WO2022121979A1 - 内环值的调整方法和装置、存储介质及电子装置 - Google Patents

内环值的调整方法和装置、存储介质及电子装置 Download PDF

Info

Publication number
WO2022121979A1
WO2022121979A1 PCT/CN2021/136759 CN2021136759W WO2022121979A1 WO 2022121979 A1 WO2022121979 A1 WO 2022121979A1 CN 2021136759 W CN2021136759 W CN 2021136759W WO 2022121979 A1 WO2022121979 A1 WO 2022121979A1
Authority
WO
WIPO (PCT)
Prior art keywords
time
domain scheduling
scheduling unit
factor
error rate
Prior art date
Application number
PCT/CN2021/136759
Other languages
English (en)
French (fr)
Inventor
李建国
刘巧艳
毛凯
冀旺旺
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 US18/036,925 priority Critical patent/US20230422220A1/en
Priority to EP21902684.6A priority patent/EP4262312A4/en
Publication of WO2022121979A1 publication Critical patent/WO2022121979A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/12—Wireless traffic scheduling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/24765—Rule-based classification
    • 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
    • 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
    • H04L1/1628—List acknowledgements, i.e. the acknowledgement message consisting of a list of identifiers, e.g. of sequence numbers
    • 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
    • H04L1/18—Automatic repetition systems, e.g. Van Duuren systems
    • H04L1/1867—Arrangements specially adapted for the transmitter end
    • 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
    • H04L1/201—Frame classification, e.g. bad, good or erased
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L5/00—Arrangements affording multiple use of the transmission path
    • H04L5/003—Arrangements for allocating sub-channels of the transmission path
    • H04L5/0053—Allocation of signaling, i.e. of overhead other than pilot signals
    • H04L5/0055—Physical resource allocation for ACK/NACK
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L5/00—Arrangements affording multiple use of the transmission path
    • H04L5/003—Arrangements for allocating sub-channels of the transmission path
    • H04L5/0058—Allocation criteria
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation
    • H04W72/044—Wireless resource allocation based on the type of the allocated resource
    • H04W72/0446—Resources in time domain, e.g. slots or frames

Definitions

  • the embodiments of the present application relate to the field of communications, and in particular, to a method and device for adjusting an inner loop value, a storage medium, and an electronic device.
  • the traditional Adaptive Modulation and Coding (AMC for short) technology is based on the user level and can only compensate for the difference of the user level, while the difference between the time slots cannot be quickly compensated by the user-level AMC. Therefore, when the user's demodulation performance varies greatly between time slots, the Block Error Rate (BLER) is convergent from the user's point of view, but the BLER on different time slots varies greatly; That is, the BLER of some time slots is very high, and the BLER of some time slots is very low, so that the spectral efficiency is not fully improved. In order to solve this technical problem, it can be solved by AMC between user time slots. However, due to the time slot dimension introduced by AMC between time slots, the number of samples for AMC learning in each time slot is reduced, which further aggravates the non-convergence problem of small packet scheduling users.
  • AMC Adaptive Modulation and Coding
  • Embodiments of the present application provide an inner loop value adjustment method and device, a storage medium, and an electronic device to at least solve the technical problem that the demodulation difference between time-domain scheduling units cannot be quickly compensated by user-level AMC.
  • a method for adjusting an inner loop value including: cyclically executing the following steps, wherein a historical inner loop adjustment factor of each time-domain scheduling unit in a radio frame is initialized to The initial inner loop adjustment factor of each time-domain scheduling unit: determine the block error rate corresponding to each time-domain scheduling unit according to the ACK/NACK information corresponding to each time-domain scheduling unit, wherein the radio frame includes N time-domain scheduling units, where N is a positive integer; each time-domain scheduling unit is determined according to the block error rate corresponding to each time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit The current inner loop adjustment factor of the unit; according to the current inner loop adjustment factor of each time domain scheduling unit, the inner loop value corresponding to each time domain scheduling unit is adjusted.
  • an apparatus for adjusting an inner loop value comprising: a processing module, wherein the processing module is configured to cyclically execute the following steps, wherein each time-domain scheduling unit in a radio frame
  • the historical inner-loop adjustment factor of is initialized as the initial inner-loop adjustment factor of each time-domain scheduling unit: according to the ACK/NACK information corresponding to each time-domain scheduling unit, it is determined that each time-domain scheduling unit corresponds to The block error rate of A loop adjustment factor, determining the current inner loop adjustment factor of each time domain scheduling unit; according to the current inner loop adjustment factor of each time domain scheduling unit, the inner loop value corresponding to each time domain scheduling unit make adjustments.
  • a computer-readable storage medium is also provided, and a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute any one of the above methods when running steps in the examples.
  • an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute any one of the above Steps in Method Examples.
  • FIG. 1 is a block diagram of a hardware structure of an optional electronic device according to an embodiment of the present application.
  • FIG. 2 is a flowchart of an adjustment method for an inner loop value according to an embodiment of the present application
  • FIG. 3 is a schematic diagram of constructing a difference absolute value matrix according to an adjustment method of an inner loop value according to an embodiment of the present application
  • FIG. 4 is a flowchart of a method for adjusting an inner loop value according to another embodiment of the present application.
  • FIG. 5 is a flowchart of a data collection stage of a method for adjusting an inner loop value according to another embodiment of the present application.
  • FIG. 6 is a flowchart of a data processing stage of a method for adjusting an inner loop value according to another embodiment of the present application.
  • FIG. 7 is a flowchart of a feature extraction and calculation stage of a method for adjusting an inner loop value according to another embodiment of the present application.
  • FIG. 8 is a flowchart of a model learning phase of a method for adjusting an inner loop value according to another embodiment of the present application.
  • FIG. 9 is a flowchart of a model application stage of a method for adjusting an inner loop value according to another embodiment of the present application.
  • FIG. 10 is a structural block diagram of an apparatus for adjusting an inner loop value according to an embodiment of the present application.
  • FIG. 1 is a block diagram of a hardware structure of an optional electronic device according to an embodiment of the present application.
  • the electronic device may include one or more (only one is shown in FIG. 1 ) processor 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above electronic device may further include a transmission device 106 and an input and output device 108 for communication functions.
  • FIG. 1 is only a schematic diagram, which does not limit the structure of the above electronic device.
  • the electronic device may also include more or fewer components than shown in FIG. 1 , or have a different configuration than that shown in FIG. 1 .
  • the memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer programs corresponding to the adjustment method of the inner loop value in the embodiment of the present application.
  • the processor 102 runs the computer program stored in the memory 104, Thereby, various functional applications and data processing are performed, that is, the above-mentioned method is realized.
  • Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory.
  • memory 104 may further include memory located remotely from processor 102, which may be connected to the electronic device through a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
  • Transmission means 106 are used to receive or transmit data via a network.
  • Specific examples of the above-mentioned network may include a wireless network or a wired network.
  • the transmission device 106 includes a network adapter (Network Interface Controller, NIC for short), which can be connected to other network devices so as to communicate with the Internet.
  • the transmission device 106 may be a radio frequency (Radio Frequency, RF for short) module, which is used to communicate with the Internet in a wireless manner.
  • RF Radio Frequency
  • FIG. 2 is a flowchart of a method for adjusting an inner loop value according to an embodiment of the present application.
  • the flow includes the following steps: cyclically execute the following Steps S202 to S206, wherein the historical inner-loop adjustment factor of each time-domain scheduling unit in the radio frame is initialized to the initial inner-loop adjustment factor of each time-domain scheduling unit:
  • Step S202 Determine the block error rate corresponding to each time-domain scheduling unit according to the ACK/NACK information corresponding to each time-domain scheduling unit, where the radio frame includes N time-domain scheduling units, and N is positive integer;
  • Step S204 according to the block error rate corresponding to each time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit, determine the current inner-loop adjustment factor of each time-domain scheduling unit;
  • Step S206 Adjust the inner loop value corresponding to each time domain scheduling unit according to the current inner loop adjustment factor of each time domain scheduling unit.
  • the following steps are performed cyclically, wherein the historical inner-loop adjustment factor of each time-domain scheduling unit in the radio frame is initialized to the initial inner-loop adjustment factor of each time-domain scheduling unit: according to each time-domain scheduling unit
  • the ACK/NACK information corresponding to the time-domain scheduling unit determines the block error rate corresponding to each time-domain scheduling unit, wherein the radio frame includes N time-domain scheduling units, and N is a positive integer;
  • the block error rate corresponding to the time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit determine the current inner-loop adjustment factor of each time-domain scheduling unit; according to each time-domain scheduling unit
  • the current inner loop adjustment factor of adjusts the inner loop value corresponding to each time domain scheduling unit.
  • the current inner-loop adjustment factor of each time-domain scheduling unit is determined, and further according to the The current inner-loop adjustment factor of each time-domain scheduling unit adjusts the inner-loop value corresponding to each time-domain scheduling unit. Therefore, it is possible to solve the problem that the demodulation difference between time-domain scheduling units cannot be resolved by user-level AMC.
  • the technical problem of fast compensation achieves fast compensation for the demodulation difference between time slots, and the technical effect of different scheduling for users in different time domain scheduling units according to the difference of demodulation performance.
  • determining the block error rate corresponding to each time-domain scheduling unit according to the ACK/NACK information corresponding to each time-domain scheduling unit includes: collecting the ACK/NACK information corresponding to each time-domain scheduling unit, and the block error rate corresponding to each time-domain scheduling unit is determined according to the ACK/NACK information corresponding to each time-domain scheduling unit.
  • the block error rate corresponding to each time-domain scheduling unit may be determined, the ACK/NACK information except conservative scheduling on each time-domain scheduling unit may be collected, and for each time-domain scheduling unit ACK/NACK are counted separately.
  • the block error rate BLER i corresponding to each time-domain scheduling unit is determined according to the following formula:
  • the block error rate corresponding to each time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit are determined to determine the block error rate of each time-domain scheduling unit.
  • the historical inner loop adjustment factor of the scheduling unit i, factor i (BLER_tar-BLER i )*step, BLER_tar is the preset block error rate target value, BLER i is the block error rate corresponding to the time domain scheduling unit i, and step is Adjust the step size, and step is greater than
  • the block error rate corresponding to each time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit are determined to determine the block error rate of each time-domain scheduling unit.
  • the current inner loop adjustment factor includes: when the range of the block error rates corresponding to the N time-domain scheduling units is greater than or equal to a preset threshold, according to the block error rate corresponding to each time-domain scheduling unit, Divide the N time-domain scheduling units into M subsets, wherein the range of the block error rate corresponding to the time-domain scheduling units in each of the M subsets is less than the preset threshold, where M is A positive integer, M is less than N; according to the block error rate corresponding to the time-domain scheduling unit of each subset, the inner loop adjustment factor corresponding to each subset is determined; according to the inner loop adjustment factor corresponding to each subset and The historical inner-loop adjustment factor of each time-domain scheduling unit determines the current inner-loop adjustment factor of each time-domain scheduling unit.
  • the determining the inner loop adjustment factor corresponding to each subset according to the block error rate corresponding to the time-domain scheduling unit of each subset includes: determining the each subset according to the following formula
  • the inner loop adjustment factor factor k corresponding to the set: factor k (BLER_tar-BLERA k )*step, where 1 ⁇ k ⁇ M, BLER_tar is the preset block error rate target value, and BLERA k is the time domain in the subset k
  • the average value of the block error rate corresponding to the scheduling unit, step is the adjustment step size, and step is greater than 0.
  • the current inner loop adjustment factor of each time domain scheduling unit is determined according to the inner loop adjustment factor corresponding to each subset and the historical inner loop adjustment factor of each time domain scheduling unit.
  • dividing the N time-domain scheduling units into M subsets according to the block error rate corresponding to each time-domain scheduling unit includes: scheduling according to the N time-domain scheduling units The block error rate set corresponding to the unit is constructed, and an N ⁇ N difference absolute value matrix A is constructed, wherein each block error rate in the block error rate set is related to each time-domain scheduling in the N time-domain scheduling units Units are in one-to-one correspondence; repeat the following steps until the value c is equal to 1, wherein the current matrix is initialized as the matrix A, and the current time-domain scheduling unit set is initialized as the set composed of the N time-domain scheduling units, so
  • the value c is initially the N: search for a c ⁇ c target sub-matrix in the current matrix, wherein each element in the target sub-matrix is smaller than the preset threshold; When the target sub-matrix is searched in the matrix, the set composed of the time-domain scheduling units corresponding to each row in the target sub-matrix is determined as a sub
  • , j 1, 2, L N.
  • the dividing the N time-domain scheduling units into M subsets according to the block error rate corresponding to each time-domain scheduling unit includes: dividing the N time-domain scheduling units The N block error rates corresponding to the unit are divided into M consecutive intervals, wherein each interval in the M intervals includes at least one block error rate, and the range of the block error rate in each interval is less than the preset threshold; dividing the time-domain scheduling units corresponding to the block error rate included in the interval k in the M intervals into the subset k of the N time-domain scheduling units, and dividing them into M subsets in total, Among them, 1 ⁇ k ⁇ M.
  • the time-domain scheduling unit corresponding to the block error rate is divided into two intervals with a smaller initial value of the block error rate in the interval.
  • the block error rate corresponding to each time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit are determined to determine the block error rate of each time-domain scheduling unit.
  • the current inner-loop adjustment factor includes: when the range of the block error rates corresponding to the N time-domain scheduling units is less than a preset threshold, determining that the current inner-loop adjustment factor of each time-domain scheduling unit is the Describe the historical inner-loop adjustment factor of each time-domain scheduling unit.
  • the historical inner-loop adjustment of each time-domain scheduling unit continues to be multiplexed. factor, that is, after the current inner-loop adjustment factor of each time-domain scheduling unit is determined, keep the historical inner-loop adjustment factor of each time-domain scheduling unit unchanged.
  • each time-domain scheduling unit is determined according to the block error rate corresponding to each time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit After the current inner-loop adjustment factor of The historical inner-loop adjustment factor of is updated to the respective current inner-loop adjustment factor of each time-domain scheduling unit.
  • the adjusting the inner loop value corresponding to each time domain scheduling unit according to the current inner loop adjustment factor of each time domain scheduling unit includes: determining that the user is in the time domain The sum of the inner loop value d i corresponding to the scheduling unit i and the current inner loop adjustment factor T_factor i of the time domain scheduling unit i; the sum of the inner loop value d i and the current inner loop adjustment factor T_factor i , It is determined as the inner loop value adjusted by the user in the time domain scheduling unit i.
  • each time-domain scheduling unit in the radio frame corresponds to each subframe in the radio frame on a one-to-one basis , where N is the number of subframes in the radio frame; when the radio frame is a radio frame in a 5G communication system, each time-domain scheduling unit in the radio frame is related to the number of subframes in the radio frame. There is a one-to-one correspondence with each time slot of , where N is the number of time slots in the radio frame.
  • the inner loop value is the channel measurement value of the base station for the user. Therefore, in the above embodiment, the inner loop value adjusted by the time domain scheduling unit may be set to determine the value of the corresponding time domain scheduling unit. Scheduling strategy, that is, in the above embodiment, the scheduling strategy corresponding to each time-domain scheduling unit can be determined according to the inner loop value adjusted by each time-domain scheduling unit, so as to realize the user's demodulation performance difference in each time-domain unit.
  • the time domain scheduling unit performs differential scheduling.
  • the range of the block error rates corresponding to the N time-domain scheduling units is: the maximum value of the block error rates corresponding to the N time-domain scheduling units and the N time-domain scheduling units The difference between the minimum values in the block error rate corresponding to the unit.
  • a method for adjusting the inner loop value is provided. As shown in FIG. 4 , the method includes the following stages:
  • Phase 1 includes the following steps:
  • Step 1 For the 4G system, initialize the inner loop conversion value of each subframe in the radio frame; for the 5G system, initialize the inner loop conversion factor of each time slot in the radio frame;
  • a subframe or a time slot may be referred to as a time-domain scheduling unit (or a time-domain scheduling unit) in a radio frame;
  • Step 2 Collect the ACK/NACK information externalized by conservative scheduling on each time-domain scheduling unit, and count them separately in each time-domain scheduling unit;
  • Step 3 Determine whether the number of samples collected on each time-domain scheduling unit reaches the minimum threshold. If not, return to step 2 and continue counting; The BLER corresponding to the domain scheduling unit is obtained, and the corresponding BLER information set on each time domain scheduling unit in the radio frame is obtained.
  • stage 3 feature extraction and calculation; wherein stage 3 includes the following steps:
  • Step 4 Calculate a statistic reflecting the divergence degree of the BLER information set according to the corresponding BLER information set on each time-domain scheduling unit in the radio frame;
  • Step 5 Determine whether the divergence degree of the corresponding BLER information set on each time-domain scheduling unit in the wireless frame is greater than the given threshold according to the statistics; if it is greater than the given threshold, perform the differential conversion between the time-domain scheduling units and go to step 6 ; Otherwise, reuse the conversion factor corresponding to each current time-domain scheduling unit, apply this conversion factor to perform the inner-loop conversion on the user on the corresponding time-domain scheduling unit, and go to step 2, and clear the historical data of the last statistics (i.e. last collected ACK/NACK information).
  • stage 4 includes the following steps:
  • Step 6 classifying each time-domain scheduling unit in the radio frame, so that the divergence degree of the BLER information corresponding to the time-domain scheduling unit in each category is less than a given threshold;
  • Step 7 Calculate the average BLER information corresponding to each category
  • Step 8 Calculate the inner loop conversion factor of the corresponding class according to the target BLER information set by the system and the average BLER information calculated by each class;
  • Step 9 Update the inner loop conversion factor of each time-domain scheduling unit according to the maintained historical inner-loop conversion factor of the time-domain scheduling unit, the set to which the time-domain scheduling unit belongs, and the conversion factor corresponding to the set, thereby obtaining each of the time-domain scheduling units in the wireless frame. Inner loop reduction factor on time-domain scheduling units.
  • stage 5 model application; wherein stage 5 includes the following steps:
  • Step 10 Output the inner loop conversion factor on each time-domain scheduling unit, and apply the factor to perform inner-loop conversion for the user on the corresponding time-domain scheduling unit.
  • Step 11 go to step 2, and clear the statistical information of the previous round, and proceed to the next round of learning.
  • the data collection stage (that is, the process of counting the number of ACK/NACK) specifically includes the following steps:
  • Step S501 clearing cached data
  • Step S502 determine whether the corresponding adjustment is conservative scheduling; if so, go to step S503; otherwise, go to step S504;
  • Step S503 discard the data, and go to step S505;
  • Step S504 collecting corresponding ACK/NACK information by time slot for uplink and downlink respectively;
  • Step S505 judging whether the statistical sample size meets the requirements; if so, jump to stage 2 for data processing, otherwise, return to step S502.
  • the data processing stage includes the following steps:
  • Step S601 determine whether to perform model learning (that is, determine whether to perform differential conversion between subframes); if so, perform step S603, otherwise, perform step S602;
  • Step S602 obtain A/N sample data corresponding to each time slot according to the collected data, and end the current data processing stage;
  • Step S603 aggregate the statistical ACK/NACK information according to the user ID
  • Step S604 calculating the BLER corresponding to each user
  • Step S605 determine whether the BLER corresponding to each user is within the corresponding convergence interval, if so, go to step S607, otherwise go to step S606;
  • Step S606 discarding the data
  • Step S607 delivering the ACK/NACK data corresponding to each user to the set corresponding to the subframe according to the subframe to which the ACK/NACK belongs;
  • Step S608 Obtain ACK/NACK sample data corresponding to each time slot.
  • the feature extraction and calculation stage includes the following steps:
  • Step S701 obtaining A/N sample data corresponding to each time slot
  • Step S702 calculating the BLER corresponding to each time slot to obtain a set BLER_SET;
  • Step S703 calculating the range of the set BLER_SET
  • Step S704 determine whether the range is greater than the given threshold TH; if so, go to step S706, otherwise go to step S705;
  • Step S705 continue to collect data, and do not correct the existing model, but use the existing model
  • Step S706 learn and correct the model.
  • model learning phase includes the following steps:
  • Step S801 obtaining the BLER corresponding to each time slot
  • Step S802 classifying the time slot
  • the range within each class is less than the threshold; the range after union of any two classes is greater than the threshold; the time-domain scheduling units (such as time slots) in any two classes have no intersection; the union of all classes is all A set of time slots.
  • Step S803 calculating the average BLER of each class
  • Step S804 calculating the correction value of each class; wherein, specifically calculating the inner loop adjustment factor of each class according to the target BLER, the average BLER of each class and the adjustment step size;
  • Step S805 revising the model, which specifically includes calculating the inner loop adjustment factor of each time slot.
  • model application phase includes the following steps:
  • Step S901 obtaining a correction value corresponding to each time slot according to the result of the model correction
  • Step S902 determining the time slot corresponding to the user scheduling (including but not limited to the corresponding air interface index);
  • Step S903 determining the inner loop correction value corresponding to the time slot; wherein, according to the index position, determine the corresponding inner loop correction value, and perform the inner loop correction;
  • Step S904 Determine the scheduling of the user according to the revised inner loop and outer loop.
  • a method for adjusting an inner loop value is provided, and the method can be applied in a Long Term Evolution (Long Term Evolution, referred to as LTE for short) system, and the time domain scheduling unit in the above embodiment is LTE Taking a subframe in a radio frame of the system as an example, the method includes the following steps:
  • Step 1 The inner loop conversion factor of each subframe of the initialized radio frame (that is, the inner loop adjustment factor in the above embodiment) is:
  • sf_factor i represents the inner loop conversion factor corresponding to the ith subframe, and N represents the number of subframes included in a radio frame;
  • Step 2 Count the number of ACK/NACK corresponding to each subframe except for conservative scheduling; wherein, the number of ACK/NACK is maintained according to the subframe corresponding to the ACK/NACK, for example: each ACK/NACK counted according to the following format Information (or called samples) for maintenance: user ID information carried in ACK/NACK: subframe to which ACK/NACK belongs;
  • format Information or called samples
  • Step 3 Determine whether the number of ACK/NACK samples collected for each subframe reaches the sample number threshold, if not, return to step 2 to continue counting; otherwise, perform the following steps:
  • Step 3.1 Aggregate the statistical ACK/NACK information according to the user ID, and calculate the BLER corresponding to each user;
  • Step 3.2 Determine whether the BLER corresponding to each user is in the corresponding convergence interval (the convergence interval is a preset interval range, such as the range from 0.08 to 0.12); if so, press the ACK/NACK data corresponding to each user as ACK/NACK
  • the home subframe is delivered to the set corresponding to the subframe, and step 3.3 is performed; if not, discard the user's data;
  • Step 3.3 Calculate the BLER corresponding to each subframe, and count the BLER information corresponding to each subframe to obtain the BLER set (BLER_SET):
  • BLER i represents the block error rate corresponding to the ith subframe
  • Step 4 Calculate the range D of the set BLER_SET:
  • the max ⁇ function means taking the maximum value
  • the min ⁇ function means taking the minimum value
  • Step 5 determine whether the range is greater than the given threshold TH (that is, the preset threshold in the above-mentioned embodiment); if D ⁇ TH, determine to perform differential conversion between subframes, and go to step 6; otherwise, reuse the current subframe.
  • Frame conversion factor sf_factor i , i 1, 2, ⁇ N, and go to step 2, and clear the historical data of the last statistics;
  • Step 6 Divide the BLER_SET set into N subsets ⁇ set k
  • Step 7 Calculate the average BLER corresponding to each of the N subsets: BLERA k , wherein, in this embodiment,
  • Step 8 Calculate the inner ring conversion factor factor k corresponding to each subset:
  • BLER_tar is the preset block error rate target value
  • step is the preset value
  • Step 9 Calculate the inner loop conversion factor corresponding to each subframe:
  • ⁇ is the filter factor
  • sf'_factor is the historical inner ring conversion factor
  • Step 11 Go to Step 2, and clear the statistical information of the previous round, and proceed to the next round of learning.
  • a method for adjusting an inner loop value is provided, and the method can be applied in a 5G NR system.
  • the time domain scheduling unit in the above embodiment is a time slot in a radio frame of the 5G NR system.
  • the method includes the following steps, wherein, when the time-domain scheduling unit is a time slot in a radio frame of the 5G NR system, the block error rate corresponding to each time-domain scheduling unit is actually the corresponding block error rate of each time slot.
  • Slot error rate :
  • Step 1 Initialize the inner loop conversion factor of each time slot in the radio frame:
  • slot_factor i represents the conversion factor corresponding to the i-th time slot
  • N represents the number of time slots in a radio frame
  • Step 2 Count the number of ACK/NACK corresponding to each time slot except conservative scheduling, and maintain the collected ACK/NACK data (or called samples) by time slot, wherein each time slot is counted separately;
  • Step 3 Determine whether the number of ACK/NACK samples collected for each time slot reaches the minimum threshold (the minimum threshold can be 100,000, that is, it is necessary to ensure that the number of ACK/NACK samples collected for each time slot reaches 100,000), if If it is not reached, go back to step 2 to continue the statistics; otherwise, calculate the BLER corresponding to each time slot, and count the BLER information corresponding to each time slot to obtain the BLER set (BLER_SET):
  • BLER i represents the slot error rate corresponding to the ith slot
  • Step 4 Calculate the range D of the set BLER_SET:
  • Step 6 Divide the BLER_SET set into M subsets ⁇ set k
  • k 1, 2, ⁇ M ⁇ , and the specific division method is as follows: initialize the time slot set St to a set composed of N time slots;
  • Step 6.1 Construct an N ⁇ N inter-slot BLER difference absolute value matrix A according to BLER_SET;
  • Step 6.2 According to the constructed difference absolute value matrix A, starting from N, the sub-matrix is dimensionally searched, wherein the sub-matrix satisfies the following conditions: each element is less than the threshold TH, and adding another dimension will no longer satisfy the sub-matrix.
  • the condition that each element of is less than the threshold TH; among them, if there are multiple sub-matrices of the same dimension that satisfy this condition, the sub-matrix with the smallest average value of all elements in the sub-matrix is selected, and the selected sub-matrix is recorded as matrix B;
  • Step 6.3 determine the time slot number corresponding to the row of matrix B, and form a subset set
  • Step 6.4 delete the time slots included in the subset set from the time slot set St, obtain the time slot set St, construct a matrix C according to the set of time slot error rates corresponding to the time slot set St, and assign the matrix C to A;
  • Step 6.5 repeat the process of steps 6.2 to 6.4, and finally get ⁇ set k
  • k 1,2, ⁇ M ⁇ ;
  • Step 7 Calculate the average BLER corresponding to each of the M subsets:
  • Step 8 Calculate the inner ring conversion factor corresponding to each subset
  • Step 9 Calculate the inner loop conversion factor corresponding to each time slot, wherein, for time slot i, if i ⁇ set k (that is, the set to which slot i belongs is i ⁇ set k ), update the corresponding subframe according to the following formula: Inner ring reduction factor:
  • slot_factor i (1- ⁇ )*slot'_factor i + ⁇ *factor k
  • ⁇ is the filter factor
  • slot'_factor i is the historical inner ring conversion factor
  • Step 11 go to step 2, and clear the statistical information of the previous round, and proceed to the next round of learning.
  • a method for adjusting an inner loop value is provided, and the method can be applied in a 5G NR system.
  • the time domain scheduling unit in the above embodiment is a time slot in a radio frame of the 5G NR system.
  • the method includes the following steps, wherein, when the time-domain scheduling unit is a time slot in a radio frame of the 5G NR system, the block error rate corresponding to each time-domain scheduling unit is actually the corresponding block error rate of each time slot.
  • Slot error rate :
  • Step 1 Initialize the inner loop conversion factor of each time slot in the radio frame:
  • slot_factor i represents the inner loop conversion factor corresponding to the i-th time slot
  • N represents the number of time slots in a radio frame
  • Step 2 Count the number of ACKs/NACKs in each time slot.
  • the ACK/NACKs in conservative scheduling are not counted, and the NACKs of the lowest-order scheduling are not counted, and the ACKs of the highest-order scheduling are not counted, that is, the ACKs of each time slot are counted.
  • /NACK does not include ACK/NACK in conservative scheduling, NACK in the lowest order and ACK in the highest order scheduling; the collected ACK/NACK data is maintained by time slot, and the ACK/NACK of each time slot is counted separately;
  • Step 3 Determine whether the number of ACK/NACK samples collected for each time slot reaches the minimum threshold (the minimum threshold can be 100,000, that is, it is necessary to ensure that the number of ACK/NACK samples collected for each time slot reaches 100,000), if If it is not reached, go back to step 2 to continue the statistics; otherwise, calculate the BLER corresponding to each time slot, and count the corresponding time slots to obtain the BLER set (BLER_SET):
  • the minimum threshold can be 100,000, that is, it is necessary to ensure that the number of ACK/NACK samples collected for each time slot reaches 100,000
  • BLER i represents the slot error rate corresponding to the ith slot
  • Step 4 Calculate the range D of the set BLER_SET:
  • Step 6 Divide the BLER_SET set into M subsets ⁇ set k
  • k 1, 2, ⁇ M ⁇ , and the specific division method is as follows:
  • Step 6.1 Sort the time slot indices (for example, time slot serial numbers) according to the descending order of the corresponding BLERs of all time slots;
  • Step 6.2 With the minimum BLER among the BLERs corresponding to all the time slots as the starting point, the maximum BLER+0.1 as the end point, and the preset block error rate interval as the step size, divide all the sorted BLERs into several intervals (the last interval The length may be less than the fixed preset block error rate interval);
  • Step 6.3 According to the several intervals obtained in step 6.2, take several intervals that are closed on the left and open on the right, and classify the corresponding time slots according to the interval to which the corresponding BLER belongs, so as to obtain the subset division ⁇ set k
  • k 1,2, ⁇ M ⁇ ;
  • Step 7 Calculate the average BLER corresponding to each of the M subsets:
  • Step 8 Calculate the inner ring reduction factor corresponding to each subset:
  • Step 9 Calculate the inner loop conversion factor corresponding to each time slot: where, for time slot i, if i ⁇ set k (that is, the set to which slot i belongs is i ⁇ set k ), update the corresponding subframe according to the following formula: Inner ring reduction factor:
  • slot_factor i (1- ⁇ )*slot'_factor i + ⁇ *factor k
  • ⁇ is the filter factor
  • slot'_factor i is the historical inner ring conversion factor
  • Step 11 go to step 2, and clear the statistical information of the previous round, and proceed to the next round of learning.
  • the method for adjusting the inner loop value in the above-mentioned embodiment may be performed repeatedly for many times.
  • determine the current inner-loop adjustment factor of each time-domain scheduling unit as the respective Historical inner-loop adjustment factor (optionally, keep the historical inner-loop adjustment factor of each time-domain scheduling unit unchanged, or update the historical inner-loop adjustment factor of each time-domain scheduling unit to the The current inner loop adjustment factor of each time domain scheduling unit), and clear the ACK/NACK information collected before, and re-execute the collection of the ACK/NACK information corresponding to each time domain scheduling unit, and determine the corresponding time domain scheduling unit.
  • the step of the block error rate of and the historical inner-loop adjustment factor of each time-domain scheduling unit after determining the current inner-loop adjustment factor of each time-domain scheduling unit, update the historical inner-loop adjustment factor of each time-domain scheduling unit to The respective current inner loop adjustment factor of each time-domain scheduling unit.
  • the historical data in the network (that is, the ACK/NACK data corresponding to each time-domain scheduling unit) is used. All the data on the data (it does not matter whether it converges or not, but the conservatively scheduled data is excluded); count the BLER on each time slot; calculate the BLER range of all time slots; thus determine whether the BLER difference between time slots If it is greater than a certain threshold, if it is smaller than the current conversion model (that is, the inner loop conversion factor corresponding to each time-domain scheduling unit); otherwise, the conversion amount of each time slot is learned according to the historical data, and the conversion model is corrected;
  • all time slots are grouped (ie subsets) according to certain rules, for example: the range in each group should be less than a threshold value , and the combined range of any two groups is greater than the threshold, the time-domain scheduling units (such as time slots) in any two groups have no intersection, and the union of all groups
  • the demodulation performance of the same user on different uplink time slots is different due to the difference of reference signals contained in different time slots or due to factors such as uplink and downlink handovers, or the same
  • the learning and correction of the difference in demodulation performance in the case of a difference in transmission/reception between time slots is involved; wherein, the difference in demodulation performance of users between different time slots is learned based on historical data, and according to Based on the learning results, relevant conversion processing is performed on the inner loop values of different time slots, so that users can be scheduled differentially due to the difference in demodulation performance between different time slots, thereby improving the spectral efficiency.
  • the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is better implementation.
  • the technical solution of the present application can be embodied in the form of a software product in essence or in a part that contributes to the prior art, and the computer software product is stored in a storage medium (such as ROM/RAM, magnetic disk, CD-ROM), including several instructions to make a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) execute the methods described in the various embodiments of this application.
  • a storage medium such as ROM/RAM, magnetic disk, CD-ROM
  • an apparatus for adjusting the inner loop value is also provided, and the apparatus is used to implement the above-mentioned embodiments and preferred implementations, and what has already been described will not be repeated.
  • the term "module” may be a combination of software and/or hardware that implements a predetermined function.
  • the apparatus described in the following embodiments is preferably implemented in software, implementations in hardware, or a combination of software and hardware, are also possible and contemplated.
  • Fig. 10 is a structural block diagram of an apparatus for adjusting an inner loop value according to an embodiment of the present application.
  • the apparatus includes: a processing module 1002, wherein the processing module is configured to cyclically execute the following steps, wherein the radio frame
  • the historical inner-loop adjustment factor of each time-domain scheduling unit in is initialized to the initial inner-loop adjustment factor of each time-domain scheduling unit:
  • the inner loop value corresponding to each time domain scheduling unit is adjusted according to the current inner loop adjustment factor of each time domain scheduling unit.
  • the following steps are performed cyclically, wherein the historical inner-loop adjustment factor of each time-domain scheduling unit in the radio frame is initialized to the initial inner-loop adjustment factor of each time-domain scheduling unit: according to each time-domain scheduling unit
  • the ACK/NACK information corresponding to the time-domain scheduling unit determines the block error rate corresponding to each time-domain scheduling unit, wherein the radio frame includes N time-domain scheduling units, and N is a positive integer;
  • the block error rate corresponding to the time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit determine the current inner-loop adjustment factor of each time-domain scheduling unit; according to each time-domain scheduling unit
  • the current inner loop adjustment factor of adjusts the inner loop value corresponding to each time domain scheduling unit.
  • the current inner-loop adjustment factor of each time-domain scheduling unit is determined, and further according to the The current inner-loop adjustment factor of each time-domain scheduling unit adjusts the inner-loop value corresponding to each time-domain scheduling unit. Therefore, it is possible to solve the problem that the demodulation difference between time-domain scheduling units cannot be resolved by user-level AMC.
  • the technical problem of fast compensation achieves fast compensation for the demodulation difference between time slots, and the technical effect of different scheduling for users in different time domain scheduling units according to the difference of demodulation performance.
  • the processing module is further configured to: when the range of the block error rate corresponding to the N time-domain scheduling units is greater than or equal to a preset threshold, determine the time-domain according to the following formula
  • the block error rate corresponding to i, step is the adjustment step size, and step is greater than 0.
  • the processing module is further configured to: in the case that the range of the block error rate corresponding to the N time-domain scheduling units is greater than or equal to a preset threshold, according to each time-domain scheduling unit
  • the block error rate corresponding to the domain scheduling unit the N time-domain scheduling units are divided into M subsets, wherein the range of the block error rate corresponding to the time-domain scheduling unit of each subset of the M subsets is less than
  • M is a positive integer, and M is less than N
  • the inner loop adjustment factor corresponding to each subset is determined;
  • the inner-loop adjustment factor corresponding to the subsets and the historical inner-loop adjustment factor of each time-domain scheduling unit determine the current inner-loop adjustment factor of each time-domain scheduling unit.
  • the determining the inner loop adjustment factor corresponding to each subset according to the block error rate corresponding to the time-domain scheduling unit of each subset includes: determining the each subset according to the following formula
  • the inner loop adjustment factor factor k corresponding to the set: factor k (BLER_tar-BLERA k )*step, where 1 ⁇ k ⁇ M, BLER_tar is the preset block error rate target value, and BLERA k is the time domain in the subset k
  • the average value of the block error rate corresponding to the scheduling unit, step is the adjustment step size, and step is greater than 0.
  • the processing module is further configured to: construct an N ⁇ N difference absolute value matrix A according to the block error rate set corresponding to the N time-domain scheduling units, wherein the block error rate Each block error rate in the rate set has a one-to-one correspondence with each time-domain scheduling unit in the N time-domain scheduling units; the following steps are repeated until the value c is equal to 1, wherein the current matrix is initialized to the Matrix A, the current time-domain scheduling unit set is initialized as a set composed of the N time-domain scheduling units, and the value c is initialized as the N: search for a c ⁇ c target sub-matrix in the current matrix, Wherein, each element in the target sub-matrix is smaller than the preset threshold; when the target sub-matrix is searched from the current matrix, the corresponding row in the target sub-matrix is The set composed of time domain scheduling units is determined as a subset; the one subset is deleted from the current time domain scheduling unit set, and the current time domain scheduling unit set after
  • the processing module is further configured to: divide the N block error rates corresponding to the N time-domain scheduling units into M consecutive intervals, wherein the Each interval includes at least one block error rate, and the range of the block error rate in each interval is less than the preset threshold; comparing the N time-domain scheduling units with the interval in the M intervals The time-domain scheduling unit corresponding to the block error rate included in k is divided into subset k, and M subsets are obtained by division, wherein 1 ⁇ k ⁇ M.
  • the processing module is further configured to: in the case that the range of the block error rate corresponding to the N time-domain scheduling units is smaller than a preset threshold, determine the each time-domain scheduling unit The current inner-loop adjustment factor of the unit is the historical inner-loop adjustment factor of each time-domain scheduling unit.
  • the processing module is further configured to: in the block error rate corresponding to each time-domain scheduling unit and the historical inner-loop adjustment factor of each time-domain scheduling unit, After determining the current inner-loop adjustment factor of each time-domain scheduling unit, when the range of the block error rate corresponding to the N time-domain scheduling units is greater than or equal to a preset threshold, the The historical inner-loop adjustment factor of the domain scheduling unit is updated to the respective current inner-loop adjustment factor of each time-domain scheduling unit.
  • the processing module is further configured to: determine the sum of the inner loop value d i corresponding to the user in the time domain scheduling unit i and the current inner loop adjustment factor T_factor i of the time domain scheduling unit i ; Determine the sum of the inner loop value d i and the current inner loop adjustment factor T_factor i as the inner loop value adjusted by the user in the time domain scheduling unit i.
  • each time-domain scheduling unit in the radio frame corresponds to each subframe in the radio frame on a one-to-one basis , where N is the number of subframes in the radio frame; when the radio frame is a radio frame in a 5G communication system, each time-domain scheduling unit in the radio frame is related to the number of subframes in the radio frame. There is a one-to-one correspondence with each time slot of , where N is the number of time slots in the radio frame.
  • the above modules can be implemented by software or hardware, and the latter can be implemented in the following ways, but not limited to this: the above modules are all located in the same processor; or, the above modules can be combined in any combination The forms are located in different processors.
  • Embodiments of the present application further provide a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.
  • the above-mentioned computer-readable storage medium may include, but is not limited to, a USB flash drive, a read-only memory (Read-Only Memory, referred to as ROM for short), and a random access memory (Random Access Memory, referred to as RAM for short) , mobile hard disk, magnetic disk or CD-ROM and other media that can store computer programs.
  • ROM Read-Only Memory
  • RAM Random Access Memory
  • Embodiments of the present application further provide an electronic device, including a memory and a processor, where a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.
  • the above-mentioned electronic device may further include a transmission device and an input-output device, wherein the transmission device is connected to the above-mentioned processor, and the input-output device is connected to the above-mentioned processor.
  • modules or steps of the present application can be implemented by a general-purpose computing device, and they can be centralized on a single computing device, or distributed in a network composed of multiple computing devices
  • they can be implemented in program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, can be performed in a different order than shown here.
  • the described steps, or they are respectively made into individual integrated circuit modules, or a plurality of modules or steps in them are made into a single integrated circuit module to realize.
  • the present application is not limited to any particular combination of hardware and software.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Quality & Reliability (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请实施例提供了一种内环值的调整方法和装置、存储介质及电子装置，上述方法包括：循环执行以下步骤，其中，无线帧中的每个时域调度单元的历史内环调整因子被初始化为每个时域调度单元的初始内环调整因子：根据每个时域调度单元对应的ACK/NACK信息，确定每个时域调度单元对应的误块率，其中，无线帧包括N个时域调度单元，N为正整数；根据每个时域调度单元对应的误块率和每个时域调度单元的历史内环调整因子，确定每个时域调度单元的当前内环调整因子；根据每个时域调度单元的当前内环调整因子，对每个时域调度单元对应的内环值进行调整。通过本申请实施例，解决了时域调度单元间的解调差异无法通过用户级的AMC来快速补偿的技术问题。

Description

内环值的调整方法和装置、存储介质及电子装置
相关申请的交叉引用
本公开基于2020年12月09日提交的发明名称为“内环值的调整方法和装置、存储介质及电子装置”的中国专利申请CN202011433224.3，并且要求该专利申请的优先权，通过引用将其所公开的内容全部并入本公开。
技术领域
本申请实施例涉及通信领域，具体而言，涉及一种内环值的调整方法和装置、存储介质及电子装置。
背景技术
随着移动通信技术的发展，用户对通信的需求也在逐渐提升，通信需求从语音为主发展到目前的数据为主，并且通信需求还在不断提升，例如通过更高的业务速率实时观看高清直播等等需求，为满足这些需求，第五代(5G)移动通信技术应运而生。
在5G标准协议中，不同时隙配置的参考信号存在差异，导致通过相同数量的资源块(Resource Block,简称为RB)传输的有效信息的数量存在差异，因而会导致用户在不同时隙间的码率存在差异，进而会引起解调性能的差异。并且由于上下行切换对于射频端的影响，也会导致时隙间的解调性能存在差异，目前的现网测试也可以证明存在这一差异。
传统的自适应编码调制技术(Adaptive Modulation and Coding，简称为AMC)是基于用户级别的，仅能补偿用户级别的差异，而时隙间的差异无法通过用户级的AMC来快速补偿。因此，当用户的解调性能在时隙间差异比较大时，从用户来看误块率(Block Error Rate，简称为BLER)是收敛的，但会出现不同时隙上的BLER差异很大；即有的时隙BLER非常高，有的时隙BLER非常低，从而导致频谱效率没有得到充分的提升。为了解决这一技术问题，可以通过用户时隙间AMC来解决。但是，由于时隙间AMC引入了时隙维度，从而导致供每个时隙进行AMC学习的样本数减少了，因此进一步加剧了小包调度用户的不收敛问题。
发明内容
本申请实施例提供了一种内环值的调整方法和装置、存储介质及电子装置，以至少解决时域调度单元间的解调差异无法通过用户级的AMC来快速补偿的技术问题。
根据本申请的一个实施例，提供了一种内环值的调整方法，包括：循环执行以下步骤，其中，无线帧中的每个时域调度单元的历史内环调整因子被初始化为所述每个时域调度单元的初始内环调整因子：根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，其中，所述无线帧包括N个时域调度单元，N为正整数；根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子；根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整。
根据本申请的另一个实施例，提供了一种内环值的调整装置，包括：处理模块，其中，所述处理模块设置为循环执行以下步骤，其中，无线帧中的每个时域调度单元的历史内环调 整因子被初始化为所述每个时域调度单元的初始内环调整因子：根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，其中，所述无线帧包括N个时域调度单元，N为正整数；根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子；根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整。
根据本申请的又一个实施例，还提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机程序，其中，所述计算机程序被设置为运行时执行上述任一项方法实施例中的步骤。
根据本申请的又一个实施例，还提供了一种电子装置，包括存储器和处理器，所述存储器中存储有计算机程序，所述处理器被设置为运行所述计算机程序以执行上述任一项方法实施例中的步骤。
附图说明
图1是根据本申请实施例的一种可选的电子装置的硬件结构框图；
图2是根据本申请实施例的内环值的调整方法的流程图；
图3是根据本申请实施例的内环值的调整方法构造差绝对值矩阵的示意图；
图4是根据本申请另一实施例的内环值的调整方法的流程图；
图5是根据本申请另一实施例的内环值的调整方法的数据收集阶段的流程图；
图6是根据本申请另一实施例的内环值的调整方法的数据处理阶段的流程图；
图7是根据本申请另一实施例的内环值的调整方法的特征提取与计算阶段的流程图；
图8是根据本申请另一实施例的内环值的调整方法的模型学习阶段的流程图；
图9是根据本申请另一实施例的内环值的调整方法的模型应用阶段的流程图；
图10是根据本申请实施例的内环值的调整装置的结构框图。
具体实施方式
下文中将参考附图并结合实施例来详细说明本申请的实施例。
需要说明的是，本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。
本申请实施例中所提供的方法实施例可以在移动终端、基站、计算机终端、服务器或者类似的运算装置中执行。以运行在电子装置上为例，图1是根据本申请实施例的一种可选的电子装置的硬件结构框图。如图1所示，电子装置可以包括一个或多个(图1中仅示出一个)处理器102(处理器102可以包括但不限于微处理器MCU或可编程逻辑器件FPGA等的处理装置)和用于存储数据的存储器104，其中，上述电子装置还可以包括用于通信功能的传输设备106以及输入输出设备108。本领域普通技术人员可以理解，图1所示的结构仅为示意，其并不对上述电子装置的结构造成限定。例如，电子装置还可包括比图1中所示更多或者更少的组件，或者具有与图1所示不同的配置。
存储器104可用于存储计算机程序，例如，应用软件的软件程序以及模块，如本申请实施例中的内环值的调整方法对应的计算机程序，处理器102通过运行存储在存储器104内的计算机程序，从而执行各种功能应用以及数据处理，即实现上述的方法。存储器104可包括高速随机存储器，还可包括非易失性存储器，如一个或者多个磁性存储装置、闪存、或者其他非易失性固态存储器。在一些实例中，存储器104可进一步包括相对于处理器102远程设 置的存储器，这些远程存储器可以通过网络连接至电子装置。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。
传输装置106用于经由一个网络接收或者发送数据。上述的网络具体实例可包括无线网络，或是有线网络。在一个实例中，传输装置106包括一个网络适配器(Network Interface Controller，简称为NIC)，其可通过其他网络设备相连从而可与互联网进行通讯。在一个实例中，传输装置106可以为射频(Radio Frequency，简称为RF)模块，其用于通过无线方式与互联网进行通讯。
在本实施例中提供了一种内环值的调整方法，图2是根据本申请实施例的内环值的调整方法的流程图，如图2所示，该流程包括如下步骤：循环执行以下步骤S202至步骤S206，其中，无线帧中的每个时域调度单元的历史内环调整因子被初始化为所述每个时域调度单元的初始内环调整因子：
步骤S202，根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，其中，所述无线帧包括N个时域调度单元，N为正整数；
步骤S204，根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子；
步骤S206，根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整。
通过本申请，循环执行以下步骤，其中，无线帧中的每个时域调度单元的历史内环调整因子被初始化为所述每个时域调度单元的初始内环调整因子：根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，其中，所述无线帧包括N个时域调度单元，N为正整数；根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子；根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整。由于根据每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，以及进而根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整，因此，可以解决时域调度单元间的解调差异无法通过用户级的AMC来快速补偿的技术问题，达到了对时隙间的解调差异的快速补偿，以及对用户在不同时域调度单元根据解调性能的差异记性差异化调度的技术效果。
需要说明的是，在上述实施例中，所述根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，包括：收集所述每个时域调度单元对应的ACK/NACK信息，并根据所述每个时域调度单元对应的ACK/NACK信息确定所述每个时域调度单元对应的误块率。
其中，在上述实施例中，可以在确定所述每个时域调度单元对应的误块率，收集各个时域调度单位上除保守调度外的ACK/NACK信息，并针对每个时域调度单位对ACK/NACK单独计数。可选地，根据以下公式确定所述每个时域调度单元对应的误块率BLER i:
Figure PCTCN2021136759-appb-000001
其中，i＝1,2,L N，ni1为时域调度单元i对应的NACK数量，ni2为时域调度单元i对应的ACK数量。
在一个示例性实施例中，所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，包括：在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，根据以下公式确定时域调度单元i的当前内环调整因子T_factor i：T_factor i＝(1-α)*T'_factor i+α*factor i，其中，1≤i≤N,α为滤波因子，0≤α≤1，T'_factor i为所述时域调度单元i的历史内环调整因子，factor i＝(BLER_tar-BLER i)*step，BLER_tar为预设误块率目标值，BLER i为所述时域调度单元i对应的误块率，step为调整步长、且step大于0。
在一个示例性实施例中，所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，包括：在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，根据所述每个时域调度单元对应的误块率，将所述N个时域调度单元划分至M个子集，其中，所述M个子集中的每个子集的时域调度单元对应的误块率的极差均小于所述预设阈值，M中为正整数，M小于N；根据所述每个子集的时域调度单元对应的误块率，确定所述每个子集对应的内环调整因子；根据所述每个子集对应的内环调整因子和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子。
在一个示例性实施例中，所述根据所述每个子集的时域调度单元对应的误块率，确定所述每个子集对应的内环调整因子，包括：根据以下公式确定所述每个子集对应的内环调整因子factor k：factor k＝(BLER_tar-BLERA k)*step，其中，1≤k≤M,BLER_tar为预设误块率目标值，BLERA k为子集k中的时域调度单元对应的误块率的平均值，step为调整步长、且step大于0。
在一个示例性实施例中，所述根据所述每个子集对应的内环调整因子和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，包括：根据以下公式确定时域调度单元i的当前内环调整因子T_factor i：T_factor i＝(1-α)*T'_factor i+α*factor k，其中，所述时域调度单元i属于所述M个子集中的第k个子集，1≤i≤N，α为滤波因子，0≤α≤1，T'_factor i为所述时域调度单元i的历史内环调整因子。
在一个示例性实施例中，所述根据所述每个时域调度单元对应的误块率，将所述N个时域调度单元划分至M个子集，包括：根据所述N个时域调度单元对应的误块率集合，构造N×N的差绝对值矩阵A，其中，所述误块率集合中的每个误块率与所述N个时域调度单元中的每个时域调度单元一一对应；重复执行以下步骤，直至数值c等于1，其中，当前矩阵被初始化为所述矩阵A，当前时域调度单元集合被初始为所述N个时域调度单元组成的集合，所述数值c被初始为所述N：在所述当前矩阵中搜索c×c的目标子矩阵，其中，所述目标子矩阵中的每个元素均小于所述预设阈值；在从所述当前矩阵中搜索到所述目标子矩阵的情况下，将所述目标子矩阵中的每一行对应的时域调度单元组成的集合确定为一个子集；从所述当前时域调度单元集合中删除所述一个子集，并将删除所述一个子集后的当前时域调度单元集合 确定为所述当前时域调度单元集合；将根据所述当前时域调度单元集合对应的误块率集合构造的(N-c)×(N-c)的差绝对值矩阵，确定为所述当前矩阵，其中，所述当前时域调度单元集合对应的误块率集合中的每个误块率与所述当前时域调度单元集合中的每个时域调度单元一一对应；在从所述当前矩阵中未搜索到所述目标子矩阵的情况下，将所述数值c减1后得到的差值确定为所述数值c；其中，在所述数值c等于1的情况下，将所述当前矩阵中的每一行对应的时域调度单元确定为一个子集。
可选地，如图3所示，根据所述N个时域调度单元对应的误块率集合，构造N×N的差绝对值矩阵A：
Figure PCTCN2021136759-appb-000002
中的每一行、每一列均与时域调度单元一一相对应，并且矩阵A中的元素A ij＝|BLER i-BLER j|，j＝1,2,L N。
在一个示例性实施例中，所述根据所述每个时域调度单元对应的误块率，将所述N个时域调度单元划分至M个子集，包括：将所述N个时域调度单元对应的N个误块率划分至M个连续的区间，其中，所述M个区间中的每个区间包括至少一个误块率，所述每个区间内的误块率的极差小于所述预设阈值；将所述N个时域调度单元中，与所述M个区间中的区间k包括的误块率对应的时域调度单元划分至子集k，共划分得到M个子集，其中，1≤k≤M。
其中，在上述实施例中，当某一个误块率为两个区间的边界值时，将该误块率对应的时域调度单元划分至两个区间中的具有较小误块率起始值的区间中。
在一个示例性实施例中，所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，包括：在所述N个时域调度单元对应的误块率的极差小于预设阈值的情况下，确定所述每个时域调度单元的当前内环调整因子为所述每个时域调度单元的历史内环调整因子。
需要说明的是，在上述实施例中，在所述N个时域调度单元对应的误块率的极差小于预设阈值的情况下，继续复用每个时域调度单元的历史内环调整因子，即在确定出每个时域调度单元的当前内环调整因子后，保持每个时域调度单元的历史内环调整因子不变。
在一个示例性实施例中，在所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子之后，所述方法还包括：在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，将所述每个时域调度单元的历史内环调整因子更新为所述每个时域调度单元各自的当前内环调整因子。
在一个示例性实施例中，所述根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整，包括：确定用户在时域调度单元i对应的内环值d i与所述时域调度单元i的当前内环调整因子T_factor i的和；将所述内环值d i与所述当前内环调整因子T_factor i的和，确定为所述用户在所述时域调度单元i调整后的内环值。
在一个示例性实施例中，在所述无线帧为4G通信系统中的无线帧的情况下，所述无线帧 中的每个时域调度单元与所述无线帧中的每个子帧一一对应，其中，N为所述无线帧中的子帧数目；在所述无线帧为5G通信系统中的无线帧的情况下，所述无线帧中的每个时域调度单元与所述无线帧中的每个时隙一一对应，其中，N为所述无线帧中的时隙数目。
需要说明的是，在上述实施例中，内环值为基站对于用户的信道测量值，因此在上述实施例中，时域调度单元调整后的内环值可以设置为确定对应时域调度单元的调度策略，即在上述实施例中，可以根据各个时域调度单元调整后的内环值确定各个时域调度单元对应的调度策略，从而实现根据用户在各个时域单元的解调性能差异对各个时域调度单元进行差异化调度。
需要说明的是，在上述实施例中，所述N个时域调度单元对应的误块率的极差为：N个时域调度单元对应的误块率中的最大值与N个时域调度单元对应的误块率中的最小值之间的差值。
以下结合一示例对上述实施例中的内环值的调整方法进行解释说明，但不用于限定本申请实施例的技术方案。
在一个示例性实施例中，提供了一种内环值的调整方法，如图4所示，该方法包括以下阶段：
阶段1、数据收集：其中，阶段1包括以下步骤：
步骤1、对于4G系统，初始化无线帧中各个子帧的内环折算值；对于5G系统，初始化无线帧中各个时隙的内环折算因子；
需要说明的是，在本申请实施例中，子帧或时隙均可以称为无线帧中的时域调度单位(或时域调度单元)；
步骤2、收集各个时域调度单位上除保守调度外化的ACK/NACK信息，在各个时域调度单位分别单独计数；
步骤3、判断各个时域调度单位上收集的样本个数是否达到最小阈值，如果未达到，返回步骤2，继续统计；否则，根据各个时域调度单位上收集的ACK/NACK信息，计算各时域调度单位对应的BLER，得到无线帧中各个时域调度单位上对应BLER信息集合。
阶段2、数据处理；
阶段3、特征提取与计算；其中，阶段3包括以下步骤：
步骤4、根据无线帧中各个时域调度单位上对应的BLER信息集合，计算反映BLER信息集合发散程度的统计量；
步骤5、根据统计量判断无线帧中各个时域调度单位上，对应的BLER信息集合发散程度是否大于给定阈值；如果大于给定阈值，进行时域调度单位之间的差异化折算转步骤6；否则，复用当前各个时域调度单位对应的折算因子，应用此折算因子在对应的时域调度单位上对用户进行内环折算，并转步骤2，以及清空上次统计的历史数据(即上次收集到的ACK/NACK信息)。
阶段4、模型学习：其中，阶段4包括以下步骤：
步骤6、对无线帧中各个时域调度单位进行分类，使得每一类中的时域调度单位对应的BLER信息的发散程度都小于给定阈值；
步骤7、计算各个分类对应的平均BLER信息；
步骤8、根据系统设置的目标BLER信息和各个类计算的平均BLER信息，计算对应类的内环折算因子；
步骤9、根据维护的时域调度单位历史内环折算因子、时域调度单位归属的集合、以及该集合对应的折算因子，更新各个时域调度单位的内环折算因子，从而得到无线帧中各个时域调度单位上的内环折算因子。
阶段5、模型应用；其中，阶段5包括以下步骤：
步骤10、输出各个时域调度单位上的内环折算因子，并应用此因子在对应的时域调度单位上对用户进行内环折算。
步骤11、转步骤2，并清空上一轮的统计信息，进行下一轮学习。
在一个示例性实施例中，如图5所示，数据收集阶段(即统计ACK/NACK数目的过程)具体包括以下步骤：
步骤S501、清空缓存数据；
步骤S502、判断对应的调整是否是保守调度；若是，执行步骤S503；否则，执行步骤S504；
步骤S503、丢弃数据，并转至步骤S505；
步骤S504、对于上下行分别按时隙收集对应的ACK/NACK信息；
步骤S505、判断统计的样本量是否满足要求；若是，跳转至阶段2进行数据处理，否则返回步骤S502。
在一个示例性实施例中，如图6所示，数据处理阶段包括以下步骤：
步骤S601、确定是否进行模型学习(即确定是否进行子帧间差异化折算)；若是，执行步骤S603，否则执行步骤S602；
步骤S602、根据收集的数据得到每个时隙对应的A/N样本数据，并结束当前数据处理阶段；
步骤S603、按用户ID对统计的ACK/NACK信息进行聚合；
步骤S604、计算每个用户对应的BLER；
步骤S605、判断各个用户对应的BLER是否在对应的收敛区间内，若是，执行步骤S607，否则执行步骤S606；
步骤S606、丢弃数据；
步骤S607、把各个用户对应的ACK/NACK数据按ACK/NACK归属的子帧，投递到子帧对应的集合中；
步骤S608、得到每个时隙对应的ACK/NACK样本数据。
在一个示例性实施例中，如图7所示，特征提取与计算阶段包括以下步骤：
步骤S701、获取每个时隙对应的A/N样本数据；
步骤S702、计算每个时隙对应的BLER，得到集合BLER_SET；
步骤S703、计算集合BLER_SET的极差；
步骤S704、判断极差是否大于给定阈值TH；若是，执行步骤S706，否则执行步骤S705；
步骤S705、继续收集数据，并且不对已有模型进行修正，而是使用已有的模型；
步骤S706、学习并修正模型。
在一个示例性实施例中，如图8所示，模型学习阶段包括以下步骤：
步骤S801、获取每个时隙对应的BLER；
步骤S802、对时隙进行分类；
其中，在分类时，按照以下分类规则进行分类：
每一类内的极差要小于阈值；任意两个类做并集后的极差大于阈值；任意两个类中的时域调度单元(例如时隙)无交集；所有类的并集是所有时隙所组成的集合。
步骤S803、计算每个类的平均BLER；
步骤S804、计算每个类的修正值；其中，具体为根据目标BLER、每一类的平均BLER以及调整步长，计算每个类的内环调整因子；
步骤S805、对模型进行修正；其中，具体为计算每个时隙的内环调整因子。
在一个示例性实施例中，如图9所示，模型应用阶段包括以下步骤：
步骤S901、根据模型修正后的结果，获取每个时隙对应的修正值；
步骤S902、确定用户调度对应的时隙(包括但不限于对应的空口索引)；
步骤S903、确定该时隙对应的内环修正值；其中，根据索引位置，确定对应的内环修正值，进行内环修正；
步骤S904、根据修正后的内环和外环确定用户的调度。
在一个示例性实施例中，提供了一种内环值的调整方法，该方法可以应用在长期演进(Long Term Evolution，简称为LTE)系统中，以上述实施例中的时域调度单元为LTE系统的无线帧中的子帧为例，该方法包括以下步骤：
步骤1、初始化无线帧的各个子帧的内环折算因子(即上述实施例中的内环调整因子)为：
sf_factor i＝0,i＝1,2,Λ N，
其中sf_factor i表示第i子帧所对应的内环折算因子，N表示一个无线帧中包括的子帧数 目；
步骤2、统计除保守调度外，各个子帧对应的ACK/NACK数目；其中，按照ACK/NACK对应的子帧对ACK/NACK数目进行维护，例如：按照以下格式对统计的每个ACK/NACK信息(或称为样本)进行维护：ACK/NACK中携带的用户ID信息：ACK/NACK归属的子帧；
步骤3、判断针对每个子帧收集的ACK/NACK样本的样本数量是否达到样本数量阈值，如果未达到，返回步骤2继续统计；否则，执行以下步骤：
步骤3.1、按用户ID对统计的ACK/NACK信息进行聚合，并计算每个用户对应的BLER；
步骤3.2、判断各个用户对应的BLER是否在对应的收敛区间(收敛区间为预设的区间范围，例如从0.08至0.12的范围)；如果在，把各个用户对应的ACK/NACK数据按ACK/NACK归属的子帧，投递到子帧对应的集合中，并执行步骤3.3；如果不在，丢弃此用户的数据；
步骤3.3、计算各子帧对应的BLER，统计各个子帧对应的BLER信息得到BLER集合(BLER_SET)：
BLER_SET＝{BLER i|i＝1,2,Λ N}，
其中，BLER i表示第i个子帧对应的误块率；
步骤4、计算集合BLER_SET的极差D：
D＝max{BLER i|i＝1,2,Λ N}-min{BLER i|i＝1,2,Λ N}；
其中，max{}函数表示取最大值，min{}函数表示取最小值。
步骤5、判断极差是否大于给定阈值TH(即上述实施例中的预设阈值)；如果D≥TH，确定进行子帧间差异化折算，转至步骤6；否则，复用当前的子帧折算因子sf_factor i,i＝1,2,Λ N，并转至步骤2，以及清空上次统计的历史数据；
步骤6、将BLER_SET集合划分为N个子集{set k|k＝1,2,Λ N}，其中set k对应N个子帧中的第i个子帧，k＝i；
步骤7、计算N个子集各自对应的平均BLER：BLERA k，其中，在本实施例中，
BLERA k＝BLER k,k＝1,2,Λ N
步骤8、计算各个子集对应的内环折算因子factor k：
factor k＝(BLER_tar-BLERA k)*step，
其中，BLER_tar为预设误块率目标值，step为预设值；
步骤9、计算各个子帧对应的内环折算因子：
sf_factor i＝(1-α)*sf'_factor i+α*factor k，
其中，α为滤波因子，sf'_factor为历史内环折算因子；
步骤10、输出各个子帧的内环折算因子sf_factor i,i＝1,2,Λ N，并应用各个子帧的内环折算因子，对对应子帧的内环值进行折算；
步骤11、转至步骤2，并清空上一轮的统计信息，进行下一轮学习。
在一个示例性实施例中，提供了一种内环值的调整方法，该方法可以应用在5G NR系统中，以上述实施例中的时域调度单元为5G NR系统的无线帧中的时隙为例，该方法包括以下步骤，其中，在时域调度单元为5G NR系统的无线帧中的时隙的情况下，每个时域调度单元对应的误块率实际为每个时隙对应的误时隙率：
步骤1、对无线帧中的各个时隙的内环折算因子进行初始化：
slot_factor i＝0,i＝1,2,Λ N
其中，slot_factor i表示第i时隙对应的折算因子，N表示一个无线帧中的时隙数目；
步骤2、统计除保守调度外，各个时隙对应的ACK/NACK数目，并按时隙对收集的ACK/NACK数据(或称为样本)进行维护，其中各个时隙单独计数；
步骤3、判断针对各个时隙收集的ACK/NACK样本的样本个数是否达到最小阈值(最小阈值可以是100000，即需要确保针对每个时隙收集的ACK/NACK样本的数量达到100000)，如果未达到，返回步骤2继续统计；否则，计算各时隙对应的BLER，并统计各个时隙对应的BLER信息得到BLER集合(BLER_SET)：
BLER_SET＝{BLER i|i＝1,2,Λ N}，
其中，BLER i表示第i个时隙对应的误时隙率；
步骤4、计算集合BLER_SET的极差D：
D＝max{BLER i|i＝1,2,Λ N}-min{BLER i|i＝1,2,Λ N}
步骤5、判断极差是否大于给定阈值TH；如果D≥TH，确定进行时隙间差异化折算，转至步骤6；否则，复用当前的时隙折算因子slot_factor i,i＝1,2,Λ N，并转至步骤2，以及清空上次统计的历史数据；
步骤6、将BLER_SET集合划分为M个子集{set k|k＝1,2,Λ M}，具体划分方法为：将时隙集合St初始化为N个时隙构成的集合；
步骤6.1、根据BLER_SET构造N×N的时隙间BLER差绝对值矩阵A；
步骤6.2、根据构造的差绝对值矩阵A，从N开始，降维搜索子矩阵，其中，子矩阵满足以下条件：每个元素都小于阈值TH,并且再增加一个维度就不再满足子矩阵中的每个元素均小于阈值TH的条件；其中，如果满足这个条件的同维子矩阵有多个，则选取子矩阵中的所有元素的平均值最小的那个子矩阵，将选择的子矩阵记为矩阵B；
步骤6.3、确定矩阵B的行对应的时隙编号，并构成子集set；
步骤6.4、从时隙集合St中删除子集set中包括的时隙，得到时隙集合St，根据时隙集合St对应的误时隙率集合构造矩阵C，将矩阵C赋值给A；
步骤6.5、重复步骤6.2到6.4的过程，最终得到{set k|k＝1,2,Λ M}；
步骤7、计算M个子集中的每个子集各自对应的平均BLER：
Figure PCTCN2021136759-appb-000003
其中，|set k|表示集合set k中的元素个数；
步骤8、计算各个子集对应的内环折算因子
factor k＝(BLER_tar-BLERA k)*step；
步骤9、计算各个时隙对应的内环折算因子，其中，对于时隙i，若i∈set k(即时隙i所归属的集合为i∈set k)，按照以下公式更新该子帧对应的内环折算因子：
slot_factor i＝(1-α)*slot'_factor i+α*factor k
其中，α为滤波因子，slot'_factor i为历史内环折算因子；
步骤10、输出各个时隙的内环折算因子slot_factor i,i＝1,2,Λ N，并应用各个子帧的因子在对应时隙上对用户进行内环折算。
步骤11、转步骤2，并清空上一轮的统计信息，进行下一轮学习。
在一个示例性实施例中，提供了一种内环值的调整方法，该方法可以应用在5G NR系统中，以上述实施例中的时域调度单元为5G NR系统的无线帧中的时隙为例，该方法包括以下步骤，其中，在时域调度单元为5G NR系统的无线帧中的时隙的情况下，每个时域调度单元对应的误块率实际为每个时隙对应的误时隙率：
步骤1、对无线帧中各个时隙的内环折算因子进行初始化：
slot_factor i＝0,i＝1,2,Λ N，
其中，slot_factor i表示第i时隙所对应的内环折算因子，N表示一个无线帧中的时隙数目；
步骤2、统计各个时隙的ACK/NACK数目，其中，保守调度中的ACK/NACK不计数，并且最低阶调度的NACK不计数、最高阶调度的ACK不计数，即统计的各个时隙的ACK/NACK中不包括保守调度中的ACK/NACK、最低阶调度的NACK以及最高阶调度的ACK；按时隙对收集的ACK/NACK数据进行维护，其中各个时隙的ACK/NACK单独计数；
步骤3、判断针对各个时隙收集的ACK/NACK样本的样本个数是否达到最小阈值(最小阈值可以是100000，即需要确保针对每个时隙收集的ACK/NACK样本的数量达到100000)，如果未达到，返回步骤2继续统计；否则，计算各时隙对应的BLER，并统计各个时隙对应得到BLER集合(BLER_SET)：
BLER_SET＝{BLER i|i＝1,2,Λ N}，
其中，BLER i表示第i个时隙对应的误时隙率；
步骤4、计算集合BLER_SET的极差D：
D＝max{BLER i|i＝1,2,Λ N}-min{BLER i|i＝1,2,Λ N}；
步骤5、判断极差是否大于给定阈值；如果D≥TH，确定进行时隙间差异化折算，转步骤6；否则，复用当前的时隙折算因子slot_factor i,i＝1,2,Λ N，并转至步骤2，以及清空上次统计的历史数据；
步骤6、将BLER_SET集合划分为M个子集{set k|k＝1,2,Λ M}，具体划分方法为：
步骤6.1、按照所有时隙对应的BLER从小到大的顺序，对时隙索引(例如时隙序号)进行排序；
步骤6.2、以所有时隙对应的BLER中的最小BLER为起点，最大BLER+0.1为终点，以预设误块率间隔为步长，将排序后的所有BLER划分若干个区间(最后一个区间的长度有可能小于固定的预设误块率间隔)；
步骤6.3、按步骤6.2划分得到的若干个区间，取左闭右开的若干个区间，把对应时隙按其对应的BLER所归属的区间进行分类，从而得到子集划分{set k|k＝1,2,Λ M}；
步骤7、计算M个子集中的每个子集各自对应的平均BLER：
Figure PCTCN2021136759-appb-000004
其中，|set k|表示集合set k中的元素个数；
步骤8、计算各个子集对应的内环折算因子：
factor k＝(BLER_tar-BLERA k)*step；
步骤9、计算各个时隙对应的内环折算因子：其中，对于时隙i，若i∈set k(即时隙i所归属的集合为i∈set k)，按照以下公式更新该子帧对应的内环折算因子：
slot_factor i＝(1-α)*slot'_factor i+α*factor k
其中，α为滤波因子，slot'_factor i为历史内环折算因子；
步骤10、输出各个时隙的内环折算因子slot_factor i,i＝1,2,Λ N，并应用各个子帧的因子在对应时隙上对用户进行内环折算；
步骤11、转步骤2，并清空上一轮的统计信息，进行下一轮学习。
需要说明的是，在上述实施例中，可以重复多次执行上述实施例中的内环值的调整方法。其中，在无线帧中的所有时域调度单元对应的误块率的极差小于预设阈值的情况下，确定每个时域调度单元的当前内环调整因子为每个时域调度单元各自的历史内环调整因子(可选地，保持每个时域调度单元的历史内环调整因子不变，或者，将所述每个时域调度单元的历史内环调整因子更新为所述每个时域调度单元各自的当前内环调整因子)，并清空之前收集到的ACK/NACK信息，以及重新执行收集每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率的步骤；在无线帧中的所有时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，在根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子之后，将所述每个时域调度单元的历史内环调整因子更新为所述每个时域调度单元各自的当前内环调整因子。
通过上述实施例，采用网络中的历史数据(即各个时域调度单元对应的ACK/NACK数据)，历史数据可以是用户历史上收敛的数据(单不包含保守调度的数据)，也可以是历史上的所有数据(不关系收敛还是不收敛，但把保守调度的数据排除在外)；统计各个时隙上的BLER； 计算所有时隙的BLER的极差；从而判断时隙间的BLER差异是不是大于一定的阈值了，如果小于，使用目前已有的折算模型(即各个时域调度单元对应的内环折算因子)；否则，根据历史数据学习各个时隙的折算量，对折算模型进行修正；其中，在对折算模型进行修正时，具体是，根据各个时隙的BLER分布情况，对所有时隙按一定规则进行分组(即子集)，例如：每个组内的极差要小于一个阈值，并且任意两个组合并后的极差大于阈值，任意两个组中的时域调度单元(例如时隙)无交集，所有组的并集是所有时隙所组成的集合；然后，统计各个子集的平均BLER；根据统计的BLER和设定的目标BLER以及对应的调整步长，计算各个时隙对应的内环折算因子；并将此内环折算因子与各个时域调度单元的历史内环折算因子进行滤波，得到针对各个时域调度单元的新的内环修正模型(即内环折算因子)，并应用此模型，继续收集历史数据，不断的对模型进行修正，最终实现用户的调度和用户所在时隙经历的信道尽可能的适配。
基于上述实施例，可以解决因不同时隙所包含的参考信号的差异或由于上下行切换等因素，导致的网络侧对同一个用户在不同上行时隙上的解调性能存在差异，或导致同一个用户对不同下行时隙的解调性能存在差异的技术问题；并且能够解决传统的AMC技术中，由于其是用户级别的，用户在不同时隙间的解调性能差异无法快速体现，因而会导致频谱效率受损的技术问题；以及能够解决时隙间AMC技术中，由于进一步离散的AMC学习的样本数据，导致小包调度用户很难收敛、新的AMC的收敛难的问题。
通过上述实施例，涉及了时隙间发送/接收存在差异的情况下，对于解调性能差异的学习与修正；其中，基于历史数据学习用户在不同时隙间的解调性能的差异，并根据学习结果，对不同时隙的内环值进行相关的折算处理，从而，使得用户在不同时隙间因解调性能的差异得到差异化调度，进而使得频谱效率得到提升。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到根据上述实施例的方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，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现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，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。
在本实施例中还提供了一种内环值的调整装置，该装置用于实现上述实施例及优选实施方式，已经进行过说明的不再赘述。如以下所使用的，术语“模块”可以实现预定功能的软件和/或硬件的组合。尽管以下实施例所描述的装置较佳地以软件来实现，但是硬件，或者软件和硬件的组合的实现也是可能并被构想的。
图10是根据本申请实施例的内环值的调整装置的结构框图，如图10所示，该装置包括：处理模块1002，其中，所述处理模块设置为循环执行以下步骤，其中，无线帧中的每个时域调度单元的历史内环调整因子被初始化为所述每个时域调度单元的初始内环调整因子：
根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，其中，所述无线帧包括N个时域调度单元，N为正整数；
根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子；
根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整。
通过本申请，循环执行以下步骤，其中，无线帧中的每个时域调度单元的历史内环调整因子被初始化为所述每个时域调度单元的初始内环调整因子：根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，其中，所述无线帧包括N个时域调度单元，N为正整数；根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子；根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整。由于根据每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，以及进而根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整，因此，可以解决时域调度单元间的解调差异无法通过用户级的AMC来快速补偿的技术问题，达到了对时隙间的解调差异的快速补偿，以及对用户在不同时域调度单元根据解调性能的差异记性差异化调度的技术效果。
在一个示例性实施例中，所述处理模块，还设置为：在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，根据以下公式确定时域调度单元i的当前内环调整因子T_factor i：T_factor i＝(1-α)*T'_factor i+α*factor i，其中，1≤i≤N,α为滤波因子，0≤α≤1，T'_factor i为所述时域调度单元i的历史内环调整因子，factor i＝(BLER_tar-BLER i)*step，BLER_tar为预设误块率目标值，BLER i为所述时域调度单元i对应的误块率，step为调整步长、且step大于0。
在一个示例性实施例中，所述处理模块，还设置为：在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，根据所述每个时域调度单元对应的误块率，将所述N个时域调度单元划分至M个子集，其中，所述M个子集中的每个子集的时域调度单元对应的误块率的极差均小于所述预设阈值，M中为正整数，M小于N；根据所述每个子集的时域调度单元对应的误块率，确定所述每个子集对应的内环调整因子；根据所述每个子集对应的内环调整因子和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子。
在一个示例性实施例中，所述根据所述每个子集的时域调度单元对应的误块率，确定所述每个子集对应的内环调整因子，包括：根据以下公式确定所述每个子集对应的内环调整因子factor k：factor k＝(BLER_tar-BLERA k)*step，其中，1≤k≤M,BLER_tar为预设误块率目标值，BLERA k为子集k中的时域调度单元对应的误块率的平均值，step为调整步长、且step大于0。
在一个示例性实施例中，所述处理模块，还设置为：根据以下公式确定时域调度单元i的当前内环调整因子T_factor i：T_factor i＝(1-α)*T'_factor i+α*factor k，其中，所述时域调度单元i属于所述M个子集中的第k个子集，1≤i≤N，α为滤波因子，0≤α≤1，T'_factor i为所述时域调度单元i的历史内环调整因子。
在一个示例性实施例中，所述处理模块，还设置为：根据所述N个时域调度单元对应的误块率集合，构造N×N的差绝对值矩阵A，其中，所述误块率集合中的每个误块率与所述N 个时域调度单元中的每个时域调度单元一一对应；重复执行以下步骤，直至数值c等于1，其中，当前矩阵被初始化为所述矩阵A，当前时域调度单元集合被初始为所述N个时域调度单元组成的集合，所述数值c被初始为所述N：在所述当前矩阵中搜索c×c的目标子矩阵，其中，所述目标子矩阵中的每个元素均小于所述预设阈值；在从所述当前矩阵中搜索到所述目标子矩阵的情况下，将所述目标子矩阵中的每一行对应的时域调度单元组成的集合确定为一个子集；从所述当前时域调度单元集合中删除所述一个子集，并将删除所述一个子集后的当前时域调度单元集合确定为所述当前时域调度单元集合；将根据所述当前时域调度单元集合对应的误块率集合构造的(N-c)×(N-c)的差绝对值矩阵，确定为所述当前矩阵，其中，所述当前时域调度单元集合对应的误块率集合中的每个误块率与所述当前时域调度单元集合中的每个时域调度单元一一对应；在从所述当前矩阵中未搜索到所述目标子矩阵的情况下，将所述数值c减1后得到的差值确定为所述数值c；其中，在所述数值c等于1的情况下，将所述当前矩阵中的每一行对应的时域调度单元确定为一个子集。
在一个示例性实施例中，所述处理模块，还设置为：将所述N个时域调度单元对应的N个误块率划分至M个连续的区间，其中，所述M个区间中的每个区间包括至少一个误块率，所述每个区间内的误块率的极差小于所述预设阈值；将所述N个时域调度单元中，与所述M个区间中的区间k包括的误块率对应的时域调度单元划分至子集k，共划分得到M个子集，其中，1≤k≤M。
在一个示例性实施例中，所述处理模块，还设置为：在所述N个时域调度单元对应的误块率的极差小于预设阈值的情况下，确定所述每个时域调度单元的当前内环调整因子为所述每个时域调度单元的历史内环调整因子。
在一个示例性实施例中，所述处理模块，还设置为：在所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子之后，在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，将所述每个时域调度单元的历史内环调整因子更新为所述每个时域调度单元各自的当前内环调整因子。
在一个示例性实施例中，所述处理模块，还设置为：确定用户在时域调度单元i对应的内环值d i与所述时域调度单元i的当前内环调整因子T_factor i的和；将所述内环值d i与所述当前内环调整因子T_factor i的和，确定为所述用户在所述时域调度单元i调整后的内环值。
在一个示例性实施例中，在所述无线帧为4G通信系统中的无线帧的情况下，所述无线帧中的每个时域调度单元与所述无线帧中的每个子帧一一对应，其中，N为所述无线帧中的子帧数目；在所述无线帧为5G通信系统中的无线帧的情况下，所述无线帧中的每个时域调度单元与所述无线帧中的每个时隙一一对应，其中，N为所述无线帧中的时隙数目。
需要说明的是，上述各个模块是可以通过软件或硬件来实现的，对于后者，可以通过以下方式实现，但不限于此：上述模块均位于同一处理器中；或者，上述各个模块以任意组合的形式分别位于不同的处理器中。
本申请的实施例还提供了一种计算机可读存储介质，该计算机可读存储介质中存储有计 算机程序，其中，该计算机程序被设置为运行时执行上述任一项方法实施例中的步骤。
在一个示例性实施例中，上述计算机可读存储介质可以包括但不限于：U盘、只读存储器(Read-Only Memory，简称为ROM)、随机存取存储器(Random Access Memory，简称为RAM)、移动硬盘、磁碟或者光盘等各种可以存储计算机程序的介质。
本申请的实施例还提供了一种电子装置，包括存储器和处理器，该存储器中存储有计算机程序，该处理器被设置为运行计算机程序以执行上述任一项方法实施例中的步骤。
在一个示例性实施例中，上述电子装置还可以包括传输设备以及输入输出设备，其中，该传输设备和上述处理器连接，该输入输出设备和上述处理器连接。
本实施例中的具体示例可以参考上述实施例及示例性实施方式中所描述的示例，本实施例在此不再赘述。
显然，本领域的技术人员应该明白，上述的本申请的各模块或各步骤可以用通用的计算装置来实现，它们可以集中在单个的计算装置上，或者分布在多个计算装置所组成的网络上，它们可以用计算装置可执行的程序代码来实现，从而，可以将它们存储在存储装置中由计算装置来执行，并且在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤，或者将它们分别制作成各个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样，本申请不限制于任何特定的硬件和软件结合。
以上所述仅为本申请的优选实施例而已，并不用于限制本申请，对于本领域的技术人员来说，本申请可以有各种更改和变化。凡在本申请的原则之内，所作的任何修改、等同替换、改进等，均应包含在本申请的保护范围之内。

Claims (14)

  1. 一种内环值的调整方法，包括：循环执行以下步骤，其中，无线帧中的每个时域调度单元的历史内环调整因子被初始化为所述每个时域调度单元的初始内环调整因子：
    根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，其中，所述无线帧包括N个时域调度单元，N为正整数；
    根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子；
    根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整。
  2. 根据权利要求1所述的方法，其中，所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，包括：
    在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，根据以下公式确定时域调度单元i的当前内环调整因子T_factor i：
    T_factor i＝(1-α)*T'_factor i+α*factor i，
    其中，1≤i≤N,α为滤波因子，0≤α≤1，T'_factor i为所述时域调度单元i的历史内环调整因子，factor i＝(BLER_tar-BLER i)*step，BLER_tar为预设误块率目标值，BLER i为所述时域调度单元i对应的误块率，step为调整步长、且step大于0。
  3. 根据权利要求1所述的方法，其中，所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，包括：
    在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，根据所述每个时域调度单元对应的误块率，将所述N个时域调度单元划分至M个子集，其中，所述M个子集中的每个子集的时域调度单元对应的误块率的极差均小于所述预设阈值，M中为正整数，M小于N；
    根据所述每个子集的时域调度单元对应的误块率，确定所述每个子集对应的内环调整因子；
    根据所述每个子集对应的内环调整因子和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子。
  4. 根据权利要求3所述的方法，其中，所述根据所述每个子集的时域调度单元对应的误块率，确定所述每个子集对应的内环调整因子，包括：
    根据以下公式确定所述每个子集对应的内环调整因子factor k：
    factor k＝(BLER_tar-BLERA k)*step，
    其中，1≤k≤M,BLER_tar为预设误块率目标值，BLERA k为子集k中的时域调度单元对应的误块率的平均值，step为调整步长、且step大于0。
  5. 根据权利要求4所述的方法，其中，所述根据所述每个子集对应的内环调整因子和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，包括：
    根据以下公式确定时域调度单元i的当前内环调整因子T_factor i：
    T_factor i＝(1-α)*T'_factor i+α*factor k，
    其中，所述时域调度单元i属于所述M个子集中的第k个子集，1≤i≤N，α为滤波因子，0≤α≤1，T'_factor i为所述时域调度单元i的历史内环调整因子。
  6. 根据权利要求3中所述的方法，其中，所述根据所述每个时域调度单元对应的误块率，将所述N个时域调度单元划分至M个子集，包括：
    根据所述N个时域调度单元对应的误块率集合，构造N×N的差绝对值矩阵A，其中，所述误块率集合中的每个误块率与所述N个时域调度单元中的每个时域调度单元一一对应；
    重复执行以下步骤，直至数值c等于1，其中，当前矩阵被初始化为所述矩阵A，当前时域调度单元集合被初始为所述N个时域调度单元组成的集合，所述数值c被初始为所述N：
    在所述当前矩阵中搜索c×c的目标子矩阵，其中，所述目标子矩阵中的每个元素均小于所述预设阈值；
    在从所述当前矩阵中搜索到所述目标子矩阵的情况下，将所述目标子矩阵中的每一行对应的时域调度单元组成的集合确定为一个子集；
    从所述当前时域调度单元集合中删除所述一个子集，并将删除所述一个子集后的当前时域调度单元集合确定为所述当前时域调度单元集合；
    将根据所述当前时域调度单元集合对应的误块率集合构造的(N-c)×(N-c)的差绝对值矩阵，确定为所述当前矩阵，其中，所述当前时域调度单元集合对应的误块率集合中的每个误块率与所述当前时域调度单元集合中的每个时域调度单元一一对应；
    在从所述当前矩阵中未搜索到所述目标子矩阵的情况下，将所述数值c减1后得到的差值确定为所述数值c；
    其中，在所述数值c等于1的情况下，将所述当前矩阵中的每一行对应的时域调度单元确定为一个子集。
  7. 根据权利要求3中所述的方法，其中，所述根据所述每个时域调度单元对应的误块率，将所述N个时域调度单元划分至M个子集，包括：
    将所述N个时域调度单元对应的N个误块率划分至M个连续的区间，其中，所述M个区间中的每个区间包括至少一个误块率，所述每个区间内的误块率的极差小于所述预设阈值；
    将所述N个时域调度单元中，与所述M个区间中的区间k包括的误块率对应的时域调度 单元划分至子集k，共划分得到M个子集，其中，1≤k≤M。
  8. 根据权利要求1所述的方法，其中，所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子，包括：
    在所述N个时域调度单元对应的误块率的极差小于预设阈值的情况下，确定所述每个时域调度单元的当前内环调整因子为所述每个时域调度单元的历史内环调整因子。
  9. 根据权利要求1所述的方法，其中，在所述根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子之后，所述方法还包括：
    在所述N个时域调度单元对应的误块率的极差大于或等于预设阈值的情况下，将所述每个时域调度单元的历史内环调整因子更新为所述每个时域调度单元各自的当前内环调整因子。
  10. 根据权利要求1所述的方法，其中，所述根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整，包括：
    确定用户在时域调度单元i对应的内环值d i与所述时域调度单元i的当前内环调整因子T_factor i的和；
    将所述内环值d i与所述当前内环调整因子T_factor i的和，确定为所述用户在所述时域调度单元i调整后的内环值。
  11. 根据权利要求1至10中任一项所述的方法，其中，在所述无线帧为4G通信系统中的无线帧的情况下，所述无线帧中的每个时域调度单元与所述无线帧中的每个子帧一一对应，其中，N为所述无线帧中的子帧数目；在所述无线帧为5G通信系统中的无线帧的情况下，所述无线帧中的每个时域调度单元与所述无线帧中的每个时隙一一对应，其中，N为所述无线帧中的时隙数目。
  12. 一种内环值的调整装置，包括：处理模块，其中，所述处理模块设置为循环执行以下步骤，其中，无线帧中的每个时域调度单元的历史内环调整因子被初始化为所述每个时域调度单元的初始内环调整因子：
    根据所述每个时域调度单元对应的ACK/NACK信息，确定所述每个时域调度单元对应的误块率，其中，所述无线帧包括N个时域调度单元，N为正整数；
    根据所述每个时域调度单元对应的误块率和所述每个时域调度单元的历史内环调整因子，确定所述每个时域调度单元的当前内环调整因子；
    根据所述每个时域调度单元的当前内环调整因子，对所述每个时域调度单元对应的内环值进行调整。
  13. 一种计算机可读存储介质，所述计算机可读存储介质中存储有计算机程序，其中，所述计算机程序被设置为运行时执行所述权利要求1至11任一项中所述的方法。
  14. 一种电子装置，包括存储器和处理器，所述存储器中存储有计算机程序，所述处理 器被设置为运行所述计算机程序以执行所述权利要求1至11任一项中所述的方法。
PCT/CN2021/136759 2020-12-09 2021-12-09 内环值的调整方法和装置、存储介质及电子装置 WO2022121979A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US18/036,925 US20230422220A1 (en) 2020-12-09 2021-12-09 Method and Apparatus for Adjusting Inner Loop Value, Storage Medium and Electronic Device
EP21902684.6A EP4262312A4 (en) 2020-12-09 2021-12-09 INNER LOOP VALUE ADJUSTMENT METHOD AND DEVICE, RECORDING MEDIUM, AND ELECTRONIC DEVICE

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011433224.3 2020-12-09
CN202011433224.3A CN114630443A (zh) 2020-12-09 2020-12-09 内环值的调整方法和装置、存储介质及电子装置

Publications (1)

Publication Number Publication Date
WO2022121979A1 true WO2022121979A1 (zh) 2022-06-16

Family

ID=81896461

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/136759 WO2022121979A1 (zh) 2020-12-09 2021-12-09 内环值的调整方法和装置、存储介质及电子装置

Country Status (4)

Country Link
US (1) US20230422220A1 (zh)
EP (1) EP4262312A4 (zh)
CN (1) CN114630443A (zh)
WO (1) WO2022121979A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103546244A (zh) * 2013-11-01 2014-01-29 武汉邮电科学研究院 一种自适应调制编码方法及装置
CN106656411A (zh) * 2015-11-04 2017-05-10 北京信威通信技术股份有限公司 Harq绑定模式下调度mcs的方法及系统
CN111355557A (zh) * 2018-12-21 2020-06-30 大唐移动通信设备有限公司 一种调整调制编码方式mcs的方法及装置
CN111447041A (zh) * 2019-01-16 2020-07-24 北京小米松果电子有限公司 调制与编码策略的控制方法、装置、存储介质和电子设备

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9584288B2 (en) * 2012-03-28 2017-02-28 Nec Corporation Communication channel quality estimating method, wireless communications system, base station, and program
US9374191B2 (en) * 2012-05-17 2016-06-21 Apple Inc. Outer loop link adaptation for device resumption
US9713189B2 (en) * 2012-11-07 2017-07-18 Telefonaktiebolaget Lm Ericsson (Publ) Multiple outer loop link adaptation
US9712306B2 (en) * 2013-01-21 2017-07-18 Apple Inc. Adaptive link adaptation for wireless communications

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103546244A (zh) * 2013-11-01 2014-01-29 武汉邮电科学研究院 一种自适应调制编码方法及装置
CN106656411A (zh) * 2015-11-04 2017-05-10 北京信威通信技术股份有限公司 Harq绑定模式下调度mcs的方法及系统
CN111355557A (zh) * 2018-12-21 2020-06-30 大唐移动通信设备有限公司 一种调整调制编码方式mcs的方法及装置
CN111447041A (zh) * 2019-01-16 2020-07-24 北京小米松果电子有限公司 调制与编码策略的控制方法、装置、存储介质和电子设备

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4262312A4 *

Also Published As

Publication number Publication date
US20230422220A1 (en) 2023-12-28
EP4262312A4 (en) 2024-06-12
EP4262312A1 (en) 2023-10-18
CN114630443A (zh) 2022-06-14

Similar Documents

Publication Publication Date Title
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
US11888703B1 (en) Machine learning algorithms for quality of service assurance in network traffic
WO2022121979A1 (zh) 内环值的调整方法和装置、存储介质及电子装置
CN115374853A (zh) 基于T-Step聚合算法的异步联邦学习方法及系统
Zhang et al. Communication-efficient federated continual learning for distributed learning system with Non-IID data
CN116050540B (zh) 一种基于联合双维度用户调度的自适应联邦边缘学习方法
EP4054134A1 (en) Congestion control method and releated device
CN106027288A (zh) 一种配电线路信息监测业务通信流量预测方法
CN112202847B (zh) 一种基于移动边缘计算的服务器资源配置方法
CN114357676A (zh) 一种针对层次化模型训练框架的聚合频率控制方法
CN113518007A (zh) 一种基于联邦学习的多物联网设备异构模型高效互学习方法
CN114781545B (zh) 一种联邦学习方法及系统
CN110913239B (zh) 一种精细化的移动边缘计算的视频缓存更新方法
CN117499297B (zh) 数据包传输路径的筛选方法及装置
CN113676357B (zh) 面向电力物联网中边缘数据处理的决策方法及其应用
CN114118748A (zh) 一种服务质量预测方法、装置、电子设备及存储介质
CN116362328A (zh) 一种基于公平性特征表示的联邦学习异构模型聚合方法
WO2023051318A1 (zh) 模型训练方法、无线资源调度方法及其装置及电子设备
CN111010695A (zh) 基于信道空闲时长预测的信道分配方法
CN114997400A (zh) 一种神经网络加速推理方法
Tian et al. Hierarchical federated learning with adaptive clustering on non-IID data
CN117479306B (zh) 一种数字孪生的数据处理方法及系统
CN115225478B (zh) 网络切片动态配置方法、装置、电子设备及存储介质
WO2024067100A1 (zh) 目标误块率控制方法及装置
WO2024124975A1 (zh) 网络质量评估方法、装置、设备、存储介质及程序产品

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: 21902684

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 18036925

Country of ref document: US

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2021902684

Country of ref document: EP

Effective date: 20230710