A method and apparatus for scheduling RI in uplink SU-MIMO

Abstract

This invention relates to the field of wireless communication technology, specifically to a method and apparatus for scheduling RI in uplink SU-MIMO, comprising: a trial condition judgment step: determining whether RI_i corresponding to UE_k satisfies the trial condition to RI_j; if yes, updating the trial time of RI_i to RI_j, maintaining the scheduling of RI_i corresponding to UE_K for a preset number of times, obtaining SE4UE_kRI_i, and proceeding to the next step; if no, maintaining the scheduling of RI_i, waiting for the trial time, and repeating this step; a trial pre-scheduling step: updating the initial MCS corresponding to UE_k under RI_j, maintaining the scheduling of RI_j corresponding to UE_K for a preset number of times, obtaining the equivalent SE4UE_kRI_j, and proceeding to the next step; a trial result judgment step: determining whether SE4UE_kRI_j is greater than or equal to SE4UE_kRI_i; if yes, maintaining the scheduling of RI_j, waiting for the trial time, and re-executing the trial condition judgment step; if no, executing a trial penalty step to increase the trial time of RI_i to RI_j. This solution can quickly select the scheduling RI and scheduling MCS that match the current channel state to improve the SE of the user terminal.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, specifically to a method and apparatus for scheduling RI in uplink SU-MIMO. Background Technology

[0002] SU-MIMO (Single User Multiple Input Multiple Output) is an important technology in wireless systems. In wireless communication systems, if the UE (User Equipment) selects a scheduling mode with a lower RI (Rank Indicator), although the UE's performance is likely to be relatively stable, the transmission rate will be limited. Similarly, if the UE selects a modulation scheme with a higher RI, although it has a higher transmission rate, the UE's performance stability will deteriorate, and a large number of bit errors may occur. The scheduling MCS (Modulation and Coding Scheme) cannot be improved, ultimately leading to a decrease in the UE's transmission efficiency.

[0003] In wireless communication systems, the actual wireless channel changes rapidly due to the influence of many external factors. The existing scheduling mode selection cannot quickly select the scheduling RI and scheduling MCS that can match the current channel state, which affects the UE's SE (Spectral Efficiency). Summary of the Invention

[0004] One of the objectives of this invention is to provide a method for scheduling RI in uplink SU-MIMO, which can quickly select the scheduling RI and scheduling MCS that match the current channel state for the UE, thereby improving the UE's SE.

[0005] The basic solution provided by this invention is: a method for scheduling RI in uplink SU-MIMO, comprising:

[0006] Trial condition judgment step: Determine whether the current UE_k corresponding to RI_i satisfies the trial condition from RI_i to RI_j. If yes, update the trial time T4UE_kRIi_jTest for the RI_i of UE_k to RI_j, UE_k maintains RI_i for a preset number of scheduling attempts, and obtains the equivalent SE corresponding to the scheduled RI_i = SE4UE_kRI_i, and executes the trial pre-scheduling step; if no, UE_k maintains the RI_i scheduling, waits for the trial time, and then executes the trial condition judgment step again.

[0007] Probing pre-scheduling step: Update the initial MCS corresponding to UE_k under RI_j, perform scheduling for a preset number of times while maintaining RI_j, and obtain the equivalent SE = SE4UE_kRI_j corresponding to the scheduled RI_j, and execute the probing result judgment step;

[0008] Probing result judgment step: Judge whether SE4UE_kRI_j is greater than or equal to SE4UE_kRI_i (where SE4RI_i combines the redundancy coefficient). If so, UE_k maintains the RI_j scheduling and waits for the probing time and then executes the probing condition judgment step again; if not, then execute the probing penalty step;

[0009] Probing penalty step: Increase the probing time from RI_i corresponding to UE_k to RI_j, and wait for the probing time and then execute the probing condition judgment step again.

[0010] Furthermore, the probing condition judgment step includes:

[0011] S1. Judge whether the current RI_i corresponding to UE_k satisfies the probing condition from RI_i to RI_j. If so, execute S2; if not, maintain the RI_i scheduling, wait for the probing time and then execute S1 again;

[0012] S2. UE_k maintains the RI_i scheduling for N4SE times, counts the MCS, SE, and responses of each scheduling, generates the statistical result of RI_i, and then calculates the equivalent SE = SE4UE_kRI_i of UE_k corresponding to RI_i for N4SE times of scheduling according to the statistical result of RI_i.

[0013] Furthermore, S1 includes:

[0014] S101. Judge whether the latest new transmission scheduling MCS = MCS4UE_k_RI_i corresponding to UE_k under RI_i satisfies the preset condition; the preset condition is: MCS4UE_k_RI_i > MCS4RIi_j1 or MCS4UE_k_RI_i < MCS4RIi_j2, where MCS4RIi_j is the MCS threshold for probing from RI_i to RI_j, j1 = i + 1, j2 = i - 1.

[0015] If so, execute S102; if not, maintain the RI_i scheduling, wait for the probing time and then execute S101 again;

[0016] S102. Judge whether the block error rate Bler value of RI_i converges. If so, execute S103; if not, maintain the RI_i scheduling, wait for the probing time and then execute S101 again;

[0017] S103. Determine whether the time limit for retrying is met. If yes, execute S2; otherwise, maintain the RI_i schedule and wait for the retry time before executing S101 again.

[0018] Further, S102 includes:

[0019] Maintain the block error rate Bler4UE_k_RI_i corresponding to UE_k, and record the latest N4Bler scheduling response for RI_i corresponding to UE_k, where the response includes: ACK and NACK; N4Bler is the scheduling number threshold used to calculate Bler;

[0020] Calculate Bler4UE_k_RI_i based on the response: Bler4UE_k_RI_i = total number of NACKs / (total number of ACKs + total number of NACKs);

[0021] Calculate the scheduling duration Tk_i of UE_k corresponding to RI_i in the latest N4Bler cycle;

[0022] Determine if Tk_i is less than TValidity4Bler. If it is, the calculated Bler4UE_k_RI_i is considered valid. If not, the calculated Bler4UE_k_RI_i is considered invalid. TValidity4Bler is the validity duration of Bler.

[0023] Determine whether the valid Bler4RI_iUE_k has converged. If yes, execute S103; otherwise, the trial condition is not met, maintain the RI_i schedule, wait for the trial time, and execute S1 again.

[0024] Where Bler4UE_k_RI_i converges as follows:

[0025] TargetBler-DeltaBler<=Bler4UE_k_RI_i= <TargetBler+DeltaBler;

[0026] Where TargetBler is the target Bler corresponding to the uplink AMC, and DeltaBler is the Bler redundancy used for convergence judgment.

[0027] Furthermore, S103 includes:

[0028] S1031. Determine whether the system time corresponding to the current UE_k scheduling RI_i meets the time limit for retrying. If yes, execute S1032; otherwise, maintain the RI_i scheduling and wait for the retry time before executing S101 again.

[0029] The time limit is as follows:

[0030] Time-T4UE_kRIi_jTest>=Time4ReAttempt0(1+N4UE_kRIi_jTestFail*TimePunishCoe4Attempt);

[0031] In the formula, Time is the system time corresponding to the current UE_k scheduling RI_i;

[0032] T4UE_kRIi_jTest is the time when UE_k last tried to probe RI_j with RI_i;

[0033] TimePunishCoe4Attempt is the penalty coefficient for the trial time interval;

[0034] N4UE_kRIi_jTestFail represents the number of times UE_k failed to probe RI_j from RI_i.

[0035] Time4ReAttempt0 is the initial time interval threshold for retrying;

[0036] S1032. Perform a trial time update, update the trial time T4UE_kRIi_jTest corresponding to RI_i of UE_k to RI_j: T4UE_kRIi_jTest = current system time, and execute S2.

[0037] Furthermore, S2 includes:

[0038] UE_k maintains RI_i scheduling N4SE times;

[0039] Statistically record the MCS for each scheduling, and based on the correspondence between MCS and SE, record the SE for each scheduling of UE_k under RI_i, SEFront_index4UE_k_RI_i, index=1、2、...N4SE, and record the response result of each scheduling of UE_k under RI_i, ANFront_index4UE_k_RI_i, index=1...N4SE;

[0040] Multiply the SEFront_index4UE_k_RI_i corresponding to the N4SE scheduling of RI_i with the ACK response by RI_i and then add them together to obtain the equivalent SE of UE_k under RI_i = SE4UE_k_RI_i.

[0041]

[0042] Furthermore, the trial pre-scheduling step includes:

[0043] S3. Update the initial MCS under the RI_j scheduling corresponding to UE_k, including:

[0044] Get the scheduling MCS corresponding to the latest scheduling of RI_i, which is LastMCS4UE_k_RI_i;

[0045] The LastSE4UE_k_RI_i corresponding to LastMCS4UE_k_RI_i is obtained by looking up the table;

[0046] Calculate the minimum SE corresponding to RI_j = MinSE4UE_k_RI_j:

[0047] MinSE4UE_k_RI_j=LastSE4UE_k_RI_i*RI_i / RI_j;

[0048] The minimum MCS that is not less than MinSE4UE_k_RI_j is obtained by reverse lookup table, and is used as the initial scheduling MCS of UE_k under RI_j = InitialMCS4UE_kRI_j;

[0049] S4, UE_k maintains RI_j and schedules N4SE times, calculates the MCS, SE, and response for each scheduling, and generates statistical results for RI_j, including:

[0050] UE_k maintains RI_j scheduling N4SE times;

[0051] Statistically record the MCS for each scheduling. Based on the correspondence between MCS and SE, record the SE for each scheduling of UE_k under RI_j: SEBack_index4UE_k_RI_j, index = 1, 2, ... N4SE. Record the response result for each scheduling of UE_k under RI_j: ANBack_index4UE_k_RI_j, index = 1... N4SE.

[0052] S5. Based on the statistical results of UE_k corresponding to RI_j, calculate the equivalent SE = SE4UE_k_RI_j corresponding to N4SE scheduling operations of UE_k under RI_j, including:

[0053] In the N4SE scheduling of RI_j corresponding to UE_k, the SEBack_index4UE_k_RI_j corresponding to the ACK scheduling is multiplied by RI_j and then added to obtain the equivalent SE of UE_k under RI_j = SE4UE_k_RI_j.

[0054]

[0055] Furthermore, the step of determining the trial result includes:

[0056] S6. Compare the equivalent SE before and after the trial. If SE4UE_k_RI_j >= SE4UE_k_RI_i * (1 + Thr4SE), then for the next scheduling of UE_k, maintain the RI_j scheduling, and at the same time update the trial time T4UE_kRIj_iTest from RI_j to RI_i for UE_k, T4UE_kRIj_iTest = the current system scheduling time, update the trial failure counter N4UE_kRIi_jTestFail from RI_i to RI_j for UE_k to 0, and execute S1;

[0057] If SE4UE_k_RI_j < SE4UE_k_RI_i * (1 + Thr4SE), then for the next scheduling of UE_k, resume the RI_i scheduling and execute the trial penalty step.

[0058] Furthermore, the trial penalty step includes:

[0059] S7. Update the trial failure counter N4UE_kRIi_jTestFail from RI_i to RI_j for UE_k, N4UE_kRIi_jTestFail = N4UE_kRIi_jTestFail + 1, and execute S1.

[0060] Beneficial effects of this solution: In this solution, through the trial condition judgment step, it is judged whether the current RI_i meets the condition for RI_j scheduling. If it does, it will not directly complete the RI_j switch, but perform the trial pre-scheduling step, and respectively obtain the equivalent SE after the actual scheduling of RI_i and RI_j, and then enter the trial result judgment step to judge whether the equivalent SE (SE4UE_k_RI_i) corresponding to RI_i is greater than or equal to the equivalent SE (SE4UE_k_RI_j) corresponding to RI_j and consider the redundancy threshold. If so, it means that after updating the scheduling to RI_j, the SE of the UE has increased and is stable, and the RI_j scheduling can be maintained, and the trial time from RI_j to RI_i is updated to prevent frequent RI switching trials; if not, it means that after updating the scheduling to RI_j, the SE has not been stably improved, so the RI_i scheduling is resumed, and through the trial penalty step, the difficulty of re-trialing from RI_i to RI_j is increased; and regardless of the judgment result, the trial condition judgment step will be executed again to quickly select the scheduling RI and scheduling MCS that can match the current channel state and maintain a stable and efficient SE in the real wireless channel. In summary, this solution can quickly select the scheduling RI and scheduling MCSI that match the current channel state to improve the SE of the UE.

[0061] The second objective of this invention is to provide a device for scheduling RI in uplink SU-MIMO, which can quickly select the scheduling RI and scheduling MCSI that match the current channel state to improve the UE's SE.

[0062] This invention provides a second basic solution: an apparatus for scheduling RIs in uplink SU-MIMO, employing the aforementioned method for scheduling RIs in uplink SU-MIMO. This solution can quickly select the scheduling RI and scheduling MCSI that match the current channel state, thereby improving the UE's SE (Search Effectiveness). Attached Figure Description

[0063] Figure 1 This is a flowchart illustrating an embodiment of a method for scheduling RI in uplink SU-MIMO according to the present invention. Detailed Implementation

[0064] The following detailed description illustrates the specific implementation method:

[0065] Note: AMC: Adaptive Modulation and Coding.

[0066] LTE: Long Term Evolution;

[0067] NR: New Radio;

[0068] MCS: Modulation and coding scheme;

[0069] CQI: Channel Quality Indicator;

[0070] ACK: Acknowledgement, confirmation;

[0071] NACK: Negative Acknowledgement;

[0072] SE: Spectral efficiency;

[0073] BLER: Block Error Rate;

[0074] UE: User Equipment;

[0075] SU-MIMO: Single User Multiple Input Multiple Output;

[0076] PMI: Precoding Matrix Indicator;

[0077] N4SE: Used to calculate the number of times an SE is scheduled;

[0078] Thr4SE: Redundancy threshold used to compare SE;

[0079] Time4ReAttempt0: The threshold for the initial time interval for retrying;

[0080] TimePunishCoe4Attempt: Penalty coefficient for trial interval; MCS4RI1_2: MCS threshold for RI1 to trial RI2;

[0081] MCS4RI2_1: The MCS threshold that RI2 probes towards RI1;

[0082] MCS4RI2_3: The MCS threshold for RI2 to probe RI3;

[0083] MCS4RI3_2: The MCS threshold that RI3 tests against RI2;

[0084] MCS4RI3_4: The MCS threshold for RI3 to probe RI4;

[0085] MCS4RI4_3: RI4: The MCS threshold that probes towards RI3;

[0086] TargetBler: The target Bler corresponding to the uplink AMC;

[0087] DeltaBler: Bler redundancy used for convergence assessment;

[0088] N4Bler: A threshold for calculating the number of times Bler can be scheduled;

[0089] TValidity4Bler: Bler validity duration.

[0090] The basic implementation examples are as follows: Figure 1 As shown: A method for scheduling RI in uplink SU-MIMO, comprising:

[0091] Trial condition judgment step: Determine whether the current UE_k corresponding to RI_i satisfies the trial condition from RI_i to RI_j. If yes, update the trial time T4UE_kRIi_jTest for the RI_i of UE_k to RI_j, UE_k maintains RI_i for a preset number of scheduling attempts, and obtains the equivalent SE corresponding to the scheduled RI_i = SE4UE_kRI_i, and executes the trial pre-scheduling step; if no, UE_k maintains the RI_i scheduling, waits for the trial time, and then executes the trial condition judgment step again.

[0092] Trial pre-scheduling steps: Update the initial MCS corresponding to UE_k under RI_j, keep RI_j scheduled for a preset number of times, and obtain the equivalent SE corresponding to scheduled RI_j = SE4UE_kRI_j, and execute the trial result judgment step;

[0093] Trial result judgment step: Determine whether SE4UE_kRI_j is greater than or equal to SE4UE_kRI_i (where SE4RI_i is combined with the redundancy coefficient). If yes, UE_k maintains RI_j scheduling and waits for the trial time before executing the trial condition judgment step again; if no, execute the trial penalty step.

[0094] Trial penalty step: Increase the trial time from RI_i to RI_j corresponding to UE_k, and wait for the trial time before executing the trial condition judgment step again.

[0095] Specifically, the trial-and-error condition judgment steps include:

[0096] S1. Determine whether the current UE_k corresponds to RI_i and satisfies the probing condition from RI_i to RI_j. If yes, execute S2; otherwise, maintain the RI_i scheduling and wait for the probing time before executing S1 again.

[0097] Specifically, in S101, determine whether the latest new transmission schedule MCS = MCS4UE_k_RI_i under RI_i corresponding to UE_k satisfies the preset condition (MCS4RIi_j). If yes, proceed to S102; otherwise, maintain the RI_i schedule and wait for the trial period before proceeding to S101 again. The preset condition is:

[0098] MCS4UE_k_RI_i>MCS4RIi_j1 or MCS4UE_k_RI_i <MCS4RIi_j2;

[0099] Where MCS4RIi_j is the MCS threshold for RI_i to probe RI_j, j1=i+1,j2=i-1;

[0100] S102. Determine whether the block error rate Bler value of RI_i has converged. If yes, execute S103; otherwise, maintain the scheduling of RI_i, wait for the trial time, and then execute S101 again.

[0101] Specifically, maintain the block error rate Bler4UE_k_RI_i corresponding to UE_k, and record the latest N4Bler scheduling response for RI_i corresponding to UE_k, where the response includes: ACK and NACK; N4Bler is the scheduling number threshold used to calculate Bler;

[0102] Calculate Bler4UE_k_RI_i based on the response: Bler4UE_k_RI_i = total number of NACKs / (total number of ACKs + total number of NACKs);

[0103] Calculate the scheduling duration Tk_i of UE_k corresponding to RI_i in the latest N4Bler cycle;

[0104] Determine if Tk_i is less than TValidity4Bler. If it is, the calculated Bler4UE_k_RI_i is considered valid. If not, the calculated Bler4UE_k_RI_i is considered invalid. TValidity4Bler is the validity duration of Bler.

[0105] Determine whether the valid Bler4RI_iUE_k has converged. If yes, execute S103; otherwise, the trial condition is not met, maintain the RI_i schedule, wait for the trial time, and execute S1 again.

[0106] Where Bler4UE_k_RI_i converges as follows:

[0107] TargetBler-DeltaBler<=Bler4UE_k_RI_i= <TargetBler+DeltaBler;

[0108] Where TargetBler is the target Bler corresponding to the uplink AMC, and DeltaBler is the Bler redundancy used for convergence judgment.

[0109] S103. Determine whether the time limit for retrying is met. If yes, execute S2; otherwise, maintain the RI_i schedule and wait for the retry time before executing S101 again.

[0110] Specifically, in S1031, determine whether the system time corresponding to RI_i retrieved by the current UE_k meets the time limit for retrying. If yes, execute S1032; otherwise, maintain the RI_i scheduling and wait for the retry time before executing S101 again.

[0111] The time limit is as follows:

[0112] Time-T4UE_kRIi_jTest>=Time4ReAttempt0(1+N4UE_kRIi_jTestFail*TimePunishCoe4Attempt);

[0113] In the formula, Time is the system time corresponding to the current UE_k scheduling RI_i; T4UE_kRIi_jTest is the time when UE_k last attempted to probe RI_j from RI_i; TimePunishCoe4Attempt is the trial interval penalty coefficient; N4UE_kRIi_jTestFail is the number of times UE_k failed to probe RI_j from RI_i; and Time4ReAttempt0 is the initial time interval threshold for re-probing.

[0114] S1032. Update the trial time, update the trial time T4UE_kRIi_jTest for RI_i corresponding to UE_k: T4UE_kRIi_jTest = current system time, and execute S2;

[0115] S2. UE_k maintains RI_i for N4SE scheduling times, counts the MCS, SE and response for each scheduling, and generates the statistical results of RI_i. Then, based on the statistical results of RI_i, calculates the equivalent SE of UE_k for N4SE scheduling times under RI_i = SE4UE_kRI_i.

[0116] Specifically, UE_k maintains RI_i scheduling N4SE times;

[0117] Statistically record the MCS for each scheduling, and based on the correspondence between MCS and SE, record the SE for each scheduling of UE_k under RI_i, SEFront_index4UE_k_RI_i, index=1、2、...N4SE, and record the response result of each scheduling of UE_k under RI_i, ANFront_index4UE_k_RI_i, index=1...N4SE;

[0118] Multiply the SEFront_index4UE_k_RI_i corresponding to the N4SE scheduling of RI_i with the ACK response by RI_i and then add them together to obtain the equivalent SE of UE_k under RI_i = SE4UE_k_RI_i.

[0119]

[0120] The trial pre-scheduling steps include:

[0121] S3. Update the initial MCS under the RI_j scheduling corresponding to UE_k;

[0122] Specifically, obtain the scheduling MCS corresponding to the latest scheduling of RI_i, which is LastMCS4UE_k_RI_i;

[0123] The LastSE4UE_k_RI_i corresponding to LastMCS4UE_k_RI_i is obtained by looking up the table;

[0124] Calculate the minimum SE corresponding to RI_j = MinSE4UE_k_RI_j:

[0125] MinSE4UE_k_RI_j=LastSE4UE_k_RI_i*RI_i / RI_j;

[0126] The smallest MCS with SE not less than MinSE4UE_k_RI_j is obtained by reverse lookup table and used as the initial scheduling MCS of UE_k under RI_j = InitialMCS4UE_kRI_j.

[0127] S4, UE_k maintains RI_j scheduling N4SE times, counts the MCS, SE and response of each scheduling, and generates the statistical results of RI_j;

[0128] Specifically, UE_k maintains RI_j scheduling N4SE times;

[0129] Statistically record the MCS for each scheduling. Based on the correspondence between MCS and SE, record the SE for each scheduling of UE_k under RI_j, SEBack_index4UE_k_RI_j, index=1、2、...N4SE. Record the response result of each scheduling of UE_k under RI_j, ANBack_index4UE_k_RI_j, index=1...N4SE.

[0130] S5. Based on the statistical results of UE_k corresponding to RI_j, calculate the equivalent SE = SE4UE_k_RI_j corresponding to N4SE scheduling of UE_k under RI_j;

[0131] Specifically, in the N4SE scheduling of UE_k corresponding to RI_j, the SEBack_index4UE_k_RI_j corresponding to the ACK scheduling is multiplied by RI_j and then added together to obtain the equivalent SE of UE_k under RI_j = SE4UE_k_RI_j.

[0132]

[0133] The steps for judging the results of the trial include:

[0134] S6. Compare the equivalent SE before and after the trial. If SE4UE_k_RI_j >= SE4UE_k_RI_i * (1 + Thr4SE), then for the next scheduling of UE_k, maintain the RI_j scheduling, and at the same time update the trial time T4UE_kRIj_iTest from RI_j to RI_i for UE_k, T4UE_kRIj_iTest = the current system scheduling time. Update the trial failure counter N4UE_kRIi_jTestFail from RI_i to RI_j for UE_k to 0, and execute S1;

[0135] If SE4UE_k_RI_j < SE4UE_k_RI_i * (1 + Thr4SE), then for the next scheduling of UE_k, resume the RI_i scheduling and execute the trial penalty step.

[0136] The trial penalty step includes:

[0137] S7. Update the trial failure counter N4UE_kRIi_jTestFail from RI_i to RI_j for UE_k, N4UE_kRIi_jTestFail = N4UE_kRIi_jTestFail + 1, and execute S1, forming a loop execution of S1 - S7.

[0138] The specific implementation process is as follows: Assume that the transmission mode corresponding to UE_k in the current system is PMI; the corresponding scheduled RI_i = 1;

[0139] The latest new transmission scheduling MCS under the corresponding RI_i = 1, MCS4UE_k_RI_1 = 24; the trial time T4UE_kRI1_2Test from RI_1 to RI_2 = NAN; the block error rate Bler4UE_k_RI_1 of UE_k corresponding to RI_1 = 10%; Specifically, the settings of each configuration parameter are as follows in the table:

[0140]

[0141]

[0142] S1. Determine whether the current UE_k corresponding to RI_1 meets the trial condition from RI_1 to RI_2. If so, execute S2; if not, maintain the RI_1 scheduling and continue with the trial condition judgment step during the next scheduling, that is, S1;

[0143] Specifically, execute S101. Since MCS4UE_k_RI_1 > MCS4RI1_2, further execute S102;

[0144] Execute S102. Since TargetBler - DeltaBler <= Bler4UE_k_RI_1 <= TargetBler + DeltaBler, further execute S103;

[0145] Execute S103. Since T4UE_kRI1_2Test = NAN, that is, when the first - trial time for UE_k corresponding to RI_1 to RI_2 has not been updated, in this case, it is default to meet the trial time limit. Update the trial time T4UE_kRI1_2Test for UE_k corresponding to RI_1 to RI_2 = the current system time, and then execute S2;

[0146] S2: UE_k maintains RI_1 scheduling for N4SE times, counts the MCS, SE, and responses for each scheduling, and generates the statistical results for RI1;

[0147] Specifically, maintain RI_1 scheduling for N4SE (50) times, then count the MCS for each scheduling, and then according to the correspondence between MCS and SE (3GPP - 214 protocol), record the SE for each scheduling of UE_k under RI_1 = SEFront_index4UE_k_RI_1, index = 1, 2,... N4SE; record the response results ANFront_index4UE_k_RI_1, index = 1... N4SE for each scheduling of UE_k under RI_1;

[0148] Then, according to the statistical results of RI_1, calculate the equivalent SE for RI_1 scheduling N4SE times = SE4RI_1;

[0149] In the N4SE - time scheduling of UE_k corresponding to RI_1, add up the products of the SEFront_index4UE_k_RI_1 corresponding to the scheduling with ACK response multiplied by RI_1 to obtain the equivalent SE for UE_k under RI_1 = SE4UE_k_RI_1;

[0150]

[0151] Assume SE4UE_k_RI_1 = 90;

[0152] S3: Update the initial MCS (InitialMCS4UE_kRI_2) for UE_k corresponding to RI_2 scheduling;

[0153] Specifically, obtain the new - transmission scheduling MCS corresponding to the latest scheduling of RI_1 = LastMCS4UE_k_RI_1;

[0154] By looking up the table (3GPP protocol-214), we can obtain SE = LastSE4UE_k_RI_1 corresponding to LastMCS4UE_k_RI_1;

[0155] Calculate the minimum SE corresponding to RI_2 = MinSE4UE_k_RI_2:

[0156] MinSE4UE_k_RI_2=LastSE4UE_k_RI_1*RI_1 / RI_2;

[0157] The minimum MCS that is not less than MinSE4UE_k_RI_2 is obtained by reverse lookup table and used as the initial scheduling MCS of UE_k under RI_2 = InitialMCS4UE_kRI_2.

[0158] S4, UE_k maintains RI_2 scheduling N4SE times, counts the MCS, SE and response of each scheduling, and generates the statistical results of RI_2;

[0159] Specifically, maintain RI_2 scheduling N4SE(50) times, then count the MCS of each scheduling, and then record the SE of UE_k in each scheduling under RI_2 according to the correspondence between MCS and SE (3GPP-214 protocol), SEBack_index4UE_k_RI_2,index=1、2、...N4SE, and record the response result of UE_k in each scheduling under RI_2 ANBack_index4UE_k_RI_2,index=1...N4SE;

[0160] S5. Based on the statistical results of RI_2, calculate the equivalent SE of RI_2 scheduling N4SE times;

[0161] In the N4SE scheduling of RI_2 corresponding to UE_k, the SEBack_index4UE_k_RI_2 corresponding to the ACK scheduling is multiplied by RI_2 and then added to obtain the equivalent SE of UE_k under RI_2 = SE4UE_k_RI_2.

[0162]

[0163] Assume SE4UE_k_RI_2 = 100;

[0164] S6. Compare the equivalent SE before and after the trial. If SE4UE_k_RI_2 >= SE4UE_k_RI_1 * (1 + Thr4SE), then maintain the RI_2 scheduling. At the same time, update the trial time T4UE_kRI2_1Test from RI_2 to RI_1 for UE_k, T4UE_kRI2_1Test = the current system scheduling time, update the trial failure counter N4UE_kRI1_2TestFail from RI_1 to RI_2 for UE_k to 0, and execute S1. If SE4UE_k_RI_2 < SE4UE_k_RI_1 * (1 + Thr4SE), then the next scheduling for UE_k resumes the RI_1 scheduling and executes the trial penalty step.

[0165] Specifically, because SE4UE_k_RI_2 > SE4UE_k_RI_1 * (1 + 3%), maintain the RI_2 scheduling. At the same time, update the trial time T4UE_kRI2_1Test from RI_2 to RI_1 for UE_k, T4UE_kRI2_1Test = the current system scheduling time, update the trial failure counter N4UE_kRI1_2TestFail from RI_1 to RI_2 for UE_k to 0, and execute S1.

[0166] This embodiment also provides a device for scheduling RI in uplink SU - MIMO, which adopts the method for scheduling RI in uplink SU - MIMO as described above.

[0167] The above are only embodiments of the present invention. Specific structures and common knowledge such as characteristics well - known in the art are not described in detail here. Those of ordinary skill in the art know all the common technical knowledge in the technical field to which the invention belongs before the application date or the priority date, can know all the existing technologies in this field, and have the ability to apply the conventional experimental means before this date. Those of ordinary skill in the art can, under the inspiration given by this application, combine their own abilities to complete and implement this solution. Some typical well - known structures or well - known methods should not be an obstacle for those of ordinary skill in the art to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, and these should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners described in the specification can be used to interpret the content of the claims.

Claims

1. A method for scheduling RI in uplink SU-MIMO, characterized in that, Comprising: Trial condition judgment step: Judge whether the current RI_i corresponding to UE_k satisfies the trial condition from RI_i to RI_j. If so, update the trial time T4UE_kRIi_jTest from RI_i to RI_j corresponding to UE_k. UE_k maintains RI_i and performs scheduling for a preset number of times, and obtains the equivalent SE = SE4UE_kRI_i corresponding to the scheduled RI_i, and execute the trial pre-scheduling step; if not, UE_k maintains the RI_i scheduling, and after waiting for the trial time, execute the trial condition judgment step again; Trial pre-scheduling step: Update the initial MCS corresponding to UE_k under RI_j. UE_k maintains RI_j and performs scheduling for a preset number of times, and obtains the equivalent SE = SE4UE_kRI_j corresponding to the scheduled RI_j, and execute the trial result judgment step; Trial result judgment step: Judge whether SE4UE_kRI_j is greater than or equal to SE4UE_kRI_i. If so, UE_k maintains the RI_j scheduling, and after waiting for the trial time, execute the trial condition judgment step again; if not, execute the trial penalty step; Trial penalty step: Increase the trial time from RI_i to RI_j corresponding to UE_k, and after waiting for the trial time, execute the trial condition judgment step again.

2. The method for scheduling RI in uplink SU-MIMO according to claim 1, characterized in that, The trial condition judgment step includes: S1. Judge whether the current RI_i corresponding to UE_k satisfies the trial condition from RI_i to RI_j. If so, execute S2; if not, maintain the RI_i scheduling, and after waiting for the trial time, execute S1 again; S2. UE_k maintains the RI_i scheduling for N4SE times, counts the MCS, SE and responses of each scheduling, generates the statistical result of RI_i, and according to the statistical result of RI_i, calculates the equivalent SE = SE4UE_kRI_i of UE_k corresponding to RI_i for N4SE times of scheduling.

3. The method for scheduling RI in uplink SU-MIMO according to claim 2, characterized in that, The said S1 includes: S101. Judge whether the latest new transmission scheduling MCS = MCS4UE_k_RI_i corresponding to UE_k under RI_i satisfies the preset condition; the preset condition is: MCS4UE_k_RI_i > MCS4RIi_j1 or MCS4UE_k_RI_i < MCS4RIi_j2, where MCS4RIi_j is the MCS threshold for the trial from RI_i to RI_j, j1 = i + 1, j2 = i - 1; If so, execute S102; if not, maintain the RI_i scheduling, and after waiting for the trial time, execute S101 again; S102. Judge whether the block error rate Bler value of RI_i converges. If so, execute S103; if not, maintain the RI_i scheduling, and after waiting for the trial time, execute S101 again; S103. Judge whether the time limit for re-trial is satisfied. If so, execute S2; if not, maintain the RI_i scheduling, and after waiting for the trial time, execute S101 again.

4. The method for scheduling RI in uplink SU-MIMO according to claim 3, characterized in that: The said S102 includes: Maintain the block error rate Bler4UE_k_RI_i corresponding to UE_k, and record the latest N4Bler scheduling response for RI_i corresponding to UE_k, where the response includes: ACK and NACK; N4Bler is the scheduling number threshold used to calculate Bler; Calculate Bler4UE_k_RI_i based on the response: Bler4UE_k_RI_i = total number of NACKs / (total number of ACKs + total number of NACKs); Calculate the scheduling duration Tk_i of UE_k corresponding to RI_i in the latest N4Bler cycle; Determine if Tk_i is less than TValidity4Bler. If it is, the calculated Bler4UE_k_RI_i is considered valid. If not, the calculated Bler4UE_k_RI_i is considered invalid. TValidity4Bler is the validity duration of Bler. Determine whether the valid Bler4RI_iUE_k has converged. If yes, execute S103; otherwise, the trial condition is not met, maintain the RI_i schedule, wait for the trial time, and execute S1 again. Where Bler4UE_k_RI_i converges as follows: TargetBler-DeltaBler<= Bler4UE_k_RI_i= <TargetBler+DeltaBler; Where TargetBler is the target Bler corresponding to the uplink AMC, and DeltaBler is the Bler redundancy used for convergence judgment.

5. The method for scheduling RI in uplink SU-MIMO according to claim 3, characterized in that: S103 includes: S1031. Determine whether the system time corresponding to the current UE_k scheduling RI_i meets the time limit for retrying. If yes, execute S1032; otherwise, maintain the RI_i scheduling and wait for the retry time before executing S101 again. The time limit is as follows: Time-T4UE_kRIi_jTest>=Time4ReAttempt0(1+N4UE_kRIi_jTestFailed TimePunishCoe4Attempt)! In the formula, Time is the system time corresponding to the current UE_k scheduling RI_i; T4UE_kRIi_jTest is the time when UE_k last tried to probe RI_j with RI_i; TimePunishCoe4Attempt is the penalty coefficient for the trial time interval; N4UE_kRIi_jTestFail represents the number of times UE_k failed to probe RI_j from RI_i. Time4ReAttempt0 is the initial time interval threshold for retrying; S1032. Update the trial time, update the trial time T4UE_kRIi_jTest corresponding to RI_i of UE_k to RI_j: T4UE_kRIi_jTest = current system time, and execute S2.

6. The method for scheduling RI in uplink SU-MIMO according to claim 2, characterized in that, The S2 includes: UE_k maintains RI_i scheduling N4SE times; Statistically analyze the MCS for each scheduling, and based on the correspondence between MCS and SE, record the SE for each scheduling of UE_k under RI_i, SEFront_index4UE_k_RI_i,index=1、2、...N4SE, and record the response result of each scheduling of UE_k under RI_i, ANFront_index4UE_k_RI_i,index=1...N4SE; Multiply the SEFront_index4UE_k_RI_i corresponding to the N4SE scheduling of RI_i with ACK by RI_i and then add them together to obtain the equivalent SE of UE_k under RI_i = SE4UE_k_RI_i. 。 7. The method for scheduling RI in uplink SU-MIMO according to claim 1, characterized in that, The trial pre-scheduling step includes: S3. Update the initial MCS under the RI_j scheduling corresponding to UE_k, including: Get the scheduling MCS corresponding to the latest scheduling of RI_i, which is LastMCS4UE_k_RI_i; The LastSE4UE_k_RI_i corresponding to LastMCS4UE_k_RI_i is obtained by looking up the table; Calculate the minimum SE corresponding to RI_j = MinSE4UE_k_RI_j: MinSE4UE_k_RI_j=LastSE4UE_k_RI_i RI_i / RI_j; The smallest MCS with SE not less than MinSE4UE_k_RI_j is obtained by reverse lookup table and used as the initial scheduling MCS of UE_k under RI_j = InitialMCS4UE_kRI_j. S4, UE_k maintains RI_j and schedules N4SE times, calculates the MCS, SE, and response for each scheduling, and generates statistical results for RI_j, including: UE_k maintains RI_j scheduling N4SE times; Statistically record the MCS for each scheduling. Based on the correspondence between MCS and SE, record the SE for each scheduling of UE_k under RI_j: SEBack_index4UE_k_RI_j, index=1, 2, ..., N4SE. Record the response result for each scheduling of UE_k under RI_j: ANBack_index4UE_k_RI_j, index=1...N4SE. S5. Based on the statistical results of UE_k corresponding to RI_j, calculate the equivalent SE = SE4UE_k_RI_j corresponding to N4SE scheduling times of UE_k under RI_j, including: In the N4SE scheduling operations corresponding to RI_j for UE_k, multiply the SEBack_index4UE_k_RI_j corresponding to the ACK scheduling by RI_j and then add them together to obtain the equivalent SE for UE_k under RI_j: SE4UE_k_RI_j. 。 8. The method for scheduling RI in uplink SU-MIMO according to claim 1, characterized in that, The steps for determining the trial result include: S6. Compare the equivalent SE before and after the trial. If SE4UE_k_RI_j>= SE4UE_k_RI_i If (1+Thr4SE) is executed, then the next scheduling of UE_k will maintain the scheduling of RI_j, and at the same time, the trial time T4UE_kRIj_i for UE_k corresponding to RI_j will be updated, T4UE_kRIj_iTest = the current system scheduling time, the trial failure counter N4UE_kRIi_j for UE_k corresponding to RI_i will be updated to 0, and S1 will be executed; If SE4UE_k_RI_j <SE4UE_k_RI_i If (1+Thr4SE), then the next scheduling UE_k will resume RI_i scheduling and execute the trial penalty step; Thr4SE refers to the redundancy threshold used to compare with SE.

9. The method for scheduling RI in uplink SU-MIMO according to claim 1, characterized in that, The trial and punishment steps include: S7. Update the failure counter N4UE_kRIi_jTestFail for RI_i to RI_j corresponding to UE_k, N4UE_kRIi_jTestFail=N4UE_kRIi_jTestFail+1, and execute S1.

10. An apparatus for scheduling RI in uplink SU-MIMO, characterized in that, The method for scheduling RI in uplink SU-MIMO as described in any one of claims 1-9 is adopted.

Images