Communication method, network node, storage medium and program product

Abstract

The embodiment of the invention provides a communication method, a network node, a storage medium and a program product, and relates to the fields of communication, artificial intelligence and the like. In the method, a first network node determines a first retransmission UE; and determining the MU pairing recommendation result by using the first AI network based on the first retransmission UE, and sending the MU pairing recommendation result to the second network node, so that the MU scheduling calculation of the second network node can be assisted, thereby reducing the calculation complexity of the MU scheduling of the second network node. Optionally, the method performed by the network node may be performed using an artificial intelligence model.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a communication method, a network node, a storage medium and a program product. Background Art

[0002] mMIMO (Massive Multi-Input Multi-Output, also known as MassiveMIMO) technology is a key technology for 5G (5th Generation) / B5G (Beyond 5th Generation). In order to support more antennas, more UEs per cell, and more cells per base station, the computational complexity of MU scheduling is greatly increased. Under the condition that the hardware conditions remain unchanged, how to reduce the computational complexity of MU scheduling has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The purpose of the embodiments of the present application is to solve the technical problem of how to reduce the computational complexity of MU scheduling.

[0004] According to one aspect of an embodiment of the present application, a method performed by a first network node in a communication system is provided, the method comprising:

[0005] Determine a first retransmission UE (User Equipment);

[0006] Determine, based on the first retransmitting UE, a MU (Multi-User) pairing recommendation result using a first AI (Artificial Intelligence) network;

[0007] The MU pairing recommendation result is sent to the second network node.

[0008] Optionally, determining a first retransmission UE includes:

[0009] determining a retransmission probability of at least one UE;

[0010] Based on the retransmission probability of at least one UE, a first retransmission UE is determined.

[0011] Optionally, determining a retransmission probability of at least one UE includes:

[0012] A retransmission probability of at least one UE is determined according to the historical MU pairing recommendation result and the target block error rate.

[0013] Optionally, determining a retransmission probability of at least one UE according to a historical MU pairing recommendation result and a target block error rate includes:

[0014] Determine the number of times at least one UE is used as an initial transmission in the historical MU pairing recommendation results;

[0015] Determining a retransmission probability of an initial transmission based on a target block error rate;

[0016] A retransmission probability of at least one UE is determined based on the number of initial transmissions and the retransmission probability of the initial transmission.

[0017] Optionally, determining a first retransmission UE based on a retransmission probability of at least one UE includes:

[0018] Determining a retransmission validity threshold corresponding to at least one UE;

[0019] A first retransmission UE is determined based on the retransmission probability of at least one UE and a retransmission validity threshold corresponding to the at least one UE.

[0020] Optionally, determining a retransmission validity threshold corresponding to at least one UE includes:

[0021] A retransmission validity threshold corresponding to at least one UE is determined according to the historical scheduling information.

[0022] Optionally, determining, according to historical scheduling information, a retransmission validity threshold corresponding to at least one UE includes:

[0023] Determine, according to the historical scheduling information, a validity index of a historical MU pairing recommendation result, where the validity index of the MU pairing recommendation result is used to measure the validity of the retransmitting UE in the historical MU pairing recommendation result;

[0024] According to the historical MU pairing recommendation result validity indicator, the second AI network is used to determine the retransmission validity threshold corresponding to at least one UE in each first time unit set in the current period.

[0025] Optionally, each first time unit set includes at least one first time unit for performing HARQ feedback on the same continuous uplink second time unit.

[0026] Optionally, the historical scheduling information includes at least one of the following:

[0027] The first time unit using multi-user MU scheduling;

[0028] The second network node determines a first time unit for the MU pairing recommendation result;

[0029] The second network node uses a first time unit of the MU pairing recommendation result provided by the first network node;

[0030] The actual usage of the MU pairing recommendation result provided by the first network node by the second network node in each first time unit set of the historical period:

[0031] The number of first time units in which the second network node uses MU scheduling in each set of first time units in the historical period.

[0032] Optionally, determining a first retransmission UE based on a retransmission probability of at least one UE and a retransmission validity threshold corresponding to at least one UE includes:

[0033] Based on a comparison result of the retransmission probability of at least one UE and a retransmission validity threshold corresponding to at least one UE, a first retransmission UE is determined.

[0034] Optionally, determining a first retransmission UE based on a comparison result of a retransmission probability of at least one UE and a retransmission validity threshold corresponding to at least one UE includes:

[0035] Sort at least one UE according to the retransmission probability from low to high;

[0036] For at least one ranked UE, the following operations are performed in sequence until a target UE is determined: determining a cumulative retransmission probability of the UE and its previous UEs, determining a corresponding retransmission validity probability based on the cumulative retransmission probability, and if the retransmission validity probability is less than a retransmission validity threshold, determining the UE as a target UE;

[0037] The target UE and the UEs thereafter are determined as the first retransmitting UEs.

[0038] Optionally, based on the first retransmitting UE, using the first AI network, determining the MU pairing recommendation result includes:

[0039] Determine at least one candidate UE according to the amount of data to be transmitted by the UE and the measurement configuration information;

[0040] Based on the first retransmitting UE and at least one candidate UE, using the first AI network, an MU pairing recommendation result is determined.

[0041] Optionally, determining at least one candidate UE according to the amount of data to be transmitted of the UE and the measurement configuration information includes:

[0042] Determine at least one candidate UE according to the amount of data to be transmitted by the UE, the measurement configuration information, and the UE scheduling parameter;

[0043] The UE scheduling parameter is determined according to the number of first time units that meet a preset condition within a preset time period including multiple first time units.

[0044] Optionally, determining at least one candidate UE according to the amount of data to be transmitted of the UE and the measurement configuration information includes:

[0045] Determine initial candidate UEs;

[0046] The candidate UEs whose amount of data to be transmitted is less than the first threshold value are deleted from the initial candidate UEs, and the candidate UEs in the measurement interval are deleted from the initial candidate UEs according to the measurement configuration information to obtain at least one candidate UE.

[0047] Optionally, determining at least one candidate UE according to the amount of data to be transmitted of the UE, the measurement configuration information, and the UE scheduling parameter includes:

[0048] Determine initial candidate UEs;

[0049] The candidate UEs whose amount of data to be transmitted is less than the first threshold value are deleted from the initial candidate UEs, and the candidate UEs in the measurement interval are deleted from the initial candidate UEs according to the measurement configuration information, and the initial candidate UEs are sorted according to the UE scheduling parameters and / or the candidate UEs whose UE scheduling parameters are greater than the fairness threshold are deleted from the initial candidate UEs to obtain at least one candidate UE.

[0050] Optionally, based on the first retransmitting UE and at least one candidate UE, using the first AI network, determining an MU pairing recommendation result includes:

[0051] Determine, using a first AI network, a MU pairing recommendation result based on a first retransmitting UE, at least one candidate UE, and at least one of the following:

[0052] Predicted signal to interference plus noise ratio SINR;

[0053] Predicted detection reference signal matrix H_SRS;

[0054] Uplink and downlink time slot configuration information;

[0055] HARQ related configuration information.

[0056] Optionally, the method further comprises:

[0057] Training a first AI network to maximize a multi-user modulation order product code rate (MU-MPR) of at least one UE;

[0058] The MU-MPR of each UE is determined according to at least one of the following:

[0059] The predicted SINR of the UE;

[0060] Multi-user beamforming weights determined based on predicted H_SRS.

[0061] Optionally, determining a first retransmission UE includes:

[0062] Obtaining computing capacity of a second network node;

[0063] When the computing capability is less than a predetermined condition, a first retransmitting UE is determined.

[0064] Optionally, the first network node is a radio access network intelligent controller RIC, and the second network node is a distribution unit DU.

[0065] According to another aspect of an embodiment of the present application, a method performed by a second network node in a communication system is provided, the method comprising:

[0066] Receiving a MU pairing recommendation result sent by the first network node;

[0067] Determine whether the current scheduling time unit uses the MU pairing recommendation result;

[0068] If yes, MU scheduling is performed based on the MU pairing recommendation result.

[0069] Optionally, determining whether the current scheduling time unit uses the MU pairing recommendation result includes:

[0070] Determine the second retransmission UE of the current scheduling time unit according to the HARQ feedback;

[0071] If the second retransmitting UE does not exist in the current scheduling time unit, or the second retransmitting UE is included in the MU pairing recommendation result corresponding to the current scheduling time unit, it is determined that the current scheduling time unit uses the MU pairing recommendation result.

[0072] Optionally, the first network node is a radio access network intelligent controller RIC, and the second network node is a distribution unit DU.

[0073] According to another aspect of an embodiment of the present application, a first network node is provided, the first network node including:

[0074] a transceiver configured to transmit and receive signals; and

[0075] The processor is coupled to the transceiver and is configured to execute the method performed by the first network node provided in the embodiment of the present application.

[0076] According to another aspect of an embodiment of the present application, a second network node is provided, the second network node including:

[0077] a transceiver configured to transmit and receive signals; and

[0078] The processor is coupled to the transceiver and is configured to execute the method performed by the second network node provided in the embodiment of the present application.

[0079] According to another aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method performed by the first network node provided in the embodiment of the present application is implemented.

[0080] According to another aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the method performed by the second network node provided in an embodiment of the present application is implemented.

[0081] According to another aspect of an embodiment of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method performed by the first network node provided in the embodiment of the present application is implemented.

[0082] According to another aspect of an embodiment of the present application, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method performed by the second network node provided in the embodiment of the present application is implemented.

[0083] The communication method, network node, storage medium and program product provided in the embodiments of the present application, the first network node determines a first retransmitting UE; and based on the first retransmitting UE, uses a first AI network to determine a MU pairing recommendation result, and sends the MU pairing recommendation result to a second network node, which can assist the MU scheduling calculation of the second network node, thereby reducing the computational complexity of the MU scheduling of the second network node. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in describing the embodiments of the present application are briefly introduced below.

[0085] Figure 1 A schematic flow chart of a method performed by a first network node in a communication system provided in an embodiment of the present application;

[0086] Figure 2 A schematic diagram of bundle division within a cycle provided in an embodiment of the present application;

[0087] Figure 3 A schematic diagram of using an AI network to determine a MU-UE list provided in an embodiment of the present application;

[0088] Figure 4 A schematic diagram of selecting a solution according to DU computing power provided in an embodiment of the present application;

[0089] Figure 5A schematic flow chart of a method performed by a second network node in a communication system provided in an embodiment of the present application;

[0090] Figure 6 A schematic diagram of a DU scheduling operation provided in an embodiment of the present application;

[0091] Figure 7 A schematic diagram of a DU operation process provided in an embodiment of the present application;

[0092] Figure 8 A schematic diagram of a MU scheduling process jointly completed by RIC and DU provided in an embodiment of the present application;

[0093] Fig. 9 A schematic diagram of a RIC periodic information collection method provided in an embodiment of the present application;

[0094] Fig.10 A schematic diagram of determining a MU-UE list for each TTI within a period provided by an embodiment of the present application;

[0095] Fig.11 A schematic diagram of a DU operation based on a MU-UE list provided in an embodiment of the present application;

[0096] Fig.12 A schematic diagram of a RTM-Net processing flow provided in an embodiment of the present application;

[0097] Fig.13 A schematic diagram of a MU scheduling solution deployed in a RIC and a gNB provided in an embodiment of the present application;

[0098] Fig.14 A schematic diagram of reducing DU complexity provided by an embodiment of the present application;

[0099] Fig.15 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0100] The following description with reference to the accompanying drawings is provided to facilitate a comprehensive understanding of the various embodiments of the present application as defined by the claims and their equivalents. This description includes various specific details to facilitate understanding but should be considered as exemplary only. Therefore, one of ordinary skill in the art will recognize that various changes and modifications can be made to the various embodiments described herein without departing from the scope and spirit of the present application. In addition, for the sake of clarity and conciseness, descriptions of well-known functions and structures may be omitted.

[0101] The terms and expressions used in the following specification and claims are not limited to their dictionary meanings, but are merely used by the inventor to enable a clear and consistent understanding of the present application. Therefore, it should be apparent to those skilled in the art that the following description of various embodiments of the present application is provided for illustration purposes only and not to limit the purpose of the present application as defined in the appended claims and their equivalents.

[0102] It should be understood that the singular forms "a", "an", and "the" may also include plural references unless the context clearly indicates otherwise. Thus, for example, reference to "a component surface" includes reference to one or more such surfaces. When we refer to an element as being "connected" or "coupled" to another element, the one element may be directly connected or coupled to the other element, or the one element and the other element may establish a connection relationship through an intermediate element. In addition, "connected" or "coupled" as used herein may include wireless connection or wireless coupling.

[0103] The term "include" or "may include" refers to the presence of corresponding disclosed functions, operations, or components that can be used in various embodiments of the present application, rather than limiting the presence of one or more additional functions, operations, or features. In addition, the term "include" or "have" may be interpreted as indicating certain characteristics, numbers, steps, operations, constituent elements, components, or combinations thereof, but should not be interpreted as excluding the possibility of the presence of one or more other characteristics, numbers, steps, operations, constituent elements, components, or combinations thereof.

[0104] The term "or" used in various embodiments of the present application includes any of the listed terms and all combinations thereof. For example, "A or B" may include A, may include B, or may include both A and B. When describing multiple (two or more) items, if the relationship between the multiple items is not clearly defined, the multiple items may refer to one, multiple, or all of the multiple items. For example, the description of "parameter A includes A1, A2, A3" may be implemented as parameter A including A1 or A2 or A3, and may also be implemented as parameter A including at least two of the three items A1, A2, and A3.

[0105] Unless defined differently, all terms (including technical terms or scientific terms) used in this application have the same meanings understood by those skilled in the art of this application. Common terms as defined in dictionaries are interpreted as having meanings consistent with the context in the relevant technical field, and should not be interpreted ideally or overly formally, unless clearly defined in this application.

[0106] At least some functions of the device or electronic device provided in the embodiments of the present application can be implemented by an AI model, such as at least one module among multiple modules of the device or electronic device can be implemented by an AI model. Functions associated with AI can be performed by non-volatile memory, volatile memory and processor.

[0107] The processor may include one or more processors. In this case, the one or more processors may be general-purpose processors, such as a central processing unit (CPU), an application processor (AP), etc., or pure graphics processing units, such as a graphics processing unit (GPU), a visual processing unit (VPU), and / or an AI-specific processor, such as a neural processing unit (NPU).

[0108] The one or more processors control the processing of input data according to predefined operating rules or artificial intelligence (AI) models stored in non-volatile memory and volatile memory. The predefined operating rules or artificial intelligence models are provided by training or learning.

[0109] Here, providing by learning means obtaining a predefined operating rule or an AI model with desired characteristics by applying a learning algorithm to a plurality of learning data. The learning can be performed in the device or electronic device itself in which the AI ​​according to the embodiment is executed, and / or can be implemented by a separate server / system.

[0110] The AI ​​model may include multiple neural network layers. Each layer has multiple weight values, and each layer performs neural network calculations by calculating between the input data of the layer (such as the calculation results of the previous layer and / or the input data of the AI ​​model) and the multiple weight values ​​of the current layer. Examples of neural networks include, but are not limited to, convolutional neural networks (CNNs), deep neural networks (DNNs), recurrent neural networks (RNNs), restricted Boltzmann machines (RBMs), deep belief networks (DBNs), bidirectional recurrent deep neural networks (BRDNNs), generative adversarial networks (GANs), and deep Q networks.

[0111] A learning algorithm is a method of using a plurality of learning data to train a predetermined target device (e.g., a robot) to enable, allow, or control the target device to make a determination or prediction. Examples of the learning algorithm include, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning.

[0112] According to the present application, at least one step in the method performed in the electronic device, such as determining the MU pairing recommendation result, determining the retransmission validity threshold, etc., can be implemented using an artificial intelligence model. The processor of the electronic device can perform a preprocessing operation on the data to convert it into a form suitable for use as an artificial intelligence model input. The artificial intelligence model can be obtained through training. Here, "obtained through training" means obtaining a predefined operation rule or artificial intelligence model configured to perform the desired feature (or purpose) by training the basic artificial intelligence model with multiple training data through a training algorithm.

[0113] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the embodiment of the present application and the technical effects produced by the technical solution of the present application are explained below through the description of several optional embodiments. It should be pointed out that the following implementation modes can refer to, draw lessons from or be combined with each other, and the same terms, similar features and similar implementation steps in different implementation modes will not be described repeatedly.

[0114] An embodiment of the present application provides a method executed by a first network node in a communication system, such as Figure 1 As shown, the method includes:

[0115] Step S101: Determine the first retransmitting UE;

[0116] The retransmission UE (RetransmissionUE, re-Tx UE) refers to a UE that needs to retransmit data due to data transmission failure, and the first retransmission UE refers to a retransmission UE predicted by the first network node, which may also be called a virtual retransmission UE (virtualRe-Tx UE).

[0117] In the embodiment of the present application, the determined first retransmission UE may be processed in the form of a list, that is, in this step, the first retransmission UE list (Re-Tx UE list) may be directly determined.

[0118] Step S102: Based on the first retransmitting UE, using the first AI network, determine the MU pairing recommendation result;

[0119] That is, in the embodiment of the present application, the determined MU pairing recommendation result includes the first retransmitting UE predicted by the first network node, ensuring that the MU pairing recommendation result provided by the first network node is valid for the final MU transmission.

[0120] For the embodiments of the present application, some implementations of this step may use the first AI network. In practical applications, those skilled in the art may set the specific AI network used in this step according to actual conditions, and the embodiments of the present application are not limited here.

[0121] In the embodiment of the present application, the MU pairing recommendation result may specifically be a MU-UE list (also referred to as a MU list), but is not limited thereto.

[0122] Step S103: Send the MU pairing recommendation result to the second network node.

[0123] In an embodiment of the present application, the first network node submits the determined MU pairing recommendation result to the second network node, and the second network node can use the MU pairing recommendation result to perform MU scheduling (MU-scheduling), such as executing MU layer decision and resource allocation process, thereby reducing the complexity of MU scheduling by the second network node.

[0124] In the embodiment of the present application, the first network node may be a RIC (Radio Access Network Intelligent Controller), and the second network node may be a DU (Distributed Unit). The following problems of the current communication system can be solved:

[0125] In the current communication system, the 5G mMIMO BS (Base Station) consists of three parts: CU (Central Unit), DU and MMU (Massive MIMO Unit). The mMIMO scheduler is a MAC (Medium Access Control) scheduler located in the DU. The key function of the mMIMO scheduler is to achieve spatial multiplexing of the TDD system through UE pairing. In order to support more antennas, more UEs per cell, and more cells per DU base station, the computational complexity of the MU scheduling of the DU mMIMO scheduler is greatly increased.

[0126] And due to the delay between DU and RIC (typical delay is: 10ms~100ms), RIC cannot obtain HARQ (Hybrid Automatic Repeat Request) ACK (ACKnowledge Character) / NACK (Negative ACKnowledgement) feedback information in time, resulting in RIC being unable to know the UE that needs to retransmit, so that the MU pairing recommendation result generated by RIC may not include the retransmitting UE. Therefore, DU will not use the MU pairing recommendation result generated by RIC for final scheduling, because the MU pairing recommendation result generated in RIC is invalid for the final MU transmission. That is, due to the HARQ retransmission delay problem, the solution of using RIC to reduce the DU calculation complexity is difficult to implement.

[0127] In response to at least one of the above-mentioned technical problems or areas that need improvement, the present application proposes an intelligent mMIMO solution based on RIC. RIC effectively reduces the computational complexity of the DU mMIMO scheduler (also referred to as the DU scheduler) by determining the MU pairing recommendation results including the retransmitting UE to meet the hardware limitations, while solving the retransmission delay problem in the existing technical solution and maintaining system performance.

[0128] Specifically, the method performed by the first network node (RIC) in the communication system provided in the embodiment of the present application determines a first retransmission UE; and based on the first retransmission UE, uses the first AI network to determine the MU pairing recommendation result to ensure that the MU pairing recommendation result generated in the first network node (RIC) is valid for the final MU transmission, and sends the MU pairing recommendation result to the second network node (DU), which can assist the scheduling calculation of the second network node (DU), thereby reducing the computational complexity of the DU mMIMO scheduler (also referred to as the DU scheduler).

[0129] In addition, the method performed by the first network node (RIC) in the communication system provided in the embodiment of the present application takes into account the problem of retransmission delay. The first network node (RIC) predicts the first retransmission UE and, based on the first retransmission UE, uses the first AI network to determine the MU pairing recommendation result to ensure that the MU pairing recommendation result generated in the first network node (RIC) is valid for the final MU transmission, thereby overcoming the retransmission delay problem that has not been solved in the prior art.

[0130] In an embodiment of the present application, an optional implementation is provided for step S101, which may specifically include: determining the retransmission probability of at least one UE; and determining a first retransmission UE based on the retransmission probability of at least one UE.

[0131] In the embodiment of the present application, the retransmission probability of the UE refers to the probability that each UE may retransmit estimated by the first network node.

[0132] Optionally, for each UE in one or more first time units within a period, the retransmission probability of the UE may be determined, and based on the determined retransmission probability of the UE, the first retransmission UE may be determined. The first time unit here may specifically refer to a TTI (Transmission Time Interval, transmission time interval), but is not limited thereto.

[0133] In an optional implementation, the retransmission probability of at least one UE may be determined according to a historical MU pairing recommendation result. As an example, the retransmission probability of at least one UE is calculated according to a MU pairing recommendation result generated by the first network node.

[0134] In the embodiment of the present application, the historical MU pairing recommendation result may correspond to a first time unit set (bundle). A bundle is a set of first time units (such as a set of TTIs), and the HARQ ACK / NACK feedback associated with this set of first time units are all located on the same continuous UL (Uplink) second time unit (for example, the second time unit here may refer to a time slot). According to the transmission timing relationship between data and HARQ ACK / NACK feedback, the first time units within a period are clustered to obtain each bundle. See Figure 2 , shows the division of bundles in a cycle. In this example, a cycle contains multiple bundles, each bundle contains multiple downlink first time units (TTIs), where the number of first time units corresponding to different bundles can be the same or different, that is, Figure 2 N1 to N3 in the above may be the same or different. In practical applications, based on different transmission timing designs, the bundles may be divided in other ways, which are not limited in the embodiments of the present application. In addition, the bundle may also be understood as a third time unit.

[0135] In an embodiment of the present application, the retransmission probability of each UE in the current bundle can be calculated based on the MU pairing recommendation result of the previous bundle. As an example, assuming that the current bundle is bundle b, the retransmission probability of each UE in each first time unit of bundle b can be calculated based on the MU pairing recommendation result of bundle b-1, but it is not limited to this. In practical applications, the retransmission probability of each UE in the current bundle can also be calculated based on one or more historical bundles, or one or more historical periods of MU pairing recommendation results.

[0136] In the embodiment of the present application, the retransmission probability of at least one UE may be determined according to the historical MU pairing recommendation result and the target block error rate (Block Error Ratio, BLER).

[0137] As an example, according to the MU pairing recommendation result of the previous bundle, the retransmission probability of the corresponding UE is calculated based on the target BLER.

[0138] In an optional implementation manner, the process of “determining the retransmission probability of at least one UE according to the historical MU pairing recommendation result and the target block error rate” may specifically include:

[0139] Step SA1: Determine the number of times at least one UE is used as an initial transmission in the historical MU pairing recommendation results;

[0140] As an example, the UE information in the MU pairing recommendation result of the previous bundle (such as the above bundle b-1) is read, and the number of times each UE in the MU pairing recommendation result of the previous bundle is used as the initial transmission is determined;

[0141] Step SA2: Determine the retransmission probability of the initial transmission based on the target block error rate;

[0142] That is, the value of the retransmission probability of the initial transmission is set based on the target BLER, wherein those skilled in the art can set the value of the target BLER according to actual needs, and the embodiments of the present application are not limited here.

[0143] Step SA3: Based on the number of initial transmissions and the retransmission probability of the initial transmission, determine the retransmission probability of at least one UE.

[0144] In an optional implementation manner, step SA3 may use the following formula 1 to estimate UE u The retransmission probability P u :

[0145]

[0146] Where n represents UE u The number of times it is used as the initial transmission in the historical MU pairing recommendation results; represents the retransmission probability of the initial transmission, The value of is set based on the target BLER.

[0147] Therefore, the retransmission probability of each UE in the read historical MU pairing recommendation result can be estimated respectively.

[0148] In an embodiment of the present application, an optional implementation method is provided for the step of "determining a first retransmission UE based on the retransmission probability of at least one UE", which may specifically include: determining a retransmission validity threshold corresponding to at least one UE; determining the first retransmission UE based on the retransmission probability of at least one UE and the retransmission validity threshold corresponding to at least one UE.

[0149] In the embodiment of the present application, the retransmission validity threshold refers to a threshold used by the first network node to evaluate the retransmission validity of each retransmitting UE.

[0150] Optionally, for a bundle within a period, a retransmission validity threshold corresponding to all UEs in the bundle may be determined, and based on the determined retransmission validity threshold of the bundle, a first retransmission UE may be determined.

[0151] In an optional implementation, a retransmission validity threshold corresponding to at least one UE is determined based on historical scheduling information (historical sch.Inf.). The historical scheduling information refers to the actual scheduling of the second network node in each first time unit in the past one or more cycles. That is, in an embodiment of the present application, the retransmission validity threshold corresponding to the UE of each bundle in the current cycle can be determined based on the historical scheduling information of the previous one or more cycles.

[0152] In the embodiment of the present application, in order to make the first retransmission UE predicted by the first network node consistent with the actual retransmission UE determined by the second network node as much as possible, an optional implementation mode (i.e., a validity strategy) is provided for the step of "determining the retransmission validity threshold corresponding to at least one UE according to the historical scheduling information", which may specifically include:

[0153] Step SB1: Determine the validity index of the historical MU pairing recommendation result according to the historical scheduling information;

[0154] The historical MU pairing recommendation result validity index is used to measure the validity of the retransmitting UE in the historical MU pairing recommendation result. Optionally, each bundle in a cycle has a corresponding historical MU pairing recommendation result validity index. In an example, the historical MU pairing recommendation result validity index corresponding to a bundle can be determined based on the historical scheduling information of the bundle, but is not limited thereto.

[0155] In an optional implementation manner, the historical scheduling information includes at least one of the following:

[0156] (1) Using the first time unit of multi-user MU scheduling;

[0157] That is, it can be determined whether each first time unit (such as TTI) is MU scheduled (MU scheduled or not), where whether it is MU scheduled refers to whether the second network node uses the MU scheduling method in the specified first time unit. If the MU scheduling method is used, MU scheduled or not is set to Y, otherwise, it is set to N.

[0158] (2) The first time unit in which the second network node determines the MU pairing recommendation result;

[0159] That is, it is possible to determine whether each first time unit (such as TTI) is re-determined by the second network node (DU re-decided result), wherein the second network node re-determined result (DU re-decided result) means that the second network node does not use the MU pairing recommendation result provided by the first network node in the specified first time unit, but re-determines a new MU pairing recommendation result by itself, assuming that it is recorded as That is, at the time corresponding to the t'th first time unit of the b'th bundle in the historical period p', the second network node re-determines the MU pairing recommendation result.

[0160] (3) the second network node uses the first time unit of the MU pairing recommendation result provided by the first network node;

[0161] That is, it can be determined whether the second network node uses the MU pairing recommendation result (MU-UE list used by DU) for each first time unit (such as TTI), wherein the second network node uses the MU pairing recommendation result means that the second network node uses the MU pairing recommendation result provided by the first network node in the specified first time unit, assuming that it is recorded as That is, at the time corresponding to the t'th first time unit of the b'th bundle in the historical period p', the second network node uses the MU pairing recommendation result provided by the first network node.

[0162] For ease of understanding, the following table 1 shows an example of the structure of the above-mentioned historical scheduling information (in actual application, it is not limited to this structure):

[0163]

[0164]

[0165] Table 1

[0166] Among them, P′ 1 ~P′ T represents the period index, B′ 1 ~B′ M represents the bundle index, t′ 1 ~t′ N : Indicates the first time unit index (such as TTI index) in each bundle. There are a total of M×N first time units in one cycle.

[0167] In Table 1, N in the MU scheduling column indicates that MU is not scheduled in the first time unit (possibly because SU (Single User)-UE has a higher PF (Proportional Fairness) metric in this first time unit, but not limited to this); Y indicates that MU scheduling is performed in the first time unit. Indicates that in the first first time unit of the Mth bundle in the first historical period, the second network node does not use the MU pairing recommendation result provided by the first network node, but re-determines the MU pairing recommendation result (possibly because a certain retransmitting UE is not in the MU pairing recommendation result provided by the first network node, but is not limited to this), other And so on. Indicates that in the second first time unit of the first bundle of the first historical period, the second network node uses the MU pairing recommendation result provided by the first network node, and the other And so on.

[0168] Furthermore, the historical scheduling information that can be determined based on the table also includes at least one of the following:

[0169] (4) actual usage of the MU pairing recommendation result provided by the first network node by the second network node in each first time unit set of the historical period;

[0170] Specifically, the actual usage may include:

[0171] 1) the number of first time units in each set of first time units in the historical period in which the second network node uses the MU pairing recommendation result provided by the first network node;

[0172] For example, in Table 1, in the first bundle of the first cycle, the first time unit in which the second network node uses the MU pairing recommendation result provided by the first network node includes t′ 2 and t′ 2 ~t′ N According to the number of the first time units, the number of first time units of the MU pairing recommendation result provided by the first network node used by the second network node in the first bundle of the first cycle can be determined, and the same applies to other bundles.

[0173] 2) It is possible to determine the first time unit in each set of first time units in the historical period in which the second network node uses the MU pairing recommendation result provided by the first network node, and the number of first time units in which the first retransmission UE provided by the first network node is greater than the number of first time units of the first retransmission UE actually used by the second network node;

[0174] In actual applications, after the first network node provides the MU pairing recommendation result to the second network node, due to some specific UE-related factors, such as PF measurement or control channel resource usage, some first retransmission UEs in the MU pairing recommendation result no longer use MU scheduling. At this time, the first retransmission UEs provided by the first network node may be more than the first retransmission UEs actually used by the second network node. The actual usage of the MU pairing recommendation result used by the second network node in each first time unit set of the historical period can be used as the actual usage.

[0175] Specifically, it is possible to determine the first time unit in each first time unit set of the historical period that satisfies the second network node to use the MU pairing recommendation result provided by the first network node, and the first time units satisfy the condition that The first retransmission UE in The number of first time units of the first retransmitting UE in . It can represent the MU pairing recommendation result provided by the first network node to the second network node at the time corresponding to the t'th first time unit of the b'th bundle in the historical period p', It can represent the MU pairing recommendation result actually used by the second network node after the first network node provides the MU pairing recommendation result at this time. From the above description, it can be known that The number of first retransmitting UEs is greater than or equal to The number of first retransmitting UEs.

[0176] (5) The number of first time units in each set of first time units in the historical period that the second network node uses MU scheduling.

[0177] For example, in Table 1, in the Mth bundle of the first cycle, the first time unit of the second network node using MU scheduling includes t′ 1 , t′ 2 and t′ 2 ~t′ N According to the number of the first time units, the number of the first time units in which the second network node uses MU scheduling in the Mth bundle of the first cycle can be determined, and the same applies to the other bundles.

[0178] In an optional implementation, taking the first time unit as TTI as an example, step SB1 can use the following formula 2 to calculate the historical MU pairing recommendation result effective indicator Valid metric based on historical scheduling information: b :

[0179] Valid metric b’ =(valid time b’ -over valid time b’ ) / sum number of TTI

[0180] ——Formula 2

[0181] Among them, valid time b’ Indicates the number of TTIs corresponding to the b'th bundle in the MU pairing recommendation result provided by the first network node used by the second network node in the historical period. b’ Indicates that the second network node uses the number of TTIs corresponding to the b'th bundle in the MU pairing recommendation result provided by the first network node during the historical period, and the TTI also satisfies The first retransmission UE in sumnumber of TTI represents the sum of TTI numbers corresponding to the b'th bundle when the second network node uses MU scheduling in the historical period.

[0182] Thus, the effective index of the historical MU pairing recommendation results corresponding to each bundle can be calculated.

[0183] Step SB2: Determine a retransmission validity threshold corresponding to at least one UE in each first time unit set in the current cycle according to the validity index of the historical MU pairing recommendation result.

[0184] As can be seen from the above introduction, each first time unit bundle includes at least one first time unit for performing HARQ feedback on the same continuous UL second time unit.

[0185] As an example, according to the historical MU pairing recommendation result validity indicators corresponding to one or more bundles, a retransmission validity threshold corresponding to at least one UE in each bundle in the current period is determined.

[0186] For the embodiments of the present application, some implementations of this step may use a second AI network. That is, based on historical scheduling information, use the second AI network to determine the retransmission validity threshold corresponding to at least one UE in each first time unit set in the current period; or use the second AI network to determine the retransmission validity threshold corresponding to at least one UE in each first time unit set in the current period based on the historical MU pairing recommendation result validity index. In practical applications, those skilled in the art can set the specific AI network used in these steps according to actual conditions, and the embodiments of the present application are not limited here.

[0187] In an optional implementation, the second AI network may adopt an ε-greedy reinforcement learning network, that is, in an embodiment of the present application, ε-greedy reinforcement learning may be used to determine and continuously optimize the retransmission validity threshold based on the effective indicators of the historical MU pairing recommendation results.

[0188] Specifically, an optional implementation method is provided for step SB2 in the embodiment of the present application, which may include the following steps:

[0189] Step SB21: Obtain a trained validity threshold set. The reward value corresponding to each validity threshold in the validity threshold set is determined based on the validity index of the historical MU pairing recommendation results:

[0190] The validity threshold set is trained in the following way:

[0191] Step SC1: Obtain a preset validity threshold set, exploration probability, and historical MU pairing recommendation result validity index;

[0192] Assume that the initial validity threshold set is For example a represents a set ; and assuming that the exploration probability is ε; the historical MU pairing recommendation result validity index is obtained from step SB1.

[0193] Step SC2: Repeat the training steps SC21 to SC24 until the reward value converges to obtain a trained validity threshold set:

[0194] Step SC21: generating random values ​​for each first time unit set of at least one historical period;

[0195] For example, for the b'th bundle of period p', a random value is generated. Optionally, the value range of the random value may be within the range of 0 to 1, or may not be limited thereto.

[0196] Step SC22: If the random value is less than the exploration probability, then select a validity threshold from the validity threshold set according to a uniform distribution;

[0197] That is, if the random value is less than ε, the validity threshold a is from the set Among them, Th_vaild b’ Indicates the validity threshold of the b'th bundle in a cycle.

[0198] Step SC23: If the random number is greater than or equal to the exploration probability, then selecting the validity threshold corresponding to the maximum reward in the validity threshold set;

[0199] That is, if the random value ≥ ε, the validity threshold Th_vaild b ,=a,a is The maximum reward in corresponds to a, where Represents the reward value corresponding to the validity threshold a in the validity threshold set.

[0200] Step SC24: updating the reward value corresponding to the effectiveness threshold selected from the effectiveness threshold set according to the historical MU pairing recommendation result effectiveness index corresponding to the historical period.

[0201] That is, the selected reward value is updated according to the historical MU pairing recommendation result validity index corresponding to the b'th bundle obtained after the second network node completes the period p' scheduling Right now:

[0202]

[0203] Among them, valid time b’ Indicates the number of TTIs corresponding to the b'th bundle in the MU pairing recommendation result provided by the first network node used by the second network node in the historical period. b’ Indicates that the second network node uses the number of TTIs corresponding to the b'th bundle in the MU pairing recommendation result provided by the first network node during the historical period, and the TTI also satisfies The first retransmission UE in sumnumber of TTI represents the sum of TTI numbers corresponding to the b'th bundle when the second network node uses MU scheduling in the historical period.

[0204] That is, the effectiveness threshold set and the reward value of the effectiveness threshold selected in the period are continuously optimized and updated according to the historical scheduling information in each period.

[0205] Therefore, a trained validity threshold set can be obtained through initialization and iteration of greedy reinforcement learning, and each validity threshold in the validity threshold set learns the corresponding reward value.

[0206] Step SB22: Determine the validity threshold corresponding to the maximum reward value in the validity threshold set as the retransmission validity threshold corresponding to at least one UE.

[0207] That is, for each bundle in the current cycle, select the retransmission validity threshold Th_valid corresponding to at least one UE in the bundle b =a, a is The maximum reward in corresponds to a. It can be understood that in the embodiment of the present application, b' represents the historical bundle (training data) and b represents the real-time bundle.

[0208] In an embodiment of the present application, after step SB22, the following step may also be included: after the scheduling of the current cycle is completed, based on the effective indicator of the MU pairing recommendation result corresponding to the current cycle, the reward value corresponding to the retransmission effectiveness threshold corresponding to at least one UE in the effectiveness threshold set is updated, that is, the effectiveness threshold set and the reward value of each effectiveness threshold in the effectiveness threshold set are continuously optimized.

[0209] In an embodiment of the present application, based on historical scheduling information, by using greedy reinforcement learning, the effectiveness strategy is determined and continuously optimized to ensure the accuracy of predicting the first retransmitting UE, thereby further ensuring the effectiveness of the MU pairing recommendation result provided by the first network node.

[0210] In an embodiment of the present application, for step S101, the first retransmission UE can be determined based on the retransmission probability of at least one UE and the retransmission validity threshold corresponding to at least one UE, that is, the first retransmission UE is determined based on the validity strategy according to the retransmission probability predicted in the first network node. Specifically, the retransmission probability of the corresponding UE can be calculated based on the target BLER according to the MU pairing recommendation result in the previous bundle, and then the first retransmission UE in the current bundle that meets the retransmission validity threshold condition can be predicted according to the validity strategy.

[0211] In an optional implementation, this step may specifically include: determining a first retransmission UE based on a comparison result of the retransmission probability of at least one UE and a retransmission validity threshold corresponding to at least one UE.

[0212] Specifically, based on the comparison result of the retransmission probability of at least one UE and the retransmission validity threshold corresponding to at least one UE, the step of determining the first retransmission UE may specifically include:

[0213] Step S1011: sorting at least one UE according to the retransmission probability from low to high;

[0214] As an example, for the UEs in the MU pairing recommendation result of the previous bundle (such as the above-mentioned bundle b-1), the UEs are sorted according to the calculated retransmission probabilities.

[0215] Step S1012: performing the following operations in sequence for at least one ranked UE until a target UE is determined: determining the cumulative retransmission probability of the UE and the UEs before it, determining the corresponding retransmission validity probability based on the cumulative retransmission probability, and if the retransmission validity probability is less than the retransmission validity threshold, determining the UE as the target UE;

[0216] The retransmission validity probability may refer to the opposite probability of the cumulative retransmission probability, or may be calculated in combination with a certain weight.

[0217] According to the sorted UEs, starting from the UE with the lowest retransmission probability, according to the validity threshold Th_valid corresponding to bundle b, b It is evaluated and determined one by one and cumulatively whether each UE is a virtual retransmission UE.

[0218] Sorting UE Retransmission probability 1 UE3 0.001 2 UE5 0.002 3 UE4 0.3 4 UE1 0.35 ... ... ...

[0219] Table 2

[0220] In an example, after sorting, at least one UE and the probability of retransmission are shown in Table 2. Assume that the validity threshold Th_valid corresponding to bundle b is b =0.8. Taking the retransmission validity probability as the opposite probability of the cumulative retransmission probability as an example, the opposite probability of the retransmission probability of the first UE is 1-0.001=0.999; the opposite probability of the sum of the retransmission probabilities of the first 2 UEs is 1-(0.001+0.002)=0.997; the opposite probability of the sum of the retransmission probabilities of the first 3 UEs is 1-(0.001+0.002+0.3)=0.697. Among them, 0.697<0.8, 0.997≥0.8. Therefore, the target UE is the third UE.

[0221] Step S1013: determine the target UE and the UEs after it as the first retransmitting UEs.

[0222] Continuing with the above example, all UEs starting from the third UE may be determined as first retransmission UEs and may be included in the first retransmission UE list.

[0223] In this way, all first retransmitting UEs (or a first retransmitting UE list) can be determined.

[0224] In an optional implementation manner, step S1012 may use the following formula 4 to calculate the retransmission validity probability values ​​one by one starting from the UE with the lowest retransmission probability:

[0225]

[0226] Where u represents the sorting number of the UE. Indicates the first retransmitting UE list corresponding to the tth first time unit (such as TTI) of the bth bundle of the corresponding period m. It can be understood that when u=1, the current The UE set includes the first UE. When u = 2, the current The UE set includes the first UE and the second UE, and so on.

[0227] In an optional implementation manner, the specific operation process of step S1011 to step S1013 can be represented by the following method:

[0228] For UEu=1, 2, ..., U, the following evaluation operations are performed in sequence: Evaluate the validity of bundle b (retransmission validity probability): It is determined that UEu is not the first retransmitting UE, that is Otherwise, start from UEu to UE U These UEs are determined as the first retransmitting UEs, and the evaluation operation for subsequent UEs is terminated.

[0229] For the embodiment of the present application, since the error probability of chase combining (CC, a retransmission method) after retransmission is relatively low, the retransmission probability of the first retransmitting UE can be omitted.

[0230] In the embodiment of the present application, the first retransmission UE list of bundle b can be calculated according to the MU pairing recommendation result of bundle b-1. In an example, it is assumed that the MU pairing recommendation result of bundle b-1 is as shown in Table 3:

[0231]

[0232] Table 3

[0233] in, represents the nth first time unit of bundle b-1 of period p; 4 thThe layer indicates that in this example, the maximum number of UEs in the MU pairing recommendation result is limited to 4. In the MU pairing recommendation result of the bundle b-1, some UEs (such as UE1 and UE2) are the first retransmission UEs of the bundle b-1, and the other UEs are the initial transmission UEs of the bundle b-1.

[0234] Based on at least one of the foregoing embodiments, the first retransmission UE list of bundle b as shown in Table 4 may be determined according to the MU pairing recommendation result of bundle b-1:

[0235]

[0236] Table 4

[0237] in, represents the nth first time unit of bundle b of period p; 4 th The layer indicates that in this example, the maximum number of UEs in the MU pairing recommendation result is limited to 4. In the first retransmission UE list of bundle b, each UE is a first retransmission UE.

[0238] It should be noted that the number of layers and corresponding UEs shown in Table 3 and Table 4 are for illustration only, and the specific first retransmission UE list shall be subject to the actual implementation, and the number of layers and corresponding UE examples in Table 3 and Table 4 should not be understood as limiting the present application. For example, the number of layers used in actual applications may be much greater than 4, and for example, one UE may correspond to multiple layers (that is, different layers may correspond to the same UE index), and for example, the number of layers corresponding to different first time units may be different, etc., which are not limited in the embodiments of the present application.

[0239] The technical solution provided in the embodiment of the present application calculates the retransmission probability of the UE based on the target BLER, and then determines the first retransmission UE according to the retransmission probability of the UE and the effectiveness strategy prediction, thereby ensuring the accuracy of the predicted first retransmission UE, and ensuring that the actual retransmission UEs of the second network node are in the MU pairing recommendation results predicted by the first network node during real-time scheduling, thereby ensuring the effectiveness of the MU pairing recommendation results.

[0240] In addition, according to the transmission timing relationship between data and HARQ ACK / NACK feedback, the first time unit (such as TTI) in a cycle is divided into bundles in order to schedule retransmission of all retransmitting UEs of a bundle in the cycle, which helps to predict the retransmitting UE and schedule the retransmitting UE in time.

[0241] In the embodiment of the present application, an optional implementation is also provided for step S102, which may specifically include:

[0242] Step S1021: determining at least one candidate UE according to the amount of data to be transmitted and measurement configuration information of the UE;

[0243] In the embodiment of the present application, the amount of data to be transmitted can be determined by predicting buffer occupancy (BUFIER Occupy, BO).

[0244] In order to obtain the UE to be scheduled, the first network node may remove the UE that cannot be scheduled in each first time unit according to the predicted BO of each UE and the UE measurement configuration information (UE meas.Conf.), thereby generating at least one candidate UE in each first time unit within a period.

[0245] The amount of data to be transmitted or the predicted BO can be provided by the service prediction xAPP (application in the first network node) of the first network node, which can come from the report data of the event or the periodic report data. The measurement configuration information (UE meas.Conf.) can come from the periodic report data of the second network node. The measurement configuration information is the configuration information provided by the second network node to configure the measurement GAP for the UE. Through the measurement configuration information, it can be known that a period of time (i.e., the measurement interval GAP) is reserved for the measurement. During this period of time, the UE will not send and receive any data, so the base station will not schedule the UE during this period of time.

[0246] Specifically, the initial candidate UEs may be determined first, then the candidate UEs whose amount of data to be transmitted is less than the first threshold value may be deleted from the initial candidate UEs, and the candidate UEs in the measurement interval may be deleted from the initial candidate UEs according to the measurement configuration information to obtain at least one candidate UE.

[0247] That is, in the embodiment of the present application, taking the first time unit as TTI as an example, at least one candidate UE may be generated using a UE scheduling restriction criterion at a predicted TTI level, the criterion including:

[0248] 1) In a given TTI, a UE without service is determined based on whether the predicted BO is less than a first threshold value, and then this UE or these UEs cannot be scheduled in this TTI;

[0249] 2) The TTI corresponding to the UE's measurement interval (Gap) is obtained according to the UE measurement configuration information (UE meas.Conf.). When the UE is in the measurement Gap period, the UE cannot be scheduled in the TTI.

[0250] As an example, Table 5 shows some examples of initial candidate UEs:

[0251] UE List TTI 1 TTI 2 TTI 3 TTI 4 TTI 5 TTI 6 TTI 7 ... TTI_M×N UE1 X1 ... X1 X1 ... X1 UE2 X1 X2 X2 X1 ... ... ... ... ... ... ... ... ... ... ... UE L X1 X1 ...

[0252] Table 5

[0253] Among them, UE1~UE L represent UE indexes, L represents the total number of UEs with services in the current cycle; TTI_1~TTI_M×N represent TTI indexes, and M×N represents the total number of TTIs in the current cycle. X1 means that the corresponding UE has no services in the corresponding TTI, which can be determined based on the predicted BO being less than the first threshold value, and these UEs are deleted from the corresponding TTI in the initial candidate UE list. X2 means that the corresponding UE is in the measurement Gap period in the corresponding TTI, which can be determined based on the measurement configuration information (UE meas.Conf.), and these UEs are deleted from the corresponding TTI in the initial candidate UE list.

[0254] It should be noted that the initial candidate UEs, UEs that cannot be scheduled in the corresponding TTI, and identifiers of the reasons for not being able to be scheduled (X1 and X2) shown in Table 5 are only schematic representations, and the specific initial candidate UE list is subject to the actual implementation. These representations in Table 5 should not be understood as limitations on the present application. For example, in actual applications, the number of UEs corresponding to different TTIs may be different; for another example, each UE can be identified as "candidate" or "non-candidate" in each TTI (technical personnel in this field can set the identification symbol according to actual conditions); for another example, the initial candidate UE list can be represented by an incremental UE list or a decremental UE list compared to the preset UE list, etc., and the embodiments of the present application are not limited here. Similarly, the at least one determined candidate UE can be processed in the form of a list, that is, in this step, the candidate UE list (Candidate UE list) can be directly determined.

[0255] Step S1022: Based on the first retransmitting UE and at least one candidate UE, using the first AI network, determine an MU pairing recommendation result.

[0256] That is, in an embodiment of the present application, the determined MU pairing recommendation result includes not only the first retransmitting UE predicted by the first network node, but also the corresponding UE to be scheduled for MU pairing in each first time unit within a cycle, thereby ensuring that the MU pairing recommendation result provided by the first network node is valid for the final MU transmission.

[0257] In the embodiment of the present application, a fairness operation may be additionally performed based on step S1021. Specifically, step S1021 may include: determining at least one candidate UE according to the amount of data to be transmitted of the UE, measurement configuration information and UE scheduling parameters.

[0258] Specifically, the initial candidate UEs can be determined first; then, the candidate UEs whose amount of data to be transmitted is less than a first threshold value are deleted from the initial candidate UEs; according to the measurement configuration information, the candidate UEs in the measurement interval are deleted from the initial candidate UEs; the initial candidate UEs are sorted according to the UE scheduling parameters and / or the candidate UEs whose UE scheduling parameters are greater than the fairness threshold are deleted from the initial candidate UEs, so as to obtain at least one candidate UE.

[0259] The UE scheduling parameter (also referred to as a normalized UE scheduling ratio) is determined according to the number of first time units that meet a preset condition within a preset time period including multiple first time units. In one example, the preset condition may refer to the amount of data to be transmitted being greater than or equal to a first threshold value (or BO being greater than or equal to a BO threshold value) and not being in a measurement interval, etc., but is not limited thereto.

[0260] In an optional implementation manner, taking the first time unit as TTI as an example, the UE scheduling parameter may be calculated using the following calculation formula 5:

[0261] Normalized UE scheduling ratioi=Scheduling TTI numberi / TTI numberwith enough BO and out of M-gap

[0262] ——Formula 5

[0263] Among them, Scheduling TTI numberi represents the number of TTIs of scheduled UEi that meet the condition that the amount of data to be transmitted of UEi is greater than or equal to the first threshold value (or BO is greater than or equal to the BO threshold value) and is not in the measurement Gap period within a statistical time (i.e., a preset time period, including multiple TTIs). TTI number with enough BO andout of M-gap is the sum of the number of TTIs corresponding to the condition that the amount of data to be transmitted of all UEs is greater than or equal to the first threshold value (or BO is greater than or equal to the BO threshold value) and is not in the measurement Gap period within a statistical time (i.e., a preset time period).

[0264] For ease of understanding, a specific example of a process for generating a candidate UE list is given below:

[0265] Step 1): Initialize a candidate UE list, in which each TTI contains L UEs (ie, UE1, UE2, ..., UEL).

[0266] Step 2): According to the predicted BO, the BO value corresponding to each UE in each TTI is obtained. If there is a UE whose corresponding BO value in a certain TTI is less than the BO threshold value (i.e., the first threshold value), and the UE is not included in the corresponding TTI in the candidate UE list, then the UE is deleted from the UE list of the corresponding TTI in the candidate UE list.

[0267] Step 3): According to the UE measurement configuration information (UE meas.Conf.), the TTI corresponding to the measurement gap of each UE is obtained. If a certain UE is in the measurement gap period of a certain TTI, and the corresponding TTI in the candidate UE list does not include the UE, then the UE is deleted from the UE list of the corresponding TTI in the candidate UE list.

[0268] Step 4) (optionally): calculate the UE scheduling parameters of each UE. If there is a UE whose UE scheduling parameters are greater than the fairness threshold, delete the UE from the candidate UE list. Specifically, the UE can be deleted from the UE list of the corresponding TTI containing the UE; and / or sort the UEs in order from small to large according to the UE scheduling parameters, and move the UE with relatively large UE scheduling parameters to the end of the queue.

[0269] Step 5): According to step 2) and step 3), using the predicted TTI-level UE scheduling restriction criteria and / or the fairness operation of step 4), UEs that cannot be scheduled in each TTI within a period are screened out, and finally a candidate UE list is generated.

[0270] In the embodiment of the present application, by adding additional operations to the candidate UE list, the candidate UE list no longer contains UEs whose data volume to be transmitted is less than the first threshold value and are in the measurement gap period, that is, the candidate UE list deletes UEs with more scheduling opportunities, avoiding UEs with good channel conditions and less multi-user interference from other UEs from being more frequently selected into the MU pairing recommendation result based on the maximum throughput criterion, so that UEs get scheduling opportunities more fairly, thereby avoiding the phenomenon that some UEs are oversaturated due to excessive resource allocation due to more scheduling opportunities, while some UEs are starved due to too few scheduling opportunities and too few resource allocations. That is, by preliminarily screening out the candidate UE list, the effectiveness of the MU pairing recommendation result is further guaranteed.

[0271] In an embodiment of the present application, an optional implementation manner is provided for step S1022, which may specifically include: determining a MU pairing recommendation result using a first AI network based on a first retransmitting UE, at least one candidate UE, and at least one of the following:

[0272] (1) Predict SINR (Signal to Interference plus Noise Ratio);

[0273] The predicted SINR may come from other xAPPs (applications in the first network node).

[0274] (2) Predict H_SRS (H_Sounding Reference Signal, detection reference signal matrix);

[0275] The predicted H_SRS may also come from other xAPPs.

[0276] (3) Uplink and downlink timeslot configuration information (UD Conf.);

[0277] (4) HARQ related configuration information (HARQ conf.).

[0278] For example, the first network node uses a first AI network (such as a deep learning neural network, but not limited to this) to determine the MU pairing recommendation result for each TTI in the bundle based on the predicted first retransmission UE list (virtual Re-Tx UE 1ist), the candidate UE list (candidate UE list), the predicted SINR and the predicted H_SRS, and finally obtains the MU pairing recommendation result for each TTI within a cycle.

[0279] For example, Figure 3 As shown, the first network node uses a first AI network (such as a deep learning neural network, including at least one hidden layer, but not limited to this) to determine the MU pairing recommendation result, the input layer includes a predicted first retransmission UE list (virtual Re-Tx UE list), a candidate UE list (candidate UE list), a predicted SINR, a predicted H_SRS, and uplink and downlink time slot configuration information (UD Conf.) and HARQ related configuration information (HARQ conf.), and the output layer includes a MU-UE list for each TTI in each bundle (ie, a MU pairing recommendation result), and finally obtains a MU-UE list for each TTI in a cycle (ie, a MU pairing recommendation result).

[0280] In an embodiment of the present application, the first AI network may determine the MU pairing recommendation result according to the maximum throughput criterion.

[0281] In an optional implementation, the first AI network may determine the MU pairing recommendation result based on the goal of maximizing the MU-MPR (Multi-User-Modulation order Product code Rate) and (sumMU-MPR) of at least one UE. That is, the embodiment of the present application also includes the step of training the first AI network to maximize the MU-MPR and of at least one UE. Optionally, the first AI network adopts a greedy search method and uses predetermined label data for training. Maximizing sum MU-MPR is used as a goal to improve the throughput of the system.

[0282] The MU-MPR of each UE is determined based on at least one of the following: the predicted SINR of the UE; and a multi-user beam forming (MU BF) weight determined based on the predicted H_SRS.

[0283] In an optional implementation manner, the MU-MPR calculation of a UEu may be calculated using the following calculation formulas 6 and 7:

[0284]

[0285]

[0286] Among them, W u Indicates the MU BF weight calculated based on the predicted H_SRS; mu_loss u Indicates the MU loss due to multi-user interference; SINR u Represents the predicted SINR of UEu.

[0287] In the embodiment of the present application, the MU pairing recommendation result for each TTI of bundle b can be calculated (as shown in Table 8 below) according to the first retransmission UE list of bundle b (as shown in Table 6 below) and the candidate UE list of bundle b (as shown in Table 7 below).

[0288]

[0289]

[0290] Table 6

[0291] Candidate UE List <![CDATA[t 1 ]]> <![CDATA[t 2 ]]> ... <![CDATA[t N ]]> UE1 X1 UE2 X1 ... X2 ... ... ... ... ... UE L X1

[0292] Table 7

[0293]

[0294] Table 8

[0295] Among them, the specific content of Table 6 can refer to the introduction of Table 4 above, and the specific content of Table 7 can refer to the introduction of Table 5 above, which will not be repeated here.

[0296] In Table 8, Indicates the nth TTI of bundle b with period p; 4 th The layer indicates that in this example, the maximum number of UEs in the MU pairing recommendation result is limited to 4. In the MU pairing recommendation result of bundle b, each UE is selected by the first AI network. 1 Each UE and t 2 UE3 is a retransmitting UE, and the other UEs are initial transmission UEs.

[0297] It should also be noted that the number of layers and corresponding UEs shown in Table 8 are for illustration only, and the specific MU pairing recommendation results are subject to actual implementation. The number of layers and corresponding UE examples in Table 8 should not be understood as limiting the present application. For example, the number of layers used in actual applications may be much greater than 4, and one UE may correspond to multiple layers. For example, the number of UEs corresponding to different TTIs may be different, and some UEs may be associated UEs, etc., which are not limited in the embodiments of the present application.

[0298] Thus, the MU pairing recommendation result of each first time unit (such as TTI) of each bundle can be obtained. The first network node integrates the MU-UE list of each first time unit of all bundles in a period, and in step S103, submits the integrated MU pairing recommendation result to the second network node.

[0299] The technical solution provided in the embodiment of the present application, in order to solve the retransmission delay problem and improve the cell throughput, provides a first network node to determine the MU pairing recommendation result including the retransmitting UE based on the AI ​​network, assists the calculation of the second network node intelligent large-scale MIMO scheduler, and ensures system performance.

[0300] In the embodiment of the present application, considering the computing power (which may be referred to as computing power) of the second network node and the complexity of MU scheduling in the second network node, the gNB (the next Generation Node B) may choose to execute the solution of the present application according to the computing power of the second network node. That is, step S101 may specifically include obtaining the computing power of the second network node; when the computing power is less than a predetermined condition, determining the first retransmitting UE. Among them, the computing power of the second network node can be judged according to the hardware information of the second network node, the historical computing information of the second network node, etc., but is not limited thereto.

[0301] In an optional solution, taking the first network node as RIC and the second network node as DU as an example, the gNB can switch between the present application solution and the existing DU solution according to the DU computing power, such as Figure 4 As shown. If the DU computing power is insufficient, the gNB switches to the solution of the present application, that is, the RIC calculates the MU-UE list (i.e., the MU pairing recommendation result) through the AI ​​network, and submits the MU-UE list to the DU, and the DU performs MU scheduling according to the MU-UE list; wherein, since the RIC determines the first retransmission UE according to the UE's retransmission probability and validity strategy, and determines the MU-UE list according to the first retransmission, it ensures that the first retransmission UE can be included in the MU-UE list. Otherwise, the gNB uses the existing basic MU scheduling solution in the DU.

[0302] Furthermore, after the MU pairing recommendation result generated by the first network node through the AI ​​network is submitted to the second network node, the second network node may also decide whether to use the MU pairing recommendation result provided by the first network node for MU scheduling according to certain principles. The principle is that for each first time unit, if there is no real retransmission UE in the first time unit or there is a real retransmission UE in the first time unit and the real retransmission UE is included (it may be all included, or a certain proportion, or a set number) in the MU-UE list provided by the first network node, the MU pairing recommendation result provided by the first network node is used for MU scheduling. For details, please refer to the introduction of the method executed on the second network node side below.

[0303] The embodiment of the present application also provides a method executed by a second network node in a communication system, such as Figure 5 As shown, the method includes:

[0304] Step S501: receiving a MU pairing recommendation result sent by a first network node;

[0305] Step S502: Determine whether the current scheduling time unit uses the MU pairing recommendation result;

[0306] The currently scheduled time unit may refer to the first time unit of the current schedule, but is not limited thereto.

[0307] In an optional implementation, the second retransmission UE of the current scheduling time unit may be determined based on HARQ (ACK / NACK) feedback, wherein the second retransmission UE refers to the retransmission UE determined by the second network node, and may also be referred to as the actual retransmission UE (actual Re-Tx UE).

[0308] Further, if there is no second retransmitting UE in the current scheduling time unit, or the second retransmitting UE is included in the MU pairing recommendation result corresponding to the current scheduling time unit, it is determined that the current scheduling time unit uses the MU pairing recommendation result provided by the first network node; otherwise, it is determined that the current scheduling time unit does not use the MU pairing recommendation result provided by the first network node.

[0309] Among them, the second retransmission UE is included in the MU pairing recommendation result corresponding to the current scheduling time unit, which may mean that all second retransmission UEs are included in the MU pairing recommendation result, or it may mean that a certain proportion of second retransmission UEs are included in the MU pairing recommendation result, or it may mean that a set number of second retransmission UEs are included in the MU pairing recommendation result. Technical personnel in this field can set it according to actual conditions, and the embodiments of the present application are not limited here.

[0310] Step S503: If yes, MU scheduling is performed based on the MU pairing recommendation result.

[0311] The MU scheduling operation includes but is not limited to MU layer decision and resource allocation process.

[0312] Similarly, the first network node may be a RIC, and the second network node may be a DU, but is not limited thereto.

[0313] As an example, Figure 6 As shown, RIC calculates the MU-UE list (i.e., MU pairing recommendation result) for each TTI in the next cycle based on the AI ​​network, and DU receives the MU-UE list for each TTI in a cycle provided by RIC. The mMIMO scheduler in DU decides whether to use the MU-UE list of the corresponding TTI provided by RIC in each scheduling time unit (such as the corresponding TTI), and performs the corresponding MU layer decision and resource allocation.

[0314] In one example, a DU (i.e., a second network node) operation process may be as follows: Figure 7 , may include the steps of:

[0315] For each scheduling time unit, DU determines the second retransmission UE (real retransmission UE) of the corresponding scheduling time unit based on HARQ ACK / NACK feedback. If there is no second retransmission UE in the time unit, the MU-UE list corresponding to the time unit provided by RIC (i.e., MU pairing recommendation result) is directly used. Otherwise, it is determined whether all the existing second retransmission UEs are included (or replaced by other inclusion conditions) in the MU-UE list corresponding to the time unit provided by RIC. If so, the MU-UE list corresponding to the time unit provided by RIC is directly used. Otherwise, the MU-UE list provided by RIC is not used, and DU generates the MU-UE list of the time unit based on the second retransmission UE. Then the MU layer decision and resource allocation operations are performed to finally complete the MU scheduling.

[0316] Based on at least one of the above embodiments, in the embodiments of the present application, by Figure 8 The complete process of the solution of the present application is shown, wherein the entire MU scheduling is jointly completed by the RIC (first network node) and the DU (second network node), the RIC determines the MU-UE list (MU pairing recommendation result) containing the retransmitting UE, and the DU scheduler completes the MU layer decision and resource allocation according to the MU-UE list. The process mainly includes:

[0317] Step 8-1 (In Figure 8 "8-" is omitted in the figure, and the subsequent steps are the same and will not be repeated): Collect information. Fig. 9 As shown, the gNB includes an array antenna, and transmits data (Transmit) and receives data (Receive) through mMIMO. RIC periodically collects information, including but not limited to receiving UE measurement configuration information (UE meas.Conf., and used to determine candidate UEs), cell configuration information (cell conf.Inf. or Cell basic conf.Inf., including but not limited to uplink and downlink time slot configuration information, HARQ related configuration information, etc., used to determine bundle related information) and historical scheduling information (historicalsch.Inf., including but not limited to scheduled UE list, ACK / NACK feedback, etc., used to calculate historical validity data and used to determine validity strategy), etc., and obtains report data from xAPP, including but not limited to predicted BO (used to generate candidate UEs), predicted SINR (used to generate MU-UE list), predicted H_SRS (used to generate MU-UE list), etc. This step is mainly to collect data reported regularly by gNB DU and RIC to prepare for calculating the MU-UE list.

[0318] Step 8-2: Generate a candidate UE list. The RIC removes the UEs that cannot be scheduled in each TTI (first time unit) according to the predicted BO of each UE and the UE measurement configuration information (UEmeas.Conf.), and generates a candidate UE list for each TTI in a cycle.

[0319] Optionally, considering the fairness of users, additional operations on fairness are given based on step 8-2, that is, the UE scheduling parameters of each UE are calculated, and if there is a UE whose UE scheduling parameters are greater than the fairness threshold, the UE is deleted from the candidate UE list.

[0320] Step 8-3: Generate a MU-UE list (for selecting MU paired UE). Optionally, Fig.10 As shown, the MU-UE list for each TTI in a period is determined according to the following three steps:

[0321] Step 8-3-1: Determine the validity strategy. Set the validity strategy based on historical scheduling information (historical sch.Inf.), where the validity strategy is to determine and continuously optimize the retransmission validity threshold (Th_vaild b ) to ensure that the MU-UE list provided by RIC can include the actual retransmission UE (actual Re-TxUE, i.e., the second retransmission UE) determined by DU.

[0322] Step 8-3-2: Determine the virtual Re-Tx UE list (i.e., the first re-transmission UE list). According to the last generated MU-UE list and the target BLER, calculate the UE retransmission probability, and then predict the virtual re-transmission UE list according to the validity strategy to ensure that the re-transmission UE can be included in the MU-UE list.

[0323] Step 8-3-3: Determine the MU-UE list. Based on the predicted virtual Re-Tx UE list, candidate UE list, predicted SINR and predicted H_SRS, RIC uses a deep learning neural network to determine the MU-UE list according to the maximum throughput criterion, and finally obtains the MU-UE list for each TTI within a cycle.

[0324] It should be noted that the above-mentioned period refers to the time interval for information interaction between RIC and DU, which is usually tens or hundreds of milliseconds. In order for RIC to predict the retransmitting UE in real time and provide a more effective MU pairing result, the embodiment of the present application divides the entire period into multiple bundles. According to the transmission timing relationship between data and HARQ ACK / NACK feedback, the TTIs within a period are clustered to obtain individual bundles. The HARQ ACK / NACK of each TTI in each bundle is fed back in the same continuous uplink time slot. Therefore, steps 8-3-2 and 8-3-3 can be processes that are executed cyclically for each bundle within a period. In addition, in order to ensure that all required retransmitting UEs in each bundle are retransmitted in the period, the embodiment of the present application sets the validity policy in step 8-3-1 for the bundle.

[0325] In practical applications, step 8-3-2 and step 8-3-3 can be specific implementation steps of RTM-Net (Maximum Throughput with Restriction Network). RTM-Net is a semi-supervised clustering machine learning interwoven with a deep learning neural network to maximize MU-MPR, and is used to determine the MU-UE list for each TTI in each bundle according to the maximum throughput criterion, through which the retransmitting UE can be included.

[0326] Through the above steps, the RIC can eventually generate the MU-UE list for each TTI in the next cycle and submit it to the DU.

[0327] The scheduler in the DU decides whether to use the MU-UE list provided by the RIC and performs the MU layer decision and resource allocation process, thus completing the MU scheduling.

[0328] In the embodiment of the present application, the DU (second network node) operates based on the MU-UE list (MU pairing recommendation result) as follows: Fig.11As shown, due to the delay between DU and near-real time (Near-RT) RIC (first network node) (usually tens or hundreds of milliseconds of DU-RIC communication cycle), RIC cannot obtain HARQ ACK / NACK information in time, and once UE feedbacks NACK, DU should immediately arrange retransmission, and DU uses the MU list provided by RIC to assist in scheduling, which can be achieved in the following way: For each scheduling time unit, DU determines the real retransmission UE based on ACK / NACK feedback. If there is no real retransmission UE or all the real retransmission UEs are included in the MU-UE list provided by RIC, the MU-UE list provided by RIC is used directly, otherwise, DU regenerates the MU-UE list based on the real retransmission UE.

[0329] As an example, assume that the MU-UE lists of bundle b and bundle b+1 provided by the RIC are as shown in Table 9:

[0330]

[0331] Table 9

[0332] in, Indicates the nth TTI of bundle b with period p; 4 th The layer indicates that in this example, the maximum number of UEs in the MU-UE list is limited to 4. For other details not detailed, please refer to the introduction to Table 8 above, which will not be repeated here.

[0333] The situation where DU uses the MU list provided by RIC to assist in scheduling is shown in Table 10:

[0334]

[0335] Table 10

[0336] Among them, for case a), in bundle b's t 1 , UE1 and UE2 have NACKs for historical uplink time slots and should be retransmitted. UE1 and UE2 are in the MU-UE list provided by the RIC, so the DU can use this MU-UE list to perform MU layer decisions.

[0337] For case b), in bundle b's t N , there is no NACK feedback. DU can directly use the MU-UE list provided by RIC to perform MU layer decision.

[0338] For case c), at t in bundle b+1 1UE6 and UE1 have NACKs for the historical uplink time slots and should be retransmitted. UE6 is not in the MU-UE list provided by RIC, so DU generates a bundle b+1 based on UE6 and UE1. 1 The MU-UE list is then used to perform MU layer decisions.

[0339] It can be understood that the specific scheduling situations of the several situations shown in Table 10 are for illustration only. The specific NACK / ACK feedback situation, the MU-UE list provided by the RIC, and the MU-UE list input to the DU MU layer decision shall be subject to the actual implementation. The scheduling example in Table 10 should not be understood as a limitation on the present application.

[0340] It should be noted that the MU layer decision and resource allocation process can adopt the existing solution, and the embodiments of the present application will not be repeated here. Because the first network node fully considers multi-user interference when generating the MU pairing recommendation result, the second network node performs the MU layer decision operation based on the MU pairing recommendation result provided by the first network node, and there is no risk of performance loss.

[0341] In an embodiment of the present application, RTM-Net is used to determine the MU pairing recommendation result (such as MU-UE list) of each first time unit (such as TTI) of each bundle, and maximize the throughput (such as the total MU-MPR sum), and the MU pairing recommendation result may include a retransmitting UE.

[0342] like Fig.12 As shown in the figure, the processing flow of RTM-Net includes:

[0343] Module 3-1( Fig.12 The "module" corresponds to Figure 8 The retransmission validity threshold and the MU-UE list of bundle b-1 generated in "step 8-", the same content will not be repeated below) are used to generate a virtual retransmission UE list of bundle b in module 3-2. Among them, the virtual retransmission UE list of bundle b, Indicates the TTI t virtual retransmission UE list of bundle b with period p. The length is the number of virtual retransmission UEs in the TTI, and the value is the UE index number.

[0344] The candidate UE list generated by module 2 and the virtual retransmission UE list of bundle b generated in module 3-2 are used to generate the MU-UE list in module 3-3, and the input of module 3-3 also includes the predicted H_SRS from other xAPPs for calculating multi-user interference, and the predicted SINR from other xAPPs for calculating MU MPR and, module 3-3 uses a deep learning neural network to generate the MU-UE list of bundle b, with the aim of maximizing throughput. Among them, in the MU-UE list of bundle b, Indicates the MU-UE list of TTI t of bundle b with period p. The length is the maximum number of UEs and the value is the UE index number.

[0345] Based on at least one of the above embodiments, see Fig.13 , shows an example of deployment of an embodiment of the present application in RIC and gNB, mainly including:

[0346] Step S13-1 (in Fig.13 "13-" is omitted in the figure, and the subsequent steps are the same and will not be repeated): Near-RT RIC collects data to generate a MU-UE list, including but not limited to receiving UE measurement configuration information (UE meas.Conf.), cell configuration information (cell conf.Inf.) and historical scheduling information (historical sch.Inf.) from the gNB;

[0347] Step S13-2: Near-RT RIC generates a candidate UE list for each TTI in a period, excluding UEs that cannot be scheduled due to no services and UEs in the corresponding TTI within the measurement interval;

[0348] Step S13-3-1: Near-RT RIC determines a validity policy for a period based on historical data and sets the validity policy (ie, retransmission validity threshold Th_valid) to ensure that the retransmitting UE is included in the MU-UE list.

[0349] Step S13-3-2: Near-RT RIC determines the virtual retransmission UE list for each TTI of the current bundle based on the MU-UE list in the previous bundle, the predicted BO from other xAPPs, and the retransmission validity threshold Th_vaild;

[0350] Step S13-3-3: Near-RT RIC uses RTM-Net to determine the MU-UE list with maximum throughput for each TTI of each bundle based on the virtual retransmission UE list, the candidate UE list, the predicted SINR from other xAPPs, and the predicted H_SRS from other xAPPs for each bundle. Based on the MU-UE list for each TTI of each bundle, a MU-UE list of periods is generated;

[0351] Step S13-4: The DU of the gNB uses the MU-UE list provided by the RIC. When the TTI meets the conditions, the scheduler makes a MU layer decision on the TTI and allocates resources. The MU scheduling result is output to the MMU. The conditions include: 1) the real retransmission UE does not exist; or 2) all real retransmission UEs are included in the MU-UE list provided by the RIC (if there are real retransmission UEs).

[0352] The technology involved in the embodiments of the present application can be applied to large-scale array antenna scenarios to perform MU pairing for UEs.

[0353] In the embodiment of the present application, the retransmission UE is determined in the first network node by predicting the retransmission probability and validity strategy of the UE, thereby ensuring that the retransmission UE is included in the MU pairing recommendation result as much as possible during the actual scheduling of the second network node, thereby ensuring the validity of the MU pairing recommendation result. Based on the predicted retransmission UE and the channel state information (SINR, H_SRS, etc.) predicted for each UE, the MU pairing recommendation result is determined using a deep learning neural network according to the criterion of maximizing throughput, thereby improving the cell throughput. In short, the complexity of MU scheduling of the second network node is reduced through the solution of the present application, while maintaining good performance (try to ensure that the retransmission UE can be included in the validity of the MU pairing recommendation result and maximize the cell throughput).

[0354] After experimental verification by the inventors of this application, the solution of this application provides a more reasonable MU-UE list (MU pairing recommendation result) to the DU scheduler through RIC, which has a gain of about 22% compared with the simplified mMIMO scheduler solution. In addition, after a certain number of iterations, the solution of this application can accurately predict the retransmitting UE and finally achieve the expected 100% effectiveness.

[0355] It can be seen that the present application solution can maintain system performance by executing part of the mMIMO scheduler function in the RIC. In addition, the present application solution solves the retransmission delay problem by predicting the retransmission UE list in the RIC, and ensures that the retransmission UE is scheduled in time by enhancing the DU operation.

[0356] like Fig.14As shown, the intelligent mMIMO solution based on RIC (first network node) provided in the embodiment of the present application is equivalent to moving a part of the basic mMIMO scheduler in the existing DU (second network node) to the RIC side, forming a scheduling scheme of the RIC joint DU mMIMO scheduler. The modules with large calculations are decoupled through RIC, which reduces the complexity of the DU, that is, by executing some mMIMO scheduler functions in the RIC, the complexity of the DU is greatly reduced while maintaining performance.

[0357] In an embodiment of the present application, a first network node (such as RIC) is provided, including a transceiver configured to send and receive signals; and a processor coupled to the transceiver and configured to implement the steps of the aforementioned first network node method embodiments.

[0358] An embodiment of the present application also provides a second network node (such as DU), including a transceiver configured to send and receive signals; and a processor coupled to the transceiver and configured to implement the steps of the aforementioned second network node method embodiments.

[0359] Among them, the detailed functional description of the first network node and the second network node and the beneficial effects produced can be specifically referred to the description of each method embodiment in the above text, and will not be repeated here.

[0360] An embodiment of the present application also provides an electronic device, which includes a processor and, optionally, may also include a transceiver and / or a memory coupled to the processor, and the processor is configured to execute the steps of the method provided in any optional embodiment of the present application.

[0361] Fig.15 A schematic diagram of the structure of an electronic device applicable to an embodiment of the present invention is shown in FIG. Fig.15 As shown, Fig.15 The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present application. Optionally, the electronic device may be a first network node, a second network node, or a third network node.

[0362] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0363] The bus 4002 may include a path to transmit information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.15 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0364] The memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store computer programs and can be read by a computer, without limitation herein.

[0365] The memory 4003 is used to store the computer program for executing the embodiment of the present application, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the computer program stored in the memory 4003 to implement the steps shown in the above method embodiment.

[0366] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps and corresponding contents of the aforementioned method embodiment can be implemented.

[0367] The embodiment of the present application also provides a computer program product, including a computer program, which can implement the steps and corresponding contents of the aforementioned method embodiment when executed by a processor.

[0368] The terms "first", "second", "third", "fourth", "1", "2", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than that shown or described in the drawings.

[0369] It should be understood that, although each operation step is indicated by arrows in the flowchart of the embodiment of the present application, the implementation order of these steps is not limited to the order indicated by the arrows. Unless clearly stated herein, in some implementation scenarios of the embodiment of the present application, the implementation steps in each flowchart can be performed in other orders according to demand. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages based on actual implementation scenarios. Some or all of these sub-steps or stages may be executed at the same time, and each sub-step or stage in these sub-steps or stages may also be executed at different times respectively. In different scenarios of execution time, the execution order of these sub-steps or stages may be flexibly configured according to demand, and the embodiment of the present application does not limit this.

[0370] The above text and drawings are provided only as examples to help readers understand the present application. They are not intended to and should not be interpreted as limiting the scope of the present application in any way. Although certain embodiments and examples have been provided, based on the contents disclosed herein, it is obvious to those skilled in the art that the embodiments and examples shown can be changed without departing from the scope of the present application, and other similar implementation means based on the technical ideas of the present application are adopted, which also fall within the protection scope of the embodiments of the present application.

Claims

1. A method performed by a first network node in a communication system, characterized in that: include: Determine a first retransmission user equipment UE; Based on the first retransmitting UE, using a first artificial intelligence AI network, determining a multi-user MU pairing recommendation result; The MU pairing recommendation result is sent to the second network node.

2. The method according to claim 1, characterized in that The determining a first retransmission UE includes: determining a retransmission probability of at least one UE; Based on the retransmission probability of at least one UE, a first retransmission UE is determined.

3. The method according to claim 2, characterized in that Determining a retransmission probability of at least one UE includes: A retransmission probability of at least one UE is determined according to the historical MU pairing recommendation result and the target block error rate.

4. The method according to claim 3, characterized in that The determining, according to the historical MU pairing recommendation result and the target block error rate, a retransmission probability of at least one UE includes: Determine a number of times at least one UE is used as an initial transmission in the historical MU pairing recommendation result; determining a retransmission probability of an initial transmission based on the target block error rate; Based on the number of initial transmissions and the retransmission probability of the initial transmissions, a retransmission probability of at least one UE is determined.

5. The method according to any one of claims 2 to 4, characterized in that: The determining, based on the retransmission probability of at least one UE, a first retransmission UE includes: Determining a retransmission validity threshold corresponding to at least one UE; A first retransmission UE is determined based on the retransmission probability of at least one UE and a retransmission validity threshold corresponding to the at least one UE.

6. The method according to claim 5, characterized in that The determining a retransmission validity threshold corresponding to at least one UE includes: A retransmission validity threshold corresponding to at least one UE is determined according to the historical scheduling information.

7. The method according to claim 6, characterized in that The determining, according to the historical scheduling information, a retransmission validity threshold corresponding to at least one UE includes: Determine, according to the historical scheduling information, a validity index of a historical MU pairing recommendation result, where the validity index of the MU pairing recommendation result is used to measure the validity of a retransmitting UE in the historical MU pairing recommendation result; According to the historical MU pairing recommendation result validity indicator, using the second AI network, determine the retransmission validity threshold corresponding to at least one UE in each first time unit set in the current period.

8. The method according to claim 7, characterized in that Each of the first time unit sets includes at least one first time unit for performing hybrid automatic repeat request HARQ feedback in the same continuous uplink second time unit.

9. The method according to any one of claims 6 to 8, characterized in that: The historical scheduling information includes at least one of the following: The first time unit using multi-user MU scheduling; The second network node determines a first time unit for the MU pairing recommendation result; The second network node uses a first time unit of the MU pairing recommendation result provided by the first network node; actual usage of the MU pairing recommendation result provided by the first network node by the second network node within each first time unit set of the historical period; The number of first time units in each set of first time units in the historical period in which the second network node uses MU scheduling.

10. The method according to any one of claims 6 to 9, characterized in that: Determining a first retransmitting UE based on a retransmission probability of at least one UE and a retransmission validity threshold corresponding to the at least one UE includes: Based on a comparison result of the retransmission probability of at least one UE and a retransmission validity threshold corresponding to at least one UE, a first retransmission UE is determined.

11. The method according to claim 10, characterized in that Determining a first retransmitting UE based on a comparison result of a retransmission probability of at least one UE and a retransmission validity threshold corresponding to at least one UE includes: Sort at least one UE according to the retransmission probability from low to high; For at least one ranked UE, the following operations are performed in sequence until a target UE is determined: determining a cumulative retransmission probability of the UE and its previous UEs, determining a corresponding retransmission validity probability based on the cumulative retransmission probability, and if the retransmission validity probability is less than the retransmission validity threshold, determining the UE as a target UE; The target UE and subsequent UEs are determined as first retransmitting UEs.

12. The method according to any one of claims 1 to 11, characterized in that: The determining, based on the first retransmitting UE, a MU pairing recommendation result using a first AI network includes: Determine at least one candidate UE according to the amount of data to be transmitted by the UE and the measurement configuration information; Based on the first retransmitting UE and the at least one candidate UE, using a first AI network, determine an MU pairing recommendation result.

13. The method according to claim 12, characterized in that Determining at least one candidate UE according to the amount of data to be transmitted by the UE and the measurement configuration information includes: Determine at least one candidate UE according to the amount of data to be transmitted by the UE, the measurement configuration information, and the UE scheduling parameter; The UE scheduling parameter is determined according to the number of first time units that meet a preset condition within a preset time period including multiple first time units.

14. The method according to claim 12, characterized in that The determining, according to the amount of data to be transmitted of the UE and the measurement configuration information, at least one candidate UE comprises: Determine initial candidate UEs; The candidate UEs whose amount of data to be transmitted is less than the first threshold value are deleted from the initial candidate UEs, and the candidate UEs in the measurement interval are deleted from the initial candidate UEs according to the measurement configuration information to obtain at least one candidate UE.

15. The method according to claim 13, characterized in that Determining at least one candidate UE according to the amount of data to be transmitted by the UE, the measurement configuration information, and the UE scheduling parameter, including: Determine initial candidate UEs; The candidate UEs whose amount of data to be transmitted is less than the first threshold value are deleted from the initial candidate UEs, and the candidate UEs in the measurement interval are deleted from the initial candidate UEs according to the measurement configuration information, and the initial candidate UEs are sorted according to the UE scheduling parameters and / or the candidate UEs whose UE scheduling parameters are greater than the fairness threshold are deleted from the initial candidate UEs to obtain at least one candidate UE.

16. A method performed by a second network node in a communication system, characterized in that: include: Receiving a multi-user MU pairing recommendation result sent by the first network node; Determine whether the current scheduling time unit uses the MU pairing recommendation result; If yes, MU scheduling is performed based on the MU pairing recommendation result.

17. A first network node, characterized in that: include: a transceiver configured to transmit and receive signals; as well as A processor is coupled to the transceiver and configured to execute the method according to any one of claims 1 to 15.

18. A second network node, characterized in that: include: a transceiver configured to transmit and receive signals; as well as A processor is coupled to the transceiver and configured to execute the method of claim 16.

19. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 16 is implemented.

20. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to claims 1 to 16 is implemented.

Images