Wireless communication methods and apparatuses, and device, chip and storage medium

Abstract

Provided in an embodiment of the present application is a wireless communication method. The method comprises: on the basis of a first AI model, a first device determining a retransmission decision corresponding to first information; and when the retransmission decision is to perform retransmission, the first device sending retransmitted data to a second device.

Description

A wireless communication method, apparatus, device, chip, and storage medium Technical Field

[0001] This application relates to the field of communication technology, specifically to a wireless communication method, apparatus, device, chip, and storage medium. Background Technology

[0002] In wireless communication systems, to overcome the impact of time-varying characteristics of wireless channels and multipath fading on signal transmission, Hybrid Automatic Repeat Request (HARQ) technology is commonly used. HARQ is a technique that combines Forward Error Correction (FEC) and ARQ methods. FEC adds redundant information, enabling the receiver to correct some errors, thereby reducing the number of retransmissions. For errors that FEC cannot correct, the receiver requests the transmitter to retransmit the data according to the ARQ mechanism. Current HARQ retransmissions can be divided into two categories: active retransmission and passive retransmission.

[0003] However, while proactive retransmission can help reduce service latency, it also leads to higher resource waste / consumption; while passive retransmission can avoid unnecessary resource consumption, feedback-based retransmission results in higher latency.

[0004] Summary of the Invention

[0005] This application provides a wireless communication method, apparatus, device, chip, and storage medium.

[0006] In a first aspect, embodiments of this application provide a wireless communication method, the method comprising: a first device determining a retransmission decision corresponding to first information based on a first artificial intelligence (AI) model; and, if the retransmission decision is to retransmit, the first device sending retransmitted data to a second device.

[0007] Secondly, embodiments of this application provide a wireless communication method, the method comprising: a second device determining feedback information and / or first indication information corresponding to fourth information based on a third AI model; wherein the feedback information is feedback information for data transmitted by a first device; the first indication information is used to indicate whether to retransmit the data; and the second device sending the feedback information and / or the first indication information to the first device.

[0008] Thirdly, embodiments of this application provide a wireless communication device, the device comprising: a first determining unit configured to determine a retransmission decision corresponding to first information based on a first AI model; and a first communication unit configured to send retransmission data to a second device when the retransmission decision is to retransmit.

[0009] Fourthly, embodiments of this application provide a wireless communication device, the device comprising: a second determining unit configured to determine feedback information and / or first indication information corresponding to fourth information based on a third AI model; wherein the feedback information is feedback information for data transmitted by a first device; the first indication information is used to indicate whether to retransmit the transmitted data; and a second communication unit configured to send the feedback information and / or the first indication information to the first device.

[0010] Fifthly, embodiments of this application provide a communication device, including: a memory for storing a computer program; a processor connected to the memory for calling and running the computer program from the memory to implement the method described in the first or second aspect; and a transceiver for receiving and sending information during the process of sending and receiving information with other devices.

[0011] Sixthly, embodiments of this application provide a chip. The chip includes: a processor for retrieving and running a computer program from a memory, causing a device on which the chip is installed to perform the method described in the first or second aspect; and a transceiver for receiving and sending information during the exchange of information with the device or the chip.

[0012] In a seventh aspect, embodiments of this application provide a computer-readable storage medium for storing a computer program that causes a computer to perform the methods described in the first or second aspect.

[0013] Eighthly, embodiments of this application provide a computer program product including computer program instructions that cause a computer to perform the method described in the first or second aspect.

[0014] Ninthly, embodiments of this application provide a computer program that, when run on a computer, causes the computer to perform the method described in the first or second aspect.

[0015] According to the method of this application embodiment, the first device determines the retransmission decision corresponding to the first information based on the first AI model; if the retransmission decision is to retransmit, the first device sends retransmission data to the second device; thus, by using the intelligence of the AI ​​model to determine the retransmission decision corresponding to the first information, a retransmission decision that is more consistent with and appropriate to the first information can be obtained, thereby helping to save transmission resource overhead. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0017] Figure 1 is a schematic diagram of an application scenario of an embodiment of this application;

[0018] Figure 2 is a flowchart illustrating the wireless communication method provided in an embodiment of this application;

[0019] Figure 3 is a schematic flowchart of the wireless communication method provided in an embodiment of this application;

[0020] Figure 4 is a schematic flowchart of the wireless communication method provided in an embodiment of this application;

[0021] Figure 5 is a flowchart illustrating the wireless communication method provided in an embodiment of this application.

[0022] Figure 6 is a flowchart illustrating the wireless communication method provided in an embodiment of this application.

[0023] Figure 7 is a flowchart illustrating the wireless communication method provided in an embodiment of this application.

[0024] Figure 8 is a flowchart illustrating the wireless communication method provided in an embodiment of this application.

[0025] Figure 9 is a flowchart illustrating the wireless communication method provided in an embodiment of this application.

[0026] Figure 10 is a flowchart illustrating the wireless communication method provided in an embodiment of this application.

[0027] Figure 11 is a flowchart illustrating the wireless communication method provided in an embodiment of this application;

[0028] Figure 12 is a schematic diagram of the structural composition of the wireless communication device provided in an embodiment of this application;

[0029] Figure 13 is a schematic diagram of the structure of the wireless communication device provided in an embodiment of this application;

[0030] Figure 14 is a schematic structural diagram of a communication device provided in an embodiment of this application;

[0031] Figure 15 is a schematic structural diagram of a chip according to an embodiment of this application;

[0032] Figure 16 is a schematic block diagram of a communication system provided in an embodiment of this application. Detailed Implementation

[0033] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0034] Figure 1 is a schematic diagram of an application scenario of an embodiment of this application.

[0035] As shown in Figure 1, the communication system 100 may include a terminal device 110 and a network device 120. The network device 120 can communicate with the terminal device 110 via an air interface. Multi-service transmission is supported between the terminal device 110 and the network device 120.

[0036] It should be understood that the embodiments of this application are only illustrated by way of example with communication system 100, but the embodiments of this application are not limited thereto. That is to say, the technical solutions of the embodiments of this application can be applied to various communication systems, such as: Long Term Evolution (LTE) system, LTE Time Division Duplex (TDD), Universal Mobile Telecommunication System (UMTS), Internet of Things (IoT) system, Narrow Band Internet of Things (NB-IoT) system, enhanced Machine-Type Communications (eMTC) system, 5G communication system (also known as New Radio (NR) communication system), 6G communication system, or future communication systems, etc.

[0037] In the communication system 100 shown in Figure 1, network device 120 may be an access network device that communicates with terminal device 110. The access network device can provide communication coverage for a specific geographical area and can communicate with terminal device 110 (e.g., UE) located within that coverage area.

[0038] Network device 120 may be an evolved Node B (eNB or eNodeB) in a Long Term Evolution (LTE) system, or a Next Generation Radio Access Network (NG RAN) device, or a base station (gNB) in an NR system, or a base station in a 6G system, or a radio controller in a Cloud Radio Access Network (CRAN), or the network device 120 may be a relay station, access point, vehicle-mounted equipment, wearable device, hub, switch, bridge, router, or network equipment in a future evolved Public Land Mobile Network (PLMN), etc.

[0039] Terminal device 110 can be any terminal device, including but not limited to terminal devices that are connected to network device 120 or other terminal devices via wired or wireless connections.

[0040] For example, terminal equipment 110 can refer to an access terminal, user equipment (UE), user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication equipment, user agent, or user device. Access terminals can be cellular phones, cordless phones, Session Initiation Protocol (SIP) phones, IoT devices, satellite handheld terminals, Wireless Local Loop (WLL) stations, Personal Digital Assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to a wireless modem, in-vehicle devices, wearable devices, terminal equipment in 5G networks, terminal equipment in 6G networks, or terminal equipment in future evolved networks, etc.

[0041] Terminal device 110 can be used for device-to-device (D2D) communication.

[0042] The communication system 100 may further include a core network device 130 that communicates with the network device 120. This core network device 130 may be a 5G core network (5G Core, 5GC) device, such as an Access and Mobility Management Function (AMF), an Authentication Server Function (AUSF), a User Plane Function (UPF), or a Session Management Function (SMF). In some embodiments, the core network device 130 may also be an Evolved Packet Core (EPC) device for an LTE network, such as a Session Management Function + Core Packet Gateway (SMF+PGW-C) device. It should be understood that SMF+PGW-C can simultaneously implement the functions of both SMF and PGW-C. During network evolution, the aforementioned core network device may also be called by other names, or new network entities may be formed by dividing the core network functions; this embodiment does not limit this.

[0043] The various functional units in the communication system 100 can also establish connections and communicate with each other through the next generation (NG) interface.

[0044] For example, terminal devices establish air interface connections with access network devices through the NR interface for transmitting user plane data and control plane signaling; terminal devices can establish control plane signaling connections with the AMF through NG interface 1 (N1); access network devices, such as next-generation radio access base stations (gNB), can establish user plane data connections with the UPF through NG interface 3 (N3); access network devices can establish control plane signaling connections with the AMF through NG interface 2 (N2); the UPF can establish control plane signaling connections with the SMF through NG interface 4 (N4); the UPF can interact with the data network for user plane data through NG interface 6 (N6); the AMF can establish control plane signaling connections with the SMF through NG interface 11 (N11); and the SMF can establish control plane signaling connections with the PCF through NG interface 7 (N7).

[0045] Figure 1 exemplarily illustrates a network device, a core network device, and two terminal devices. Optionally, the communication system 100 may include multiple network devices, and each network device may include other numbers of terminal devices within its coverage area. This application embodiment does not limit this.

[0046] It should be noted that Figure 1 is merely an example illustrating the system to which this application applies. Of course, the method shown in the embodiments of this application can also be applied to other systems. Furthermore, the terms "system" and "network" are often used interchangeably in this document. The term "and / or" in this document merely describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship. It should also be understood that "instruction" mentioned in the embodiments of this application can be a direct instruction, an indirect instruction, or an indication of a related relationship. For example, A instructing B can mean that A directly instructs B, for example, B can be obtained through A; it can also mean that A indirectly instructs B, for example, A instructs C, B can be obtained through C; or it can mean that there is a related relationship between A and B. It should also be understood that "correspondence" mentioned in the embodiments of this application can indicate a direct or indirect correspondence between two things, or an related relationship between two things, or a relationship of instruction and being instructed, configuration and being configured, etc. It should also be understood that the "predefined" or "predefined rules" mentioned in the embodiments of this application can be implemented by pre-storing corresponding codes, tables, or other means that can be used to indicate relevant information in the device (e.g., including terminal devices and network devices), and this application does not limit the specific implementation method. For example, predefined can refer to those defined in a protocol. It should also be understood that in the embodiments of this application, the "protocol" can refer to standard protocols in the field of communication, such as LTE protocol, NR protocol, and related protocols applied to future communication systems, and this application does not limit this.

[0047] To facilitate understanding of the technical solutions of the embodiments of this application, the relevant technologies of the embodiments of this application are described below. The following relevant technologies are optional solutions and can be combined with the technical solutions of the embodiments of this application in any way, and they all fall within the protection scope of the embodiments of this application.

[0048] 1. MAC HARQ

[0049] HARQ is a technique that combines FEC and ARQ methods. FEC adds redundant information, enabling the receiver to correct some errors and thus reduce the number of retransmissions. For errors that FEC cannot correct, the receiver requests the sender to retransmit the data according to the ARQ mechanism. As shown in Table 1, current HARQ retransmissions can be divided into two categories:

[0050] Table 1 HARQ Mechanism Analysis

[0051] 2. RLC ARQ

[0052] Acknowledgement Mode (AM) is the primary operating mode for Downlink Shared Channel (DL-SCH) and Uplink Shared Channel (UL-SCH). It supports segmentation, deduplication, and retransmission of erroneous data. The AM entity's transmitter supports retransmission of RLC SDUs or RLC SDU segments (ARQ): RLC PDUs sent to the MAC layer are simultaneously placed in a retransmission buffer to support potential retransmissions. If an ACK corresponding to the entire RLC SDU is received, the corresponding RLC PDU is removed from the retransmission buffer; if a NACK corresponding to the entire or partial RLC SDU is received, the entire or partial SDU is retransmitted. As shown in Table 2, current HARQ retransmissions can be divided into two categories:

[0053] Table 2 Analysis of ARQ Mechanism

[0054] 3. PDCP ARQ

[0055] PDCP retransmission is currently limited to mobile processes. On one hand, the sending end actively retransmits, and on the other hand, the receiving end avoids unnecessary retransmissions through SR. Overall, its mechanism can be compared to RLC ARQ.

[0056] PDCP Duplication: PDCP duplication is currently more based on active type mechanisms.

[0057] Table 3 Analysis of PDCP duplication mechanism

[0058] 4. Artificial Intelligence / Machine Learning

[0059] In Release 18, 3GPP explored how artificial intelligence / machine learning can enhance the functionality and performance of the 5G air interface. This research was driven by the following three use cases:

[0060] (1) Positioning;

[0061] (2) Beam management;

[0062] (3) Channel status information reporting.

[0063] In addition, it includes some research on the lifecycle management of artificial intelligence, including performance testing and evaluation.

[0064] Based on the findings of Release 18, Release 19 will define a comprehensive framework for the application of artificial intelligence / machine learning in the air interface and will standardize related use cases such as positioning and beam management. Furthermore, 3GPP will continue to explore the availability of artificial intelligence / machine learning in other use cases, such as mobility.

[0065] In this embodiment of the application, an AI-assisted retransmission mechanism is established, and the specific implementation includes:

[0066] - For uplink (or downlink), optimize the active retransmission mechanism based on the UE (or network) one-sided model;

[0067] - For uplink (or downlink), optimize the passive retransmission mechanism based on the network (or UE) one-sided model;

[0068] - For uplink (or downlink), optimize the active + passive retransmission mechanism based on the dual-side model of UE and network.

[0069] The following text does not distinguish between one-sided and two-sided models, and describes the uplink and downlink schemes.

[0070] This application provides a wireless communication method, apparatus, device, chip, and storage medium. Regarding the active retransmission mechanism, in this method, a first device determines a retransmission decision corresponding to first information based on a first AI model; if the retransmission decision is to retransmit, the first device sends retransmitted data to a second device. Thus, by utilizing the intelligence of the AI ​​model to determine the retransmission decision corresponding to the first information, a retransmission decision that is more consistent and appropriate to the first information can be obtained, thereby helping to save transmission resource overhead.

[0071] Regarding the passive retransmission mechanism, in this method, the second device determines the feedback information and / or the first indication information corresponding to the fourth information based on the third AI model; the second device sends the feedback information and / or the first indication information to the first device; wherein, the feedback information is feedback information for the data sent by the first device; the first indication information is used to indicate whether to retransmit the data; in this way, by utilizing the intelligence of the AI ​​model, feedback information and / or the first indication information that are more consistent with the fourth information can be obtained, thereby helping to reduce service transmission latency.

[0072] To facilitate understanding of the technical solutions of the embodiments of this application, the technical solutions of this application are described in detail below through specific embodiments. The above-mentioned related technologies are optional solutions and can be arbitrarily combined with the technical solutions of the embodiments of this application, all of which fall within the protection scope of the embodiments of this application. The embodiments of this application include at least some of the following contents.

[0073] Figure 2 is a flowchart illustrating a wireless communication method provided in an embodiment of this application. As shown in Figure 2, the method may include the following steps:

[0074] S201, the first device determines the retransmission decision corresponding to the first information based on the first AI model;

[0075] S202, if the retransmission is determined to be a retransmission, the first device sends the retransmission data to the second device.

[0076] In other embodiments, if the retransmission decision is not to retransmit, or the number of retransmissions in the retransmission decision is 0, the first device sends new data to the second device.

[0077] It is understood that the wireless communication method shown in Figure 2 is an active retransmission mechanism, which is applicable to both downlink and uplink data transmission.

[0078] In some embodiments, the retransmission decision includes whether to retransmit and / or the number of retransmissions.

[0079] For example, in some embodiments, the retransmission decision includes one or more of the following decisions:

[0080] A. The decision to initiate retransmission using HARQ at the MAC layer;

[0081] B. Decision on ARQ active retransmission at the RLC layer;

[0082] C.PDCP ARQ / duplication decision.

[0083] In this embodiment, the first information is not limited; it refers to information that affects the determination of the retransmission decision, i.e., information related to the retransmission decision. In some embodiments, the first information is QoS-related information. Further, in some embodiments, the first information includes one or more of the following:

[0084] (1) QoS requirements related information;

[0085] (2) Channel measurement information;

[0086] (3) Caching related information;

[0087] (4) Second information sent by the second device; wherein the second information includes one or more of the following: feedback information on the data sent by the first device, time-related information on the transmission of the feedback information, first indication information on the data sent by the first device, and time-related information on the transmission of the first indication information; the first indication information is used to indicate whether to retransmit the data.

[0088] (5) QoS performance;

[0089] (6) Third information; wherein the third information is information determined based on one or more of the second information, QoS performance, QoS requirement information, channel measurement information and buffer information.

[0090] In some embodiments, QoS requirement-related information includes terminal-side QoS requirement-related information and / or network-side QoS requirement-related information.

[0091] For example, in an uplink-oriented active retransmission mechanism, the first device is a terminal device and the second device is a network device. The QoS requirements related to the terminal side may include the uplink service pattern; the QoS requirements related to the network side may include latency and / or rate, etc.

[0092] For example, in a downlink-oriented active retransmission mechanism, the first device is a network device and the second device is a terminal device. The QoS requirements related to the terminal side may include latency and / or rate, etc.; the QoS requirements related to the network side may include downlink service patterns.

[0093] In some embodiments, the channel measurement information includes uplink channel measurement information and / or downlink channel measurement information.

[0094] In this embodiment of the application, the QoS performance in the first information can be understood as QoS performance-related information, which may include terminal-side QoS performance (i.e., terminal-side QoS performance-related information) and / or network-side QoS performance (i.e., network-side QoS performance-related information).

[0095] In some embodiments, the QoS performance in the first information includes QoS performance generated based on the second information (such as terminal-side and / or network-side QoS performance), and / or QoS performance generated based on retransmission decisions (such as terminal-side and / or network-side QoS performance).

[0096] For example, in an uplink-oriented active retransmission mechanism, the first device is a terminal device and the second device is a network device. The QoS performance on the terminal side may include one or more of power consumption and resource consumption; the QoS performance on the network side may include latency and / or rate, etc.

[0097] For example, in a downlink-oriented active retransmission mechanism, the first device is a network device and the second device is a terminal device. The QoS performance on the terminal side may include latency and / or rate, etc.; the QoS performance on the network side may include one or more of power consumption and resource consumption.

[0098] It should be noted that the QoS performance (such as terminal-side and / or network-side QoS performance) generated based on the second information can be information obtained by the second device based on the third AI model, or information obtained based on other methods (such as retransmission mechanisms such as MAC HARQ, RLC ARQ, and / or PDCP ARQ mentioned in the related technologies above). Similarly, the QoS performance (such as terminal-side and / or network-side QoS performance) generated based on the retransmission decision can be information obtained by the first device based on the first AI model, or information obtained based on other methods (such as retransmission mechanisms such as MAC HARQ, RLC ARQ, and / or PDCP ARQ mentioned in the related technologies above).

[0099] In some embodiments, the third information may include one or more of the following:

[0100] (1) The weighted average of one or more parameters in the second information;

[0101] (2) The weighted average of one or more parameters in QoS performance;

[0102] (3) The weighted average of one or more parameters in the QoS requirement information;

[0103] (4) The weighted average of one or more parameters in the channel measurement information;

[0104] (5) The weighted average of one or more parameters in the cache-related information.

[0105] In other embodiments, the third information may also include a normalized value of one or more of the following: second information, QoS performance, QoS requirement-related information, channel measurement information, and cache-related information.

[0106] In some other embodiments, the third information may also include a weighted average of one or more of the following: second information, QoS performance, QoS requirement-related information, channel measurement information, and buffer-related information.

[0107] In one possible implementation, the aforementioned first information can be used as input information for the first AI model, which then determines the retransmission decision corresponding to the first information based on the input first information.

[0108] In some embodiments, the model parameters of the first AI model are obtained by iteratively updating the first information at different times. The basis for updating the model parameters of the first AI model is the QoS performance generated based on the retransmission decision. The cutoff condition for iterative updates is that the QoS performance generated based on the retransmission decision or the parameter value determined based on the QoS performance meets the predefined QoS performance index.

[0109] Furthermore, in some embodiments, the model parameters of the first AI model are updated based on the QoS performance generated by the retransmission decision, wherein the retransmission decision includes the retransmission decision corresponding to the first information of the first AI model.

[0110] In some embodiments, the wireless communication method provided in this embodiment further includes: the first device sending a retransmission decision to the second device.

[0111] It is understandable that the first device sends a retransmission decision to the second device for two reasons. First, it is to facilitate the second device (for example, in the case of model management of the first AI model on the second device side) to decide whether to replace the first AI model with the second AI model. Second, it can also serve as one of the bases for the second device to determine the passive retransmission decision.

[0112] For example, in some embodiments, the first device sends a retransmission decision to the second device via one or more of the following signaling:

[0113] (1)UCI or DCI;

[0114] (2) MAC CE;

[0115] (3) RLC data PDU or RLC control PDU;

[0116] (4) PDCP data PDU or PDCP control PDU;

[0117] (5) RRC.

[0118] In some embodiments, the first device periodically sends a retransmission decision to the second device. In other embodiments, the first device sends a retransmission decision to the second device when a first condition is met.

[0119] For example, in some embodiments, the first condition includes one or more of the following:

[0120] (1) The retransmission decision is different from the previous retransmission decision;

[0121] (2) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold.

[0122] For example, a retransmission decision that differs from a previous retransmission decision can be that the current retransmission decision is different from the previous retransmission decision. The difference could be that the current retransmission decision is to retransmit (or not to retransmit), while the previous retransmission decision was to not retransmit (or to retransmit); and / or, the number of retransmissions in the current retransmission decision is different from the number of retransmissions in the previous retransmission decision.

[0123] Figure 3 is a schematic flowchart of the wireless communication method provided in an embodiment of this application. As shown in Figure 3, the method may include the following steps:

[0124] S301, the second device determines the feedback information and / or the first indication information corresponding to the fourth information based on the third AI model; wherein, the feedback information is feedback information for the data sent by the first device; the first indication information is used to indicate whether to retransmit the data;

[0125] S302, the second device sends feedback information and / or first instruction information to the first device.

[0126] It should be noted that the feedback information determined by the third AI model regarding the data transmitted by the first device can be for data that the first device has not transmitted, data that has not been received during transmission, or data that has been received during transmission. Furthermore, for data that has been received during transmission, this feedback information can be determined before the verification result of the transmitted data is obtained.

[0127] It is understood that the wireless communication method shown in Figure 3 is a passive retransmission mechanism, which is applicable to both downlink and uplink data transmission.

[0128] For S302, further, in some embodiments, S302 includes: the second device sending feedback information and / or first indication information to the first device according to the transmission time related information; wherein, the transmission time related information includes the first transmission time of the feedback information and / or the second transmission time of the first indication information.

[0129] In some embodiments, both the first transmission time and / or the second transmission time occur before the verification result of the transmitted data is obtained. This helps to reduce service transmission latency.

[0130] In this application, the method for determining the first transmission time and / or the second transmission time is not limited. In some embodiments, the first transmission time and / or the second transmission time are pre-configured. In other embodiments, the first transmission time and / or the second transmission time are information determined by a third AI model using fourth information.

[0131] It is understandable that, for passive retransmission mechanisms, the third AI model can be used to determine the timing of advance transmission of feedback information and / or first indication information; thus, it is beneficial to reduce service transmission latency.

[0132] For example, in some embodiments, the first transmission time and / or the second transmission time satisfy one or more of the following:

[0133] (1) Before the timer expires; the timer is used to trigger the sending of feedback information;

[0134] (2) Before receiving the transmitted data;

[0135] (3) After receiving the transmitted data and before obtaining the decoding result of the transmitted data.

[0136] In some embodiments, the feedback information includes ACK information or NACK information, wherein the NACK information refers to feedback information for data loss or receiving incorrect data; the ACK information can be feedback information for receiving correct data, or it can be feedback information for data loss or receiving incorrect data (i.e., false ACK information).

[0137] It's understandable that a third AI model can identify false ACK messages (i.e., messages indicating incorrect data reception), and sending false ACK messages can avoid meaningless retransmissions when latency is too high, thereby reducing service transmission latency. On the other hand, sending NACK messages in advance can help the sender perform rapid retransmissions.

[0138] Furthermore, in some embodiments, the feedback information includes MAC HARQ-ACK information and / or RLC ARQ SR; and / or, the first indication information includes MAC HARQ retransmission scheduling and / or PDCP duplication retransmission configuration.

[0139] It can be understood that MAC HARQ-ACK information is an acknowledgment message regarding the reception of transmitted data. This acknowledgment message includes ACK information or NACK information. The ACK information can be feedback information for receiving correct data, or it can be feedback information for data loss or receiving incorrect data (i.e., false ACK information). The NACK information can be information for data loss or receiving incorrect data.

[0140] In some embodiments, the RLC ARQ SR includes ACK information or NACK information, wherein the NACK information refers to feedback information for data loss or receiving incorrect data; the ACK information can be feedback information for receiving correct data, or it can be feedback information for data loss or receiving incorrect data (i.e., false ACK information).

[0141] In this application embodiment, the fourth information is not limited; this information is information that affects the determination result of the feedback information and / or the first indication information. In some embodiments, the fourth information is QoS-related information. Further, in some embodiments, the fourth information includes one or more of the following:

[0142] (1) QoS requirements related information;

[0143] (2) Channel measurement information;

[0144] (3) Caching related information;

[0145] (4) The retransmission decision sent by the first device;

[0146] (5) QoS performance;

[0147] (6) Downlink data reception status;

[0148] (7) Fifth information; wherein the fifth information is information determined based on one or more of the following: QoS requirement-related information, channel measurement information, buffer-related information, retransmission decision sent by the first device, QoS performance, and downlink data reception status.

[0149] In some embodiments, QoS requirement-related information includes terminal-side QoS requirement-related information and / or network-side QoS requirement-related information.

[0150] For example, in an uplink-oriented passive retransmission mechanism, the first device is a terminal device and the second device is a network device. The QoS requirements related to the terminal side may include the uplink service pattern; the QoS requirements related to the network side may include latency and / or rate, etc.

[0151] For example, in a downlink-oriented passive retransmission mechanism, the first device is a network device and the second device is a terminal device. The QoS requirements related to the terminal side may include latency and / or rate, etc.; the QoS requirements related to the network side may include downlink service patterns.

[0152] In some embodiments, the channel measurement information includes uplink channel measurement information and / or downlink channel measurement information.

[0153] In this embodiment of the application, the QoS performance in the fourth information can be understood as QoS performance-related information, which may include terminal-side QoS performance (i.e., terminal-side QoS performance-related information) and / or network-side QoS performance (i.e., network-side QoS performance-related information).

[0154] In some embodiments, the QoS performance in the fourth information includes QoS performance generated based on the second information (such as terminal-side and / or network-side QoS performance), and / or QoS performance generated based on retransmission decisions (such as terminal-side and / or network-side QoS performance).

[0155] For example, in an uplink-oriented passive retransmission mechanism, the first device is a terminal device and the second device is a network device. The QoS performance on the terminal side may include one or more of power consumption and resource consumption; the QoS performance on the network side may include latency and / or rate, etc.

[0156] For example, in a downlink-oriented passive retransmission mechanism, the first device is a network device and the second device is a terminal device. The QoS performance on the terminal side may include latency and / or rate, etc.; the QoS performance on the network side may include one or more of power consumption and resource consumption.

[0157] It should be noted that the QoS performance based on the second information can be information obtained by the second device based on the third AI model, or information obtained based on other methods (such as retransmission mechanisms such as MAC HARQ, RLC ARQ, and / or PDCP ARQ mentioned in the related technologies above). Similarly, the QoS performance based on the retransmission decision can be information obtained by the first device based on the first AI model, or information obtained based on other methods (such as retransmission mechanisms such as MAC HARQ, RLC ARQ, and / or PDCP ARQ mentioned in the related technologies above).

[0158] In some embodiments, the fifth information may include one or more of the following:

[0159] (1) The weighted average of one or more parameters in the QoS requirement information;

[0160] (2) The weighted average of one or more parameters in the channel measurement information;

[0161] (3) The weighted average of one or more parameters in the cache-related information;

[0162] (4) The weighted average of one or more parameters in QoS performance;

[0163] (5) The weighted average of one or more parameters in the downlink data reception status.

[0164] In other embodiments, the fifth information may also include a normalized value of one or more of the following: retransmission decision, QoS performance, downlink data reception status, QoS requirement-related information, channel measurement information, and buffer-related information.

[0165] In some embodiments, the downlink data reception status includes one or more of the following reception statuses:

[0166] (1) MAC layer HARQ reception status;

[0167] (2) ARQ reception status at the RLC layer;

[0168] (3) PDCP layer PDU reception status.

[0169] In some other embodiments, the fifth information may also include a weighted average of one or more of the following: retransmission decision, QoS performance, downlink data reception status, QoS requirement-related information, channel measurement information, and buffer-related information.

[0170] In one possible implementation, the aforementioned fourth information can be used as input information for the third AI model, which then uses the input fourth information to determine the corresponding feedback information and / or first indication information.

[0171] In some embodiments, the wireless communication method provided in this embodiment further includes: a second device sending second information to a first device; wherein the second information includes one or more of the following:

[0172] (1) Feedback information;

[0173] (2) The first time the feedback information is sent;

[0174] (3) First instruction information;

[0175] (4) The second time of sending the first instruction information.

[0176] It is understandable that the second device sends the second information to the first device. On the one hand, for the model management of the third AI model on the first device side, it is to facilitate the first device to decide whether to replace the third AI model with the fourth AI model. On the other hand, it can also serve as one of the bases for the first device to make an active retransmission decision.

[0177] In some embodiments, the model parameters of the third AI model are obtained by iteratively updating the fourth information at different times. The basis for updating the model parameters of the third AI model is the QoS performance generated based on the second information. The cutoff condition for iterative updates is that the QoS performance generated based on the second information or the parameter value determined based on the QoS performance meets the predefined QoS performance indicators.

[0178] Furthermore, in some embodiments, the model parameters of the third AI model are updated based on the QoS performance generated by the second information, wherein the second information includes the second information corresponding to the fourth information of the third AI model.

[0179] In some embodiments, the second device sends second information to the first device via one or more of the following signaling methods:

[0180] (1)UCI or DCI;

[0181] (2) MAC CE;

[0182] (3) RLC data PDU or RLC control PDU;

[0183] (4) PDCP data PDU or PDCP control PDU;

[0184] (5) RRC.

[0185] Furthermore, in some embodiments, the second device periodically sends second information to the first device. In other embodiments, the second device sends the second information to the first device when a fourth condition is met.

[0186] For example, in some embodiments, the fourth condition includes one or more of the following:

[0187] (1) The cumulative number of times the same feedback information is sent exceeds or falls below the corresponding threshold;

[0188] (2) The cumulative number of times the same first instruction information is sent exceeds or falls below the corresponding threshold.

[0189] It is understandable that identical feedback information refers to feedback information for the same sent data / data packets, and the content of these feedback information is the same, such as all being ACK information or all being NACK information.

[0190] Similarly, the same first indication information refers to the first indication information for the same transmitted data / data packet, and the content of these indication information is the same, such as the same PDCP duplication retransmission configuration or the same MAC HARQ retransmission schedule.

[0191] For the data retransmission mechanism, the first AI model can be applied on the first device, that is, for uplink (or downlink), a one-sided model based on the terminal device (or network device) can be used to optimize the active retransmission mechanism.

[0192] For the data retransmission mechanism, a third AI model can be used on the second device, that is, for uplink (or downlink), a one-sided model based on network device (or terminal device) can be used to optimize the passive retransmission mechanism.

[0193] For data retransmission mechanisms, a first AI model can be applied on the first device and a third AI model can be applied on the second device. That is, for uplink (or downlink), based on the dual-side models of terminal devices and network devices, the active retransmission mechanism and the passive retransmission mechanism can be optimized.

[0194] The above-described combination of one or more embodiments is applicable to communication systems where the first device is a terminal device and the second device is a network device; it is also applicable to communication systems where the first device is a network device and the second device is a terminal device. The above-described combination of one or more embodiments' active retransmission mechanism is applicable to both uplink and downlink active retransmission mechanisms. The above-described combination of one or more embodiments' passive retransmission mechanism is applicable to both uplink and downlink passive retransmission mechanisms. The following description does not distinguish between single-sided and double-sided models, but rather describes the relevant schemes for uplink and downlink active and passive retransmission mechanisms.

[0195] The following embodiments, including Example 1 and its further or additional embodiments, describe an uplink-oriented active retransmission mechanism; Example 2 and its further or additional embodiments describe an uplink-oriented passive retransmission mechanism; Example 3 and its further or additional embodiments describe a downlink-oriented active retransmission mechanism; and Example 4 and its further or additional embodiments describe a downlink-oriented passive retransmission mechanism.

[0196] The following describes and illustrates further or additional embodiments of Examples 1 to 4, mainly from three aspects: model inference, model updating, and model management. Model inference refers to the model usage phase. For example, in Example 1, the terminal device uses a first AI model to determine the retransmission decision for uplink data; similarly, in Example 2, the network device uses a third AI model to determine second information such as feedback information and / or first indication information for uplink data. Model updating refers to the model training phase. For example, in Example 1, the model parameters of the first AI model are updated; similarly, in Example 2, the model parameters of the third AI model are updated. Model management refers to replacing the model. For example, in Example 1, the first AI model is replaced with a second AI model, and the terminal device determines the retransmission decision based on the second AI model; similarly, in Example 2, the third AI model is updated to a fourth AI model, and the network device determines the second information based on the fourth AI model.

[0197] Example 1: Uplink-oriented active retransmission mechanism

[0198] In Embodiment 1 and its further or additional embodiments, the embodiment corresponding to FIG2 and its further or additional embodiments are described, wherein the first device is a terminal device and the second device is an active retransmission mechanism of a network device.

[0199] Figure 4 is a schematic flowchart of the wireless communication method provided in an embodiment of this application. As shown in Figure 4, the method may include the following steps:

[0200] S401, The terminal device determines the retransmission decision corresponding to the first information based on the first AI model;

[0201] S402: If a retransmission is determined to be a retransmission, the terminal device sends retransmission data to the network device.

[0202] In other embodiments, when the retransmission decision is not to retransmit, or the number of retransmissions in the retransmission decision is 0, the terminal device sends new uplink data to the network device.

[0203] In some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device receiving one or more of the following information sent by the network device:

[0204] (1) Information related to QoS requirements on the network side;

[0205] (2) Uplink channel measurement information;

[0206] (3) Network-side QoS performance.

[0207] In some embodiments, the terminal device may obtain the above information from the network device according to per-flow, per-bearer, per-LCH, and / or per-LCG. The above information may be information corresponding to per-flow, per-bearer, per-LCH, and / or per-LCG respectively.

[0208] In an embodiment where the first information includes cache-related information, in one example, the cache-related information includes the terminal-side uplink data cache information.

[0209] In some embodiments, the first information includes information specific to the terminal device, or the first information includes information specific to the terminal device and information from other terminal devices different from the terminal device.

[0210] It is understandable that, regardless of whether it is the online usage / model inference phase or the online or offline training / update phase of the first AI model, the type of information used to determine the retransmission decision is the same (both are primary information). However, the retransmission decision at different times is based on the primary information at different times. In one possible implementation, the primary information can be used as input information for the first AI model, and the first AI model obtains the corresponding retransmission decision based on the primary information.

[0211] It should be noted that in this embodiment, the terminal device receiving one or more of the above-mentioned information sent by the network device does not limit the first information to including one or more of these information. Further, in some embodiments of Embodiment 1, the first information includes one or more of the following:

[0212] (1) QoS requirements related to the terminal side; for example, uplink service pattern;

[0213] (2) Network-side QoS requirements related information; for example, latency and / or rate, etc.

[0214] (3) Uplink channel measurement information; for example, RSRP, RSRQ and / or SINR of uplink signals;

[0215] (4) Downlink channel measurement information; for example, RSRP, RSRQ and / or SINR of downlink signals;

[0216] (5) Terminal-side uplink data cache information; for example, BSR / DSR. Furthermore, the terminal-side uplink data cache information may include information on the amount of cached data acquired for different data layers.

[0217] (6) Second information sent by the network device; wherein the second information includes one or more of the following: feedback information for data sent to the terminal device, time-related information for the transmission of feedback information, first indication information for data sent to the terminal device, and time-related information for the transmission of the first indication information; the first indication information is used to indicate whether to retransmit the data.

[0218] (7) Terminal-side QoS performance; for example, power consumption and / or resource consumption, etc.

[0219] (8) Network-side QoS performance; for example, latency and / or speed;

[0220] (9) Third information; wherein the third information is information determined based on one or more of the above information.

[0221] In this embodiment, the task of updating the model parameters of the first AI model used by the terminal device can be performed on either the network side or the terminal side. In the network-side solution, after determining the new model parameters based on the first information, the network device configures them to the terminal device via instruction information.

[0222] That is, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device receiving second indication information sent by the network device, the second indication information being used to indicate that the model parameters of the first AI model be updated to new model parameters.

[0223] For network-side update tasks, in some embodiments, the network device receives one or more of the following information sent by the terminal device:

[0224] (1) QoS requirements related to the terminal side; for example, uplink service pattern;

[0225] (2) Terminal-side uplink data cache information; for example, BSR / DSR. Furthermore, the terminal-side uplink data cache information may include information on the amount of cached data acquired for different data layers.

[0226] (3) Downlink channel measurement information; for example, RSRP, RSRQ and / or SINR of downlink signals.

[0227] In some embodiments, the network device may obtain the above information from the terminal device side according to per-flow, per-bearer, per-LCH, and / or per-LCG. The above information may be information corresponding to per-flow, per-bearer, per-LCH, and / or per-LCG respectively.

[0228] It should be noted that for the network-side update task, the network device receiving one or more of the aforementioned information sent by the terminal device does not limit the first information to including only one or more of these information. As mentioned above, the first information may include one or more of the enumerated information above.

[0229] For the update task on the network side, the network device determines the retransmission decision corresponding to the first information based on the first AI model, and sends the retransmission decision to the terminal device through the eighth indication information. The terminal device executes the retransmission decision indicated by the eighth indication information. Based on the terminal-side QoS performance and / or network-side QoS performance generated by the retransmission decision, the network device determines the first difference of the first AI model, and determines the new model parameters of the first AI model based on the first difference.

[0230] For example, the first difference could be the difference between the terminal-side QoS performance and / or the network-side QoS performance and a predefined learning target.

[0231] In one possible implementation, the first AI model is a reinforcement learning model. The network device can determine the corresponding reward based on the terminal-side QoS performance and / or network-side QoS performance generated by the retransmission decision, as well as the predefined learning objectives (such as terminal-side QoS performance indicators and / or network-side QoS performance indicators). Based on the reward, new model parameters of the first AI model are determined.

[0232] In some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device sending the terminal-side QoS performance generated based on the retransmission decision to the network device. This facilitates the network device in determining new model parameters for the first AI model based on the terminal-side QoS performance generated by the retransmission decision.

[0233] In some embodiments, the new model parameters are related to the QoS performance resulting from retransmission decisions; the QoS performance resulting from retransmission decisions includes the terminal-side QoS performance and / or the network-side QoS performance based on retransmission decisions.

[0234] Optionally, the network device may determine new model parameters for the first AI model if the terminal-side QoS performance resulting from the retransmission decision exceeds or falls below a corresponding threshold, and / or if the network-side QoS performance resulting from the retransmission decision exceeds or falls below a corresponding threshold.

[0235] It is understandable that for the uplink-oriented active retransmission mechanism, the update / training task of the model parameters of the first AI model used by the terminal device is completed by the network device. This is mainly because the network device has stronger capabilities and storage capacity than the terminal device, making it easier to update or train the model.

[0236] Of course, the terminal device can also update the model parameters of the first AI model independently, that is, the task of updating / training the model parameters of the first AI model can also be done on the terminal side.

[0237] That is, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device updating the model parameters of the first AI model to new model parameters based on the QoS performance generated by the retransmission decision; wherein the QoS performance generated based on the retransmission decision includes terminal-side QoS performance and / or network-side QoS performance.

[0238] Furthermore, in some embodiments, the terminal device updates the model parameters of the first AI model to new model parameters based on the QoS performance generated by the retransmission decision, including: the terminal device determining a first difference of the first AI model based on the QoS performance generated by the retransmission decision; and the first device updating the model parameters of the first AI model to new model parameters based on the first difference.

[0239] It is understandable that for an uplink-oriented active retransmission mechanism, the terminal device automatically updates or trains the model parameters of the first AI model. In this way, for the first information, which includes a lot of terminal-side information, the amount of information that the network device needs to transmit is less, which helps to save the signaling overhead incurred in obtaining the first information.

[0240] Considering the generalization problem of the first AI model in use, that is, the first AI model suitable for scenario one may exhibit poor QoS performance in scenario two. Here, scenarios mainly refer to physical environment, wireless channel conditions, etc. In view of this, this application embodiment provides a model management mechanism. For the uplink-oriented active retransmission mechanism, model management can be set on the network side or the terminal side. Model management is aimed at: whether to replace the first AI model with the second AI model.

[0241] For network-side model management solutions, in some embodiments, the wireless communication method provided in this embodiment further includes: a terminal device receiving third indication information sent by a network device, the third indication information being used to indicate that the first AI model is replaced with a second AI model.

[0242] In some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device reporting one or more of the following second conditions to the network device:

[0243] (1) The uplink data cache information on the terminal side exceeds or falls below the corresponding threshold;

[0244] (2) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0245] (3) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0246] (4) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold;

[0247] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0248] It is understandable that the terminal device will report one or more of the information that meets the second condition to the network device, so that the network device can monitor the usage of the first AI model and replace the first AI model in a timely manner.

[0249] For solutions where model management is on the terminal side, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device replacing the first AI model with the second AI model when one or more of the following second conditions are met:

[0250] (1) The uplink data cache information on the terminal side exceeds or falls below the corresponding threshold;

[0251] (2) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0252] (3) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0253] (4) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold;

[0254] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0255] For a model management solution on the terminal side, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device receiving fourth indication information sent by the network device, the fourth indication information being used to indicate the second condition.

[0256] For solutions where model management is performed on the terminal side, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device reporting model replacement-related information for the first AI model to the network device.

[0257] Furthermore, in some embodiments, model replacement-related information includes a second condition that triggers the replacement of the first AI model with the second AI model.

[0258] Example 2: Uplink-oriented passive retransmission mechanism

[0259] In Embodiment 2 and its further or additional embodiments, the embodiment corresponding to FIG3 and its further or additional embodiments are described, in which the first device is a terminal device and the second device is a passive retransmission mechanism of a network device. Embodiments 1 and 2 both optimize the QoS performance of the uplink services of the terminal device.

[0260] Figure 5 is a schematic flowchart of the wireless communication method provided in an embodiment of this application. As shown in Figure 5, the method may include the following steps:

[0261] S501, the network device determines the feedback information and / or the first indication information corresponding to the fourth information based on the third AI model; wherein, the feedback information is feedback information for the data sent by the terminal device; the first indication information is used to indicate whether to retransmit the data;

[0262] S502, the network device sends feedback information and / or first instruction information to the terminal device.

[0263] The following describes and illustrates further or additional embodiments of Embodiment 2.

[0264] In some embodiments, the wireless communication method provided in this embodiment further includes: the network device receiving one or more of the following information sent by the terminal device:

[0265] (1) QoS requirements related to the terminal side;

[0266] (2) Downlink channel measurement information;

[0267] (3) Terminal-side QoS performance;

[0268] (4) Uplink data cache information on the terminal side.

[0269] In some embodiments, the network device may obtain the above information from the terminal device side according to per-flow, per-bearer, per-LCH, and / or per-LCG. The above information may be information corresponding to per-flow, per-bearer, per-LCH, and / or per-LCG respectively.

[0270] It is understandable that, regardless of whether it is the online usage / model inference phase or the online or offline training / update phase of the third AI model, the type of information used to determine the feedback information and / or the first indication information is the same (all are fourth information). However, the feedback information and / or the first indication information at different times are based on the fourth information at different times. In one possible implementation, the fourth information can be used as the input information of the third AI model, and the third AI model obtains the corresponding feedback information and / or the first indication information based on the fourth information.

[0271] It should be noted that in this embodiment, the network device receiving one or more of the above-mentioned information sent by the terminal device does not limit the fourth information to including one or more of these information. Further, in some embodiments of Embodiment 2, the fourth information includes one or more of the following:

[0272] (1) QoS requirements related to the terminal side; for example, uplink service pattern;

[0273] (2) Network-side QoS requirements related information; for example, latency and / or rate, etc.

[0274] (3) Uplink channel measurement information; for example, RSRP, RSRQ and / or SINR of uplink signals;

[0275] (4) Downlink channel measurement information; for example, RSRP, RSRQ and / or SINR of downlink signals;

[0276] (5) Terminal-side uplink data cache information; for example, BSR / DSR. Furthermore, the terminal-side uplink data cache information may include information on the amount of cached data acquired for different data layers.

[0277] (6) Retransmission decision sent by the terminal device;

[0278] (7) Terminal-side QoS performance; for example, power consumption and / or resource consumption, etc.

[0279] (8) Network-side QoS performance; for example, latency and / or speed;

[0280] (9) Fifth information; wherein the fifth information is information determined based on one or more of the above information.

[0281] In this embodiment of the application, the task of updating the model parameters of the third AI model used by the network device can be performed on the network side or on the terminal side.

[0282] In schemes where the update task is performed on the network side, i.e., the network device autonomously updates the model parameters of the third AI model, in some embodiments, the network device updates the model parameters of the third AI model to new model parameters based on the QoS performance generated by the second information; wherein, the QoS performance generated by the second information includes terminal-side QoS performance and / or network-side QoS performance.

[0283] Furthermore, in some embodiments, the network device updates the model parameters of the third AI model to new model parameters based on the QoS performance generated by the second information, including: the network device determining a second difference of the third AI model based on the QoS performance generated by the second information; and the network device updating the model parameters of the third AI model to new model parameters based on the second difference.

[0284] For example, the second difference could be the difference between the terminal-side QoS performance and / or the network-side QoS performance and a predefined learning target.

[0285] In one possible implementation, the third AI model is a reinforcement learning model. The network device can determine the corresponding reward based on the terminal-side QoS performance and / or network-side QoS performance generated by the second information, as well as predefined learning objectives (such as terminal-side QoS performance indicators and / or network-side QoS performance indicators). Based on the reward, new model parameters of the third AI model are determined.

[0286] In schemes where the update task is performed on the network side, in some embodiments, the wireless communication method provided in this embodiment further includes: the network device receiving terminal-side QoS performance generated based on second information sent by the terminal device. This facilitates the network device in determining new model parameters for the third AI model based on the terminal-side QoS performance generated by the second information.

[0287] Considering the generalization problem of the third AI model in use, that is, the third AI model suitable for scenario one may exhibit poor QoS performance in scenario two. Here, scenarios mainly refer to physical environment, wireless channel conditions, etc. In view of this, this application provides a model management mechanism. For the uplink-oriented passive retransmission mechanism, model management can be set on the network side or the terminal side. Model management is aimed at: whether to replace the third AI model with the fourth AI model.

[0288] For network-side model management solutions, in some embodiments, the wireless communication method provided in this embodiment further includes: the network device replacing the third AI model with the fourth AI model when the second condition and / or the fifth condition are met.

[0289] In some embodiments, the second condition includes one or more of the following:

[0290] (1) The uplink data cache information on the terminal side exceeds or falls below the corresponding threshold;

[0291] (2) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0292] (3) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0293] (4) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold;

[0294] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0295] In some embodiments, the wireless communication method provided in this embodiment further includes: a network device sending fourth indication information to a terminal device, the fourth indication information being used to indicate relevant information for reporting the second condition.

[0296] For terminal devices, they can send relevant information that meets the second condition to network devices if one or more of the second conditions are met.

[0297] In some embodiments, the fifth condition includes one or more of the following:

[0298] (1) Uplink channel measurement information exceeds or falls below the corresponding threshold;

[0299] (2) The QoS requirements on the network side exceed or fall below the corresponding threshold;

[0300] (3) The cumulative number of times the same feedback information is sent exceeds or falls below the corresponding threshold;

[0301] (4) The cumulative number of times the same first instruction information is sent exceeds or falls below the corresponding threshold;

[0302] (5) The QoS performance on the network side exceeds or falls below the corresponding threshold.

[0303] Example 3: Downlink-oriented active retransmission mechanism

[0304] In Embodiment 3 and further or additional embodiments of Embodiment 3, the embodiment corresponding to FIG2 and further or additional embodiments of the embodiment corresponding to FIG2 are described, wherein the first device is a network device and the second device is an active retransmission mechanism of a terminal device.

[0305] Figure 6 is a flowchart illustrating the wireless communication method provided in an embodiment of this application. As shown in Figure 6, the method may include the following steps:

[0306] S601, the network device determines the retransmission decision corresponding to the first information based on the first AI model;

[0307] S602: When a retransmission is determined to be a retransmission, the network device sends retransmitted data to the terminal device.

[0308] The following describes and illustrates further or additional embodiments of Embodiment 3.

[0309] In some embodiments, the wireless communication method provided in this embodiment further includes: the network device receiving one or more of the following information sent by the terminal device:

[0310] (1) QoS requirements related to the terminal side;

[0311] (2) Downlink channel measurement information;

[0312] (3) QoS performance on the terminal side.

[0313] In some embodiments, the network device may obtain the above information from the terminal device side according to per-flow, per-bearer, per-LCH, and / or per-LCG. The above information may be information corresponding to per-flow, per-bearer, per-LCH, and / or per-LCG respectively.

[0314] In an embodiment where the first information includes cache-related information, in one example, the cache-related information includes network-side downlink data cache information.

[0315] It is understandable that, regardless of whether it is the online usage / model inference phase or the online or offline training / update phase of the first AI model, the type of information used to determine the retransmission decision is the same (both are primary information). However, the retransmission decision at different times is based on the primary information at different times. In one possible implementation, the primary information can be used as input information for the first AI model, and the first AI model obtains the corresponding retransmission decision based on the primary information.

[0316] It should be noted that in this embodiment, the network device receiving one or more of the above-mentioned information sent by the terminal device does not limit the first information to including one or more of these information. Further, in some embodiments of Embodiment 3, the first information includes one or more of the following:

[0317] (1) QoS requirements related to the terminal side; for example, downlink service pattern;

[0318] (2) Network-side QoS requirements related information; for example, latency and / or rate, etc.

[0319] (3) Uplink channel measurement information; for example, RSRP, RSRQ and / or SINR of uplink signals;

[0320] (4) Downlink channel measurement information; for example, RSRP, RSRQ and / or SINR of downlink signals;

[0321] (5) Downlink data caching information on the network side; for example, data volume information can be obtained for caching at different data layers;

[0322] (6) Second information sent by the terminal device; wherein the second information includes one or more of the following: feedback information for data sent by the network device, time-related information for the feedback information, first indication information for data sent by the network device, and time-related information for the first indication information; the first indication information is used to indicate whether to retransmit the data.

[0323] (7) Terminal-side QoS performance; for example, power consumption and / or resource consumption, etc.

[0324] (8) Network-side QoS performance; for example, latency and / or speed;

[0325] (9) Third information; wherein the third information is information determined based on one or more of the above information.

[0326] In this embodiment, the task of updating the model parameters of the first AI model used by the network device can be performed on either the network side or the terminal side. In the network-side solution, the network device autonomously updates the model parameters of the first AI model.

[0327] That is, in some embodiments, the wireless communication method provided in this embodiment further includes: the network device updating the model parameters of the first AI model to new model parameters based on the QoS performance generated by the retransmission decision; wherein the QoS performance generated by the retransmission decision includes terminal-side QoS performance and / or network-side QoS performance.

[0328] Furthermore, in some embodiments, the network device updates the model parameters of the first AI model to new model parameters based on the QoS performance generated by the retransmission decision, including: the network device determining a first difference of the first AI model based on the QoS performance generated by the retransmission decision; and the network device updating the model parameters of the first AI model to new model parameters based on the first difference.

[0329] For example, the first difference could be the difference between the terminal-side QoS performance and / or the network-side QoS performance and a predefined learning target.

[0330] In one possible implementation, the first AI model is a reinforcement learning model. The network device can determine the corresponding reward based on the terminal-side QoS performance and / or network-side QoS performance generated by the retransmission decision, as well as the predefined learning objectives (such as terminal-side QoS performance indicators and / or network-side QoS performance indicators). Based on the reward, new model parameters of the first AI model are determined.

[0331] Considering the generalization problem of the first AI model in use, that is, the first AI model suitable for scenario one may exhibit poor QoS performance in scenario two. Here, scenarios mainly refer to physical environment, wireless channel conditions, etc. In view of this, this application provides a model management mechanism. For the downlink-oriented active retransmission mechanism, model management can be set on the network side or the terminal side. Model management is aimed at: whether to replace the first AI model with the second AI model.

[0332] For network-side model management solutions, in some embodiments, the wireless communication method provided in this embodiment further includes: the network device replacing the first AI model with the second AI model when the second condition and / or the third condition are met.

[0333] For network-side model management solutions, in some embodiments, the second condition includes one or more of the following:

[0334] (1) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0335] (2) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0336] (3) The QoS performance of the terminal side exceeds or falls below the corresponding threshold.

[0337] Regarding the second condition, the network device can configure a reporting mechanism for the terminal device, the purpose of which is to instruct the terminal device to report one or more pieces of information that meet the second condition to the network device.

[0338] In some embodiments, the wireless communication method provided in this embodiment further includes: a network device sending fifth indication information to a terminal device, the fifth indication information being used to indicate a second condition.

[0339] In some embodiments, the third condition includes one or more of the following:

[0340] (1) The downlink data cache information on the network side exceeds or falls below the corresponding threshold;

[0341] (2) Uplink channel measurement information exceeds or falls below the corresponding threshold;

[0342] (3) The QoS requirements related to the network side exceed or fall below the corresponding threshold;

[0343] (4) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold;

[0344] (5) The QoS performance on the network side exceeds or falls below the corresponding threshold.

[0345] Example 4: Downlink-oriented passive retransmission mechanism

[0346] In Embodiment 4 and its further or additional embodiments, the passive retransmission mechanism of the embodiment corresponding to FIG3 and its further or additional embodiments corresponding to FIG3 is described, wherein the first device is a network device and the second device is a terminal device. Embodiments 3 and 4 both optimize the QoS performance of downlink services of the network device.

[0347] Figure 7 is a schematic flowchart of the wireless communication method provided in an embodiment of this application. As shown in Figure 7, the method may include the following steps:

[0348] S701, the terminal device determines the feedback information and / or the first indication information corresponding to the fourth information based on the third AI model; wherein, the feedback information is feedback information for the data sent by the network device; the first indication information is used to indicate whether to retransmit the data;

[0349] S702, the terminal device sends feedback information and / or first instruction information to the network device.

[0350] The following describes and illustrates further or additional embodiments of Embodiment 4.

[0351] In some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device receiving one or more of the following information sent by the network device:

[0352] (1) Information related to QoS requirements on the network side;

[0353] (2) Uplink channel measurement information;

[0354] (3) Downlink data caching information on the network side;

[0355] (4) Network-side QoS performance.

[0356] In some embodiments, the terminal device may obtain the above information from the network device according to per-flow, per-bearer, per-LCH, and / or per-LCG. The above information may be information corresponding to per-flow, per-bearer, per-LCH, and / or per-LCG respectively.

[0357] In some embodiments, the fourth information includes information specific to the terminal device, or the fourth information includes information specific to the terminal device and information from other terminal devices different from the terminal device.

[0358] It is understandable that, regardless of whether it is the online usage / model inference phase or the online or offline training / update phase of the third AI model, the type of information used to determine the feedback information and / or the first indication information is the same (all are fourth information). However, the feedback information and / or the first indication information at different times are based on the fourth information at different times. In one possible implementation, the fourth information can be used as the input information of the third AI model, and the third AI model obtains the corresponding feedback information and / or the first indication information, etc., based on the fourth information.

[0359] It should be noted that in this embodiment, the terminal device receiving one or more of the above-mentioned information sent by the network device does not limit the fourth information to including one or more of these information. Further, in some embodiments of embodiment four, the fourth information includes one or more of the following:

[0360] (1) QoS requirements related to the terminal side; for example, downlink service pattern;

[0361] (2) Network-side QoS requirements related information; for example, latency and / or rate, etc.

[0362] (3) Uplink channel measurement information; for example, RSRP, RSRQ and / or SINR of uplink signals;

[0363] (4) Downlink channel measurement information; for example, RSRP, RSRQ and / or SINR of downlink signals;

[0364] (5) Downlink data caching information on the network side; for example, data volume information can be obtained for caching at different data layers;

[0365] (6) Second information sent by the network device; wherein the second information includes one or more of the following: feedback information for data sent to the terminal device, time-related information for the transmission of feedback information, first indication information for data sent to the terminal device, and time-related information for the transmission of the first indication information; the first indication information is used to indicate whether to retransmit the data.

[0366] (7) Retransmission decisions sent by network devices;

[0367] (8) Terminal-side QoS performance; for example, power consumption and / or resource consumption, etc.

[0368] (9) Network-side QoS performance; for example, latency and / or speed;

[0369] (10) Third information; wherein the third information is information determined based on one or more of the above information.

[0370] In this embodiment, the update task of the model parameters of the third AI model used by the terminal device can be performed on either the network side or the terminal side. In the network-side solution, after determining the new model parameters based on the fourth information, the network device configures them to the terminal device via instruction information.

[0371] That is, for the scheme where the update task is on the network side, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device receiving a fifth indication information sent by the network device, the fifth indication information being used to indicate that the model parameters of the third AI model be updated to new model parameters.

[0372] For network-side update schemes, in some embodiments, the new model parameters are related to the QoS performance generated based on the second information.

[0373] For network-side update tasks, in some embodiments, the QoS performance generated based on the second information includes terminal-side QoS performance and / or network-side QoS performance generated based on the second information.

[0374] For solutions where the update task is performed on the network side, in some embodiments, the terminal device sends one or more of the following information to the network device:

[0375] (1) Terminal-side QoS performance based on the second information;

[0376] (2) QoS requirements related to the terminal side;

[0377] (3) Downlink channel measurement information;

[0378] (4) Downlink data reception status.

[0379] For solutions where the update task is performed on the terminal side, i.e., the terminal device autonomously updates the model parameters of the third AI model, in some embodiments, the terminal device updates the model parameters of the third AI model to new model parameters based on the QoS performance generated by the second information; wherein, the QoS performance generated by the second information includes terminal-side QoS performance and / or network-side QoS performance.

[0380] Furthermore, in some embodiments, the terminal device updates the model parameters of the third AI model to new model parameters based on the QoS performance generated by the second information, including: the terminal device determining a second difference of the third AI model based on the QoS performance generated by the second information; and the terminal device updating the model parameters of the third AI model to new model parameters based on the second difference.

[0381] For example, the second difference could be the difference between the terminal-side QoS performance and / or the network-side QoS performance and a predefined learning target.

[0382] In one possible implementation, the third AI model is a reinforcement learning model. The terminal device can determine the corresponding reward based on the terminal-side QoS performance and / or network-side QoS performance generated by the second information, as well as the predefined learning objectives (such as terminal-side QoS performance indicators and / or network-side QoS performance indicators). Based on the reward, new model parameters of the third AI model are determined.

[0383] In solutions where the update task is performed on the terminal side, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device receiving network-side QoS performance data generated based on second information sent by the network device. This facilitates the terminal device in determining new model parameters for the third AI model based on the network-side QoS performance generated by the second information.

[0384] Considering the generalization problem of the third AI model in use, that is, the third AI model suitable for scenario one may exhibit poor QoS performance in scenario two. Here, scenarios mainly refer to physical environment, wireless channel conditions, etc. In view of this, this application provides a model management mechanism. For the downlink-oriented passive retransmission mechanism, model management can be set on the network side or the terminal side. Model management is aimed at: whether to replace the third AI model with the fourth AI model.

[0385] For network-side model management solutions, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device receiving a sixth indication information sent by the network device, the sixth indication information being used to indicate that the third AI model is replaced with the fourth AI model.

[0386] In some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device reporting one or more of the following sixth conditions to the network device:

[0387] (1) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0388] (2) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0389] (3) The cumulative number of times the same feedback information is sent exceeds or falls below the corresponding threshold;

[0390] (4) The cumulative number of times the same first instruction information is sent exceeds or falls below the corresponding threshold;

[0391] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0392] It is understandable that the terminal device will report one or more of the information in the sixth condition to the network device, so that the network device can monitor the usage of the third AI model and replace the third AI model in a timely manner.

[0393] For solutions where model management is on the terminal side, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device replacing the third AI model with the fourth AI model when one or more of the following sixth conditions are met:

[0394] (1) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0395] (2) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0396] (3) The cumulative number of times the same feedback information is sent exceeds or falls below the corresponding threshold;

[0397] (4) The cumulative number of times the same first instruction information is sent exceeds or falls below the corresponding threshold;

[0398] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0399] For a model management solution on the terminal side, the terminal device receives the seventh indication information sent by the network device. The seventh indication information is used to indicate the sixth condition.

[0400] For solutions where model management is performed on the terminal side, in some embodiments, the wireless communication method provided in this embodiment further includes: the terminal device reporting model replacement-related information for the third AI model to the network device.

[0401] Furthermore, in some embodiments, model replacement-related information includes a sixth condition that triggers the replacement of the third AI model with the fourth AI model.

[0402] The wireless communication method provided in the embodiments of this application has been described above. To facilitate understanding of the embodiments of this application, the following uses a UE and a network device as examples to introduce possible implementation schemes of the wireless communication method applicable to the embodiments of this application.

[0403] Option 1: Uplink-oriented UE-side model (see Figure 8), i.e., uplink-oriented active retransmission mechanism.

[0404] 1. Model Training

[0405] For model training, the model can be trained on the network side or on the terminal side.

[0406] For network-side model training, the network needs to collect one or more of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as input for model training:

[0407] (1) Cache-related information, such as BSR / DSR, and further, it can include information on the amount of cached data for different data layers;

[0408] (2) QoS requirements related to the UE side (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service patterns and other information;

[0409] (3) DL measurement information, including RSRP, RSRQ, SINR and other information;

[0410] Optionally, the input for model training can be an indication based on at least one of the above-mentioned criteria.

[0411] For network-side model training, the output of the model training (i.e., retransmission decision) includes one or more of the following information:

[0412] A. The decision to initiate retransmission using HARQ at the MAC layer;

[0413] B. Decision on ARQ active retransmission at the RLC layer;

[0414] C.PDCP ARQ / duplication decision.

[0415] For model training on the network side, the network needs to collect at least one of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as a reward for model training:

[0416] QoS results / performance, such as the QoS performance (i.e., QoS performance) obtained by the UE based on the above "MAC layer HARQ active retransmission decision, RLC layer ARQ active retransmission decision and / or PDCP duplication decision", such as latency, power consumption, resource consumption, etc., mainly refers to the QoS results on the UE side.

[0417] Optionally, the reward for model training can be an indication determined based on one or more of the QoS results mentioned above.

[0418] It's understandable to place model training on the network side, considering that the network side currently has stronger computing power and storage than the UE side, making it easier to perform model training.

[0419] For model training on the UE side (including the server connected to the UE), the UE needs to collect one or more of the following information from the network side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as input for model training:

[0420] (1) Network-side QoS requirements related information (some of this information can be obtained from the UE itself, and this part of the information does not need to be obtained from the network side), such as UL service pattern and other information;

[0421] (2) UL measurement information, including RSRP, RSRQ, SINR and other information.

[0422] For model training on the UE side, the output of model training (i.e., retransmission decision) includes one or more of the following information:

[0423] A. The decision to initiate retransmission using HARQ at the MAC layer;

[0424] B. Decision on ARQ active retransmission at the RLC layer;

[0425] C.PDCP ARQ / duplication decision.

[0426] For model training on the UE side (including the server connected to the UE), the UE needs to collect at least one of the following information from the network side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as a reward for model training:

[0427] QoS results / performance, such as the QoS performance obtained by the UE based on the above "MAC layer HARQ active retransmission decision, RLC layer ARQ active retransmission decision, PDCP duplication decision", such as latency, power consumption, resource consumption, etc., mainly refers to the QoS results on the network side.

[0428] Optionally, the reward for model training can be an indication determined based on one or more of the QoS results mentioned above.

[0429] Optionally, the above model training can be iterated simultaneously with model inference, i.e., model training can be performed online.

[0430] It is understandable that model training is placed on the UE side, considering that the UE can obtain more information and the amount of information that needs to be transmitted is less, but it requires higher computing and storage capabilities from the UE.

[0431] 2. Model Derivation / Model Inference

[0432] For a trained model, during the inference phase, the UE needs to collect the model's input information. In addition to the information obtainable by the UE itself, the UE also needs to obtain one or more of the following relevant information from the network side:

[0433] (1) Network-side QoS requirements related information (some of this information can be obtained from the UE itself, and this part of the information does not need to be obtained from the network side), such as UL service pattern and other information;

[0434] (2) UL measurement information, including RSRP, RSRQ, SINR and other information;

[0435] Optionally, the relevant information here may be specific to the UE or may include information about other UEs.

[0436] Optionally, the relevant information here can be direct information, or information that is further determined / derived from the information mentioned above.

[0437] It is understandable that the UE uses a trained model to help it decide whether and how to perform autonomous retransmission.

[0438] Optionally, the UE may report the results of the model inference to the network, including one or more of the following retransmission decisions (whether to retransmit, number of retransmissions, etc.):

[0439] A. MAC HARQ's decision to automate retransmission;

[0440] B.RLC ARQ's decision to autonomously retransmit;

[0441] C. PDCP duplication decision to autonomously retransmit.

[0442] From a signaling perspective, information regarding these retransmission decisions can be reported through one or more of the following signaling methods:

[0443] UCI;

[0444] MAC CE;

[0445] RLC data PDU or RLC control PDU;

[0446] PDCP data PDU or PDCP control PDU;

[0447] RRC.

[0448] The reporting methods can be:

[0449] (1) Periodic reporting;

[0450] (2) Event-triggered reporting, for example, when these decisions (whether to retransmit automatically, the number of retransmissions) change, or when the number of retransmissions is higher or lower than a certain threshold, a report is made.

[0451] Understandably, this reporting allows the network to decide whether to adjust the model configuration (such as changing the model) and make decisions on passive retransmission based on the results of autonomous retransmission.

[0452] 3. Model Management

[0453] For network-based model management, the network can be configured with relevant measurement reporting mechanisms to monitor model-permitted conditions, such as requiring the UE to report one or more of the following information:

[0454] (1) When cache-related information exceeds or falls below a certain threshold, such as BSR / DSR, further, data volume information can be obtained for caches of different data layers;

[0455] (2) When the QoS requirements of the UE side exceed or fall below a certain threshold (some of this information can be obtained from the CN side and the UE does not need to report this information), such as UL service patterns, etc.

[0456] (3) When the DL measurement information exceeds or falls below a certain threshold, including RSRP, RSRQ, SINR and other information;

[0457] (4) When the MAC layer HARQ active retransmission decision, the RLC layer ARQ active retransmission decision, or the PDCP ARQ / duplication decision (such as the number of retransmissions) exceeds or falls below a certain threshold.

[0458] (5) When the QoS result exceeds or falls below a certain threshold, such as the QoS performance obtained by the UE based on the above "MAC layer HARQ active retransmission decision, RLC layer ARQ active retransmission decision, PDCP duplication decision", such as latency, power consumption, resource consumption, etc. This mainly refers to the QoS result on the UE side.

[0459] If the network needs to change the model configuration, the network sends an instruction message to the UE.

[0460] It is understandable that the above model management scheme facilitates real-time network monitoring of model usage and timely adjustment of model parameters.

[0461] For UE-based model management, the network pre-configures the relevant triggering conditions for model configuration changes, including one or more of the following:

[0462] (1) When cache-related information exceeds or falls below a certain threshold, such as BSR / DSR, further, data volume information can be obtained for caches of different data layers;

[0463] (2) When the QoS requirements of the UE side exceed or fall below a certain threshold (some of this information can be obtained from the CN side and the UE does not need to report this information), such as UL service patterns, etc.

[0464] (3) When the DL measurement information exceeds or falls below a certain threshold, including RSRP, RSRQ, SINR and other information;

[0465] (4) When the MAC layer HARQ active retransmission decision, the RLC layer ARQ active retransmission decision, and the PDCP ARQ / duplication decision exceed or fall below a certain threshold.

[0466] (5) When the QoS result exceeds or falls below a certain threshold, such as the QoS performance obtained by the UE based on the above "MAC layer HARQ active retransmission decision, RLC layer ARQ active retransmission decision, PDCP duplication decision", such as latency, power consumption, resource consumption, etc. This mainly refers to the QoS result on the UE side.

[0467] When a certain triggering condition is met, the UE reports model change information to the network. Optionally, the reporting includes the triggering condition for the model change.

[0468] It is understandable that a solution where model management is on the UE side allows for advance configuration of model parameter changes when network connectivity is poor, enabling the UE to adjust model parameters autonomously when relevant situations occur.

[0469] Option 2: Uplink-oriented network-side model (see Figure 9), i.e., uplink-oriented passive retransmission mechanism.

[0470] 1. Model Training

[0471] For network-side model training, the network needs to collect one or more of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as input for model training:

[0472] (1) Cache-related information, such as BSR / DSR, and further, data volume information can be obtained for different data layers;

[0473] (2) QoS requirements related to the UE side (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service patterns and other information;

[0474] (3) DL measurement information, including RSRP, RSRQ, SINR and other information.

[0475] Optionally, the input for model training can be an indication based on at least one of the above-mentioned criteria.

[0476] For network-side model training, the network needs to collect one or more of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as input information for the model:

[0477] A. The decision to initiate retransmission using HARQ at the MAC layer;

[0478] B. Decision on ARQ active retransmission at the RLC layer;

[0479] C.PDCP ARQ / duplication decision.

[0480] For network-side model training, the output of the model training includes one or more of the following information:

[0481] A. Send MAC HARQ retransmission schedule in advance (e.g., retransmission schedule and new transmission schedule are sent simultaneously or consecutively, i.e., the retransmission schedule is sent to the UE before the new transmission reception result is available).

[0482] B. Send MAC HARQ-ACK information in advance;

[0483] C. Send an RLC ARQ SR in advance (if no polling or data is received from the sender), optionally including an ACK for no received or incorrectly received data (note that the AI-based SR triggering mechanism can work simultaneously with or replace the traditional non-AI SR triggering mechanism).

[0484] D. PDCP duplication retransmission configuration.

[0485] For model training on the network side, the network needs to collect at least one of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as a reward for model training:

[0486] QoS results / performance, such as the QoS performance (i.e., QoS performance) obtained by the UE based on the above configuration of "sending MAC HARQ retransmission scheduling in advance, sending MAC HARQ-ACK information in advance, sending RLC ARQ SR in advance, and PDCP duplication retransmission", such as latency, power consumption, resource consumption, etc. Here, we mainly focus on the QoS results on the UE side.

[0487] Optionally, the reward for model training can be an indication determined based on one or more of the QoS results mentioned above.

[0488] It's understandable to place model training on the network side, considering that the network side currently has stronger computing power and storage than the UE side, making it easier to perform model training.

[0489] 2. Model Derivation / Model Inference

[0490] For a trained model, during the inference phase, the network needs to collect model input information. In addition to the information available to the network itself, the network also needs to obtain one or more of the following relevant information from the user experience (UE):

[0491] (1) Cache-related information, such as BSR / DSR, and further, data volume information can be obtained for different data layers;

[0492] (2) QoS requirements related to the UE side (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service patterns and other information;

[0493] (3) DL measurement information, including RSRP, RSRQ, SINR and other information.

[0494] (4) MAC layer HARQ active retransmission decision;

[0495] (5) Decision on ARQ active retransmission in RLC layer;

[0496] (6) PDCP ARQ / duplication determination.

[0497] It is understandable that the network uses a trained model to help it decide whether and how to perform passive retransmission.

[0498] Optionally, the network may send the model derivation results to the UE, including at least one of the following (whether to retransmit, number of retransmissions, etc.):

[0499] A. Send MAC HARQ retransmission schedule in advance (e.g., retransmission schedule and new transmission schedule are sent simultaneously or consecutively, i.e., the retransmission schedule is sent to the UE before the new transmission reception result is available).

[0500] B. Send MAC HARQ-ACK information in advance;

[0501] C. Send an RLC ARQ SR in advance (if no polling or data is received from the sender), optionally including an ACK for no received or incorrectly received data (note that the AI-based SR triggering mechanism can work simultaneously with or replace the traditional non-AI SR triggering mechanism).

[0502] D. PDCP duplication retransmission configuration.

[0503] It's understandable that sending a NACK in advance can help the sender retransmit quickly, while sending a fake ACK can avoid meaningless retransmissions when the latency is too high.

[0504] 3. Model Management

[0505] For network-based model management, the network can be configured with relevant measurement reporting mechanisms to monitor model-permitted conditions, such as requiring the UE to report one or more of the following information:

[0506] (1) When cache-related information exceeds or falls below a certain threshold, such as BSR / DSR, further, data volume information can be obtained for caches of different data layers;

[0507] (2) When the QoS requirements of the UE side exceed or fall below a certain threshold (some of this information can be obtained from the CN side and the UE does not need to report this information), such as UL service patterns, etc.

[0508] (3) When the DL measurement information exceeds or falls below a certain threshold, including RSRP, RSRQ, SINR and other information;

[0509] (4) When the MAC layer HARQ active retransmission decision, the RLC layer ARQ active retransmission decision, and the PDCP ARQ / duplication decision exceed or fall below a certain threshold.

[0510] (5) When the QoS result exceeds or falls below a certain threshold, such as the QoS performance / performance obtained by the UE based on the above configuration of "early sending MAC HARQ retransmission scheduling, early sending RLC ARQ SR, PDCP duplication retransmission", such as latency, power consumption, resource consumption, etc., this mainly refers to the QoS result on the UE side.

[0511] Understandably, reporting the above information facilitates real-time network monitoring of model usage and timely adjustment of model parameters.

[0512] It should be noted that the model described in Scheme 2 is either the first AI model or the second AI model, and describes the inference process, training process and model management process for the first AI model or the second AI model.

[0513] Option 3: Downlink-oriented network-side model (see Figure 10), i.e., downlink-oriented active retransmission mechanism.

[0514] 1. Model Training

[0515] For network-side model training, the network needs to collect one or more of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as input for model training:

[0516] (1) QoS requirements related to the UE side (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service patterns and other information;

[0517] (2) DL measurement information, including RSRP, RSRQ, SINR and other information.

[0518] Optionally, the input for model training can be an indication based on at least one of the above-mentioned criteria.

[0519] For network-side model training, the output of the model training includes one or more of the following information:

[0520] A. The decision to initiate retransmission using HARQ at the MAC layer;

[0521] B. Decision on ARQ active retransmission at the RLC layer;

[0522] C.PDCP ARQ / duplication decision.

[0523] For model training on the network side, the network needs to collect at least one of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as a reward for model training:

[0524] QoS results, such as the QoS performance obtained by the UE based on the above "MAC layer HARQ active retransmission decision, RLC layer ARQ active retransmission decision, PDCP duplication decision", such as latency, power consumption, resource consumption, etc., mainly refer to the QoS results on the UE side.

[0525] Optionally, the reward for model training can be an indication determined based on one or more of the QoS results mentioned above.

[0526] It's understandable that if model training is done on the network side, considering that the network side currently has stronger computing power and storage than the UE side, it would be easier to do model training.

[0527] 2. Model Derivation / Model Inference

[0528] For a trained model, during the inference phase, the network needs to collect model input information. In addition to the information available to the network itself, the network also needs to obtain one or more of the following relevant information from the user experience (UE):

[0529] (1) QoS requirements related to the UE side (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service patterns and other information;

[0530] (2) DL measurement information, including RSRP, RSRQ, SINR and other information.

[0531] Understandably, this involves using a pre-trained model to help the network decide whether and how to perform autonomous retransmission.

[0532] Optionally, the network may send the model derivation results to the UE, including at least one of the following (whether to retransmit, number of retransmissions, etc.):

[0533] A. MAC HARQ's decision to automate retransmission;

[0534] B.RLC ARQ's decision to autonomously retransmit;

[0535] C. PDCP duplication decision to autonomously retransmit.

[0536] From a signaling perspective, this information can be reported through one or more of the following signaling methods:

[0537] DCI;

[0538] MAC CE;

[0539] RLC data PDU or RLC control PDU;

[0540] PDCP data PDU or PDCP control PDU;

[0541] RRC.

[0542] The reporting methods can be:

[0543] (1) Periodic indication;

[0544] (2) Event-triggered indications, such as when these decisions (whether to retransmit autonomously, the number of retransmissions) change, or when the number of retransmissions is higher or lower than a certain threshold.

[0545] Understandably, this information allows the UE to use the results of autonomous retransmission as input to the dual-side model to derive the passive retransmission mode.

[0546] 3. Model Management

[0547] For network-based model management, the network can be configured with relevant measurement reporting mechanisms to monitor model-permitted conditions, such as requiring the UE to report one or more of the following information:

[0548] (1) When the QoS requirements of the UE side exceed or fall below a certain threshold (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service pattern information;

[0549] (2) When the DL measurement information exceeds or falls below a certain threshold, including RSRP, RSRQ, SINR and other information;

[0550] (3) When the QoS result exceeds or falls below a certain threshold, such as the QoS performance obtained by the UE based on the above "MAC layer HARQ active retransmission decision, RLC layer ARQ active retransmission decision, PDCP duplication decision", such as latency, power consumption, resource consumption, etc., this mainly refers to the QoS result on the UE side.

[0551] It is understandable that reporting the above information facilitates real-time network monitoring of model usage and timely adjustment of model parameters.

[0552] Option 4: Downlink-oriented UE-side model (see Figure 11), i.e., downlink-oriented passive retransmission mechanism.

[0553] 1. Model Training

[0554] For network-side model training, the network needs to collect one or more of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as input for model training:

[0555] (1) QoS requirements related to the UE side (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service patterns and other information;

[0556] (2) DL measurement information, including RSRP, RSRQ, SINR and other information;

[0557] (3) The reception status of DL data, including the reception status of MAC layer HARQ, RLC layer ARQ, and PDCP layer PDU.

[0558] Optionally, the input for model training can be an indication based on at least one of the above-mentioned criteria.

[0559] For network-side model training, the output of the model training includes one or more of the following:

[0560] A. Send MAC HARQ retransmission schedule in advance;

[0561] B. Send MAC HARQ-ACK information in advance;

[0562] C. Send an RLC ARQ SR in advance (if no polling or data is received from the sender), which may include an ACK for data that was not received or was received incorrectly;

[0563] D. PDCP duplication retransmission configuration.

[0564] For model training on the network side, the network needs to collect at least one of the following information from the UE side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as a reward for model training:

[0565] QoS results, such as the QoS performance obtained by the UE based on the above configuration of "early sending MAC HARQ retransmission scheduling, early sending MAC HARQ-ACK information, early sending RLC ARQ SR, and PDCP duplication retransmission", include latency, power consumption, and resource consumption. This mainly refers to the QoS results on the UE side.

[0566] Optionally, the reward for model training can be an indication determined based on one or more of the QoS results mentioned above.

[0567] It's understandable to place model training on the network side, considering that the network side currently has stronger computing power and storage than the UE side, making it easier to perform model training.

[0568] For model training on the UE side (including the server connected to the UE), the UE needs to collect one or more of the following information from the network side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as input for model training:

[0569] (1) The data volume of the network-side DL buffer. Furthermore, the data volume information can be obtained for the caches of different data layers.

[0570] (2) Network-side QoS requirements related information (some of this information can be obtained from the UE itself, and this part of the information does not need to be obtained from the network side), such as UL service patterns and other information;

[0571] (3) UL measurement information, including RSRP, RSRQ, SINR and other information.

[0572] For network-side model training, the output of the model training includes one or more of the following:

[0573] A. Send MAC HARQ retransmission schedule in advance;

[0574] B. Send MAC HARQ-ACK information in advance;

[0575] C. Send an RLC ARQ SR in advance (if no polling or data is received from the sender), which may include an ACK for data that was not received or was received incorrectly;

[0576] D. PDCP duplication retransmission configuration.

[0577] For model training on the UE side (including the server connected to the UE), the UE needs to collect at least one of the following information from the network side (per-flow, per-bearer, per-LCH, per-LCG, etc.) as a reward for model training:

[0578] QoS results, such as the QoS performance obtained by the UE based on the above configuration of "early sending MAC HARQ retransmission scheduling, early sending MAC HARQ-ACK information, early sending RLC ARQ SR, and PDCP duplication retransmission", include latency, power consumption, and resource consumption. This mainly refers to the QoS results on the network side.

[0579] Optionally, the reward for model training can be an indication based on at least one of the above-mentioned criteria.

[0580] Optionally, the above model training can be iterated simultaneously with model derivation, i.e., model training can be performed online.

[0581] It's understandable that if model training is placed on the UE side, considering that the UE can obtain more information, the amount of information that needs to be transmitted is less, but it requires higher computing and storage capabilities from the UE.

[0582] 2. Model Derivation / Model Inference

[0583] For a trained model, during the inference phase, the UE needs to collect model input information. In addition to the information obtainable by the UE itself, the UE also needs to obtain one or more of the following information from the network:

[0584] (1) The data volume of the network-side DL buffer. Furthermore, the data volume information can be obtained for the caches of different data layers.

[0585] (2) Network-side QoS requirements related information (some of this information can be obtained from the UE itself, and this part of the information does not need to be obtained from the network side), such as UL service patterns and other information;

[0586] (3) UL measurement information, including RSRP, RSRQ, SINR and other information.

[0587] Optionally, the relevant information here may be specific to the UE or may include information about other UEs.

[0588] Optionally, the relevant information here can be direct information, or information that is further determined / derived from the information mentioned above.

[0589] Understandably, this involves using a pre-trained model to help the UE decide whether and how to perform passive retransmission.

[0590] Optionally, the UE may report the results of the model derivation to the network, including at least one of the following:

[0591] A. Send MAC HARQ retransmission schedule in advance;

[0592] B. Send MAC HARQ feedback (i.e., MAC HARQ-ACK information) in advance;

[0593] C. Send an RLC ARQ SR in advance (if no polling or data is received from the sender), optionally including an ACK for no received or incorrectly received data (note that the AI-based SR triggering mechanism can work simultaneously with or replace the traditional non-AI SR triggering mechanism).

[0594] D. PDCP duplication retransmission configuration.

[0595] Understandably, this reporting allows the network to decide whether to adjust the model configuration or make active retransmission decisions based on the results of passive retransmission.

[0596] 3. Model Management

[0597] For network-based model management, the network can be configured with relevant measurement reporting mechanisms to monitor model-permitted conditions, such as requiring the UE to report one or more of the following information:

[0598] (1) When the QoS requirements of the UE side exceed or fall below a certain threshold (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service pattern information;

[0599] (2) When the DL measurement information exceeds or falls below a certain threshold, including RSRP, RSRQ, SINR and other information;

[0600] (3) When sending MAC HARQ feedback in advance, send RLC ARQ SR in advance (if no polling is received from the sender, no data is received from the sender), optionally, which may include a decision on whether the ACK for not received or received incorrect data exceeds or falls below a certain threshold.

[0601] (4) When the QoS result exceeds or falls below a certain threshold, such as the QoS performance / performance obtained by the UE based on the above configuration of "sending MAC HARQ retransmission scheduling in advance, sending MAC HARQ feedback in advance, sending RLC ARQ SR in advance, and PDCP duplication retransmission", such as latency, power consumption, resource consumption, etc., this mainly refers to the QoS result on the UE side.

[0602] If the network needs to change the model configuration, the network sends an instruction message to the UE.

[0603] It is understandable that the UE reports the above information to facilitate real-time network monitoring of model usage and timely adjustment of model parameters.

[0604] For UE-based model management, the network pre-configures the relevant triggering conditions for model configuration changes, including one or more of the following:

[0605] (1) When the QoS requirements of the UE side exceed or fall below a certain threshold (some of this information can be obtained from the CN side and does not need to be reported by the UE), such as UL service pattern information;

[0606] (2) When the DL measurement information exceeds or falls below a certain threshold, including RSRP, RSRQ, SINR and other information;

[0607] (3) When sending MAC HARQ feedback in advance, send RLC ARQ SR in advance (if no polling is received from the sender, no data is received from the sender), optionally, which may include a decision on whether the ACK for not being received or receiving erroneous data exceeds or falls below a certain threshold.

[0608] (4) When the QoS result exceeds or falls below a certain threshold, such as the QoS performance / performance obtained by the UE based on the above configuration of "sending MAC HARQ retransmission scheduling in advance, sending MAC HARQ feedback in advance, sending RLC ARQ SR in advance, and PDCP duplication retransmission", such as latency, power consumption, resource consumption, etc., this mainly refers to the QoS result on the UE side.

[0609] When a certain triggering condition is met, the UE reports model change information to the network. Optionally, the reporting includes the triggering condition for the model change.

[0610] It is understandable that model management is on the UE side, which makes it easier to configure changes to model parameters in advance when the network connection is poor, and the UE can adjust the model parameters autonomously when relevant situations occur.

[0611] It should be noted that the models described in Scheme 1 and Scheme 3 are either the first AI model or the second AI model, and the inference process, training process, and model management process for the first AI model or the second AI model are described. The models described in Scheme 2 and Scheme 4 are either the third AI model or the fourth AI model, and the inference process, training process, and model management process for the third AI model or the fourth AI model are described.

[0612] The preferred embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this application, various simple modifications can be made to the technical solutions of this application, and these simple modifications all fall within the protection scope of this application. For example, the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this application will not describe the various possible combinations separately. Furthermore, various different embodiments of this application can also be arbitrarily combined, as long as they do not violate the spirit of this application, they should also be considered as the content disclosed in this application. Moreover, without conflict, the various embodiments and / or the technical features in the various embodiments described in this application can be arbitrarily combined with the prior art, and the resulting technical solutions should also fall within the protection scope of this application.

[0613] It should also be understood that in the various method embodiments of this application, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. Furthermore, in the embodiments of this application, the terms "downlink" and "uplink" are used to indicate the transmission direction of signals or data. "Downlink" indicates that the transmission direction of signals or data is a first direction from the site to the user equipment in the cell, and "uplink" indicates that the transmission direction of signals or data is a second direction from the user equipment in the cell to the site. For example, "downlink signal" indicates that the transmission direction of the signal is the first direction. Additionally, in the embodiments of this application, the term "and / or" is merely a description of the association relationship between related objects, indicating that three relationships can exist. Specifically, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0614] Based on the foregoing embodiments, this application provides corresponding wireless communication devices.

[0615] Figure 12 is a schematic diagram of the structure of a wireless communication device provided in an embodiment of this application. Applied to a first device, as shown in Figure 12, the wireless communication device 1200 (hereinafter referred to as device 1200) includes:

[0616] The first determining unit 1201 is configured to determine the retransmission decision corresponding to the first information based on the first AI model;

[0617] The first communication unit 1202 is configured to send retransmission data to the second device when a retransmission decision is made.

[0618] In some embodiments, the retransmission decision includes whether to retransmit and / or the number of retransmissions.

[0619] In some embodiments, the first information includes one or more of the following:

[0620] (1) QoS requirements related information;

[0621] (2) Channel measurement information;

[0622] (3) Caching related information;

[0623] (4) Second information sent by the second device; wherein the second information includes one or more of the following: feedback information on the data sent by the first device, time-related information on the transmission of the feedback information, first indication information on the data sent by the first device, and time-related information on the transmission of the first indication information; the first indication information is used to indicate whether to retransmit the data.

[0624] (5) QoS performance;

[0625] (6) Third information; wherein the third information is information determined based on one or more of the second information, QoS performance, QoS requirement information, channel measurement information and buffer information.

[0626] Furthermore, in some embodiments, QoS requirement-related information includes terminal-side QoS requirement-related information and / or network-side QoS requirement-related information.

[0627] Furthermore, in some embodiments, the channel measurement information includes uplink channel measurement information and / or downlink channel measurement information.

[0628] Furthermore, in some embodiments, QoS performance includes terminal-side QoS performance and / or network-side QoS performance.

[0629] In some embodiments, the first communication unit 1202 is further configured to send a retransmission decision to the second device.

[0630] Furthermore, in some embodiments, the first communication unit 1202 is configured to send a retransmission decision to the second device via one or more of the following signaling:

[0631] (1)UCI or DCI;

[0632] (2) MAC CE;

[0633] (3) RLC data PDU or RLC control PDU;

[0634] (4) PDCP data PDU or PDCP control PDU;

[0635] (5) RRC.

[0636] For example, in some embodiments, the first communication unit 1202 is configured to periodically send retransmission decisions to the second device.

[0637] For example, in some embodiments, the first communication unit 1202 is configured to send a retransmission decision to the second device when a first condition is met.

[0638] For example, the first condition includes one or more of the following:

[0639] (1) The retransmission decision is different from the previous retransmission decision;

[0640] (2) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold.

[0641] In some embodiments, the first determining unit 1201 is further configured to update the model parameters of the first AI model to new model parameters based on the QoS performance generated by the retransmission decision; wherein the QoS performance generated by the retransmission decision includes terminal-side QoS performance and / or network-side QoS performance.

[0642] Furthermore, in some embodiments, the first determining unit 1201 is configured to: determine a first difference of the first AI model based on the QoS performance generated by the retransmission decision; and update the model parameters of the first AI model to new model parameters based on the first difference.

[0643] In some of the above embodiments, the first device is a terminal device and the second device is a network device. Hereinafter, the embodiments where the first device is a terminal device and the second device is a network device, specifically those involving active retransmission, will be referred to as Embodiment Five.

[0644] Example 5:

[0645] In some embodiments, the apparatus 1200 applied to the terminal device, its first communication unit 1202, is further configured to receive one or more of the following information sent by the network device:

[0646] (1) Information related to QoS requirements on the network side;

[0647] (2) Uplink channel measurement information;

[0648] (3) Network-side QoS performance.

[0649] Furthermore, in some embodiments, cache-related information includes uplink data cache information on the terminal side.

[0650] Furthermore, in some embodiments, the first information includes information specific to the terminal device, or the first information includes information specific to the terminal device and information from other terminal devices different from the terminal device.

[0651] In some embodiments, the apparatus 1200 applied to the terminal device, wherein its first communication unit 1202 is further configured to receive second indication information sent by the network device, the second indication information being used to indicate that the model parameters of the first AI model be updated to new model parameters.

[0652] In some embodiments, the apparatus 1200 applied to the terminal device, wherein its first communication unit 1202 is further configured to send terminal-side QoS performance to the network device.

[0653] In some embodiments, the new model parameters are related to the QoS performance resulting from the retransmission decision; the QoS performance resulting from the retransmission decision includes the terminal-side QoS performance resulting from the retransmission decision and / or the network-side QoS performance resulting from the retransmission decision.

[0654] In some embodiments, the apparatus 1200 applied to the terminal device, wherein its first communication unit 1202 is further configured to receive third indication information sent by the network device, the third indication information being used to indicate that the first AI model be replaced with the second AI model.

[0655] In some embodiments, the apparatus 1200 applied to the terminal device, its first communication unit 1202, is further configured to report information satisfying one or more of the following second conditions to the network device:

[0656] (1) The uplink data cache information on the terminal side exceeds or falls below the corresponding threshold;

[0657] (2) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0658] (3) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0659] (4) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold;

[0660] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0661] In some embodiments, the apparatus 1200 applied to the terminal device, its first communication unit 1202, is further configured to replace the first AI model with a second AI model when one or more of the following second conditions are met:

[0662] The uplink data cache information on the terminal side exceeds or falls below the corresponding threshold;

[0663] The terminal-side QoS requirements are either higher or lower than the corresponding threshold.

[0664] Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0665] The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold;

[0666] The terminal-side QoS performance exceeds or falls below the corresponding threshold;

[0667] The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0668] In some embodiments, the apparatus 1200 applied to the terminal device, wherein its first communication unit 1202 is further configured to receive fourth indication information sent by the network device, the fourth indication information being used to indicate the second condition.

[0669] In some embodiments, the apparatus 1200 applied to the terminal device, wherein its first communication unit 1202 is further configured to report model replacement information for the first AI model to the network device.

[0670] In some embodiments, model replacement-related information includes a second condition that triggers the replacement of the first AI model with the second AI model.

[0671] In some of the above embodiments, the first device is a network device and the second device is a terminal device. The embodiments where the first device is a network device and the second device is a terminal device, specifically those involving active retransmission, will be referred to as Embodiment Six.

[0672] Example 6:

[0673] In some embodiments, the first device is a network device and the second device is a terminal device.

[0674] In some embodiments, the apparatus 1200 applied to the network device, its first communication unit 1202, is further configured to receive one or more of the following information sent by the terminal device:

[0675] (1) QoS requirements related to the terminal side;

[0676] (2) Downlink channel measurement information;

[0677] (3) QoS performance on the terminal side.

[0678] In some embodiments, cache-related information includes network-side downlink data cache information.

[0679] In some embodiments, the apparatus 1200 applied to the network device, wherein its first communication unit 1202 is further configured to replace the first AI model with the second AI model when a second condition and / or a third condition is met.

[0680] In some embodiments, the second condition includes one or more of the following:

[0681] (1) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0682] (2) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0683] (3) The QoS performance of the terminal side exceeds or falls below the corresponding threshold.

[0684] In some embodiments, the apparatus 1200 applied to the network device, wherein its first communication unit 1202 is further configured to send fifth indication information to the terminal device, the fifth indication information being used to indicate the second condition.

[0685] In some embodiments, the third condition includes one or more of the following:

[0686] (1) The downlink data cache information on the network side exceeds or falls below the corresponding threshold;

[0687] (2) Uplink channel measurement information exceeds or falls below the corresponding threshold;

[0688] (3) The QoS requirements related to the network side exceed or fall below the corresponding threshold;

[0689] (4) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold;

[0690] (5) The QoS performance on the network side exceeds or falls below the corresponding threshold.

[0691] Figure 13 is a schematic diagram of the structure of a wireless communication device provided in an embodiment of this application, applied to a second device. As shown in Figure 13, the wireless communication device 1300 (hereinafter referred to as device 1300) includes:

[0692] The second determining unit 1301 is configured to determine feedback information and / or first indication information corresponding to the fourth information based on the third AI model; wherein the feedback information is feedback information for the data transmitted by the first device; and the first indication information is used to indicate whether to retransmit the data.

[0693] The second communication unit 1302 is configured to send feedback information and / or first indication information to the first device.

[0694] In some embodiments, the second communication unit 1302 is configured to send feedback information and / or first indication information to the first device according to transmission time-related information; wherein, the transmission time-related information includes the first transmission time of the feedback information and / or the second transmission time of the first indication information.

[0695] In some embodiments, the first transmission time and / or the second transmission time both occur before the verification result of the transmitted data is obtained.

[0696] In some embodiments, the first transmission time and / or the second transmission time are pre-configured.

[0697] In other embodiments, the first transmission time and / or the second transmission time are determined by a third AI model using fourth information.

[0698] In some embodiments, the first transmission time and / or the second transmission time satisfy one or more of the following:

[0699] (1) Before the timer expires; the timer is used to trigger the sending of feedback information;

[0700] (2) Before receiving the transmitted data;

[0701] (3) After receiving the transmitted data and before obtaining the decoding result of the transmitted data.

[0702] In some embodiments, the feedback information includes ACK information, which is information regarding data loss or receiving incorrect data.

[0703] In some embodiments, the fourth information includes one or more of the following:

[0704] (1) QoS requirements related information;

[0705] (2) Channel measurement information;

[0706] (3) Caching related information;

[0707] (4) The retransmission decision sent by the first device;

[0708] (5) QoS performance;

[0709] (6) Downlink data reception status;

[0710] (7) Fifth information; wherein the fifth information is information determined based on one or more of the following: QoS requirement-related information, channel measurement information, buffer-related information, retransmission decision sent by the first device, QoS performance, and downlink data reception status.

[0711] In some embodiments, QoS requirement-related information includes terminal-side QoS requirement-related information and / or network-side QoS requirement-related information.

[0712] In some embodiments, the channel measurement information includes uplink channel measurement information and / or downlink channel measurement information.

[0713] In some embodiments, QoS performance includes terminal-side QoS performance and / or network-side QoS performance.

[0714] In some embodiments, the second communication unit 1302 is further configured to send second information to the first device; wherein the second information includes one or more of the following:

[0715] (1) Feedback information;

[0716] (2) The first time the feedback information is sent;

[0717] (3) First instruction information;

[0718] (4) The second time of sending the first instruction information.

[0719] In some embodiments, the second communication unit 1302 is configured to send second information to the first device via one or more of the following signaling:

[0720] (1)UCI or DCI;

[0721] (2) MAC CE;

[0722] (3) RLC data PDU or RLC control PDU;

[0723] (4) PDCP data PDU or PDCP control PDU;

[0724] (5) RRC.

[0725] In some embodiments, the second communication unit 1302 is configured to periodically send second information to the first device.

[0726] In other embodiments, the second communication unit 1302 is configured to send second information to the first device when a fourth condition is met.

[0727] In some embodiments, the fourth condition includes one or more of the following:

[0728] (1) The cumulative number of times the same feedback information is sent exceeds or falls below the corresponding threshold;

[0729] (2) The cumulative number of times the same first instruction information is sent exceeds or falls below the corresponding threshold.

[0730] In some embodiments, the second determining unit 1301 is further configured to update the model parameters of the third AI model to new model parameters based on the QoS performance generated by the second information; wherein the QoS performance generated by the second information includes terminal-side QoS performance and / or network-side QoS performance.

[0731] In some embodiments, the second determining unit 1301 is configured to: determine a second difference of the third AI model based on the QoS performance generated by the second information; and update the model parameters of the third AI model to new model parameters based on the second difference.

[0732] In some embodiments, the feedback information includes MAC HARQ-ACK information and / or RLC ARQ SR; and / or, the first indication information includes MAC HARQ retransmission scheduling and / or PDCP duplication retransmission configuration.

[0733] In some embodiments, the downlink data reception status includes one or more of the following reception statuses:

[0734] (1) MAC layer HARQ reception status;

[0735] (2) ARQ reception status at the RLC layer;

[0736] (3) PDCP layer PDU reception status.

[0737] In some of the above embodiments, the first device is a terminal device and the second device is a network device. Hereinafter, the embodiments where the first device is a terminal device and the second device is a network device, specifically the passive retransmission embodiments, will be referred to as Embodiment Seven.

[0738] Example 7:

[0739] In some embodiments, the second device is a network device and the first device is a terminal device.

[0740] In some embodiments, the second communication unit 1302 of the apparatus 1300 applied to the network device is further configured to receive one or more of the following information sent by the terminal device:

[0741] (1) QoS requirements related to the terminal side;

[0742] (2) Downlink channel measurement information;

[0743] (3) Terminal-side QoS performance;

[0744] (4) Uplink data cache information on the terminal side.

[0745] In some embodiments, the second determining unit 1301 of the apparatus 1300 applied to the network device is further configured to replace the third AI model with the fourth AI model if the second condition and / or the fifth condition are met.

[0746] In some embodiments, the second condition includes one or more of the following:

[0747] (1) The uplink data cache information on the terminal side exceeds or falls below the corresponding threshold;

[0748] (2) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0749] (3) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0750] (4) The number of retransmissions in the retransmission decision exceeds or falls below the corresponding threshold;

[0751] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0752] In some embodiments, the second communication unit 1302 of the apparatus 1300 applied to the network device is further configured to send fourth indication information to the terminal device, the fourth indication information being used to indicate relevant information for reporting the second condition.

[0753] In some embodiments, the fifth condition includes one or more of the following:

[0754] Uplink channel measurement information exceeds or falls below the corresponding threshold;

[0755] (1) The QoS requirements on the network side exceed or fall below the corresponding threshold;

[0756] (2) The cumulative number of times the same feedback information is sent exceeds or falls below the corresponding threshold;

[0757] (3) The cumulative number of times the same first instruction information is sent exceeds or falls below the corresponding threshold;

[0758] (4) The QoS performance on the network side exceeds or falls below the corresponding threshold.

[0759] In some of the above embodiments, the first device is a network device and the second device is a terminal device. Hereinafter, the embodiments where the first device is a network device and the second device is a terminal device, specifically the passive retransmission embodiments, will be referred to as Embodiment Eight.

[0760] Example 8:

[0761] In some embodiments, the second device is a terminal device and the first device is a network device.

[0762] In some embodiments, the second communication unit 1302 of the apparatus 1300 applied to the terminal device is further configured to receive one or more of the following information sent by the network device:

[0763] (1) Information related to QoS requirements on the network side;

[0764] (2) Uplink channel measurement information;

[0765] (3) Downlink data caching information on the network side;

[0766] (4) Network-side QoS performance.

[0767] In some embodiments, the fourth information includes information specific to the terminal device, or the fourth information includes information specific to the terminal device and information from other terminal devices different from the terminal device.

[0768] In some embodiments, the second communication unit 1302 of the device 1300 applied to the terminal device is further configured to receive fifth indication information sent by the network device, the fifth indication information being used to indicate that the model parameters of the third AI model be updated to new model parameters.

[0769] In some embodiments, the new model parameters are related to the QoS performance generated based on the second information.

[0770] In some embodiments, the QoS performance generated based on the second information includes the terminal-side QoS performance generated based on the second information and / or the network-side QoS performance generated based on the second information.

[0771] In some embodiments, the second communication unit 1302 of the apparatus 1300 applied to the terminal device is further configured to send one or more of the following information to the network device:

[0772] (1) Terminal-side QoS performance;

[0773] (2) QoS requirements related to the terminal side;

[0774] (3) Downlink channel measurement information;

[0775] (4) Downlink data reception status.

[0776] In some embodiments, the second communication unit 1302 of the device 1300 applied to the terminal device is further configured to receive a sixth instruction information sent by the network device, the sixth instruction information being used to instruct the third AI model to be replaced with a fourth AI model.

[0777] In some embodiments, the second communication unit 1302 of the apparatus 1300 applied to the terminal device is further configured to report information satisfying one or more of the following sixth conditions to the network device:

[0778] (1) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0779] (2) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0780] (3) The cumulative number of times the same feedback information is sent exceeds or falls below the corresponding threshold;

[0781] (4) The cumulative number of times the same first instruction information is sent exceeds or falls below the corresponding threshold;

[0782] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0783] In some embodiments, the second determining unit 1301 of the apparatus 1300 applied to the terminal device is further configured to replace the third AI model with the fourth AI model if one or more of the following sixth conditions are met:

[0784] (1) The QoS requirements on the terminal side exceed or fall below the corresponding threshold;

[0785] (2) Downlink channel measurement information exceeds or falls below the corresponding threshold;

[0786] (3) The cumulative number of times the same feedback information is sent exceeds or falls below the corresponding threshold;

[0787] (4) The cumulative number of times the same first instruction information is sent exceeds or falls below the corresponding threshold;

[0788] (5) The QoS performance on the terminal side exceeds or falls below the corresponding threshold.

[0789] In some embodiments, the second communication unit 1302 of the device 1300 applied to the terminal device is further configured to receive seventh indication information sent by the network device, the seventh indication information being used to indicate a sixth condition.

[0790] In some embodiments, the second communication unit 1302 of the device 1300 applied to the terminal device is further configured to report model replacement information for the third AI model to the network device.

[0791] In some embodiments, model replacement information includes a sixth condition that triggers the replacement of the third AI model with the fourth AI model.

[0792] Those skilled in the art should understand that the description of the wireless communication device in the embodiments of this application can be understood with reference to the description of the wireless communication method in the embodiments of this application.

[0793] Figure 14 is a schematic structural diagram of a communication device 1400 provided in an embodiment of this application. This communication device can be a terminal device or a network device. The communication device 1400 shown in Figure 14 includes a processor 1410, which can call and run computer programs from memory to implement the methods in the embodiments of this application.

[0794] Optionally, as shown in FIG14, the communication device 1400 may further include a memory 1420. The processor 1410 may retrieve and run computer programs from the memory 1420 to implement the methods described in the embodiments of this application.

[0795] The memory 1420 can be a separate device independent of the processor 1410, or it can be integrated into the processor 1410.

[0796] Optionally, as shown in FIG14, the communication device 1400 may further include a transceiver 1430, and the processor 1410 may control the transceiver 1430 to communicate with other devices. Specifically, it may send information or data to other devices or receive information or data sent by other devices.

[0797] The transceiver 1430 may include a transmitter and a receiver. The transceiver 1430 may further include an antenna, and the number of antennas may be one or more.

[0798] Optionally, the communication device 1400 may specifically be a network device in the embodiments of this application, and the communication device 1400 may implement the corresponding processes implemented by the network device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0799] Optionally, the communication device 1400 may specifically be a terminal device in the embodiments of this application, and the communication device 1400 may implement the corresponding processes implemented by the terminal device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0800] Figure 15 is a schematic structural diagram of a chip according to an embodiment of this application. The chip 1500 shown in Figure 15 includes a processor 1510, which can call and run computer programs from memory to implement the methods in the embodiments of this application.

[0801] Optionally, as shown in FIG15, chip 1500 may further include memory 1520. Processor 1510 may retrieve and run computer programs from memory 1520 to implement the methods in the embodiments of this application.

[0802] The memory 1520 can be a separate device independent of the processor 1510, or it can be integrated into the processor 1510.

[0803] Optionally, the chip 1500 may also include an input interface 1530. The processor 1510 can control the input interface 1530 to communicate with other devices or chips; specifically, it can acquire information or data sent by other devices or chips.

[0804] Optionally, the chip 1500 may also include an output interface 1540. The processor 1510 can control the output interface 1540 to communicate with other devices or chips, specifically, to output information or data to other devices or chips.

[0805] Optionally, the chip can be applied to the network device in the embodiments of this application, and the chip can implement the corresponding processes implemented by the network device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0806] Optionally, the chip can be applied to the terminal device in the embodiments of this application, and the chip can implement the corresponding processes implemented by the terminal device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0807] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0808] This application also provides a computer storage medium storing one or more programs, which can be executed by one or more processors to implement the methods in this application.

[0809] Figure 16 is a schematic block diagram of a communication system 1600 provided in an embodiment of this application. As shown in Figure 16, the communication system 1600 includes a terminal device 1610 and a network device 1620.

[0810] The terminal device 1610 can be used to implement the corresponding functions implemented by the terminal device in the above method, and the network device 1620 can be used to implement the corresponding functions implemented by the network device in the above method. For the sake of brevity, they will not be described in detail here.

[0811] It should be understood that the processor in the embodiments of this application may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method embodiments can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor described above can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor can be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0812] It is understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DR RAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0813] It should be understood that the above-described memory is exemplary and not a limiting description. For example, the memory in the embodiments of this application may also be static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DR RAM), etc. That is to say, the memory in the embodiments of this application is intended to include, but is not limited to, these and any other suitable types of memory.

[0814] This application also provides a computer-readable storage medium for storing computer programs.

[0815] Optionally, the computer-readable storage medium can be applied to the network device in the embodiments of this application, and the computer program causes the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0816] Optionally, the computer-readable storage medium can be applied to the terminal device in the embodiments of this application, and the computer program causes the computer to execute the corresponding processes implemented by the terminal device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0817] This application also provides a computer program product, including computer program instructions.

[0818] Optionally, the computer program product can be applied to the network device in the embodiments of this application, and the computer program instructions cause the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of this application. For the sake of brevity, they will not be described in detail here.

[0819] Optionally, the computer program product can be applied to the terminal device in the embodiments of this application, and the computer program instructions cause the computer to execute the corresponding processes implemented by the terminal device in the various methods of the embodiments of this application. For the sake of brevity, they will not be described in detail here.

[0820] This application also provides a computer program.

[0821] Optionally, the computer program can be applied to the network device in the embodiments of this application. When the computer program is run on a computer, it causes the computer to execute the corresponding processes implemented by the network device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0822] Optionally, the computer program can be applied to the terminal device in the embodiments of this application. When the computer program is run on the computer, it causes the computer to execute the corresponding processes implemented by the terminal device in the various methods of the embodiments of this application. For the sake of brevity, it will not be described in detail here.

[0823] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0824] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0825] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0826] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0827] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0828] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0829] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

A method of wireless communication, the method comprising: determining, by a first device, a retransmission decision corresponding to first information according to a first AI model; in a case where the retransmission decision is retransmission, transmitting, by the first device, retransmission data to a second device. The method of claim 1, wherein, The retransmission decision comprises whether to retransmit and / or a number of retransmissions. The method of claim 1, wherein, The first information comprises one or more of the following: QoS requirement related information; channel measurement information; buffer related information; second information transmitted by the second device; wherein the second information comprises one or more of feedback information for transmission data of the first device, transmission time related information of the feedback information, first indication information for the transmission data of the first device, and transmission time related information of the first indication information; the first indication information is used to indicate whether to retransmit the transmission data; QoS performance; third information; wherein the third information is information determined based on one or more of the second information, the QoS performance, the QoS requirement related information, the channel measurement information, and the buffer related information. The method of claim 3, wherein, The QoS requirement related information comprises terminal side QoS requirement related information and / or network side QoS requirement related information. The method of claim 3, wherein, The channel measurement information comprises uplink channel measurement information and / or downlink channel measurement information. The method of claim 3, wherein, The QoS performance comprises terminal side QoS performance and / or network side QoS performance. The method of any one of claims 1 to 6, wherein, The method further comprises: transmitting, by the first device, the retransmission decision to the second device. According to claim 7, wherein transmitting, by the first device, the retransmission decision to the second device through one or more of the following signaling: UCI or DCI; MAC CE; RLC data PDU or RLC control PDU; PDCP data PDU or PDCP control PDU; RRC. According to claim 7 or 8, wherein transmitting, by the first device, the retransmission decision to the second device periodically. According to claim 7 or 8, wherein transmitting, by the first device, the retransmission decision to the second device in a case where a first condition is met. The method of claim 10, wherein, The first condition comprises one or more of the following: The retransmission decision is different from a previous retransmission decision; The number of retransmissions in the retransmission decision exceeds or is lower than a corresponding threshold. The method of any one of claims 3-11, wherein The first device is a terminal device, and the second device is a network device. The method of claim 12, wherein, The method further comprises: receiving, by the terminal device, one or more of the following information transmitted by the network device: network side QoS requirement related information; uplink channel measurement information; network side QoS performance. The method according to claim 12 or 13, wherein The buffer related information comprises terminal side uplink data buffer information. The method of any one of claims 12 to 14, wherein, The first information comprises information for the terminal device, or the first information comprises information for the terminal device and information for other terminal devices different from the terminal device. The method of any one of claims 12 to 15, wherein, The method further comprises: The terminal device receives second indication information sent by the network device, where the second indication information is used to indicate that the model parameters of the first AI model are updated to new model parameters. The method of claim 16, wherein, The method further includes: The terminal device sends terminal-side QoS performance to the network device. The method according to claim 16 or 17, wherein The new model parameters are related to QoS performance generated based on the retransmission decision; the QoS performance generated based on the retransmission decision includes terminal-side QoS performance generated based on the retransmission decision and / or network-side QoS performance generated based on the retransmission decision. The method of any one of claims 12 to 15, wherein, The method further includes: The terminal device receives third indication information sent by the network device, where the third indication information is used to indicate that the first AI model is replaced by a second AI model. The method of claim 19, wherein, The method further includes: The terminal device reports, to the network device, information that meets one or more of the following second conditions: Terminal-side uplink data buffer information exceeds or is lower than a corresponding threshold; Terminal-side QoS requirement related information exceeds or is lower than a corresponding threshold; Downlink channel measurement information exceeds or is lower than a corresponding threshold; The number of retransmissions in the retransmission decision exceeds or is lower than a corresponding threshold; Terminal-side QoS performance exceeds or is lower than a corresponding threshold. The method of any one of claims 12 to 18, wherein, The method further includes: The terminal device replaces the first AI model by a second AI model in a case that one or more of the following second conditions are met: Terminal-side uplink data buffer information exceeds or is lower than a corresponding threshold; Terminal-side QoS requirement related information exceeds or is lower than a corresponding threshold; Downlink channel measurement information exceeds or is lower than a corresponding threshold; The number of retransmissions in the retransmission decision exceeds or is lower than a corresponding threshold; Terminal-side QoS performance exceeds or is lower than a corresponding threshold. The method according to claim 20 or 21, wherein The method further includes: The terminal device receives fourth indication information sent by the network device, where the fourth indication information is used to indicate the second conditions. The method of claim 21 or 22, wherein, The method further includes: The terminal device reports, to the network device, model replacement related information for the first AI model. The method of claim 23, wherein, The model replacement related information includes second conditions that trigger the replacement of the first AI model by the second AI model. The method of any one of claims 3-11, wherein The first device is a network device, and the second device is a terminal device. The method of claim 25, wherein, The method further includes: The network device receives one or more of the following information sent by the terminal device: Terminal-side QoS requirement related information; Downlink channel measurement information; Terminal-side QoS performance. The method of claim 25 or 26, wherein, The buffer related information includes network-side downlink data buffer information. The method of any one of claims 12-15, 25-27, wherein, The method further includes: The first device updates model parameters of the first AI model to new model parameters based on QoS performance generated based on the retransmission decision; where the QoS performance generated based on the retransmission decision includes terminal-side QoS performance and / or network-side QoS performance. The method of claim 28, wherein, The first device updates model parameters of the first AI model to new model parameters based on QoS performance generated based on the retransmission decision, including: The first device determines a first difference of the first AI model based on QoS performance generated based on the retransmission decision; The first device updates model parameters of the first AI model to new model parameters based on the first difference. The method of any one of claims 25 to 29, wherein, The method further includes: The network device replaces the first AI model with a second AI model if a second condition and / or a third condition is met. The method of claim 30, wherein, The second condition includes one or more of the following: Terminal-side QoS requirement related information exceeds or is below a corresponding threshold; Downlink channel measurement information exceeds or is below a corresponding threshold; Terminal-side QoS performance exceeds or is below a corresponding threshold. The method of claim 30 or 31, wherein, The method further includes: The network device sends fifth indication information to the terminal device, the fifth indication information being used to indicate the second condition. The method of any one of claims 30 to 32, wherein, The third condition includes one or more of the following: Network-side downlink data buffer information exceeds or is below a corresponding threshold; Uplink channel measurement information exceeds or is below a corresponding threshold; Network-side QoS requirement related information exceeds or is below a corresponding threshold; The number of retransmissions in the retransmission decision exceeds or is below a corresponding threshold; Network-side QoS performance exceeds or is below a corresponding threshold. A wireless communication method, the method comprising: A second device determines feedback information corresponding to fourth information and / or first indication information according to a third AI model; wherein the feedback information is feedback information for transmitted data of a first device; and the first indication information is used to indicate whether to retransmit the transmitted data; The second device sends the feedback information and / or the first indication information to the first device. The method of claim 34, wherein, The second device sends the feedback information and / or the first indication information to the first device, comprising: The second device sends the feedback information and / or the first indication information to the first device according to transmission time related information; wherein the transmission time related information includes a first transmission time of the feedback information and / or a second transmission time of the first indication information. The method of claim 35, wherein, The first transmission time and / or the second transmission time are before a check result of the transmitted data is obtained. The method of claim 35 or 36, wherein, The first transmission time and / or the second transmission time are preconfigured. The method of claim 37, wherein, The first transmission time and / or the second transmission time are determined according to the third AI model through fourth information. The method of any one of claims 35 to 38, wherein, The first transmission time and / or the second transmission time meet one or more of the following: Before a timer expires; the timer is used to trigger sending the feedback information; Before the transmitted data is received; After the transmitted data is received and before a decoding result of the transmitted data is obtained. The method of any one of claims 35-39, wherein The feedback information includes ACK information, the ACK information being information for data loss or received error data. The method of any one of claims 34-40, wherein The fourth information includes one or more of the following information: QoS requirement related information; Channel measurement information; Buffer related information; Retransmission decision of the first device; QoS performance; Reception of downlink data; and fifth information; wherein the fifth information is information determined based on one or more of the QoS requirement related information, the channel measurement information, the buffer related information, the retransmission decision sent by the first device, the QoS performance, and the reception status of the downlink data. The method of claim 41, wherein, The QoS requirement related information comprises terminal side QoS requirement related information and / or network side QoS requirement related information. The method of claim 41, wherein, The channel measurement information comprises uplink channel measurement information and / or downlink channel measurement information. The method of any one of claims 41 to 43, wherein, The QoS performance comprises terminal side QoS performance and / or network side QoS performance. The method of any one of claims 34 to 44, wherein, The method further comprises: The second device sends second information to the first device; wherein the second information comprises one or more of the following information: The feedback information; The first sending time of the feedback information; The first indication information; The second sending time of the first indication information. According to the method of claim 45, wherein The second device sends the second information to the first device through one or more of the following signaling: UCI or DCI; MAC CE; RLC data PDU or RLC control PDU; PDCP data PDU or PDCP control PDU; RRC. According to the method of claim 45 or 46, wherein The second device periodically sends the second information to the first device. According to the method of claim 45 or 46, wherein The second device sends the second information to the first device in the case that a fourth condition is met. The method of claim 48, wherein, The fourth condition comprises one or more of the following: The cumulative sending number of the same feedback information exceeds or is lower than a corresponding threshold; The cumulative sending number of the same first indication information exceeds or is lower than a corresponding threshold. The method of any one of claims 41 to 49, wherein The second device is a network device, and the first device is a terminal device. The method of claim 50, wherein, The method further comprises: The network device receives one or more of the following information sent by the terminal device: Terminal side QoS requirement related information; Downlink channel measurement information; Terminal side QoS performance; Terminal side uplink data buffer information. The method of claim 50 or 51, wherein, The method further comprises: The network device replaces the third AI model with a fourth AI model in the case that a second condition and / or a fifth condition is met. The method of claim 52, wherein, The second condition comprises one or more of the following: The terminal side uplink data buffer information exceeds or is lower than a corresponding threshold; The terminal side QoS requirement related information exceeds or is lower than a corresponding threshold; The downlink channel measurement information exceeds or is lower than a corresponding threshold; The number of retransmissions in the retransmission decision exceeds or is lower than a corresponding threshold; The terminal side QoS performance exceeds or is lower than a corresponding threshold. The method of claim 52 or 53, wherein, The method further comprises: The network device sends fourth indication information to the terminal device, wherein the fourth indication information is used to indicate the information related to the second condition. The method of any one of claims 52 to 54, wherein, The fifth condition comprises one or more of the following: The uplink channel measurement information exceeds or is lower than a corresponding threshold; The network side QoS requirement related information exceeds or is lower than a corresponding threshold; The cumulative sending number of the same feedback information exceeds or is lower than a corresponding threshold; The cumulative sending number of the same first indication information exceeds or is lower than a corresponding threshold; The network side QoS performance exceeds or is lower than a corresponding threshold. The method of any one of claims 41 to 49, wherein, The second device is a terminal device, and the first device is a network device. The method of claim 56, wherein, The method further comprises: The terminal device receives one or more of the following information sent by the network device: Network side QoS requirement related information; Uplink channel measurement information; Network side downlink data buffer information; Network side QoS performance. The method of any one of claims 56 or 57, wherein, The fourth information includes information for the terminal device, or the fourth information includes information for the terminal device and information for other terminal devices different from the terminal device. The method of any one of claims 56-58, wherein The method further comprises: The terminal device receives fifth indication information sent by the network device, and the fifth indication information is used to indicate that the model parameters of the third AI model are updated to new model parameters. The method of claim 59, wherein, The new model parameters are related to the QoS performance generated based on the second information. The method of claim 60, wherein, The QoS performance generated based on the second information includes terminal side QoS performance generated based on the second information and / or network side QoS performance generated based on the second information. The method of any one of claims 59 to 61, wherein The method further comprises: The terminal device sends one or more of the following information to the network device: Terminal side QoS performance; Terminal side QoS requirement related information; Downlink channel measurement information; The reception situation of downlink data. The method of any one of claims 45 to 58, wherein, The method further comprises: The second device updates the model parameters of the third AI model to new model parameters based on the QoS performance generated based on the second information, wherein the QoS performance generated based on the second information includes terminal side QoS performance and / or network side QoS performance. The method of claim 63, wherein, The second device updates the model parameters of the third AI model to new model parameters based on the QoS performance generated based on the second information, comprising: The second device determines a second difference value of the third AI model based on the QoS performance generated based on the second information; The second device updates the model parameters of the third AI model to new model parameters based on the second difference value. The method of any one of claims 56 to 64, wherein The method further comprises: The terminal device receives sixth indication information sent by the network device, and the sixth indication information is used to indicate that the third AI model is replaced by a fourth AI model. The method of claim 65, wherein, The method further comprises: The terminal device reports one or more of the following sixth conditions to the network device: Terminal side QoS requirement related information exceeds or is lower than a corresponding threshold; Downlink channel measurement information exceeds or is lower than a corresponding threshold; The cumulative sending number of the same feedback information exceeds or is lower than a corresponding threshold; The cumulative sending number of the same first indication information exceeds or is lower than a corresponding threshold; Terminal side QoS performance exceeds or is lower than a corresponding threshold. The method of any one of claims 56 to 64, wherein The method further comprises: The terminal device replaces the third AI model with a fourth AI model when one or more of the following sixth conditions are met: The terminal device replaces the third AI model with a fourth AI model when one or more of the following sixth conditions are met: terminal-side QoS requirement related information exceeds or is below a corresponding threshold; downlink channel measurement information exceeds or is below a corresponding threshold; cumulative sending times of the same feedback information exceed or are below a corresponding threshold; cumulative sending times of the same first indication information exceed or are below a corresponding threshold; terminal-side QoS performance exceeds or is below a corresponding threshold. The method of claim 68, wherein, The method further includes: The terminal device receives seventh indication information sent by the network device, and the seventh indication information is used to indicate the sixth condition. The method of claim 67 or 68, wherein, The method further includes: The terminal device reports model replacement related information of the third AI model to the network device. The method of claim 69, wherein, The model replacement related information includes a sixth condition triggering replacement of the third AI model with the fourth AI model. The method of any one of claims 34-70, wherein The feedback information includes MAC HARQ-ACK information and / or RLC ARQ SR; and / or, the first indication information includes MAC HARQ retransmission scheduling and / or PDCP duplication retransmission configuration. The method as claimed in claims 34 to 71, wherein, The reception situation of the downlink data includes one or more of the following reception situations: MAC layer HARQ reception situation; RLC layer ARQ reception situation; PDCP layer PDU reception situation. A wireless communication device, the device comprising: A first determining unit configured to determine a retransmission decision corresponding to first information according to a first AI model; A first communication unit configured to send retransmission data to a second device in the case of a retransmission decision. A wireless communication device, the device comprising: A second determining unit configured to determine feedback information and / or first indication information corresponding to fourth information according to a third AI model; wherein, The feedback information is feedback information for transmitted data of a first device; and the first indication information is used to indicate whether to retransmit the transmitted data; A second communication unit configured to send the feedback information and / or the first indication information to the first device. A communication device, the communication device comprising: A memory for storing a computer program; A processor connected with the memory, for calling and running the computer program from the memory, to implement the method of any one of claims 1 to 33, or the method of any one of claims 34 to 72; A transceiver for receiving and sending information in the process of transceiving information with other devices. A chip, the chip comprising: A processor for calling and running a computer program from a memory, so that a device installed with the chip executes the method of any one of claims 1 to 33, or the method of any one of claims 34 to 72; A transceiver for receiving and sending information in the process of transceiving information with devices or chips. A computer readable storage medium for storing a computer program, the computer program causing a computer to execute the method of any one of claims 1 to 33, or the method of any one of claims 34 to 72. A computer program product comprising computer program instructions causing a computer to perform the method of any one of claims 1 to 33, or the method of any one of claims 34 to 72. A computer program causing a computer to perform the method of any one of claims 1 to 33, or the method of any one of claims 34 to 72.

Images