，具体涉及一种外环值确定方法、装置、设备及存储介质。
背景技术
由于无线信道变化快的特点，如果采用固定的调制编码方式，很难充分的利用频谱资源。因此在无线通信技术中引入了自适应编码调制技术(Adaptive Modulation and Coding，AMC)用来提升频谱效率。
传统的AMC为了实现调制编码策略和用户信道条件的适配，需要确认响应/否认响应(Acknowledge/Non-Acknowledge，ACK/NACK)来调整外环值，以实现网络设置的目标可靠性要求。然而，上述收敛过程往往需要用到用户设备几十次甚至上百次传输机会；就会导致有些用户设备还没收敛，信息传输已经结束，或收敛了但是前面几十甚至上百次的传输没有采用最佳的调制编码策略，制约了频谱效率的提升。
在第五代移动通信中由于引入了Massive Mimo技术，信息传输充分利用了空域资源；另外由于传输带宽更大；这些因素会导致在相同数据包大小的条件下，5G对数据包的调度次数会更少，进一步加剧了小包用户不收敛引起的资源浪费和用户感知降低的问题。
发明内容
本申请提供用于确定外环值的方法、装置、设备和存储介质。
第一方面，本申请实施例提供一种外环值确定方法，包括：基于用户设备的当前特征数据确定预训练的外环初始化模型；基于用户设备的当前空口测量值和所述外环初始化模型确定用户设备的初始化外环值。
第二方面，本申请实施例提供一种外环值确定装置，包括：模型确定模块，被配置为基于用户设备的当前特征数据确定预训练的外环初始化模型；外环值确定模块，被配置为基于用户设备的当前空口测量值和所述外环初始化模型确定用户设备的初始化外环值。
第三方面，本申请实施例提供一种设备，包括：一个或多个处理器；存储器，用于存储一个或多个程序；当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现如本申请实施例中的任意一种方法。
第四方面，本申请实施例提供了一种存储介质，所述存储介质存储有计算机程序，所述计算机程序被处理器执行时实现本申请实施例中的任意一种方法。
关于本申请的以上实施例和其他方面以及其实现方式，在附图说明、具体实施方式和权利要求中提供更多说明。
附图说明
图1是本申请实施例提供的一种外环值确定方法的流程图；
图2是本申请实施例提供的离线状态下外环初始化模型的学习的流程图；
图3是本申请实施例提供的宫格划分的示意图；
图4是本申请实施例提供的栅格划分的示意图；
图5是本申请实施例提供的模型应用方法的流程图；
图6是本申请实施例提供的离线学习到在线学习的触发方法的流程图；
图7是本申请实施例提供的外环初始化模型在线学习方法的流程图；
图8是本申请实施例提供的外环初始化模型修正方法的流程图；
图9是本申请实施例提供的外环值确定的整体流程图；
图10是本申请实施例提供的一种外环值确定装置的结构示意图；
图11是本申请提供的一种设备的结构示意图。
具体实施方式
为使本申请的目的、技术方案和优点更加清楚明白，下文中将结合附图对本申请的实施例进行详细说明。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互任意组合。
在附图的流程图示出的步骤可以在诸如一组计算机可执行指令的计算机系统中执行。并且，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
随着移动通信技术的发展，人们对通信的需求也在逐渐提升，从语音为主的通信发展到今天的数据为主的通信，特别是第四代移动通信逐渐改变了我们的生活方式。如今，人们通信的需求还在不断提升。如，通过更高的业务速率，来实时观看高清直播；在虚拟环境下和远在千里之外的亲人，面对面交流；如身临其境，近在咫尺；目前的网络已经带给人们不错的视觉、听觉体验，但人们还想有触觉；在交通运输业、公共服务业、制造业等垂直行业物联网的需求、远程医疗的需求等。为满足这些需求，出现了第五代移动通信技术。
未来的5G通信，业务的多样性以及通信方式的多样性，势必增加无线方案设计的复杂度。再以原来的规划思维，来解决目前或将来遇到问题的复杂度可能是指数级的增长，导致问题无法解决。但未来的通信，会产生大量的数据，基于这些经验数据可以通过机器学习的方式，学习相关模型，利用经验模型做预测、做策略等等。
由于无线信道相对有线信道具有更窄的相干带宽，更短的相干时间，是无线信道的显著特点。由于无线信道变化快的特点，如果采用固定的调制编码方式，很难充分的利用频谱资源。因此在无线通信技术中引入了自适应编码调制技术用来提升频谱效率。自适应编码调制的基本思路是：自适应于信道条件采用合适的信道调制编码策略，进而提升了频谱效率和用户感知。
传统的自适应调制编码技术为了实现调制编码策略和用户信道条件的适配，需要根据用户反馈的ACK/NACK信息来调整外环，以实现系统设置的目标可靠性要求。然而，上述收敛过程往往需要用到用户设备几十次甚至上百次传输机会；就会导致有些用户设备还没收敛传输已经结束，或收敛了但是前面几十甚至上百次的传输没有采用最佳的调制编码策略，进而使得频谱效率提升受到了制约。另外，由于没有采用最佳的调制编码方案，也会影响用户对于业务的感知。
在4G、5G网络中都会存在大量小包用户，即业务量小的用户，小包用户业务的特点是 制约自适应编码调制方案提升频谱效率的关键因素。因为是小包，没有更多的学习机会使得自适应编码过程收敛。另外，因为5G采用更大的带宽，更多的天线充分利用了空间维度，使得在相同业务量的条件下，5G对用户的传输机会会更少。
针对上述问题，本申请提供了如下技术方案。
在一个实施例中，本申请提供一种外环值确定方法，图1是本申请实施例提供的一种外环值确定方法的流程图，本实施例可适用于自适应编码调制技术的情况，该方法可以由本申请实施例提供的外环值确定装置来执行，该装置可采用软件和/或硬件的方式实现。
如图1所示，本申请实施例提供的外环值确定方法主要包括步骤S11和S12。
S11、基于用户设备的当前特征数据确定预训练的外环初始化模型。
S12、基于用户设备的当前空口测量值和所述外环初始化模型确定用户设备的初始化外环值。
需要说明的是，预训练的外环初始化模型是在离线状态下训练的外环初始化模型，本申请中的离线模型均可以是指预训练的外环初始化模型。
在一个实施方式中，所述外环初始化模型包括第一外环模型和第二外环模型；所述基于用户设备的当前特征数据确定预训练的外环初始化模型，包括:基于用户设备的当前特征数据以及宫格划分策略确定所述用户设备所属宫格；获取所述用户设备所属宫格对应的第一外环模型和第二外环模型。
在一个实施方式中，所述基于用户设备的当前空口测量值和所述外环初始化模型确定用户设备的初始化外环值，包括：基于所述当前空口测量值和所述第一外环模型确定第一外环值；基于所述当前空口测量值和所述第二外环模型确定第二外环值；基于所述第一外环值和所述第二外环值确定初始化外环值。
在一个实施方式中，所述基于所述当前空口测量值和所述第一外环模型确定第一外环值，包括：基于所述当前空口测量值和所述第一外环模型确定所述当前空口测量值对应的调制与编码策略MCS；其中，所述第一外环模型是设定目标BLER区间下空口测量值与MCS的对应关系模型；基于确定的MCS和参考解调曲线获取目标BLER对应的系统映射SINR；将所述系统映射SINR与所述当前空口测量值的差值确定为第一外环值。
在一个实施方式中，所述基于所述当前空口测量值和所述第二外环模型确定第二外环值，包括：基于所述当前空口测量值，通过所述第二外环模型获取第二外环值；其中，所述第二外环模型是设定目标BLER区间下空口测量值与外环值的对应关系模型。
在一个实施方式中，所述基于用户设备的当前特征数据确定预训练的外环初始化模型之前，还包括：基于所述用户设备的历史特征数据训练外环初始化模型。
在一个实施方式中，所述基于所述用户设备的历史特征数据训练外环初始化模型，包括：基于所述历史特征数据进行宫格划分和栅格划分；确定每个栅格中的确认响应/否认响应ACK/NACK信息数目和外环值；基于每个栅格中的ACK/NACK信息数目计算每个栅格的误块率；基于所述每个栅格的误块率和外环值确定外环初始化模型。
在一个实施方式中，基于所述历史特征数据进行宫格划分和栅格划分，包括：基于所述历史特征数据对网络中的各个小区进行宫格划分；对于每个宫格按照调度的MCS和对应的空口测量值进行栅格划分。
在一个实施方式中，所述确定每个栅格中的ACK/NACK信息数目和外环值，包括：获取解调对应的确认响应/否认响应(ACK/NACK)信息；将所述ACK/NACK信息以及对应的外环 信息投递到对应的栅格中；统计每个栅格中ACK/NACK信息数目以及外环值。
在一个实施方式中，所述基于所述每个栅格的BLER和外环值确定外环初始化模型，包括：在每个宫格中，基于每个栅格的BLER确定设定目标BLER区间下空口测量值对应的MCS以及空口测量值对应的外环值；基于所述空口测量值和所述MCS的对应关系确定第一外环模型；基于所述空口测量值和所述外环值的对应关系确定第二外环模型。
在一个实施方式中，所述基于每个栅格的误块率确定设定目标误块率BLER区间下空口测量值对应的MCS以及空口测量值对应的外环值，包括：对于一个MCS，如果满足设定误块率BLER区间的栅格数量大于预设数值，根据每个栅格内的ACK/NACK信息数目计算每个栅格的权重因子；按照所述每个栅格的权重因子对各栅格对应的空口测量值进行加权平均计算，得到所述MCS对应的空口测量值；按照所述每个栅格的权重因子对各栅格对应的外环值进行加权平均计算，得到所述MCS对应的外环值；根据所述MCS对应的空口测量值和所述MCS对应的外环值确定所述空口测量值对应的外环值。
在一个实施方式中，所述基于每个栅格的BLER确定设定目标BLER区间下空口测量值对应的MCS以及空口测量值对应的外环值，包括：对于一个MCS，如果满足设定误块率区间的栅格数量小于预设数值，查找该MCS在参考解调曲线上对应的目标BLER区间所对应的空口测量值区间；将查找的空口测量值区间中的中值与系统默认初始化外环值的差值，作为该MCS对应的空口测量值；将系统默认初始化外环值作为空口测量值对应的外环值。
在一个实施方式中，所述基于所述用户设备的当前空口测量值和所述外环初始化模型确定所述用户设备的初始化外环值之后，所述方法还包括：获取应用模型前第一网络性能指标和使用外环初始化模型后的第二网络性能指标；在所述第一网络性能指标和所述第二网络性能指标满足预设条件的情况下，启动在线学习，得到外环在线模型；使用外环在线模型确定用户设备新的初始化外环值。
需要说明的是，外环在线模型是在在线状态下训练的外环初始化模型，本申请中的在线模型均可以是指外环在线模型。
在一个实施方式中，所述方法还包括：基于宫格中用户首次使用外环模型后的ACK/NACK信息对对应宫格的外环模型进行修正；所述外环模型包括预训练的外环初始化模型或外环在线模型。
在一个实施方式中，所述基于宫格中用户首次使用外环模型后的ACK/NACK信息对对应宫格的外环模型进行修正，包括:针对每个宫格，计算ACK/NACK信息中否认响应NACK所占比例；在所述NACK所占比例大于目标数值的情况下，按相应的策略减少所述外环模型的外环值；在所述NACK所占比例小于目标数值的情况下，按相应的策略增加所述外环模型的外环值。
在一个实施例中，提供一种离线状态下外环初始化模型的学习方法。如图2所示，离线状态下外环初始化模型的学习方法主要包括如下步骤：
S21、根据历史特征数据进行宫格划分。
根据收集到的网络中用户的历史特征数据进行宫格的划分。
具体的，历史特征数据包括：用户传输模式，调度的空口测量值SINR，空分标记；空分流数；所支持的最大调制阶数(64QAM,256QAM)；对应的解调ACK/NACK；外环信息等。所述用户传输模式包括：单端口、闭环复用RI＝1，闭环复用RI＝2。
如图3所示，根据设置的干扰水平(高干扰、低干扰)和路损水平划分门限，将小区 划分为是四个宫格，分别对应低干扰低路损宫格1、高干扰低路损宫格2、低干扰高路损宫格3、高干扰高路损宫格4。
S22、对于每个宫格，按调度的MCS和对应的空口测量值SINR进行栅格划分。
如图4所示，MCS以1阶为颗粒度，作为横轴，空口测量SINR值以0.5dB为颗粒度，作为纵轴、进行二维划分。
S23、循环离线抓取新的特征数据。
S24、根据抓取到的新的特征数据，确定新的特征数据归属的宫格，以及对应宫格下的栅格。
S25、将实际解调对应的ACK/NACK信息和外环信息投递到所在宫格的对应栅格。
S26、判断抓取的新的特征数据是否处理完毕，若是，则执行S27，若否，则执行S23。
S27、统计计算每个宫格中每个栅格下的BLER以及外环值。
具体的，根据每个宫格中的每个栅格统计的ACK/NACK数目计算对应栅格的BLER，和对应的外环值。如果对应的栅格样本量低于一定阈值，此栅格标记为无效栅格记为NULL。
S28、根据宫格中的每个栅格下统计的BLER和对应的外环值确定设定目标BLER区间下空口测量值SINR与调度MCS之间对应关系模型和空口测量值SINR与外环值的对应关系模型。
具体方法为，在宫格中，针对每一个MCS查找对应的BLER满足设定BLER区间的栅格，从而确定MCS对应的空口测量值SINR，以及空口测量SINR对应的外环值。
S281、如果对于一个MCS满足设定BLER区间的栅格的个数大于1个，计算每个栅格的权重因子。
计算方法为：满足条件的栅格所含样本的数量(总的ACK/NACK数目)与所有满足条件的栅格的样本数量之和的比值。
S282、按权重因子对各个栅格对应的空口测量值SINR加权平均，对各个栅格对应的外环值进行加权平均。
S283、从而确定空口测量SINR下，对应的满足目标BLER区间的，调度的MCS，以及空口测量SINR对应的外环值。
S284、如果对于一个MCS没有满足设定BLER区间的栅格，找此MCS在系统参考解调曲线上(一般是仿真出来的曲线)目标BLER区间所对应的SINR值的区间，然后取SINR区间的中值减去系统默认初始化外环值，为该MCS对应的空口测量值SINR，默认初始化外环值为其对应的外环值。
S285、通过上述过程就可以确定，每个空口测量值SINR对应的满足目标BLER区间的调度MCS，以及每个空口测量值SINR对应的外环值，即学习到的模型为，设定目标BLER区间下空口测量值SINR与调度MCS之间的关系模型记为：
Figure PCTCN2021100443-appb-000001
设定目标BLER区间空口测量值SINR与外环值ΔSINR关系模型记为：
Figure PCTCN2021100443-appb-000002
在一个实施例中，提供模型应用方法。如图5所示，模型应用方法主要包括如下步骤：
S31、根据用户对应的当前特征数据确定归属宫格，调度对应宫格的模型。
具体的，根据用户对应的当前特征数据以及划分宫格的策略判断用户设备所属的宫格，获取对应宫格的模型
Figure PCTCN2021100443-appb-000003
和
Figure PCTCN2021100443-appb-000004
S32、根据用户设备对应的空口测量值SINR0，由模型
Figure PCTCN2021100443-appb-000005
确定对应的MCS，由MCS和系统参考的BLER曲线获取在目标BLER区间内对应的SINR的中值记为 SINR1，由SINR0和SINR1计算第一外环值D_SINR0＝SINR1-SINR0。
S33、由用户设备对应的空口测量值SINR0，通过模型
Figure PCTCN2021100443-appb-000006
获取对应的第二外环值记为：D_SINR1。
S34、根据第一外环值D_SINR0和第二外环值D_SINR1计算用户设备的初始化外环值：D_SINR_Init。
在一个实施例中，提供离线学习到在线学习的触发方法。如图6所示，离线学习到在线学习的触发方法主要包括如下步骤：
S41、统计模型应用前对应网络中SE与平均空口测量值SINR的对应关系集合，以及用户级的BLER落在目标BLER区间的比例。
S42、统计模型应用后对应网络中SE与平均空口测量值SINR的对应关系集合，以及用户级的BLER落在目标BLER区间的比例。
S43、判断模型应用前和模型应用后对应关系集合是否有交集，若有交集，则执行S44，若否，则执行S45。
S44、根据交集部分和收敛比例判断是否进行在线学习，若是，则执行S46，若否，则执行S47。
S45、根据收敛比例判断是否进行在线学习，若是，则执行S46，若否，则执行S47。
S46、触发在线学习。
S47、继续使用离线模型。
在一个实施例中，提供外环初始化模型在线学习方法。如图7所示，外环模型在线学习方法主要包括如下步骤：
S51、根据新抓取的特征数据的分布确定特征分割点，进而进行宫格划分。
S52、对于每个宫格，按照SINR和调度MCS进行栅格划分。
S53、在线获取特征数据。
S54、根据数据特征确定数据归属的宫格，以及对应的宫格下的栅格。
S55、将此条数据对应的ACK/NACK信息以及外环信息，投递到归属的栅格中。
S56、判断样本量是否满足要求，如果满足要求，执行S57，如果不满足要求，执行S53。
S57、统计计算每个宫格中栅格下的BLER以及外环值。
S58、为每个MCS选取目标BLER区间内对应的SINR和外环值。
S59、确定在线学习的SINR和调度MCS的关系模型和SINR与外环值的关系模型。
在一个实施例中，提供外环初始化模型修正方法。如图8所示，外环初始化模型修正方法主要包括如下步骤：
S61、根据宫格缓存的采用模型初始化外环调度对应的ACK/NACK信息，计算NACK的比例。
S62、根据计算的NACK的比例和目标BLER之间的关系计算模型调整量。
S63、采用上述调整量对模型进行修正。
S64、修正后的模型传递给模型应用模块。
在一个实施例中，提供外环值确定的整体流程图，如图9所示，主要包括：离线学习，模型应用、模型评估、在线学习、模型应用、效果评价和模型修改7个步骤。
在一个实施例中，提供一种外环值确定方法。外环值确定方法主要包括如下步骤：
S81、根据历史特征数据进行宫格划分。
根据收集到的网络中用户的历史特征数据进行宫格的划分。
具体的，历史特征数据包括：用户传输模式，调度的空口测量值SINR，空分标记；空分流数；所支持的最大调制阶数(64QAM,256QAM)；对应的解调ACK/NACK；外环信息等。所述用户传输模式包括：单端口、闭环复用RI＝1，闭环复用RI＝2。
在非空分情况下：按传输模式三种(单端口、闭环复用RI＝1，闭环复用RI＝2)、干扰水平两种(高干扰、低干扰)、所支持的最大调制阶数(64QAM,256QAM)、RB个数(大小RB)分为24个宫格。在空分情况下：按照空分流数(2、3、4、5，以及大于等于6，分为5种)，干扰水平两种(高干扰、低干扰)、所支持的最大调制阶数(64QAM,256QAM)、RB个数(大小RB)分为40个宫格。在一些示例中，宫格级划分还可以包含用户的芯片类型信息、解调算法信息、用户的波束位置信息等等。需要说明的是，上述宫格划分方法仅为示例性说明，而非限定。
S82、对于每个宫格，按调度的MCS和对应的空口测量值SINR进行栅格划分。
例如，MCS以1阶为颗粒度，作为横轴，SINR值以0.5dB为颗粒度，作为纵轴、进行二维划分。
S83、根据抓取的新的特征数据，确定数据归属的宫格，以及对应宫格下的栅格，将实际解调对应的ACK/NACK信息，和外环信息投递到所在宫格的对应栅格。
具体的，根据抓取的新的特征数据判断是否空分，如果非空分，按照传输模式、干扰水平、所支持的最大调制阶数配置信息，以及调度的RB个数，找到此用户数据归属的宫格，并根据调度的MCS和空口测量值SINR将其对应的实际解调对应的ACK/NACK信息，和外环信息投递到所在宫格的对应栅格中，在每个栅格中记录ACK/NACK信息的数目(两个分别记)和外环值。
如果空分，按空分流数、干扰水平、所支持的最大调制阶数配置信息、以及调度的RB个数，找到此用户数据归属的宫格；并根据调度的MCS和空口测量值SINR将其对应的实际解调对应的ACK/NACK信息，和外环信息投递到所在宫格的对应栅格中，在每个栅格中记录ACK/NACK的数目(两个分别记)和外环值。
S84、统计计算每个宫格中栅格下的BLER以及外环值。
具体的，根据每个宫格中的每个栅格统计的ACK/NACK数目计算对应栅格的BLER，和对应栅格的外环值。如果对应的栅格样本量低于一定阈值，此栅格标记为无效栅格记为NULL。
S85、根据宫格中的每个栅格下统计的BLER和对应的外环值确定设定目标BLER区间下空口测量值SINR与调度MCS之间的对应关系模型和设定目标BLER区间下空口测量值SINR与外环值对应关系模型。
具体方法为，在宫格中，针对每一个MCS查找对应的BLER满足设定BLER区间的栅格，从而确定MCS对应的空口测量值SINR，以及空口测量值SINR对应的外环值。
S851、如果对于一个MCS满足设定BLER区间的栅格的个数大于1个，计算每个栅格的权重因子。
计算方法为：满足条件的栅格所含样本的数量(总的ACK/NACK数目)与所有满足条件的栅格的样本数量之和的比值。
S852、按权重因子对各个栅格对应的空口测量值SINR加权平均，对各个栅格对应的外 环值进行加权平均。
S853、从而确定该MCS下，满足目标BLER区间的，空口测量值SINR，以及外环值。
S854、如果对于一个MCS没有满足设定BLER区间的栅格，找此MCS在系统参考解调曲线上(一般是仿真出来的曲线)目标BLER区间所对应的解调值SINR的区间，然后取区间的中值减去系统默认初始化外环，为该MCS对应的空口测量SINR值，默认初始化外环值为其对应的外环值。
S855、通过上述过程就可以确定，每个空口测量值SINR对应的满足目标BLER区间的调度MCS，以及每个空口测量值SINR对应的外环值，即学习到的模型为：
设定目标BLER区间空口测量值SINR与调度MCS之间的关系模型记为：
Figure PCTCN2021100443-appb-000007
设定目标BLER区间空口测量值SINR与外环值ΔSINR的关系模型记为：
Figure PCTCN2021100443-appb-000008
其中，[0.08,0.12]为目标BLER区间，
Figure PCTCN2021100443-appb-000009
表示满足目标BLER区间的空口测量值与调度MCS的组合；
Figure PCTCN2021100443-appb-000010
表示满足目标BLER区间的空口测量值与外环值的组合；BLER(·)代表对应组合下的BLER；
Figure PCTCN2021100443-appb-000011
表示线下学习到的第j个宫格对应的空口信道质量与调度MCS之间的关系模型；
Figure PCTCN2021100443-appb-000012
表示线下学习到的第j个宫格对应的空口信道质量与外环值之间的关系模型；j＝1,…64；N表示系统所能支持的MCS个数。
S86、外环初始化模型应用前统计对应网络中SE与空口测量SINR的关系集合为：
Figure PCTCN2021100443-appb-000013
以及统计用户级别的BLER落在目标BLER区间的比例记为R0；其中K 0为样本点的个数，Set_p 0表示应用模型前的网络性能集合。
S87、外环初始化模型应用。
S871、根据用户对应的特征数据，以及划分宫格的策略判断用户所属的宫格，获取对应宫格的模型
Figure PCTCN2021100443-appb-000014
和模型
Figure PCTCN2021100443-appb-000015
S872、根据用户对应的当前空口测量值SINR0，由模型
Figure PCTCN2021100443-appb-000016
确定对应的MCS，由MCS和系统参考的BLER曲线获取在目标BLER区间内对应的SINR的中值记为SINR1，由SINR0和SINR1计算外环值D_SINR0＝SINR1-SINR0。
S873、由用户对应的当前空口测量值SINR，通过模型
Figure PCTCN2021100443-appb-000017
获取对应的外环值记为：D_SINR1。
S874、计算给用户初始化的外环值为：
D_SINR_Init＝min{max{(D_SINR0+D_SINR1)/2，-V_limt}，V_limt}
其中，V_limt为保护界限值，其值大于0，默认为10。
S88、应用模型后统计对应网络的SE与平均空口测量SINR的关系集合为：
Figure PCTCN2021100443-appb-000018
以及统计用户级别的BLER落在目标BLER区间的比例记为R1；其中K 1为样本点的个数，Set_p 1表示应用模型前的网络性能集合。
S89、判断Set_p 0与Set_p 1中的空口信道质量SINR是否有交集。
S891、如果有交集，计算交集部分对应的性能差异记为：{D_SE i|i＝1,2,…M}
其中D_SE i表示对应上述交集中的第i个元素对应的应用模型后的SE与应用模型前的SE的差。
S8911、计算{D_SE i|i＝1,2,…M}元素小于0的比例R2，以及集合中元素的平均值D_SE_mean。
S8912、若D_SE_mean≥Th 0且R 2≥Th 1且R 0≥R 1或D_SE_mean＜Th 0，则执行S810。否则，继续使用当前模型进行新接入用户的外环初始化。
S892、如果无交集，若R 0≥R 1，则执行S810。否则，继续使用当前模型进行新接入用户的外环初始化。
S810、停止当前模型的使用，启动在线学习。
S8101、在线收集用户级别与调度信息相关的数据，具体包含用户传输模式：单端口、闭环复用RI＝1，闭环复用RI＝2；干扰水平；调度MCS；分配RB个数；空口测量SINR；空分标记；空分流数；所支持的最大调制阶数(64QAM,256QAM)；对应的解调ACK/NACK；外环信息；并初始化模型更新标记Model_change_index＝0；并初始话模型性能评估集合AI_ΔSINR_int_set为空。
S8102、根据当前小区用于宫格划分的特征量的分布选取适当的分割点，如选取分布的50％的分位点做为门限，进行宫格划分。
S8103、采用S82到S86学习宫格级给定目标BLER区间[0.08,0.12]下，空口测量值SINR与调度MCS之间的关系模型记为：
Figure PCTCN2021100443-appb-000019
以空口测量值SINR与外环ΔSINR的关系模型：
Figure PCTCN2021100443-appb-000020
Figure PCTCN2021100443-appb-000021
外环值的组合；BLER(·)代表对应组合下的BLER；
Figure PCTCN2021100443-appb-000022
表示第j个宫格对应的空口测量值SINR与调度MCS之间的关系模型；
Figure PCTCN2021100443-appb-000023
表示第j个宫格对应的空口测量值SINR与外环值之间的关系模型；j＝1,…64；N表示系统所能支持的MCS个数。
S811、应用在线模型。
S8111、根据用户对应的特征数据，以及划分宫格的策略判断用户所属的宫格，获取对应宫格的模型
Figure PCTCN2021100443-appb-000024
和模型
Figure PCTCN2021100443-appb-000025
S8112、根据用户对应的空口测量SINR0，由模型
Figure PCTCN2021100443-appb-000026
确定对应的MCS，由MCS和系统参考的BLER曲线获取在目标BLER区间内对应的SINR的中值记为SINR1，由SINR0和SINR1计算外环值D_SINR0＝SINR1-SINR0。
S8113、由用户对应的空口测量SINR，通过模型
Figure PCTCN2021100443-appb-000027
获取对应的外环值记为：D_SINR1。
S8114、计算给用户初始化的外环为：
D_SINR_Init＝min{max{(D_SINR0+D_SINR1)/2，-V_limt}，V_limt}
其中，V_limt为保护界限值，其值大于0，默认为10。
S8115、判断模型更新标记Model_change_index是否为真，若为真，初始化 AI_ΔSINR_int_set为空，Model_change_index＝0记录该用户以此D_SINR_Init外环调度时，对应解调信息ACK/NACK(对于保守调度这里不作记录)计入AI_ΔSINR_int_set中。否则，记录该用户以此D_SINR_Init外环调度时，对应解调信息ACK/NACK(对于保守调度这里不作记录)计入AI_ΔSINR_int_set中。
S812、判断|AI_ΔSINR_int_set|≥Th_A/N_NUM是否成立。
若成立，根据AI_ΔSINR_int_set中的元素计算NACK的比例为:NACK_R；根据如下关系式计算模型调整量：
Figure PCTCN2021100443-appb-000028
其中，BLER_tart为设置的目标BLER,λ为模型调整步长。
对于模型
Figure PCTCN2021100443-appb-000029
采用如下式子更新模型中的信道质量维度的量：
SINR i＝max{min(SINR i-δ,SINR i+β),SINR i-β}，得到新模型：
Figure PCTCN2021100443-appb-000030
Figure PCTCN2021100443-appb-000031
得到新模型：
Figure PCTCN2021100443-appb-000032
设置Model_change_index＝1，并返回执行S811，
否则，模型不做更新，返回执行S811。
在一个实施例中，提供一种离线的初始化外环模型确定外环值的方法。离线的初始化外环模型确定外环值方法主要包括如下步骤：
S91、根据收集到的网络中用户的特征数据，进行宫格划分。
具体的，用户的特征数据包含用户传输模式：单端口、闭环复用RI＝1，闭环复用RI＝2；干扰水平；调度MCS；分配RB个数；空口测量SINR；空分标记；空分流数；所支持的最大调制阶数(64QAM,256QAM)；对应的解调ACK/NACK；外环信息。
不失一般性，在非空分情况下：按传输模式三种(单端口、闭环复用RI＝1，闭环复用RI＝2)、干扰水平两种(高干扰、低干扰)、所支持的最大调制阶数(64QAM,256QAM)、RB个数(大小RB)分为24个宫格；在空分情况下：按照空分流数(2、3、4、5，以及大于等于6，分为5种)，干扰水平两种(高干扰、低干扰)、所支持的最大调制阶数(64QAM,256QAM)、RB个数(大小RB)分为40个宫格，宫格级划分还可以包含用户的芯片类型信息、解调算法信息、用户的波束位置信息等等。
S92、对于每个宫格按调度的MCS和对应的空口测量信道质量值SINR进行栅格划分。
如，MCS以1阶为颗粒度，作为横轴，SINR值以0.5dB为颗粒度，作为纵轴，进行二维划分。
S93、根据用户的新传特征数据，判断是否空分，如果非空分，按照传输模式、干扰水平、所支持的最大调制阶数配置信息，以及调度的RB个数，找到此用户数据归属的宫格；并根据调度的MCS和空口信道质量SINR将其对应的实际解调对应的ACK/NACK，和外环信息投递到，对应宫格的对应栅格中，在每个栅格中记录ACK/NACK的数目(两个分别记)和外环值；如果空分，按空分流数、干扰水平、所支持的最大调制阶数配置信息、以及调度的RB个数，找到此用户数据归属的宫格；并根据调度的MCS和空口信道质量SINR将其对应的实际解调对应的ACK/NACK信息，和外环信息投递到，对应宫格的对应栅格中，在每个栅格中记录ACK/NACK的数目(两个分别记)和外环值。
S94、根据每个宫格中的每个栅格统计的ACK/NACK数目计算对应各个栅格的BLER，和各个栅格对应的外环值，如果对应的栅格样本量低于一定阈值，此栅格标记为无效栅格记为NULL。
S95、根据宫格中的每个栅格下的BLER和对应的外环值学习解调曲线。
具体方法为，在宫格中，针对每一个MCS找对应的BLER满足设定BLER区间的栅格，从而确定栅格对应的空口测量值SINR，以及栅格对应的外环值；并初始化模型更新标记Model_change_index＝0；并初始化模型性能评估集合AI_ΔSINR_int_set为空。
S951、如果对于一个MCS满足设定BLER区间的栅格的个数大于1个，计算每个栅格的权重因子，计算方法为：满足条件的栅格所含样本的数量(总的ACK/NACK数目)与所有满足条件的栅格的样本数量之和的比值；
S952、按权重对各个栅格对应的空口测量值SINR值加权平均，对各个栅格对应的外环值进行加权平均；
S953、从而确定空口测量值SINR下，对应的满足目标BLER区间的，调度MCS，以及对应外环值。
S954、如果对于一个MCS没有满足设定BLER区间的栅格，找此MCS在系统参考解调曲线上(一般是仿真出来的曲线)目标BLER区间所对应的SINR的区间，然后取SINR区间的中值减去系统默认初始化外环，为该MCS对应的空口测量值SINR，默认初始化外环为其对应的外环。
S955、通过上述过程就可以确定，每个测量的空口测量值SINR对应的满足目标BLER区间的调度MCS，以及空口测量值SINR对应的外环值，即学习到的模型为：
设定目标BLER区间空口测量值SINR与调度MCS之间的关系模型记为：
Figure PCTCN2021100443-appb-000033
以及，设定目标BLER区间空口测量值SINR与外环ΔSINR的关系模型：
Figure PCTCN2021100443-appb-000034
Figure PCTCN2021100443-appb-000035
的组合；BLER(·)代表对应组合下的BLER；
Figure PCTCN2021100443-appb-000036
表示线下学习到的第j个宫格对应的空口测量值SINR与调度MCS之间的关系模型；
Figure PCTCN2021100443-appb-000037
表示线下学习到的第j个宫格对应的空口测量值SINR与外环值之间的关系模型；j＝1,…64；N表示系统所能支持的MCS个数；
S96、模型应用前对应网络中SE与空口测量值SINR的关系集合为：
Figure PCTCN2021100443-appb-000038
以及统计用户级别的BLER落在目标BLER区间的比例记为R0；其中K 0为样本点的个数，Set_p 0表示应用模型前的网络性能集合。
S97、模型应用
S971、根据用户对应的特征数据，以及划分宫格的策略判断用户所属的宫格，获取对应宫格的模型
Figure PCTCN2021100443-appb-000039
和模型
Figure PCTCN2021100443-appb-000040
S972、根据用户对应的当前空口测量值SINR0，由模型
Figure PCTCN2021100443-appb-000041
确定对应的MCS，由MCS和系统参考的BLER曲线获取在目标BLER区间内对应的SINR的中值记为SINR1，由SINR0和SINR1计算外环值D_SINR0＝SINR1-SINR0
S973、由用户对应的当前空口测量值SINR，通过模型
Figure PCTCN2021100443-appb-000042
获取对应的外环值记为：D_SINR1；
S974、计算给用户初始化的外环为：
D_SINR_Init＝min{max{(D_SINR0+D_SINR1)/2，-V_limt}，V_limt}
其中，V_limt为保护界限值，其值大于0，默认为10；
S975、判断模型更新标记Model_change_index是否为真；若为真，初始化AI_ΔSINR_int_set为空，Model_change_index＝0记录该用户以此D_SINR_Init外环调度时，对应解调信息ACK/NACK(对于保守调度这里不作记录)计入AI_ΔSINR_int_set中。
否则，记录该用户以此D_SINR_Init外环调度时，对应解调信息ACK/NACK(对于保守调度这里不作记录)计入AI_ΔSINR_int_set中
S98、应用模型后统计对应网络的SE与空口测量值SINR的关系集合为：
Figure PCTCN2021100443-appb-000043
以及统计用户级别的BLER落在目标BLER区间的比例记为R1；其中K 1为样本点的个数，Set_p 1表示应用模型前的网络性能集合；
S99、判断Set_p 0与Set_p 1中的空口信道质量SINR是否有交集
S991、如果有交集，计算交集部分对应的性能差异记为：{D_SE i|i＝1,2,…M}；其中，D_SE i表示对应上述交集中的第i个元素对应的应用模型后的SE与应用模型前的SE的差；
S992、计算{D_SE i|i＝1,2,…M}元素小于0的比例R2，以及集合中元素的平均值D_SE_mean。
S993、若D_SE_mean≥Th 0且R 2≥Th 1且R 0≥R 1或D_SE_mean＜Th 0，则转S910。
否则，继续使用当前模型进行新接入用户的外环初始化。
S994、如果无交集，若R 0≥R 1，转S910。
否则，继续使用当前模型进行新接入用户的外环初始化。
S910、启动模型修正。
判断|AI_ΔSINR_int_set|≥Th_A/N_NUM是否成立，若成立，根据AI_ΔSINR_int_set中的元素计算NACK的比例为:NACK_R；根据如下关系式计算模型调整量：
Figure PCTCN2021100443-appb-000044
其中，BLER_tart为设置的目标BLER,λ为模型调整步长。
对于模型
Figure PCTCN2021100443-appb-000045
采用如下式子更新模型中的信道质量维度的量：
SINR i＝max{min(SINR i-δ,SINR i+β),SINR i-β}，得到新模型：
Figure PCTCN2021100443-appb-000046
对于模型
Figure PCTCN2021100443-appb-000047
采用如下式子更新模型中的信道质量维度的量：
Figure PCTCN2021100443-appb-000048
得到新模型：
Figure PCTCN2021100443-appb-000049
设置Model_change_index＝1，并转S97。
否则，模型不做更新，转S97。
在一个实施例中，提供修正在线模型的方法。
S101、根据历史特征数据进行宫格划分。
根据收集到的网络中用户的历史特征数据进行宫格的划分。
具体的，历史特征数据包括：用户传输模式，调度的空口测量值SINR，空分标记；空分流数；所支持的最大调制阶数(64QAM,256QAM)；对应的解调ACK/NACK；外环信息等。所述用户传输模式包括：单端口、闭环复用RI＝1，闭环复用RI＝2。
在非空分情况下：按传输模式三种(单端口、闭环复用RI＝1，闭环复用RI＝2)、干扰水平两种(高干扰、低干扰)、所支持的最大调制阶数(64QAM,256QAM)、RB个数(大小RB)分为24个宫格。在空分情况下：按照空分流数(2、3、4、5，以及大于等于6，分为5种)，干扰水平两种(高干扰、低干扰)、所支持的最大调制阶数(64QAM,256QAM)、RB个数(大小RB)分为40个宫格。在一些示例中，宫格级划分还可以包含用户的芯片类型信息、解调算法信息、用户的波束位置信息等等。需要说明的是，上述宫格划分方法仅为示例性说明，而非限定。
S102、对于每个宫格，按调度的MCS和对应解调的空口测量值SINR进行栅格划分。
例如，MCS以1阶为颗粒度，作为横轴，SINR值以0.5dB为颗粒度，作为纵轴、进行二维划分。
S103、根据抓取的新的特征数据，确定数据归属的宫格，以及对应宫格下的栅格，将实际解调对应的ACK/NACK信息，和外环信息投递到所在宫格的对应栅格。
具体的，根据抓取的新的特征数据判断是否空分，如果非空分，按照传输模式、干扰水平、所支持的最大调制阶数配置信息，以及调度的RB个数，找到此用户数据归属的宫格，并根据调度的MCS和空口测量值SINR将其对应的实际解调对应的ACK/NACK信息，和外环信息投递到所在宫格的对应栅格中，在每个栅格中记录ACK/NACK信息的数目(两个分别记)和外环值。
如果空分，按空分流数、干扰水平、所支持的最大调制阶数配置信息、以及调度的RB个数，找到此用户数据归属的宫格；并根据调度的MCS和空口测量值SINR将其对应的实际解调对应的ACK/NACK信息，和外环信息投递到所在宫格的对应栅格中，在每个栅格中记录ACK/NACK的数目(两个分别记)和外环值。
S104、统计计算每个宫格中栅格下的BLER以及外环值。
具体的，根据每个宫格中的每个栅格统计的ACK/NACK数目计算各个栅格对应的BLER，和各个栅格对应的外环值。如果对应的栅格样本量低于一定阈值，此栅格标记为无效栅格记为NULL。
S105、根据宫格中的每个栅格下的BLER和外环值确定空口测量值SINR与调度MCS之间的对应关系模型和空口测量值SINR与外环值的对应关系模型。
具体方法为，在宫格中，针对每一个MCS查找对应的BLER满足设定BLER区间的栅格，从而确定栅格对应的空口测量值SINR，以及栅格对应的外环值。
S1051、如果对于一个MCS满足设定BLER区间的栅格的个数大于1个，计算每个栅格的权重因子。
计算方法为：满足条件的栅格所含样本的数量(总的ACK/NACK数目)与所有满足条件的栅格的样本数量之和的比值。
S1052、按权重因子对各个栅格对应的空口测量值SINR加权平均，对各个栅格对应的外环值进行加权平均。
S1053、从而确定空口测量SINR下，对应的满足目标BLER区间的，调度MCS，以及对应外环值。
S1054、如果对于一个MCS没有满足设定BLER区间的栅格，找此MCS在系统参考解调曲线上(一般是仿真出来的曲线)目标BLER区间所对应的解调SINR值的区间，然后取区间的中值减去系统默认初始化外环，为该MCS对应的空口测量SINR值，默认初始化外环为其对应的外环。
S1055、通过上述过程就可以确定，每个空口测量值SINR对应的满足目标BLER区间的调度MCS，以及空口测量值SINR对应的外环值，即学习到的模型为：
设定目标BLER区间空口测量值SINR与调度MCS之间的关系模型记为：
Figure PCTCN2021100443-appb-000050
设定目标BLER区间空口测量值SINR与外环值ΔSINR的关系模型记为：
Figure PCTCN2021100443-appb-000051
其中，[0.08,0.12]为目标BLER区间，
Figure PCTCN2021100443-appb-000052
表示满足目标BLER区间的空口测量值与调度MCS的组合；
Figure PCTCN2021100443-appb-000053
表示满足目标BLER区间的空口测量值与外环值的组合；BLER(·)代表对应组合下的BLER；
Figure PCTCN2021100443-appb-000054
表示线下学习到的第j个宫格对应的空口测量值SINR与调度MCS之间的关系模型；
Figure PCTCN2021100443-appb-000055
表示线下学习到的第j个宫格对应的空口测量值SINR与外环值之间的关系模型；j＝1,…64；N表示系统所能支持的MCS个数。
S106、外环初始化模型应用前统计对应网络中SE与平均空口测量SINR的关系集合为：
Figure PCTCN2021100443-appb-000056
以及统计用户级别的BLER落在目标BLER区间的比例记为R0；其中K 0为样本点的个数，Set_p 0表示应用模型前的网络性能集合。
S107、外环初始化模型应用。
S1071、根据用户对应的特征数据，以及划分宫格的策略判断用户所属的宫格，获取对应宫格的模型
Figure PCTCN2021100443-appb-000057
和模型
Figure PCTCN2021100443-appb-000058
S1072、根据用户对应的当前空口测量值SINR0，由模型
Figure PCTCN2021100443-appb-000059
确定对应的MCS，由MCS和系统参考的BLER曲线获取在目标BLER区间内对应的SINR的中值记为SINR1，由SINR0和SINR1计算外环值D_SINR0＝SINR1-SINR0。
S1073、由用户对应的当前空口测量值SINR，通过模型
Figure PCTCN2021100443-appb-000060
获取对应的外环值记为：D_SINR1。
S1074、计算给用户初始化的外环值为：
D_SINR_Init＝min{max{(D_SINR0+D_SINR1)/2，-V_limt}，V_limt}
其中，V_limt为保护界限值，其值大于0，默认为10。
S1010、应用模型后统计对应网络的SE与空口测量值SINR的关系集合为：
Figure PCTCN2021100443-appb-000061
以及统计用户级别的BLER落在目标BLER区间的比例记为R1；其中K 1为样本点的个数，Set_p 1表示应用模型前的网络性能集合。
S109、判断Set_p 0与Set_p 1中的空口信道质量SINR是否有交集
S1091、如果有交集，计算交集部分对应的性能差异记为：
{D_SE i|i＝1,2,…M}
其中D_SE i表示对应上述交集中的第i个元素对应的应用模型后的SE与应用模型前的SE的差。
S10911、计算{D_SE i|i＝1,2,…M}元素小于0的比例R2，以及集合中元素的平均值D_SE_mean。
S10912、若D_SE_mean≥Th 0且R 2≥Th 1且R 0≥R 1或D_SE_mean＜Th 0，则执行S1010。否则，继续使用当前模型进行新接入用户的外环初始化
S1092、如果无交集，若R 0≥R 1，则执行S1010。否则，继续使用当前模型进行新接入用户的外环初始化。
S1010、停止当前模型的使用，启动在线学习
S10101、在线收集用户级别与调度信息相关的数据，具体包含用户传输模式：单端口、闭环复用RI＝1，闭环复用RI＝2；干扰水平；调度MCS；分配RB个数；空口测量SINR；空分标记；空分流数；所支持的最大调制阶数(64QAM,256QAM)；对应的解调ACK/NACK；外环信息；并初始化模型更新标记Model_change_index＝0；并初始话模型性能评估集合AI_ΔSINR_int_set为空。
S10102、根据当前小区用于宫格划分的特征量的分布选取适当的分割点，如选取分布的50％的分位点做为门限，进行宫格划分。
S10103、采用S102到S106学习宫格级给定目标BLER区间[0.08,0.12]下，空口测量值SINR与调度MCS之间的关系模型记为：
Figure PCTCN2021100443-appb-000062
以空口测量值SINR与外环ΔSINR的关系模型：
Figure PCTCN2021100443-appb-000063
其中[0.08,0.12]为目标BLER区间，
Figure PCTCN2021100443-appb-000064
表示满足目标BLER区间的空口测量值与调度MCS的组合；
Figure PCTCN2021100443-appb-000065
表示满足目标BLER区间的空口测量值与外环值的组合；BLER(·)代表对应组合下的BLER；
Figure PCTCN2021100443-appb-000066
表示第j个宫格对应的空口测量值与调度MCS之间的关系模型；
Figure PCTCN2021100443-appb-000067
表示第j个宫格对应的空口测量值与外环值之间的关系模型；j＝1,…64；N表示系统所能支持的MCS个数。
S1011、应用在线模型。
S10111、根据用户对应的特征数据，以及划分宫格的策略判断用户所属的宫格，获取对应宫格的模型
Figure PCTCN2021100443-appb-000068
和模型
Figure PCTCN2021100443-appb-000069
S10112、根据用户对应的当前空口测量值SINR0，由模型
Figure PCTCN2021100443-appb-000070
确定对应的MCS，由MCS和系统参考的BLER曲线获取在目标BLER区间内对应的SINR的中值记为SINR1，由SINR0和SINR1计算外环值D_SINR0＝SINR1-SINR0。
S10113、由用户对应的当前空口测量值SINR，通过模型
Figure PCTCN2021100443-appb-000071
获取对 应的外环值记为：D_SINR1。
S10114、计算给用户初始化的外环为：
D_SINR_Init＝min{max{(D_SINR0+D_SINR1)/2，-V_limt}，V_limt}
其中，V_limt为保护界限值，其值大于0，默认为10；
S1012、应用模型后统计对应网络的SE与平均空口测量SINR的关系集合为：
Figure PCTCN2021100443-appb-000072
以及统计用户级别的BLER落在目标BLER区间的比例记为R1；其中K 1为样本点的个数，Set_p 1表示应用模型前的网络性能集合；
S1013、判断Set_p 0与Set_p 1中的空口信道质量SINR是否有交集
S10131、如果有交集。
计算交集部分对应的性能差异记为：{D_SE i|i＝1,2,…M}
其中，D_SE i表示对应上述交集中的第i个元素对应的应用模型后的SE与应用模型前的SE的差。
S101311、计算{D_SE i|i＝1,2,…M}元素小于0的比例R2，以及集合中元素的平均值D_SE_mean。
S101312、若D_SE_mean≥Th 0且R 2≥Th 1且R 0≥R 1；或D_SE_mean＜Th 0，采用小区默认外环，停止应用模型；否则，继续使用当前模型进行新接入用户的外环初始化
S10132、如果无交集，若R 0≥R 1，采用小区默认外环，停止应用模型；否则，继续使用当前模型进行新接入用户的外环初始化。
在一个实施例中，本申请提供一种外环值确定方法，图10是本申请实施例提供的一种外环值确定装置的结构示意图，本实施例可适用于自适应编码调制技术的情况，该装置可采用软件和/或硬件的方式实现。
如图1所示，本申请实施例提供的外环值确定方法主要包括步骤模型确定模块101和外环值确定模块102。
模型确定模块101，被配置为基于用户设备的当前特征数据确定预训练的外环初始化模型；
外环值确定模块102，被配置为基于用户设备的当前空口测量值和所述外环初始化模型确定用户设备的初始化外环值，所述外环初始化模型包括第一外环模型和第二外环模型。
在一个实施方式中，所模型确定模块101，被配置为基于用户设备的当前特征数据以及宫格划分策略确定所述用户设备所属宫格；获取所述用户设备所属宫格对应的第一外环模型和第二外环模型。
在一个实施方式中，外环值确定模块102，被配置为基于所述当前空口测量值和所述第一外环模型确定第一外环值；基于所述当前空口测量值和所述第二外环模型确定第二外环值；基于所述第一外环值和所述第二外环值确定初始化外环值。
在一个实施方式中，所述基于所述当前空口测量值和第一外环模型确定第一外环值，包括：基于所述当前空口测量值和所述第一外环模型确定所述当前空口测量值对应的调制 与编码策略MCS；其中，所述第一外环模型是设定目标误块率BLER区间下空口测量值与MCS的对应关系模型；基于确定的MCS和参考解调曲线获取目标BLER对应的系统映射SINR；将所述系统映射SINR与所述当前空口测量值的差值确定为第一外环值。
在一个实施方式中，所述基于所述当前空口测量值和所述第二外环模型确定第二外环值，包括：基于所述当前空口测量值，通过所述第二外环模型获取第二外环值；其中，所述第二外环模型是设定目标BLER区间下空口测量值与外环值的对应关系模型。
在一个实施方式中，所述装置还包括：模型训练模块，被配置为在所述基于用户设备的当前特征数据确定预训练的外环初始化模型之前，基于所述用户设备的历史特征数据训练外环初始化模型。
在一个实施方式中，模型训练模块，被配置为基于所述历史特征数据进行栅格划分；确定每个栅格中的确认响应/否认响应ACK/NACK信息数目和外环值；基于每个栅格中的ACK/NACK信息数目计算每个栅格的误块率；基于所述每个栅格的误块率和外环值确定外环初始化模型。
在一个实施方式中，基于所述历史特征数据进行栅格划分，包括：基于所述历史特征数据对网络中的各个小区进行宫格划分；对于每个宫格按照调度的MCS和对应的空口测量值进行栅格划分。
在一个实施方式中，所述确定每个栅格中的ACK/NACK信息数目，包括：获取解调对应的确认响应/否认响应ACK/NACK信息；将所述ACK/NACK信息以及对应的外环信息投递到对应的栅格中；统计每个栅格中ACK/NACK信息数目以及外环值。
在一个实施方式中，所述基于所述每个栅格的误块率和外环值确定外环初始化模型，包括：在每个宫格中，基于每个栅格的误块率确定设定目标误块率BLER区间下空口测量值对应的MCS以及空口测量值对应的外环值；基于所述空口测量值和所述MCS的对应关系确定第一外环模型；基于所述空口测量值和所述外环值的对应关系确定第二外环模型。
在一个实施方式中，所述基于每个栅格的误块率确定设定目标误块率BLER区间下空口测量值对应的MCS以及空口测量值与对应的外环值，包括：对于一个MCS，如果满足设定误块率区间的栅格数量大于预设数值，计算每个栅格的权重因子；按照所述每个栅格的权重因子对各栅格对应的空口测量值进行加权平均计算，得到所述MCS对应的空口测量值；按照所述每个栅格的权重因子对各栅格对应的外环值进行加权平均计算，得到所述MCS对应的外环值；根据所述MCS对应的空口测量值和所述MCS对应的外环值确定所述空口测量值对应的外环值。
在一个实施方式中，所述基于每个栅格的误块率确定设定目标误块率BLER区间下空口测量值对应的MCS以及空口测量值与对应的外环值，包括：对于一个MCS，如果满足设定误块率区间的栅格数量小于预设数值，查找该MCS在参考解调曲线上对应的目标误块率区间所对应的空口测量值区间；将空口测量值区间中的中值与系统默认初始化外环值的差值，作为该MCS对应的空口测量值；将系统默认初始化外环值作为空口测量值对应的外环值。
在一个实施方式中，所述装置还包括：在线模型学习模块，被配置为所述基于所述用户设备的当前空口测量值和所述外环初始化模型确定所述用户设备的初始化外环值之后，获取应用模型前第一网络性能指标和使用外环初始化模型后的第二网络性能指标；在所述第一网络性能指标和所述第二网络性能指标满足预设条件的情况下，启动在线学习，得到 外环在线模型；使用外环在线模型确定用户设备新的初始化外环值。
在一个实施方式中，所述装置还包括：在线模型调整模块，被配置为基于宫格中用户首次使用外环模型后的ACK/NACK信息对对应宫格的外环模型进行修正；所述外环模型包括外环初始化模型或外环在线模型。
在一个实施方式中，所述基于使用在线模型后的ACK/NACK信息对所述在线模型进行调整，包括:针对每个宫格，计算ACK/NACK信息中否认响应NACK所占比例；在所述NACK所占比例大于目标数值的情况下，按相应的策略减少所述外环模型的外环值；
在所述NACK所占比例小于目标数值的情况下，按相应的策略增加所述外环模型的外环值。
本实施例中提供的外环值确定装置可执行本申请任意实施例所提供的外环值确定方法，具备执行该方法相应的功能模块和有益效果。未在本实施例中详尽描述的技术细节，可参见本申请任意实施例所提供的外环值确定法。
值得注意的是，上述外环值确定装置的实施例中，所包括的各个单元和模块只是按照功能逻辑进行划分的，但并不局限于上述的划分，只要能够实现相应的功能即可；另外，各功能单元的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。
本申请实施例还提供一种设备，图11是本申请提供的一种设备的结构示意图，如图11所示，该设备包括处理器111、存储器112、输入装置113、输出装置114；设备中处理器111的数量可以是一个或多个，图11中以一个处理器111为例；设备中的处理器111、存储器112、输入装置113和输出装置114可以通过总线或其他方式连接，图11中以通过总线连接为例。
存储器112作为一种计算机可读存储介质，可用于存储软件程序、计算机可执行程序以及模块，如本申请实施例中的外环值确定方法对应的程序指令/模块(例如，外环值确定装置中的模型确定模块101和外环值确定模块102)。处理器111通过运行存储在存储器112中的软件程序、指令以及模块，从而执行设备的各种功能应用以及数据处理，即实现本申请实施例提供的任一方法。
存储器112可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序；存储数据区可存储根据设备的使用所创建的数据等。此外，存储器112可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他非易失性固态存储器件。在一些实例中，存储器112可进一步包括相对于处理器111远程设置的存储器，这些远程存储器可以通过网络连接至设备。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
输入装置113可用于接收输入的数字或字符信息，以及产生与设备的用户设置以及功能控制有关的键信号输入。输出装置114可包括显示屏等显示设备。
本申请实施例还提供一种包含计算机可执行指令的存储介质，所述计算机可执行指令在由计算机处理器执行时用于执行一种外环值确定方法，包括：
基于用户设备的当前特征数据确定预训练的外环初始化模型；
基于用户设备的当前空口测量值和所述外环初始化模型确定用户设备的初始化外环值。
当然,本申请实施例所提供的一种包含计算机可执行指令的存储介质,其计算机可执行指令不限于如上所述的操作,还可以执行本申请任意实施例所提供的外环值确定方法中 的相关操作。
本申请实施例提供的外环值确定方法、装置、设备及存储介质，通过基于用户设备的当前特征数据确定预训练的外环初始化模型；基于用户设备的当前空口测量值和外环初始化模型确定用户设备的初始化外环值的技术方案，解决小包用户不收敛引起的资源浪费和用户感知降低的问题，实现提高自适应调制编码技术收敛速度。
通过以上关于实施方式的描述，所属领域的技术人员可以清楚地了解到，本申请可借助软件及必需的通用硬件来实现，当然也可以通过硬件实现，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对本领域技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如计算机的软盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、闪存(FLASH)、硬盘或光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述的方法。
以上所述，仅为本申请的示例性实施例而已，并非用于限定本申请的保护范围。
本领域内的技术人员应明白，术语用户终端涵盖任何适合类型的无线用户设备，例如移动电话、便携数据处理装置、便携网络浏览器或车载移动台。
一般来说，本申请的多种实施例可以在硬件或专用电路、软件、逻辑或其任何组合中实现。例如，一些方面可以被实现在硬件中，而其它方面可以被实现在可以被控制器、微处理器或其它计算装置执行的固件或软件中，尽管本申请不限于此。
本申请的实施例可以通过移动装置的数据处理器执行计算机程序指令来实现，例如在处理器实体中，或者通过硬件，或者通过软件和硬件的组合。计算机程序指令可以是汇编指令、指令集架构(ISA)指令、机器指令、机器相关指令、微代码、固件指令、状态设置数据、或者以一种或多种编程语言的任意组合编写的源代码或目标代码。
本申请附图中的任何逻辑流程的框图可以表示程序步骤，或者可以表示相互连接的逻辑电路、模块和功能，或者可以表示程序步骤与逻辑电路、模块和功能的组合。计算机程序可以存储在存储器上。存储器可以具有任何适合于本地技术环境的类型并且可以使用任何适合的数据存储技术实现，例如但不限于只读存储器(ROM)、随机访问存储器(RAM)、光存储器装置和系统(数码多功能光碟DVD或CD光盘)等。计算机可读介质可以包括非瞬时性存储介质。数据处理器可以是任何适合于本地技术环境的类型，例如但不限于通用计算机、专用计算机、微处理器、数字信号处理器(DSP)、专用集成电路(ASIC)、可编程逻辑器件(FGPA)以及基于多核处理器架构的处理器。
通过示范性和非限制性的示例，上文已提供了对本申请的示范实施例的详细描述。但结合附图和权利要求来考虑，对以上实施例的多种修改和调整对本领域技术人员来说是显而易见的，但不偏离本申请的范围。因此，本申请的恰当范围将根据权利要求确定。

Claims (18)

  1. 一种外环值确定方法，包括:
    基于用户设备的当前特征数据确定预训练的外环初始化模型；
    基于所述用户设备的当前空口测量值和所述外环初始化模型确定所述用户设备的初始化外环值。
  2. 根据权利要求1所述的方法，其中，所述外环初始化模型包括第一外环模型和第二外环模型；所述基于用户设备的当前特征数据确定预训练的外环初始化模型，包括:
    基于所述用户设备的当前特征数据以及宫格划分策略确定所述用户设备所属宫格；
    获取所述用户设备所属宫格对应的第一外环模型和第二外环模型。
  3. 根据权利要求2所述的方法，其中，所述基于所述用户设备的当前空口测量值和所述外环初始化模型确定所述用户设备的初始化外环值，包括：
    基于所述当前空口测量值和所述第一外环模型确定第一外环值；
    基于所述当前空口测量值和所述第二外环模型确定第二外环值；
    基于所述第一外环值和所述第二外环值确定初始化外环值。
  4. 根据权利要求3所述的方法，其中，所述基于所述当前空口测量值和所述第一外环模型确定第一外环值，包括：
    基于所述当前空口测量值和所述第一外环模型确定所述当前空口测量值对应的调制与编码策略MCS；其中，所述第一外环模型是设定目标误块率BLER区间下空口测量值与MCS的对应关系模型；
    基于确定的MCS和参考解调曲线获取所述目标BLER对应的系统映射SINR；
    将所述系统映射SINR与所述当前空口测量值的差值确定为第一外环值。
  5. 根据权利要求3所述的方法，其中，所述基于所述当前空口测量值和所述第二外环模型确定第二外环值，包括：
    基于所述当前空口测量值，通过所述第二外环模型获取第二外环值；其中，所述第二外环模型是设定目标BLER区间下空口测量值与外环值的对应关系模型。
  6. 根据权利要求1所述的方法，其中，所述基于用户设备的当前特征数据确定预训练的外环初始化模型之前，还包括：
    基于所述用户设备的历史特征数据训练外环初始化模型。
  7. 根据权利要求6所述的方法，其中，所述基于所述用户设备的历史特征数据训练外环初始化模型，包括：
    基于所述历史特征数据进行栅格划分；
    确定每个栅格中的确认响应/否认响应ACK/NACK信息数目和外环值；
    基于每个栅格中的ACK/NACK信息数目计算每个栅格的BLER；
    基于所述每个栅格的BLER和外环值确定外环初始化模型。
  8. 根据权利要求7所述的方法，其中，所述基于所述历史特征数据进行栅格划分，包括：
    基于所述历史特征数据对网络中的各个小区进行宫格划分；
    对于每个宫格按照调度的MCS和对应的空口测量值进行栅格划分。
  9. 根据权利要求7所述的方法，其中，所述确定每个栅格中的ACK/NACK信息数目和外环值，包括：
    获取解调对应的ACK/NACK信息；
    将所述ACK/NACK信息以及对应的外环信息投递到对应的栅格中；
    统计每个栅格中ACK/NACK信息数目以及外环值。
  10. 根据权利要求7所述的方法，其中，所述基于所述每个栅格的BLER和外环值确定外环初始化模型，包括：
    在每个宫格中，基于每个栅格的BLER确定设定目标BLER区间下空口测量值对应的MCS以及空口测量值对应的外环值；
    基于所述空口测量值和所述MCS的对应关系确定第一外环模型；
    基于所述空口测量值和所述外环值的对应关系确定第二外环模型。
  11. 根据权利要求10所述的方法，其中，所述基于每个栅格的BLER确定设定目标BLER区间下空口测量值对应的MCS以及空口测量值对应的外环值，包括：
    对于一个MCS，如果满足设定目标BLER区间的栅格数量大于预设数值，根据每个栅格内的ACK/NACK信息数目计算每个栅格的权重因子；
    按照所述每个栅格的权重因子对各栅格对应的空口测量值进行加权平均计算，得到所述MCS对应的空口测量值；
    按照所述每个栅格的权重因子对各栅格对应的外环值进行加权平均计算，得到所述MCS对应的外环值；
    根据所述MCS对应的空口测量值和所述MCS对应的外环值确定所述空口测量值对应的外环值。
  12. 根据权利要求10所述的方法，其中，所述基于每个栅格的BLER确定设定目标BLER区间下空口测量值对应的MCS以及空口测量值对应的外环值，包括：
    对于一个MCS，如果满足设定目标BLER区间的栅格数量小于预设数值，查找该MCS在参考解调曲线上对应的目标BLER区间所对应的空口测量值区间；
    将查找的空口测量值区间中的中值与系统默认初始化外环值的差值，作为该MCS对应的空口测量值；
    将系统默认初始化外环值作为空口测量值对应的外环值。
  13. 根据权利要求1所述的方法，其中，所述基于所述用户设备的当前空口测量值和所述外环初始化模型确定所述用户设备的初始化外环值之后，还包括:
    获取应用模型前第一网络性能指标和使用外环初始化模型后的第二网络性能指标；
    在所述第一网络性能指标和所述第二网络性能指标满足预设条件的情况下，启动在线学习，得到外环在线模型；
    使用外环在线模型确定用户设备新的初始化外环值。
  14. 根据权利要求1或13所述的方法，其中，所述方法还包括：
    基于宫格中用户首次使用外环模型后的ACK/NACK信息对对应宫格的外环模型进行修正；所述外环模型包括预训练的外环初始化模型或外环在线模型。
  15. 根据权利要求14所述的方法，其中，所述基于宫格中用户首次使用外环模型后的ACK/NACK信息对对应宫格的外环模型进行修正，包括:
    针对每个宫格，计算ACK/NACK信息中否认响应NACK所占比例；
    在所述NACK所占比例大于目标数值的情况下，按相应的策略减少所述外环模型的外环值；
    在所述NACK所占比例小于目标数值的情况下，按相应的策略增加所述外环模型的外环值。
  16. 一种外环值确定装置，包括:
    模型确定模块，被配置为基于用户设备的当前特征数据确定预训练的外环初始化模型；
    外环值确定模块，被配置为基于用户设备的当前空口测量值和所述外环初始化模型确定用户设备的初始化外环值。
  17. 一种设备，包括：
    一个或多个处理器；
    存储器，用于存储一个或多个程序；
    当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现如权利要求1-15任一项所述的方法。
  18. 一种存储介质，存储有计算机程序，其中，所述计算机程序被处理器执行时实现权利要求1-15任一项所述的方法。
PCT/CN2021/100443 2020-06-24 2021-06-16 外环值确定方法、装置、设备及存储介质 WO2021259114A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US18/002,920 US20230246729A1 (en) 2020-06-24 2021-06-16 Outer ring value determination method and apparatus, and device and storage medium
KR1020227045932A KR20230016679A (ko) 2020-06-24 2021-06-16 외부 루프값 결정 방법, 장치, 디바이스 및 저장 매체
EP21829532.7A EP4170941A4 (en) 2020-06-24 2021-06-16 METHOD AND DEVICE FOR DETERMINING THE OUTER RING VALUE AS WELL AS DEVICE AND STORAGE MEDIUM

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202010591941.2A CN113839740A (zh) 2020-06-24 2020-06-24 外环值确定方法、装置、设备及存储介质
CN202010591941.2 2020-06-24

Publications (1)

Publication Number Publication Date
WO2021259114A1 true WO2021259114A1 (zh) 2021-12-30

Family

ID=78964807

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/100443 WO2021259114A1 (zh) 2020-06-24 2021-06-16 外环值确定方法、装置、设备及存储介质

Country Status (5)

Country Link
US (1) US20230246729A1 (zh)
EP (1) EP4170941A4 (zh)
KR (1) KR20230016679A (zh)
CN (1) CN113839740A (zh)
WO (1) WO2021259114A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117134858A (zh) * 2022-05-20 2023-11-28 中兴通讯股份有限公司 外环初始值的调整方法、设备及计算机可读存储介质
CN115208511B (zh) * 2022-05-27 2024-04-26 浪潮通信技术有限公司 调制与编码策略的配置方法、装置及电子设备
CN117375777A (zh) * 2022-06-27 2024-01-09 中兴通讯股份有限公司 内环确定方法、内环模型训练方法、电子设备、可读介质
CN117856967A (zh) * 2022-09-30 2024-04-09 中兴通讯股份有限公司 目标误块率确定方法及装置
CN117856968A (zh) * 2022-09-30 2024-04-09 中兴通讯股份有限公司 目标误块率控制方法及装置

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130182569A1 (en) * 2012-01-18 2013-07-18 Texas Instruments Incorporated Link Adaptation for LTE Uplink
WO2016075517A1 (en) * 2014-11-14 2016-05-19 Telefonaktiebolaget L M Ericsson (Publ) Statistical model based control signal outer-loop adjustment
WO2018137204A1 (zh) * 2017-01-25 2018-08-02 华为技术有限公司 外环链路自适应的调整方法和装置
CN109412996A (zh) * 2018-12-11 2019-03-01 深圳大学 链路自适应方法、电子装置及计算机可读存储介质

Family Cites Families (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7170876B2 (en) * 2002-04-30 2007-01-30 Qualcomm, Inc. Outer-loop scheduling design for communication systems with channel quality feedback mechanisms
CN100373805C (zh) * 2005-01-13 2008-03-05 中兴通讯股份有限公司 移动通信系统的绝对外环门限自适应设置方法
CN101534176B (zh) * 2008-03-11 2012-12-19 中兴通讯股份有限公司 一种自适应的天线模式控制方法及装置
US20130072250A1 (en) * 2011-03-30 2013-03-21 Qualcomm Incorporated Outage based outer loop power control for wireless communications systems
WO2013143069A1 (en) * 2012-03-27 2013-10-03 Nec (China) Co., Ltd. Method and apparatus for outer loop link adaptation for a wireless communication system
CN103516463B (zh) * 2012-06-20 2017-05-31 南京中兴新软件有限责任公司 一种进行外环自适应调制与编码调整的方法和基站
EP3014799B1 (en) * 2013-06-25 2018-08-08 Telefonaktiebolaget LM Ericsson (publ) Methods and devices for link adaptation
US9379842B2 (en) * 2013-10-30 2016-06-28 Telefonaktiebolaget Lm Ericsson (Publ) Outer-loop adjustment for wireless communication link adaptation
WO2015165515A1 (en) * 2014-04-30 2015-11-05 Huawei Technologies Co., Ltd. Computer program product and apparatus for fast link adaptation in a communication system
EP3384618A1 (en) * 2015-12-01 2018-10-10 Telefonaktiebolaget LM Ericsson (publ) Methods and devices for addressing passive intermodulation in wireless communication
CN105577324B (zh) * 2015-12-22 2018-11-27 京信通信系统（中国）有限公司 通信链路自适应调整方法和系统
US11457453B2 (en) * 2016-06-07 2022-09-27 Telefonaktiebolaget Lm Ericsson (Publ) Outer-loop adjustment for link adaptation

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130182569A1 (en) * 2012-01-18 2013-07-18 Texas Instruments Incorporated Link Adaptation for LTE Uplink
WO2016075517A1 (en) * 2014-11-14 2016-05-19 Telefonaktiebolaget L M Ericsson (Publ) Statistical model based control signal outer-loop adjustment
WO2018137204A1 (zh) * 2017-01-25 2018-08-02 华为技术有限公司 外环链路自适应的调整方法和装置
CN109412996A (zh) * 2018-12-11 2019-03-01 深圳大学 链路自适应方法、电子装置及计算机可读存储介质

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
HUAWEI ET AL.: "LLS evaluations and observations in uRLLC", 3GPP TSG RAN WG1 MEETING #95 R1-1812666, 16 November 2018 (2018-11-16), XP051478909 *
See also references of EP4170941A4 *

Also Published As

Publication number Publication date
EP4170941A4 (en) 2024-04-10
US20230246729A1 (en) 2023-08-03
EP4170941A1 (en) 2023-04-26
CN113839740A (zh) 2021-12-24
KR20230016679A (ko) 2023-02-02

Similar Documents

Publication Publication Date Title
WO2021259114A1 (zh) 外环值确定方法、装置、设备及存储介质
Elgendy et al. Joint computation offloading and task caching for multi-user and multi-task MEC systems: reinforcement learning-based algorithms
CN111447083B (zh) 动态带宽和不可靠网络下的联邦学习架构及其压缩算法
TW202131661A (zh) 用於網路最佳化的裝置及方法、以及非暫時性電腦可讀媒體
WO2021026944A1 (zh) 基于粒子群和神经网络的工业无线流媒体自适应传输方法
CN111431941A (zh) 一种基于移动边缘计算的实时视频码率自适应方法
CN115884094B (zh) 一种基于边缘计算的多场景协作优化缓存方法
CN105379412A (zh) 一种控制多个无线接入节点的系统和方法
CN113989561A (zh) 基于异步联邦学习的参数聚合更新方法、设备及系统
CN104509019A (zh) 无线通信网络传输的基于表的链路自适应
CN112636891B (zh) 资源调度参数的调整方法、装置、存储介质及电子装置
WO2021083230A1 (zh) 功率调节方法和接入网设备
CN114302407A (zh) 网络决策方法及装置、电子设备和存储介质
CN109688065B (zh) 参数的处理方法、装置及存储介质
CN113543160B (zh) 5g切片资源配置方法、装置、计算设备及计算机存储介质
CN111556531B (zh) 一种微蜂窝无线网中的协作缓存优化方法
CN114745079B (zh) 一种自适应调制编码方法、接入网设备及存储介质
CN117528658A (zh) 基于联邦深度强化学习的边缘协作缓存方法及系统
CN117031977A (zh) 一种智能家居控制方法及系统
CN115835242A (zh) 面向群智感知的通感算联合优化方法、设备及储存介质
WO2023273891A1 (zh) 外环初始值的调整方法、设备及计算机可读存储介质
WO2023020086A1 (zh) 网络优化方法、装置、电子设备和存储介质
WO2021139826A1 (zh) 非周期csi请求的响应方法、系统、电子设备和介质
CN110730463B (zh) 一种双层异构缓存网络的最优概率缓存方法
CN113055212B (zh) 策略的推送方法、策略的执行方法、装置、设备及介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21829532

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 20227045932

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021829532

Country of ref document: EP

Effective date: 20230123