SDT determination method, model training method, communication device, and medium

Abstract

The present application relates to the technical field of communications, and discloses an SDT determination method, a model training method, a communication device, and a medium. The SDT determination method in embodiments of the present application comprises: a first device acquires second information on the basis of first information by using a first AI model, wherein the first information is input information required by the first AI model for model inference, and the second information is used for determining at least one of the following: whether to trigger SDT, and an SDT resource.

Description

SDT determination method, model training method, communication equipment and media

[0001] Cross-references to related applications

[0002] This application claims priority to Chinese Patent Application No. 202411893197.6, filed on December 20, 2024, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application belongs to the field of communication technology, specifically relating to an SDT determination method, a model training method, a communication device, and a medium. Background Technology

[0004] Small Data Transmission (SDT) is a process that allows terminals in a disconnected state to transmit data or signaling. SDT transmission can include terminal-triggered (Mobile-Originated Small, MO)-SDT and network-triggered (Mobile-Terminate, MT)-SDT.

[0005] In related technologies, in MO-SDT, the terminal needs to configure or reserve relevant SDT resources before entering the disconnected state. Then, after the terminal is in the disconnected state, if the uplink data to be transmitted meets the transmission conditions, such as the amount of data to be transmitted not exceeding the first threshold, the terminal can use the configured SDT resources for SDT transmission. This results in a situation where even if SDT resources are configured, they are not used, thus wasting uplink resources.

[0006] In MT-SDT, when a terminal receives a paging message instructing it to receive subsequent MT-SDT messages, if the downlink (DL) reference signal receiving power (RSRP) is greater than or equal to a second threshold, the terminal can initiate MT-SDT transmission and provide feedback on the recovery reason related to MT-SDT. If the DLRSRP is less than the second threshold, the terminal can ignore the paging message, resulting in a waste of downlink transmission resources.

[0007] Therefore, the SDT transmission method in related technologies may lead to a waste of resources. Summary of the Invention

[0008] This application provides an SDT determination method, a model training method, a communication device, and a medium, which can solve the problem of resource waste that may occur in the SDT transmission method in related technologies.

[0009] In a first aspect, an SDT (Simplified Derivative Transmission) method is provided, executed by a first device. The method includes: using a first Artificial Intelligence (AI) model, based on first information, acquiring second information, wherein the first information is input information required for model inference by the first AI model; and wherein the second information is used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources.

[0010] Secondly, a model training method is provided, executed by a second device. The method includes: training an AI model based on fourth information to obtain a first AI model, the first AI model being used for uplink transmission parameter inference; wherein the fourth information includes at least one of the following: a training dataset related to SDT transmission, a tag set related to SDT transmission, an optimization algorithm or index for AI model training, a loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0011] Thirdly, an SDT determination device is provided, the SDT determination device comprising: a processing module; the processing module being configured to use a first AI model and, based on first information, obtain second information, wherein the first information is input information required for the first AI model to perform model inference; wherein the second information is used to determine at least one of the following: whether to trigger SDT transmission, and SDT transmission resources.

[0012] Fourthly, a model training apparatus is provided, which may include a processing module. The processing module is used to train an AI model based on fourth information to obtain a first AI model, which is used for uplink transmission parameter inference; wherein the fourth information includes at least one of the following: a training dataset related to SDT transmission, a label set related to SDT transmission, an optimization algorithm or index for AI model training, a loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0013] Fifthly, a model training apparatus is provided, the apparatus being configured to perform the steps of the method described in the first aspect, or to implement the steps of the method described in the second aspect.

[0014] In a sixth aspect, a terminal is provided, the terminal including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the first or second aspect.

[0015] In a seventh aspect, a terminal is provided, including a processor and a communication interface, wherein the processor is used to use a first AI model and, based on first information, obtain second information, the first information being input information required for the first AI model to perform model inference; wherein the second information is used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources; or, the processor is used to train an AI model based on fourth information to obtain a first AI model, the first AI model being used for uplink transmission parameter inference; wherein the fourth information includes at least one of the following: a training dataset related to SDT transmission, a label set related to SDT transmission, an optimization algorithm or index for AI model training, a loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0016] Eighthly, a network-side device is provided, the network-side device including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the first aspect.

[0017] In a ninth aspect, a network-side device is provided, including a processor and a communication interface, wherein the processor is used to use a first AI model and, based on first information, obtain second information, the first information being input information required for the first AI model to perform model inference; wherein the second information is used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources; or, the processor is used to train an AI model based on fourth information to obtain a first AI model, the first AI model being used for uplink transmission parameter inference; wherein the fourth information includes at least one of the following: a training dataset related to SDT transmission, a label set related to SDT transmission, an optimization algorithm or index for AI model training, a loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0018] In a tenth aspect, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect, or implement the steps of the method described in the second aspect.

[0019] Eleventhly, a wireless communication system is provided, comprising: a terminal and a network-side device, wherein the terminal can be used to perform the steps of the method as described in the first aspect or the second aspect, and the network-side device can be used to perform the steps of the method as described in the second aspect or the second aspect.

[0020] In a twelfth aspect, a chip is provided, the chip including a processor and a communication interface coupled to the processor, the processor being configured to run programs or instructions to implement the method as described in the first aspect, or to implement the method as described in the second aspect.

[0021] In a thirteenth aspect, a computer program / program product is provided, which is stored in a storage medium and is executed by at least one processor to implement the method as described in the first aspect, or to implement the steps of the method as described in the second aspect.

[0022] In this embodiment, a first AI model can be used to obtain second information based on first information. The first information is the input information required for the first AI model to perform model inference. The second information can be used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources. Through this scheme, since the second information can be obtained through the first AI model inference, and the second information is used to determine whether SDT transmission will be triggered or whether SDT transmission resources need to be configured, SDT resources can be avoided when SDT transmission is not needed, thus avoiding waste of resources and configuration overhead. Attached Figure Description

[0023] Figure 1 is a schematic diagram of a possible architecture of a communication system provided in an embodiment of this application;

[0024] Figure 2A is a flowchart illustrating the small data transmission process based on random access;

[0025] Figure 2B is a schematic diagram of the data transmission process based on CG resources;

[0026] Figure 2C is a schematic diagram of the data transmission process based on RACH resources;

[0027] Figure 3A is a schematic diagram of a neural network structure;

[0028] Figure 3B is a schematic diagram of a possible structure of a neuron in a neural network;

[0029] Figure 3C is a schematic diagram of the specific operational framework for model lifecycle management;

[0030] Figure 4 is a flowchart illustrating the SDT determination method provided in an embodiment of this application;

[0031] Figure 5 is a flowchart illustrating the model training method provided in the embodiments of this application;

[0032] Figure 6 is a schematic diagram of the SDT determination device provided in an embodiment of this application;

[0033] Figure 7 is a schematic diagram of the model training device provided in an embodiment of this application;

[0034] Figure 8 is a schematic diagram of the structure of the communication device provided in an embodiment of this application;

[0035] Figure 9 is a schematic diagram of the structure of a terminal provided in an embodiment of this application;

[0036] Figure 10 is a schematic diagram of the structure of a network-side device provided in an embodiment of this application. Detailed Implementation

[0037] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0038] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, the first object can be one or more. Furthermore, "or" in this application indicates at least one of the connected objects. For example, the scope of protection for "A or B" covers at least three scenarios: Scenario 1: including A but not B; Scenario 2: including B but not A; Scenario 3: including both A and B. In addition, the terms "A and / or B," "at least one of A and B," and "at least one of A or B" also cover at least the above three scenarios. The character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0039] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as one in which the sender explicitly informs the receiver of specific information, the operation to be performed, or the requested result, etc., in the instruction sent. An indirect instruction can be understood as one in which the receiver determines the corresponding information based on the instruction sent by the sender, or makes a judgment and determines the operation to be performed or the requested result, etc., based on the judgment result.

[0040] It is worth noting that the technologies described in this application are not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, but can also be used in other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency-Division Multiple Access (SC-FDMA), or other systems. The terms "system" and "network" in this application are often used interchangeably, and the described technologies can be used with the systems and radio technologies mentioned above, as well as with other systems and radio technologies. The following description describes a New Radio (NR) system for illustrative purposes, and the term NR is used in most of the following description; however, these technologies can also be applied to systems other than NR systems, such as 6th generation (6G) radio systems. th Generation 6G communication system.

[0041] Figure 1 shows a block diagram of a wireless communication system applicable to an embodiment of this application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 can be a mobile phone, tablet computer, laptop computer, notebook computer, personal digital assistant (PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile internet device (MID), augmented reality (AR), virtual reality (VR) device, robot, wearable device, flight vehicle, vehicle user equipment (VUE), shipboard equipment, pedestrian user equipment (PUE), smart home (home devices with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), game console, personal computer (PC), ATM, or self-service machine, etc. Wearable devices include: smartwatches, smart bracelets, smart earphones, smart glasses, smart jewelry (smart bracelets, smart chains, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among these, in-vehicle devices can also be referred to as in-vehicle terminals, in-vehicle controllers, in-vehicle modules, in-vehicle components, in-vehicle chips, or in-vehicle units, etc. It should be noted that the specific type of terminal 11 is not limited in this application embodiment. Network-side equipment 12 may include access network equipment or core network equipment, wherein access network equipment may also be referred to as Radio Access Network (RAN) equipment, radio access network function, or radio access network unit. Access network equipment may include base stations, Wireless Local Area Network (WLAN) access points (APs), or Wireless Fidelity (WiFi) nodes, etc.In this context, a base station may be referred to as a NodeB (NB), Evolved NodeB (eNB), Next Generation NodeB (gNB), New Radio NodeB (NRNodeB), Access Point, Relay Base Station (RBS), Serving Base Station (SBS), Base Transceiver Station (BTS), Radio Base Station, Radio Transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home NodeB (HNB), Home Evolved NodeB, Transmit / Receive Point (TRP), or any other suitable term in the relevant field, as long as the same technical effect is achieved. The base station is not limited to any specific technical terminology. It should be noted that in this application embodiment, only a base station in an NR system is used as an example for introduction, and the specific type of base station is not limited.

[0042] Core network equipment, also known as core network nodes, core network functions, or core network elements, includes, but is not limited to, at least one of the following: Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), and Unified Data Warehouse (UDM). The core network equipment includes ataRepository (UDR), Home Subscriber Server (HSS), Centralized Network Configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (Local NEF, or L-NEF), Binding Support Function (BSF), Application Function (AF), Location Management Function (LMF), Gateway Mobile Location Centre (GMLC), and Network Data Analytics Function (NWDAF). It should be noted that this application only uses core network equipment in the NR system as an example and does not limit the specific type of core network equipment. If the name of the core network equipment mentioned in this application changes in subsequent protocol versions (e.g., 6G), it will still be within the scope of protection of this application.

[0043] Optionally, the core network equipment can be implemented by one or more functional modules in a single device, or by multiple devices working together; this application does not specifically limit this. It is understood that the aforementioned functional modules can be network elements in hardware devices, software functional modules running on dedicated hardware, or virtualized functional modules instantiated on a platform (e.g., a cloud platform).

[0044] The following is a description of the nouns or terms used in this application.

[0045] 1. Random Access Channel (RA) technology: There are many purposes for Random Access Channel (RACH).

[0046] For example, random access triggered by Physical Downlink Control Channel (PDCCH) orders is mainly for user equipment (terminal) terminals to obtain uplink time synchronization.

[0047] For example, when a terminal establishes an initial wireless link, it can obtain a user identifier, such as a Cell-Radio Network Temporary Identifier (C-RNTI), through a random access procedure.

[0048] 2. Random access procedure: In related technologies, the random access procedure may include: a contention-based random access (CBRA) procedure, also called a 4-step RACH; or a contention-free random access (CFRA) procedure.

[0049] The CBRA process is as follows: The terminal randomly selects a preamble from a contention-based preamble pool shared with other terminals in the cell and transmits it on the Physical Random Access Channel (PRACH); after detecting the preamble, the network-side device sends message 2 (Msg2), i.e., the Random Access Response (RAR). The response (RAR) message contains the number of the preamble detected by the network and the uplink radio resources allocated to the terminal to send message 3 (Msg3). After receiving Msg2, if the terminal confirms that at least one of the preamble numbers carried in Msg2 is consistent with the number of the preamble it sent, it sends Msg3 containing contention resolution information according to the uplink radio resources indicated by Msg2. After receiving Msg3, the network-side device can send a message (Msg4) containing contention resolution information. When the terminal receives Msg4, it confirms that the contention resolution information in Msg4 is consistent with the contention resolution information sent by the terminal in Msg3, thus completing the 4-step random access.

[0050] It is understandable that network-side devices can include the following information in the RAR message: Uplink (UL) grant information, which can be used to indicate Msg3 Physical Uplink Shared Channel (PUSCH) scheduling information, Random Access Channel preamble identity document (RACH preamble identity document, RAPID), Temporary Cell (TC) - RNTI, Timing Advance (TA), and other information.

[0051] If the network-side device does not receive Msg3 PUSCH, it can schedule the retransmission of Msg3PUSCH in the PDCCH scrambled by TC-RNTI.

[0052] During the CBRA process, different terminals randomly select preambles for transmission. Thus, different terminals may select the same preamble for transmission on the same time-frequency radio resources (e.g., random access timing (RACH Occasion, RO) resources). This situation can be understood as a preamble conflict between terminals.

[0053] During CFRA (Content-Based Random Access) processing, the preamble can be allocated by network-side equipment, such as the base station. This preamble is called a Dedicated Random Access preamble. During CFRA, the preamble is provided to the terminal via Radio Resource Control (RRC) signaling or PDCCH order signaling, thus eliminating preamble conflicts. When dedicated random access preamble resources are insufficient, the base station can notify the terminal to initiate CBRA-based access.

[0054] In NR, 2-step RACH is introduced. The first step is for the terminal to send MsgA to the network-side device. After receiving MsgA, the network-side device can send MsgB to the terminal; if the terminal does not receive MsgB within a certain time, it will increment a counter counting the number of MsgA transmissions and retransmit MsgA. If the counter counting the number of MsgA transmissions reaches a certain threshold, the terminal can switch from the 2-step random access procedure to the 4-step random access procedure. MsgA includes a MsgA preamble and a MsgA PUSCH. The MsgA preamble is transmitted on the RO resources used for 2-step RACH, and the PUSCH can be transmitted on the MsgA PUSCH resources associated with transmitting the MsgA preamble and RO resources. The MsgA PUSCH resources are a set of PUSCH resources configured relative to each PRACH slot, including time-frequency resources and demodulation reference signal (DMRS) resources.

[0055] 3. SDT Transmission: Small data transmission is a process that allows a non-connected terminal (e.g., idle or inactive) to transmit at least one of the data and signaling simultaneously without transitioning to the RRC connected state (RRC_CONNECTED). SDT transmission avoids excessive signaling overhead caused by RRC state transitions and RRC connection establishment processes due to data transmission.

[0056] The key feature of the small data transmission scheme is that the terminal's current Data Radio Bearer (DRB) is in a suspended state, not a released state. Therefore, before sending a Status (ST) ResumeReq message, the terminal can restore the DRB and then use RRC signaling to piggyback the small data. At this time, a terminal in a disconnected state can transmit data on the DRB just like a terminal in a connected state. This avoids state transitions and achieves efficient small data transmission with minimal signaling overhead.

[0057] Data to be transmitted in small data transfers can be carried on a dedicated traffic channel (DTCH). This data can be multiplexed with an uplink recovery terminal's RRC connection status request message (RRCConnectionResumeReq terminal st message) before transmission. Similarly, if there is a reply downlink message, it can also be carried on the DTCH and multiplexed with a downlink RRC connection release message (RRCConnectionRelease message) before transmission. Because small data transfers use DRB transmission, necessary security protections can be implemented for the data, such as data encryption and integrity protection.

[0058] Figure 2A is a flowchart illustrating the small data transmission process based on random access. Specifically, Figure 2A shows a flowchart illustrating the transmission of small data on the Msg3 PUSCH based on the 4-step RACH process. In actual implementation, small data can also be transmitted on the MsgA PUSCH based on the 2-step RACH process, or on PUSCH resources scheduled by a Configured Grant (CG) configured in the RRC inactive state. Small data transmission in the 2-step RACH and 4-step RACH processes is referred to as RACH-based small data transmission, and small data transmission on PUSCHs scheduled by CG is referred to as CG-based small data transmission.

[0059] Small data transfer based on RACH and small data transfer based on CG can be configured on the initial bandwidth part (BWP).

[0060] The RACH and CG resources of the SDT can be configured on at least one of the Supplementary Uplink (SUL) and Non-Supplementary Uplink (NUL) carriers. When the terminal receives an RRC release message with a pause indication, the CG resources of the SDT are only valid within the terminal's Primary Cell (PCell). CG resources are associated with one or more Synchronization Signal Blocks (SSBs).

[0061] For RACH, network-side devices can configure 2-step and / or 4-step RA resources for MO-SDT. If the MT-SDT procedure is initiated via RACH, the terminal can use RACH resources not configured for SDT. CFRA does not support SDT transmission over RACH.

[0062] 3.1. Subsequent transmissions of SDT

[0063] When using CG resources for the initial SDT transmission, if the terminal does not receive an acknowledgment message (dynamic UL authorization or DL ​​allocation) from the network-side device before the configured timer expires, the terminal can perform an autonomous retransmission of the initial transmission. After the initial PUSCH transmission, the handling of subsequent transmissions depends on the resource type used to initiate the SDT process:

[0064] 1) When using CG resources, network-side devices can use dynamic authorization to schedule subsequent UL transmissions, or perform transmissions in the CG resource scenario shown in Figure 2B. DL transmissions are scheduled using dynamic allocation. The terminal can only initiate subsequent UL transmissions after receiving confirmation (dynamic UL authorization or DL ​​allocation) from the network-side device for the initial PUSCH transmission. For subsequent UL transmissions, the terminal cannot initiate retransmissions using CG resources.

[0065] Specifically, the transmission scheme for CG resources shown in Figure 2B includes:

[0066] T1: If the signal strength of at least one SSB exceeds the RSRP threshold (cg-SDT-RSRP-ThresholdSSB) used for SSB selection, and neither the CG-based SDT (CGT) nor the Real-Time (RT) pre-configured resource-based SDT (cg-SDT-RT) is running, then it is assumed that the Network Device Interface (NDI) protocol has been switched, and the Hybrid Automatic Repeat reQuest (HARQ) information related to CG is passed to the HARQ entity for subsequent operations. Otherwise, it is determined that the CG resource is unavailable.

[0067] Conversely, if the signal strength of at least one SSB exceeds the RSRP threshold, a scheduling request (SR) and / or random access can be triggered.

[0068] T2: If cg-SDT-RT is not running, and the HARQ process is the same as the initial transmission, and no C-RNTIPDCCH is received, then CG and the associated HARQ information are passed to the HARQ entity (even if there is no SSB with SS-RSRP higher than the reference signal receive power threshold used to select the SSB).

[0069] T3: Stop CGT; only if C-RNTIPDCCH is received can subsequent transmissions be initiated via CG or Dynamic Grant (DG). Otherwise, the terminal will return to the idle state after CGT timeout.

[0070] T4: A new transfer will only occur in subsequent transfers if at least one SSB's SS-RSRP is higher than cg-SDT-RSRP-ThresholdSSB. Otherwise, the CG resource is unusable. Conversely, SR and / or RA can be triggered.

[0071] 2) When using RACH resources, network-side devices can use dynamic UL authorization and DL allocation to schedule subsequent UL and DL transmissions after the RA process is completed. See Figure 2C for the configuration method. Specifically, as shown in Figure 2C:

[0072] i.SDT data 1 can trigger RA-SDT transmission. SDT trigger condition evaluation can include: RA-SDT and CGSDT.

[0073] Choose between SDT-DRB and SRB1 recovery.

[0074] ii. Start the T319a timer after transmitting the Common Control Channel (CCCH) message;

[0075] iii. Receiving the RRC release message indicates that the SDT process has been successfully completed; that is, the SDT program may not be able to complete successfully when the T319a timer expires.

[0076] 3.2. Triggering conditions and process of SDT transmission:

[0077] Small data transmissions can be initiated by the terminal (i.e., MO-SDT, Mobile Originated SDT) or by network-side equipment (i.e., MT-SDT, Mobile Terminated SDT).

[0078] A. The conditions for starting SDT include at least one of the following:

[0079] Scenario 1: Recovery procedure initiated by the upper layer (i.e., MO-SDT):

[0080] i. SDT configuration (sdt-config) is configured, such as SIB1 containing SDT configuration commands (sdt-ConfigCommon);

[0081] ii. All pending data in the UL is mapped to radio bearers (RBs) configured for the SDT;

[0082] iii. When a Reduced Capability (RedCap) terminal does not have an SSB configured with Remaining Minimum System Information (RMSI) in its initial downlink BWP (referred to as CD-SSB), and an SSB without RMSI (referred to as NCD-SSB), i.e. (ncd-SSB-RedCapInitialBWP-SDT).

[0083] iiii, if the amount of UL data to be processed for all RBs configured for SDT is less than or equal to the threshold;

[0084] iiiii, the RSRP reference for downlink path loss is higher than the threshold.

[0085] Scenario 2: For recovery procedures initiated in response to RAN paging (i.e., MT-SDT):

[0086] The paging message stored by the UE contained an MT-SDT indication.

[0087] The RSRP reference for downlink path loss is higher than the threshold.

[0088] B. The choice between CG-SDT and RA-SDT.

[0089] The terminal selects between CG-SDT and RA-SDT according to the following rules.

[0090] If CG-SDT resources are configured, TA is valid, and at least one CG-SDT-configured SSB with SS-RSRP higher than the RSRP threshold is available, then CG-SDT is executed.

[0091] Otherwise, if RA-SDT resources are available, then RA-SDT is executed;

[0092] If no RA-SDT resource is available, then perform a traditional RA to restore the RRC connection.

[0093] 4. Artificial Intelligence or Machine Learning (ML)

[0094] AI has been widely applied in various fields. Integrating artificial intelligence into wireless communication networks to significantly improve technical indicators such as throughput, latency, and user capacity is an important task for future wireless communication networks. AI modules can be implemented in various ways, such as neural networks, decision trees, support vector machines, and Bayesian classifiers.

[0095] 4.1. Neural Networks

[0096] Figure 3A is a schematic diagram of a neural network structure. As can be seen, a neural network can include an input layer, a hidden layer, and an output layer.

[0097] The neural network is composed of neurons, as shown in Figure 3B. Here, z represents a neuron, a1, a2, ..., aK are the inputs, w is the weight (multiplicative coefficient), b is the bias (additive coefficient), and σ(.) is the activation function. Common activation functions include Sigmoid, tanh, ReLU (Rectified Linear Unit), etc.

[0098] The parameters of a neural network are optimized using gradient optimization algorithms. Gradient optimization algorithms are a class of algorithms that minimize or maximize an objective function (sometimes called a loss function), which is often a mathematical combination of model parameters and data.

[0099] For example, given data X and its corresponding label Y, a neural network model f(.) can be constructed. With the model, the predicted output f(X) can be obtained from the input X, and the difference between the predicted value and the true value (f(X)-Y) can be calculated; this is the loss function. Our goal is to find suitable weights W and biases b that minimize the value of the aforementioned loss function. The smaller the loss value, the closer the model's prediction is to the true situation, such as the ground truth.

[0100] Optimization algorithms can be based on the error back propagation (BP) algorithm. The basic idea of ​​the BP algorithm is that the learning process consists of two processes: forward propagation of the signal and backward propagation of the error. During forward propagation, the input sample is introduced from the input layer, processed layer by layer by the hidden layers, and then propagated to the output layer. If the actual output of the output layer does not match the expected output, the process transitions to the error back propagation stage. Error back propagation involves propagating the output error back to the input layer layer by layer through the hidden layers in a certain form, distributing the error to all units in each layer, thereby obtaining the error signal of each unit. This error signal serves as the basis for adjusting the weights of each unit. This process of adjusting the weights of each layer through forward and backward propagation is repeated continuously. This continuous adjustment of weights is the learning and training process of the network. This process continues until the error of the network output is reduced to an acceptable level, or until the predetermined number of learning iterations is reached.

[0101] Common optimization algorithms include: gradient descent subtraction, stochastic gradient descent (SGD), mini-batch gradient descent, momentum method, Newton's method (specifically, stochastic gradient descent with momentum), adaptive gradient descent (Adagrad), adaptive learning rate adjustment (Adadelta), root mean square prop (RMSprop), and adaptive momentum estimation (Adam).

[0102] During error backpropagation, the optimization algorithms described above calculate the gradient based on the error / loss obtained from the loss function with respect to the current neuron, add the learning rate, previous gradients / derivatives / partial derivatives, etc., and then pass the gradient to the previous layer.

[0103] Generally speaking, the AI ​​algorithms and AI models selected will vary depending on the type of problem being solved.

[0104] The main method for improving the performance of 5G networks using AI is to enhance or replace existing algorithms or processing modules through neural network-based algorithms and AI models. In specific scenarios, neural network-based algorithms and AI models can achieve better performance than deterministic algorithms.

[0105] Commonly used neural networks include deep neural networks, convolutional neural networks, and recurrent neural networks. AI tools can be used to build, train, and validate neural networks.

[0106] 4.2. Background of Fine-tuning

[0107] In practice, due to the insufficient size of real-time acquired datasets, directly training neural networks often fails to achieve convergence. A common approach is to pre-train the network using a large amount of offline collected data until it converges. Then, fine-tuning is performed on the pre-trained neural network parameters using real-time acquired data to adapt the network to the real-world environment. Fine-tuning can be considered a training process that uses the parameters of the pre-trained neural network as initialization. During the fine-tuning stage, parameters of some layers can be frozen; generally, layers closer to the input are frozen, while layers closer to the output are activated. This ensures that the network can still converge. The smaller the amount of data during the fine-tuning stage, the more layers should be frozen, with only a small number of layers near the output being fine-tuned.

[0108] 4.3. Generalization Problem of Neural Networks

[0109] Generalization refers to the ability of a neural network to produce reasonable outputs on data not encountered during its training (learning) process. To address the generalization problem caused by the variable wireless transmission environment, neural network-based wireless communication systems offer two solutions. The first is to train different neural networks under different transmission conditions, obtaining multiple sets of network parameters, and then switching these parameters as the actual environment changes. The second is to train a single, common neural network based on mixed data, where the network parameters do not change with the environment. Each approach has its advantages and disadvantages: the first approach performs excellently under different transmission conditions, but requires storing multiple network parameters and switching them as needed (incurring signaling overhead and frequent switching issues); the second approach only requires storing one set of neural network parameters without switching, but cannot achieve optimal performance under every transmission condition. The construction method of the mixed dataset affects the performance of the second approach.

[0110] 4.4. Labels

[0111] In machine learning and deep learning, a label typically refers to an identifier or annotation of the true category or target value of a data sample. Labels are used to represent the information that the model should learn and predict, for example:

[0112] (1) Labels in classification tasks: In classification tasks, labels indicate which category a data sample belongs to. For example, for image classification, each image sample has a label that indicates the category of the object or scene contained in the image, such as "dog" or "cat".

[0113] (2) Labels in object detection: In object detection tasks, labels typically include the object's location information (bounding box) and category information. Each label identifies a target object in the image, including its location and category.

[0114] (3) Labels in regression tasks: In regression tasks, labels typically represent the continuous or real-valued target to be predicted. For example, the label in a house price prediction task could be the actual selling price of a house.

[0115] (4) Labels in sequence labeling: In natural language processing, labels in sequence labeling tasks are usually used for tasks such as part-of-speech tagging and named entity recognition. The labels are used to represent the attributes or categories of each word or character in the text sequence.

[0116] As is understandable, labels are a crucial component in supervised learning tasks, used to train machine learning models. Models learn patterns and regularities by comparing their labels to true labels in order to make predictions or classifications on unseen data. The quality and accuracy of the labels are critical to the model's performance.

[0117] 4.5. Life Cycle Management (AI LCM)

[0118] The lifecycle management of AI models / ML models can include multiple AI functional modules, which may include: model training, model management, model inference, model monitoring, and model updates.

[0119] Figure 3C is a schematic diagram of the specific operational framework for model lifecycle management. Specifically, as shown in Figure 3C:

[0120] Model training

[0121] This function performs AI model training, validation, and testing, generating model performance metrics that can be used as part of the model testing process. If needed, it also handles data preparation based on the training data provided by the data collection function, such as data preprocessing and cleaning, formatting, and transformation.

[0122] Training / Update Model: If a model storage function is available, it is used to transfer trained, validated, and tested AI models to the model storage function, or to transfer updated versions of the model to the model storage function.

[0123] Model Management

[0124] This module monitors the operation of AI models or the deployment of AI functions (e.g., model selection / deactivation / switching / rollback) and provides feedback on model monitoring performance. It is also responsible for making decisions based on data received from the data collection and inference modules to ensure correct inference operations.

[0125] Management instructions: Information input from the model management function to the model inference function. This information may include selecting / (deactivating) the AI ​​model or AI / ML-based functions to activate / switch the model, reverting to non-AI / ML operations (i.e., operations independent of the inference process), etc.

[0126] Model transfer request: Used to request a model from the model storage function.

[0127] Performance Feedback / Retraining Request: Information required for model training function input, such as for model (re)training or updating purposes.

[0128] Model Inference

[0129] The data (i.e., inference data) provided by the data collection function is used as input to provide the output of the applied AI model. If necessary, the inference function is also responsible for data preparation based on the inference data provided by the data collection function (e.g., data preprocessing and cleaning, formatting and transformation).

[0130] Inference output: Data used by management functions to monitor the performance of AI models or AI / ML functions.

[0131] It is understood that in the embodiments of this application, the AI ​​model may also be referred to as an AI unit, AI structure, etc.; or the AI ​​model may also refer to a processing unit that can implement specific algorithms, formulas, processing flows, capabilities, etc. related to AI; or the AI ​​model may be a processing method, algorithm, function, module or unit for a specific dataset; or the AI ​​model may be a processing method, algorithm, function, module or unit running on AI-related hardware such as a graphics processing unit (GPU), a neural processing unit (NPU), a tensor processing unit (TPU), or an application-specific integrated circuit (ASIC). This invention does not specifically limit this.

[0132] The SDT determination method, model training method, communication equipment, and medium provided in this application will be described in detail below with reference to the accompanying drawings and through some embodiments and application scenarios.

[0133] This application provides an SDT determination method. Figure 4 shows a flowchart of the SDT determination method provided in this application. As shown in Figure 4, the SDT determination method provided in this application may include step 401.

[0134] Step 401: The first device uses the first AI model to obtain the second information based on the first information.

[0135] The first information is the input information required for the first AI model to perform model inference, such as measurement information related to channel quality. The second information can be used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources.

[0136] In some embodiments of this application, the first device may be a terminal, a network-side device, or an AI-related server. The network-side device may include, but is not limited to, any of the following: access network device or core network device.

[0137] For example, access network equipment can be a base station, TRP, or cell.

[0138] It is understood that AI in this application can also be referred to as machine learning. The first AI model may include a variety of possible implementations, such as those based on neural networks, decision trees, support vector machines, Bayesian classifiers, etc., and this application does not make any specific limitations on them.

[0139] It is understood that the small data transmission in this application may include the transmission of at least one of the following information for a terminal in a non-connected state: feedback information, response information, control information, or small data, normal data.

[0140] In some embodiments of this application, SDT transmission resources may include at least one of the following resources for SDT transmission: time-domain resources, time-frequency resources, frequency, sequence, port, reference signal, signal or channel-related resources associated with SDT transmission.

[0141] For example, the aforementioned time-frequency resources may include at least one of time-domain resources and frequency-domain resources.

[0142] For example, the signals associated with SDT transmission may include: SSB and Channel State Information-Reference Signal (CSI-RS).

[0143] In some embodiments of this application, the reference signal used for SDT transmission may include at least one of the following: DMRS signal, Tracking Reference Signal (TRS), Phase-tracking Reference Signal (PTRS), SSB, CSI-RS, TRS, downlink broadcast signal, synchronization signal, downlink reference signal, and uplink reference signal.

[0144] The synchronization signal may include at least one of the primary synchronization signal (PSS) and the secondary synchronization signal (SSS).

[0145] For example, the uplink reference signal can be a sounding reference signal (SRS).

[0146] In some embodiments of this application, the second information may include at least one of the following: SDT transmission prediction result, connected-state transmission indication, resource configuration of triggered SDT transmission, type of triggered SDT transmission, terminal state prediction result, uplink UL data volume prediction result, downlink measurement volume prediction result, candidate beams for SDT transmission, and reference signals corresponding to the candidate beams. The terminal state prediction result can be used to indicate whether the terminal enters a first state or the probability of entering the first state within at least one time period.

[0147] In some embodiments of this application, the first state may include an idle state, an inactive state, a disconnected state, a connected state, or a power-saving state.

[0148] In some embodiments of this application, the disconnected state can be an RRC-idle state, an RRC-inactive state, or a standby state introduced in 6G or future mobile communication systems. Specifically, the disconnected state can be the state after RRC release and before random access, or other disconnected states before random access.

[0149] In some embodiments of this application, the type of SDT transmission triggered above may include at least one of the following: RA-SDT, Configured Grant (CG)-SDT, paging SDT, RACH-less Dynamic Grant (DG)-SDT, paging-based downlink SDT, and paging-directly scheduled downlink SDT.

[0150] In some embodiments of this application, the paging-based downlink SDT can also be referred to as MDT.

[0151] In some embodiments of this application, paging SDT, also known as paging-based SDT, can be a downlink small data transmission directly performed via paging the Physical Downlink Control Channel (PDCCH) or paging the Physical Downlink Shared Channel (PDSCH); or paging SDT can be a downlink small data transmission scheduled via paging the PDCCH or paging the PDSCH.

[0152] In some embodiments of this application, RACH-less DG-SDT can be referred to as RACH-less SDT, or uplink SDT without Msg1 transmission. The PUSCH or associated UL grant used to transmit the SDT is obtained through the DG. For example, it can be allocated via RRC signaling or via the PDCCH channel.

[0153] The SDT or SDT type in the embodiments of this application may also represent at least one of the following: data transmission not exceeding a certain data size, data transmission under a specific terminal state (such as a specific RRC state, a specific terminal transmit power state, or a specific terminal transmit mode), data transmission based on RACH, data transmission based on resources configured with configuration authorization, and data transmission triggered or scheduled based on paging.

[0154] In some embodiments of this application, the resource configuration of the SDT transmission triggered above may include at least one of the following: the time-domain resource location of the SDT transmission, the frequency-domain resource location of the SDT transmission, the number of symbols occupied by the SDT transmission, and the number of resource blocks (RBs) occupied by the SDT transmission.

[0155] In some embodiments of this application, the above-described SDT transmission prediction results can be used to indicate at least one of the following 1) to 3):

[0156] 1) At least within a time period, whether SDT transmission is performed, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission.

[0157] 2) Whether SDT transmission is performed at least at one time point, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission.

[0158] 3) The time period or time point during which SDT transmission or the first type of SDT transmission is not performed, and the time period or time point during which SDT transmission or the first type of SDT transmission is performed, wherein the first type includes at least one of the following: Random Access RA-SDT, Configuration Authorization CG-SDT, Paging SDT, Dynamic Authorization DG-SDT without Random Access RACH-less, Paging-based Downlink SDT, and Paging Direct Scheduling Downlink SDT.

[0159] It can be understood that the time points in 1) to 3) above are time points after the current time point, that is, future time points; the time periods in 1) to 3) above are time periods after the current time point, that is, future time periods.

[0160] For example, in 1) above, the above SDT transmission prediction result can indicate: there is SDT transmission in at least one time period, or there is SDT transmission in at least one time period, or the probability of SDT transmission in at least one time period is 80%, or the probability of paging-based SDT transmission in at least one time period is 75%, or the probability of no SDT transmission in at least one time period is 50%, or the probability of no paging-based SDT transmission in at least one time period is 89%.

[0161] In some embodiments of this application, the time point can be at least one of the following: an absolute time point, or a time point M durations after a reference time point.

[0162] In some embodiments of this application, the reference time point may include at least one of the following: the time point when the second information was acquired, the time point when the terminal last entered the non-connected state, the time point when the terminal last received the RRC release message, the time point when the terminal last received the AI ​​configuration, and the time point when the terminal last experienced a system message update.

[0163] Among them, AI configuration can be used for the configuration of the first AI model inference.

[0164] In some embodiments of this application, the time period may include at least one of the following: N units of time starting from the reference time point, Q units of time between the time point of acquisition of the second information and the reference time point, and W units of time from the first time point to the second time point.

[0165] Among them, M, N, Q, and W are greater than or equal to 0, and the values ​​of M, N, and W are obtained by reasoning from the first AI model.

[0166] In some embodiments of this application, the first time point and the second time point are inferred through a first AI model. The definitions of the first time point and the second time point are the same as those for the relevant definitions of time points in the above embodiments.

[0167] It is understandable that the above-mentioned unit duration can also be referred to as unit time.

[0168] In some embodiments of this application, the unit of time may be any of the following: second (s), millisecond (ms), transmission time interval (TTI), slot, subframe, frame, or symbol.

[0169] In some embodiments of this application, the above-mentioned connection state transmission indication can be used to trigger the terminal to directly enter the connection state transmission.

[0170] In some embodiments of this application, the uplink transmission data volume prediction result may include at least one of the following:

[0171] i. At least one value of the amount of data to be transmitted uplink within at least one time period, or the probability associated with the at least one value;

[0172] iii. At least one range of values ​​for the amount of data to be transmitted uplink within at least one time period, or the probability associated with such at least one range of values;

[0173] iiii. The probability of having at least one time period with business demand or associated with at least one time period.

[0174] In some embodiments of this application, the downlink measurement prediction results may include at least one set of measurement prediction values ​​for a first signal; wherein the first signal includes a downlink broadcast signal, a synchronization signal, or a reference signal.

[0175] For example, downlink measurement prediction results may include measurements such as RSRP, RSRQ, or RSSI on a future SSB.

[0176] For example, downlink measurement prediction results may include measurements such as RSRP, RSRQ, or RSSI on a future CSI-RS.

[0177] For example, downlink measurement prediction results may include measurements such as RSRP, RSRQ, or RSSI on a future TRS.

[0178] For example, downlink measurement prediction results may include the Synchronization Signal-Reference Signal Received Power (SS-RSRP) of at least one or a set of SSBs associated with CG-SDT in the future.

[0179] It is understandable that, for the potential data transmission in the aforementioned disconnected state, if the first AI model can be used to predict the future potential SDT transmission mode, the terminal can avoid unnecessary measurements, avoid wasting uplink and downlink resources and configuration overhead, and further improve the efficiency of data transmission.

[0180] For example, if the first AI model can predict in advance that no SDT transmission will be triggered within a certain period before the RRC release, then SDT transmission resources do not need to be configured; or, if the first AI model can predict in advance the type of SDT transmission that will be triggered within a certain period before the RRC release, then only specific types of SDT resources can be configured, thereby reducing the waste of resources and configuration overhead.

[0181] For example, if the network-side device can use the first AI model to predict that the DL RSRP will be low in the future, such as below a certain threshold, the network can avoid triggering MT-SDT transmission, thereby avoiding the waste of downlink data transmission resources for paging and scheduling.

[0182] Thus, since the first AI model can be used to predict potential future SDT methods, whether SDT transmission should be performed, or the time period for SDT transmission, unnecessary measurements can be avoided, as well as the waste of uplink and downlink resources and configuration overhead, which is also beneficial to the efficiency of data transmission.

[0183] Thus, since the second information can include at least one of the following: SDT transmission prediction results, connected-state transmission indication, resource configuration for triggered SDT transmission, type of triggered SDT transmission, terminal state prediction results, uplink UL data volume prediction results, downlink measurement volume prediction results, candidate beams for SDT transmission, and reference signals corresponding to those candidate beams, it facilitates accurate configuration or reservation of SDT resources by network-side equipment and avoids unnecessary measurements by the terminal. This avoids resource waste and saves signaling overhead.

[0184] In some embodiments of this application, the first information mentioned above may include at least one of the following: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device or satellite device, panel orientation information of the terminal, network-side device or satellite device, device information of the terminal, location-related information of the terminal, network-side device or satellite device, motion-related information of the terminal, network-side device or satellite device, perception information of the terminal, network-side device or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information;

[0185] The first object includes any one of the following: beam, reference signal, transmit / receive point (TRP) currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and bandwidth portion (BWP) of the terminal.

[0186] In some embodiments of this application, the beam can be at least one of the following: the beam currently accessed by the terminal, two beams adjacent to the current beam of the terminal, or at least one beam whose included angle with the current beam of the terminal is less than or equal to a preset angle.

[0187] It should be noted that the above-mentioned cell can also be replaced with TRP, network node, carrier, bandwidth, subband, or BWP.

[0188] In some embodiments of this application, the reference signal may include at least one of the following: one or a set of downlink broadcast signals; one or a set of synchronization signals; one or a set of downlink reference signals; and one or a set of uplink reference signals.

[0189] For example, the first object can be a downlink broadcast signal, synchronization signal, SSB, CSI-RS, or TRS.

[0190] For example, the first object can be a set of uplink reference signals, such as SRS.

[0191] In some embodiments of this application, the above-mentioned measurement-related information may include at least one of the following: current measurement value, measurement value change information over a time period, and historical measurement value.

[0192] The measured value change information includes at least one of the following: an indication of an increase or decrease in the measured value, the amount of change in the measured value, the percentage change in the measured value, an indication of whether the amount of change in the measured value is greater than a first threshold, and the number of times or duration that the amount of change in the measured value is greater than the first threshold.

[0193] In some embodiments of this application, the above-mentioned measurement values ​​may include at least one of the following: signal strength information, signal quality information, interference signal strength information, TA information, and link quality information.

[0194] For example, signal strength information may include RSRP, RSSI, etc.

[0195] For example, signal quality information may include at least one of the following: RSRQ, signal-to-noise ratio (SNR), signal-to-interference-plus-noise ratio (SINR), latency, channel quality indicator (CQI), block error rate (BLER), and bit error rate (BER).

[0196] In some embodiments of this application, the TA information described above can be absolute TA or relative TA.

[0197] In some embodiments of this application, the link quality information may include at least one of the following: CQI, Rank Indicator (RI), and windband-SINR.

[0198] In some embodiments of this application, the historical measurement value of the first object can be the measurement value of the first object at a certain time point before the current time point or within a certain period of time. For a description of the measurement value, please refer to the relevant description in the above embodiments.

[0199] In some embodiments of this application, "distance or path loss information between the terminal and the network-side device" may include at least one of the following: the distance between the terminal and the network-side device to which the terminal is currently hosted or accessed, and path loss information between the terminal and the network-side device to which the terminal is currently hosted or accessed.

[0200] For example, the distance between the terminal and the TRP that the terminal is currently accessing or residing in.

[0201] For example, path loss information between the terminal and the TRP that the terminal is currently accessing or residing in.

[0202] It is understood that the solutions related to "network-side devices accessed by the terminal, such as TRP, cell, and base station" in this application are also applicable to the network-side devices where the terminal is currently camped.

[0203] In some embodiments of this application, the frequency domain resource information may include at least one of the following: frequency band information, frequency zone information, frequency point information, carrier frequency information, frequency layer information, and BWP information.

[0204] In some embodiments of this application, the frequency domain resource information mentioned above may include frequency domain resource information of a terminal, network-side device, or satellite device.

[0205] For example, the aforementioned frequency domain resource information may include at least one of the following: the frequency band, frequency point, carrier frequency, frequency layer, and BWP in which the cell actually operates.

[0206] In some embodiments of this application, the above-mentioned reference signal configuration may include at least one of the following: a reference signal pattern, a reference signal time-domain position or index, a reference signal frequency-domain position or index, and a reference signal transmission power.

[0207] For example, the reference signal configuration may include the frequency domain location of at least one of PSS, SSS, CSI-RS, DMRS, SSB, TRS, and SRS.

[0208] For example, the reference signal configuration may include the time-domain location of at least one of PSS, SSS, CSI-RS, DMRS, SSB, TRS, and SRS.

[0209] For example, the reference signal configuration may include the transmission power of at least one of PSS, SSS, CSI-RS, DMRS, SSB, TRS, and SRS.

[0210] In some embodiments of this application, the beam configuration described above may include at least one of the following: the number of beams, beam direction, beam, or the association between a beam and a reference signal corresponding to a beam transmitted or received by a terminal; the number of beams, beam direction, beam, or the association between a beam and a reference signal corresponding to a beam transmitted or received by a network-side device; and one or a set of reference signal indices.

[0211] One or a set of beam indices; one or a set of beam direction information; one or a set of transmission configuration indication (TCI) status information.

[0212] For example, the beam configuration described above can be a reference signal index, beam index, or beam direction actually transmitted between the terminal and the network nodes of the cell. The network nodes of the cell can be the access network equipment to which the cell belongs, such as a base station.

[0213] In some embodiments of this application, the network identifier may include at least one of the following: TRP identifier, TRP group identifier, cell identifier, cell group identifier, timing advance TAG group identifier, tracking area identifier, and radio access network notification area RNA identifier.

[0214] In some embodiments of this application, the transmission power information of the network-side device may include, but is not limited to, at least one of the following: transmission power information of the core network device, transmission power information of the base station, transmission power information of the TRP, transmission power information of the satellite, and transmission power information of the server.

[0215] In some embodiments of this application, the panel orientation information of the network-side device may include the panel orientation information of at least one of the base station, TRP, and satellite.

[0216] In some embodiments of this application, the device information of the terminal may include at least one of the following: power consumption, power consumption, supported operators, supported network types, terminal type, and power class.

[0217] For example, the type of terminal may include, but is not limited to, at least one of the following: Internet of Things (IoT) terminal, NR terminal, Non-Terrestrial Networks (NTN) terminal, smartphone, handheld radio.

[0218] In some embodiments of this application, the aforementioned location-related information may include at least one of the following:

[0219] Geographic location information of terminals, network-side devices or satellite devices; geographic location change information of terminals, network-side devices or satellite devices; cell reference location or movement trajectory of terminals in non-terrestrial network scenarios; ephemeris information of satellite devices; and distribution information of terminals.

[0220] In some embodiments of this application, the aforementioned geographic location change information may include at least one of the following:

[0221] An indication of whether a geographical location has changed within a certain period of time;

[0222] An indication of whether the change in geographical location exceeds a first threshold within a certain time period;

[0223] The change between the geographic location information at time T+h and the location information at time T, where T and h are greater than 0;

[0224] The number of times or the duration of change in geographic location information exceeding the first change threshold within a given time period;

[0225] The number of times or duration during which the change in geographic location information is less than or equal to the second change threshold within a given time period.

[0226] In some embodiments of this application, the geographic location information of the terminal, network-side device, or satellite device can be at least one of the following:

[0227] The geographical coordinates of the terminal, network-side equipment, or satellite equipment, such as GPS coordinates;

[0228] Approximate location range information of the terminal, network-side equipment, or satellite equipment, such as which street or square the terminal is located in;

[0229] The location information of the terminal relative to the currently camped cell, the currently accessed cell, a certain TRP or a group of TRPs, such as the terminal being due east of the camped cell;

[0230] Location information of network-side devices relative to reference devices.

[0231] In some embodiments of this application, non-terrestrial network scenarios may include geostationary orbit (GEO) and low-Earth orbit (LEO) scenarios.

[0232] In the LEO scenario, the cells on Earth move as the satellite moves.

[0233] In the GEO scenario, as the satellite moves, the cell on Earth is fixed, and a reference position can be considered to exist.

[0234] In some embodiments of this application, the distribution information of the terminals may include the number of terminals in different regions, that is, the distribution information may include the number of terminals in different regions.

[0235] For example, the number of cells the terminal is stationed in, the number of cells it accesses, the number of a certain TRP or a group of TRPs.

[0236] In some embodiments of this application, the above-mentioned motion-related information may include at least one of the following:

[0237] Information on the direction of movement or changes in the direction of movement of network-side devices;

[0238] Information on the terminal's direction of motion or changes in direction of motion;

[0239] Information on the direction of motion or changes in the direction of motion of satellite equipment;

[0240] Information on the movement speed or changes in movement speed of network-side devices;

[0241] Terminal movement speed or movement speed change information

[0242] Information on the speed of satellite equipment or changes in its speed.

[0243] In some embodiments of this application, the direction of movement can be an absolute direction, such as 40 degrees east of south; or the direction of movement can be a relative direction, such as the direction relative to a certain base station or reference node.

[0244] In some embodiments of this application, the speed of motion can be absolute speed or relative speed.

[0245] In some embodiments of this application, the aforementioned motion direction change information includes at least one of the following:

[0246] The change in direction of motion per unit time, such as a change of 5 degrees in direction of movement per unit time;

[0247] The difference between the initial and final values ​​of the direction of motion over a given time period;

[0248] The percentage change in the direction of motion over a given time period compared to the initial value;

[0249] The difference between the maximum and minimum values ​​in the direction of motion over a given time period;

[0250] An indication of whether the value of the direction of motion increases or decreases over a period of time;

[0251] Whether the change in the direction of motion is greater than or equal to the fourth threshold within a certain time period;

[0252] The direction of motion at time P+i is greater than 0, compared to the indication of whether the velocity at time P increases or decreases.

[0253] Within a certain time period, the number of times or the duration of movement in the direction of motion is greater than or equal to the seventeenth threshold;

[0254] The number of times or duration in which the direction of movement is less than or equal to the eighteenth threshold within a certain time period.

[0255] In some embodiments of this application, the aforementioned motion speed change information includes at least one of the following:

[0256] The change in speed per unit time, for example, the speed of movement increases by 3 m / s per unit time;

[0257] The difference between the initial and final values ​​of the velocity over a period of time, for example: the difference between the velocity at time t1 and the velocity at time t1+a, where a is greater than 0;

[0258] The percentage change in velocity over a given time period relative to the initial value;

[0259] The difference between the maximum and minimum values ​​of motion speed over a given time period;

[0260] An indication of whether the speed of movement increases or decreases over a period of time;

[0261] Whether the change in velocity over a given time period is greater than or equal to the nineteenth threshold;

[0262] The velocity at time Q+j is an indicator of whether the velocity at time Q has increased or decreased, where Q and j are greater than 0;

[0263] Within a given time period, the number of times or the duration of movement speed being greater than or equal to the twentieth threshold;

[0264] The number of times or duration during which the movement speed is less than or equal to the twenty-first threshold within a certain time period.

[0265] In some embodiments of this application, the sensing information of the terminal, network-side device, or satellite device may include at least one of the following: information obtained by the terminal through sensing, information obtained by the network-side device through sensing, and information obtained by the satellite device through sensing.

[0266] In some embodiments of this application, the sensing information may include communication environment information, scene information, channel state information, the number of terminals, etc. The channel state information may indicate at least one of the following: line of sight (LOS), non-line of sight (NLOS), the presence of obstacles, and the number of obstacles.

[0267] For example, a terminal, network-side device, or satellite device can sense whether there are obstacles or the number of obstacles under a specific beam.

[0268] For example, terminals, network-side devices, or satellite devices can sense the number of terminals or devices within the coverage area of ​​a specific beam.

[0269] In some embodiments of this application, weather information or weather change information may include at least one of the following:

[0270] Weather or weather change information of the terminal's environment;

[0271] Weather or weather change information of the environment in which the network-side devices are located.

[0272] In some embodiments of this application, the security-related information of data or services may include at least one of the following: the security level requirements of the data or services to be transmitted, and whether the data or services to be transmitted involve confidentiality.

[0273] In some embodiments of this application, historical transmission-related information may include at least one of the following: historical service or load information, historical transmission performance parameters, historical modulation and coding information, historical link quality information, historical beam information, transmission time of the most recent downlink data, transmission time of the most recent uplink data, number of downlink data transmissions within a historical time period, number of uplink data transmissions within a historical time period, most recent transmission time of SDT data, most recent transmission time of SDT data in disconnected state, number of SDT data transmissions within a historical time period, and number of SDT data transmissions in disconnected state within a historical time period.

[0274] In some embodiments of this application, the aforementioned historical service or load information may include at least one of the following: service information or load information of network-side devices within a historical time period; load information corresponding to one or more measurement objects of network-side devices within a historical time period, wherein the measurement object may be the SSB, beam, TCI, TRP, carrier, or frequency point of the network-side device; historical service information of the terminal; and historical load information of the terminal.

[0275] In some embodiments of this application, the above-mentioned business information may include at least one of the following: business type, business cycle, business pattern, size of business packet, packet delay budget (PDB) of the business, and distribution of business packet size.

[0276] In some embodiments of this application, the load information may include at least one of the following: uplink or downlink traffic volume, number of users, maximum uplink or downlink traffic volume, and maximum number of users.

[0277] In some embodiments of this application, the aforementioned historical transmission performance parameters may include at least one of the following:

[0278] Historical average transmission performance parameters of network-side devices;

[0279] The average, maximum, or minimum transmission performance parameters of the network-side device over a period of time prior to the current point in time;

[0280] Historical average transmission performance parameters of the terminal;

[0281] The terminal's average, maximum, or minimum transmission performance parameters over a period of time prior to the current point in time.

[0282] In some embodiments of this application, the transmission performance parameters may include at least one of the following: rate, throughput, and transmission delay.

[0283] For example, taking the network-side device as the cell currently accessed by the terminal, the aforementioned historical transmission performance parameters may include at least one of the following: the historical average throughput of the cell currently accessed by the terminal; the average throughput of the cell currently accessed by the terminal over a period of time prior to the current time; the average rate of the cell currently accessed by the terminal over a period of time prior to the current time; the maximum instantaneous rate or minimum instantaneous rate of the cell currently accessed by the terminal over a period of time prior to the current time; the historical average throughput of the terminal; the average throughput of the terminal over a period of time prior to the current time; the average rate of the terminal over a period of time prior to the current time; the historical average throughput of the terminal; the historical average throughput of the terminal over a period of time prior to the current time ... The maximum or minimum instantaneous rate over a previous period of time; the average latency of historical data transmission in the cell currently accessed by the terminal; the average latency of data transmission in the cell currently accessed by the terminal over a previous period of time at the current point in time; the maximum or minimum latency of data transmission in the cell currently accessed by the terminal over a previous period of time at the current point in time; the average latency of historical data transmission of the terminal; the average latency of data transmission of the terminal over a previous period of time at the current point in time; the maximum or minimum latency of data transmission of the terminal over a previous period of time at the current point in time, wherein the data transmission level can be data stream, data packet, transport block (TB), or Media Access Control (MAC) Protocol Data Unit (PDU).

[0284] In some embodiments of this application, historical modulation and coding information may include at least one of the following: historical average modulation and coding information of the network-side device; average modulation and coding information, maximum modulation and coding information, or minimum modulation and coding information of the network-side device within a time period prior to the current time point; historical average modulation and coding information of the terminal; average modulation and coding information, maximum modulation and coding information, or minimum modulation and coding information of the terminal within a time period prior to the current time point.

[0285] In some embodiments of this application, the modulation and coding information may include at least one of the following: modulation and coding strategy (MCS) level, modulation method, and code rate.

[0286] For example, taking the cell currently accessed by the terminal as an example, the aforementioned historical modulation and coding information may include at least one of the following:

[0287] The historical average MCS level, modulation scheme, and code rate of the cell currently accessed by the terminal;

[0288] The average MCS level, modulation scheme, or code rate of the cell currently accessed by the terminal over a period of time prior to the current point in time;

[0289] The maximum or minimum MCS level, modulation scheme or code rate of the cell currently accessed by the terminal within a period of time prior to the current point in time;

[0290] The terminal's historical average MCS level, modulation scheme, or code rate;

[0291] The terminal's average MCS level, modulation scheme, or code rate over a period of time prior to the current point in time;

[0292] The terminal's maximum or minimum MCS level, modulation scheme, or bit rate during a period of time preceding the current point in time.

[0293] In some embodiments of this application, the aforementioned historical link quality information may include at least one of the following:

[0294] The average, maximum, or minimum quality parameters of the network-side devices over a period of time prior to the current point in time;

[0295] Historical average quality parameters of the terminal;

[0296] The terminal's average, maximum, or minimum mass parameters over a period of time prior to the current point in time.

[0297] For example, taking the network-side device as the cell currently accessed by the terminal, the aforementioned historical link quality information may include at least one of the following:

[0298] At least one of the historical average CQI, CSI-SINR, RI, and windband-SINR of the cell currently accessed by the terminal;

[0299] The average CQI, CSI-SINR, RI, and windband-SINR of the currently accessing cell over a period of time prior to the current point in time;

[0300] The maximum CQI, CSI-SINR, RI, windband-SINR, minimum CQI, CSI-SINR, RI, and windband-SINR of the currently accessed cell within a period of time preceding the current point in time;

[0301] At least one of the terminal's historical average CQI, CSI-SINR, RI, and windband-SINR;

[0302] The terminal's average CQI, CSI-SINR, RI, and windband-SINR over a period of time preceding the current point in time;

[0303] The terminal has at least one of the following parameters during a period of time prior to the current point in time: maximum CQI, CSI-SINR, RI, minimum CQI, CSI-SINR, RI, and windband-SINR.

[0304] In some embodiments of this application, the aforementioned historical beam information may include at least one of the following: the terminal's historical average beam direction; the terminal's average beam direction over a period of time prior to the current time point; and the beam direction or beam index most frequently used by the terminal over a period of time prior to the current time point.

[0305] In some embodiments of this application, "the transmission time of the most recent downlink data" may include at least one of the following: the time when the network-side device last sent downlink data; the time when the terminal last received downlink data, i.e., the arrival time of the most recent downlink data.

[0306] In some embodiments of this application, "the transmission time of the most recent uplink data" may include at least one of the following:

[0307] The time when the network-side device last received uplink data, that is, the arrival time of the most recent uplink data, such as the time when the terminal last received uplink data sent by the terminal in the cell it is currently accessing;

[0308] The time when the terminal last sent downlink data.

[0309] In some embodiments of this application, "the most recent transmission time of SDT data" can be at least one of the following: the time when the terminal last sent small data in idle or inactive state; the time when the network-side device last sent small data in idle or inactive state; the time when the terminal last received small data in idle or inactive state; and the time when the network-side device last received small data in idle or inactive state.

[0310] In some embodiments of this application, the aforementioned "transmission time" can be a specific time accurate to a certain precision. For example, accurate to the second, such as 13:25:38.

[0311] In some embodiments of this application, the aforementioned "transmission time" can be a time range.

[0312] For example, 1 PM to 2 PM, morning, afternoon, daytime or nighttime, etc.

[0313] In some embodiments of this application, the aforementioned "transmission time" may be timing information obtained by the terminal or network-side device through other radio access technologies (RAT). The RAT may be Bluetooth, Wi-Fi, or 3G, 4G, or 5G technologies, etc.

[0314] In some embodiments of this application, "the number of times SDT data is sent within a historical time period" may include at least one of the following: the number of times the terminal sends SDT data within a historical time period; the number of times the network-side device sends SDT data within a historical time period; and the sum of the number of times the terminal and the network-side device send SDT data within a historical time period.

[0315] In some embodiments of this application, "the number of times SDT data is sent in a non-connected state within a historical time period" may include at least one of the following: the number of times the terminal sends SDT data in a non-connected state within a historical time period; the number of times the network-side device sends SDT data in a non-connected state within a historical time period; and the sum of the number of times the terminal and the network-side device send SDT data in a non-connected state within a historical time period.

[0316] In some embodiments of this application, the model inference triggering condition for inference using the first AI model may include at least one of the following conditions a1 to a14:

[0317] a1. Obtain the first configuration;

[0318] a2. Periodic triggering;

[0319] a3. Receive activation information for activating the inference of the first AI model;

[0320] a4. Inference-related timer timed out;

[0321] a5. Before the terminal enters the idle or inactive state;

[0322] a6. The duration of no data transmission is greater than or equal to the first duration;

[0323] a7. The cumulative amount of data to be transmitted within a certain time period is greater than the second threshold;

[0324] a8. The cumulative amount of data to be transmitted within a time period is less than or equal to the second threshold.

[0325] a9. RRC state transition related timer timeout;

[0326] a10. The remaining duration of the timer related to the RRC state transition is less than or equal to the third threshold;

[0327] a11. After receiving the RRC release instruction from the upper layer;

[0328] a12. Receive at least a portion of the information in the first information;

[0329] a13. Within a time period, the measured value or the change in measured value of the first object satisfies the condition.

[0330] a14. Businesses with target types exist;

[0331] a15. Obtain the first configuration and receive the relevant instruction; for example, the relevant instruction may be a DCI activation instruction.

[0332] In some embodiments of this application, the first configuration may include at least one of the following: an AI model for inference or an identifier of the AI ​​model; the application scope of the AI ​​model inference; the cycle of the AI ​​model inference; the effective duration of the AI ​​model inference; at least one of the inference triggering condition or inference stopping condition of the AI ​​model inference; the configuration of the AI ​​model; a switch to activate the AI ​​model inference; and a switch to deactivate the AI ​​model inference.

[0333] In some embodiments of this application, the application scope of AI model inference may include at least one of the following: the frequency domain range to which AI model inference can be applied; the cell or cell list to which AI model inference can be applied; the range of the terminal's location; and the distance range between the terminal and network-side devices, such as base stations.

[0334] In some embodiments of this application, the configuration of the AI ​​model may include at least one of the following: at least some of the input information of the AI ​​model, such as the first information described above; at least some of the output information of the AI ​​model, such as the second information described above; and an indication of whether joint inference between the network and the terminal is supported or required.

[0335] In some embodiments of this application, the inference-related timer can be an inference startup timer. For example, the inference startup timer can be started upon receiving the first configuration.

[0336] In some embodiments of this application, "the measured value or change in the measured value of the first object satisfies a condition within a time period" includes at least one of the following:

[0337] Within a certain time period, at least one measurement or change in measurement of the first object is greater than or equal to the eleventh threshold.

[0338] Within a certain time period, the rate of change of at least one measurement of the first object is greater than or equal to the twelfth threshold;

[0339] Within a certain time period, at least one measurement of the first object is greater than or equal to the thirteenth threshold, or the duration for which the at least one measurement is greater than or equal to the fourteenth threshold is greater than the second duration;

[0340] Within a certain time period, the duration for which at least one measurement of the first object is less than or equal to the fifteenth threshold or the duration for which the at least one measurement is less than or equal to the sixteenth threshold is greater than the third duration.

[0341] For example, taking the first object as a beam, if at least one of the signal strength information, signal quality information, interference signal strength information, TA information, and link quality information of a beam is greater than or equal to the eleventh threshold within a time period, then the measurement value of the beam is considered to meet the condition.

[0342] For example, taking the first object as a beam, if at least one of the RSRQ, SNR, SINR, BLER, BER and CQI of a beam is greater than or equal to the eleventh threshold within a time period, then the measurement value of the beam is considered to meet the condition.

[0343] Thus, since the model reasoning triggering conditions for reasoning using the first AI model include at least one of a1 to a15 above, the accuracy and flexibility of reasoning using the first AI model can be improved.

[0344] In some embodiments of this application, the inference stopping conditions for using the first AI model include at least one of the following b1 to b6: b1, the terminal enters an idle state, an inactive state, a disconnected state, a power-saving state, or a connected state; b2, the inference-related timer times out; b3, the terminal receives an RRC release message; b4, RACH transmission is triggered; b5, SDT transmission is triggered; b6, the predicted measurement value of the first object inferred by the first AI model is greater than or equal to a fourth threshold; b7, the predicted measurement value of the first object inferred by the first AI model is less than or equal to a fifth threshold.

[0345] In some embodiments of this application, the inference-related timer timeout can be: inference timer timeout.

[0346] For example, the inference timer can be started after the first AI model begins inference. If the inference timer times out, it means that the inference stop condition has been met.

[0347] Thus, since the reasoning stopping condition for using the first AI model can include at least one of the six conditions, the first device can stop using the first AI model for model reasoning when at least one of the six conditions is met, thereby improving the flexibility of stopping reasoning.

[0348] In some embodiments of this application, step 401 described above can be implemented by step 401A described below.

[0349] Step 401A: When the reasoning triggering condition and the second condition are met, the first device uses the first AI model to obtain the second information based on the first information;

[0350] The second condition may include at least one of the following: the inference requirement matches the inference capability of the first AI model; or the inference instruction is received from the network-side device.

[0351] In some embodiments of this application, the inference requirement can be a requirement for the first device to perform inference. For example, it can be the inference requirement of the first device, the inference requirement of the terminal, or the inference requirement of the network-side device.

[0352] For example, suppose a terminal needs to determine whether to take a measurement within at least one time period in the future, the terminal can request a first device to obtain second information based on the first information through a first AI model.

[0353] In some embodiments of this application, before the first device performs AI model training, supervision, and inference, it is necessary to define the relevant capabilities of the first device and provide a method for determining these capabilities. Introducing a mechanism for defining and determining relevant capabilities helps each node flexibly implement its corresponding features and facilitates feature implementation between nodes. It can be understood that the terminal, server, base station, and satellite are each considered a node.

[0354] In some embodiments of this application, the first device may have AI-related capabilities for determining SDT transmission based on AI assistance;

[0355] The AI-related capabilities may include at least one of the following: the ability to support the training of a first AI model; the ability to support inference using an AI model; and the ability to send at least one type of auxiliary information, which can be used for the training, inference, or supervision of the first AI model.

[0356] For example, the aforementioned auxiliary information can be input information that assists in the training, inference, or supervision of the first AI model, such as at least a portion of the information in the aforementioned first information.

[0357] In some embodiments of this application, the aforementioned AI-related capabilities can be determined by at least one of the following methods: the type of the first device; indication by one or more reference signals; or capability information of the AI-related capabilities carried by fifth information, which may include at least one of the following: uplink control information, RRC signaling, and AI-related interface messages between the first device and the terminal, network-side device, or satellite device.

[0358] In some embodiments of this application, the type of the first device may differ, and the AI-related capabilities of the first device may also differ.

[0359] For example, taking the first device as a terminal, the aforementioned AI-related capabilities may include at least one of the following: the terminal has the ability to support the training of the first AI model; the terminal has the ability to support AI inference using the first AI model; the terminal has the ability to support reporting one or more auxiliary information for the training / inference of the first AI model.

[0360] For example, taking the first device as an access network device or a core network device, the above-mentioned AI-related capabilities may include at least one of the following: the base station, TRP, or core network device has the ability to support the training of the first AI model; the base station, TRP, or core network device has the ability to support AI inference using the first AI model; the base station, TRP, or core network device has the ability to support indicating one or more auxiliary information for the training / inference of the first AI model.

[0361] For example, taking the first device as an AI-related server, the aforementioned AI-related capabilities may include at least one of the following: the AI-related server has the ability to support the training of a first AI model; the AI-related server has the ability to support AI inference using the first AI model; the AI-related server has the ability to support reporting one or more auxiliary information for the training or inference of the first AI model.

[0362] In some embodiments of this application, taking the determination of AI-related capabilities based on the type of the first device as an example, then:

[0363] If the first device is a terminal, then the AI-related capabilities depend on the type of terminal.

[0364] For example, different AI-related capabilities can be introduced for different terminal types (such as RedCap and IoT).

[0365] If the first device is a network-side device, then the AI-related capabilities depend on the network type.

[0366] For example, different AI-related capabilities can be introduced for different network types (such as NTN and Terrestrial Network (TN)).

[0367] For example, taking the aforementioned AI-related capabilities as an example where a reference signal is one or more reference signals:

[0368] Network-side devices can use specific resources of PRACH to instruct terminals to have the ability to train the first AI model or to use the first AI model for inference.

[0369] For example, the specific resource could be a random access opportunity (RACH Occasion RO), a PRACH Occasion, or a preamble.

[0370] As can be understood, both RACH Occasion (RO) and PRACH Occasion refer to the time-frequency resources that can be used to send RACH or preamble.

[0371] For example, taking the capability information of AI-related abilities carried by the fifth piece of information as an example, this capability information can be carried by at least one of the following pieces of information:

[0372] Uplink control information, such as physical layer control information, such as Uplink Control Information (UCI) information reported by the terminal to the network;

[0373] RRC signaling;

[0374] Specific interface messages between the terminal and the server;

[0375] Specific interface messages between the terminal and the access network equipment;

[0376] Specific interface messages between the server and access network equipment or core network equipment;

[0377] The aforementioned specific interface messages can be messages related to a specific AI model or messages related to all AI models.

[0378] In some embodiments of this application, if the terminal performs training or inference of the first AI model, the type of the terminal needs to be considered, as different types of terminals may have different AI model training or inference capabilities.

[0379] For example, the input information used for model training or inference differs for different types of terminals.

[0380] For example, less input information should be used for model training on terminal devices with weaker capabilities.

[0381] For example, the labels used for model training differ for different types of terminals.

[0382] For example, the model training is performed differently for different types of terminals.

[0383] For example, for terminals with weaker capabilities, we can consider performing model training only on the network side, or performing only a small part of the joint model training on the terminal side (e.g., model training involving user privacy data can be performed on the terminal side).

[0384] For example, the models used for model training differ for different types of terminals.

[0385] For example, overly complex artificial intelligence models may not be applicable to devices with weaker capabilities.

[0386] It is understood that the RACH, PRACH, or preamble in the above embodiments can also be any module that includes at least one of synchronization signals, random access signals or channels, uplink control signals or channels, or other control channels used for access. Random access timing (RACH Occasion, RO) and PRACH Occasion both refer to the time-frequency resources required to transmit RACH or preamble.

[0387] It is understood that the SSB or synchronization signal in the embodiments of this application may also be referred to as any module that includes a synchronization signal, a broadcast signal, a broadcast channel, other system messages, a downlink broadcast channel, and a control channel of the downlink broadcast channel.

[0388] In the SDT determination method provided in this application embodiment, since the second information can be obtained by reasoning using the first AI model, and the second information is used to determine whether SDT transmission will be triggered or whether SDT transmission resources need to be configured, SDT resources or SDT measurements are avoided when SDT transmission is not required, thereby avoiding waste of resources and configuration overhead.

[0389] In some embodiments of this application, after step 401 above, the SDT determination method provided in the embodiments of this application may further include the following step 402.

[0390] Step 402: The first device performs the first operation based on the second information.

[0391] The first operation may include at least one of the following: sending third information to a second device, which may be a device for training the first AI model, and the third information may be used for fine-tuning or updating the first AI model; reverting to a non-AI SDT transmission determination process; triggering a switch or update of the AI ​​model; triggering a change processing of the AI ​​model input information; triggering AI model training; triggering fine-tuning processing of the AI ​​model; triggering AI model supervision; not configuring or reserving the first SDT resource in advance; configuring or reserving only the second SDT resource; not configuring SDT-related fields in the system message block SIB1; not configuring SDT-related fields in the RRC release message; performing SDT transmission at the first resource location; stopping measurement for at least one time period; stopping measurement on the reference signal; stopping measurement on the reference signal for at least one time period; increasing the measurement period; performing continuous measurement for at least one time period; and performing measurement only on at least one beam or the reference signal corresponding to the at least one beam.

[0392] In some embodiments of this application, the second device can be a terminal, a network-side device, or an AI-related server. The network-side device may include: an access network device or a core network device specifically used for model training.

[0393] In some embodiments of this application, the second device is different from the first device.

[0394] For example, the first device is a base station, and the second device is a terminal.

[0395] For example, the first device is a base station, and the second device is another base station.

[0396] For example, the first device is a server, and the second device is a terminal.

[0397] For example, the first device is a terminal, and the second device is another terminal.

[0398] In some embodiments of this application, the aforementioned third information may include at least one of the following: first information, second information, and the current status information of the first device.

[0399] For example, the aforementioned status information can indicate the current state of the first device, such as connected state or power-saving state.

[0400] In some embodiments of this application, step 402 described above can be implemented by step 402A described below.

[0401] Step 402A: If the first condition is met, the first device performs the first operation based on the second information.

[0402] The first condition may include at least one of the following:

[0403] The reasoning time using the first AI model exceeds K time intervals, and the second information still does not meet the performance requirements.

[0404] Inference using the first AI model failed;

[0405] Inference was successful using the first AI model;

[0406] The number of times the first AI model is used for inference reaches U, where U is a positive integer.

[0407] Thus, the first condition may include conditions that characterize the inference performance of the first AI model, such as the inference time exceeding K durations, the second information still not meeting the performance requirements, the inference using the first AI model failing, the inference using the first AI model succeeding, or the number of inferences using the first AI model reaching U times. That is, when the inference performance of the first AI model meets the inference performance conditions, the first device performs the first operation based on the first information, thereby improving the accuracy of the first operation performed by the first device.

[0408] It is understandable that before using a specific AI model for specific reasoning, it is necessary to train the model to generate a specific AI model with corresponding functions, such as the first AI model.

[0409] In some embodiments of this application, the entity performing the training of the first AI model may include at least one of the following:

[0410] Model training is performed on the terminal side;

[0411] Model training is performed on the network side, which may include at least one of a base station, a TRP, or a core network device (such as a core network device specifically designed for model training).

[0412] Model training is performed on the server side. This server can be a device specifically designed for training, inference, or providing AI-related information, or it can be a third-party server, or a device provided by an over-the-top (OTT) service provider, a third-party service provider, or the internet. These servers can be collectively referred to as: AI-related servers.

[0413] In some embodiments of this application, the base station for training the first AI model may be the base station where the current terminal is stationed, accessing, or reselecting the target of the target-switching terminal.

[0414] In some embodiments of this application, the training of the first AI model can be performed on the terminal side, the network side, or the server side. Alternatively, the first AI model can be jointly trained by at least two of the terminal side, the network side, and the server side.

[0415] In some embodiments of this application, partial model training is performed only on the terminal side, network side, or server side, or joint training is performed on the terminal side, network side, and server side.

[0416] In some embodiments of this application, when the terminal side, network side, and server side jointly train the first AI model, i.e., when performing joint training, the devices involved in the joint training can interact with each other to exchange training-related information, thereby improving training performance. Specifically, the interaction of training-related information may include at least one of the following:

[0417] The terminal side can report at least part of the output information of the AI ​​model training to the network side or server side, and then the network side or server side will use the information reported by the terminal side (i.e. the output information of the terminal side model training) as one of the input contents of its own model training.

[0418] The network side sends at least a portion of the output information of the model training to the terminal or server side, and the terminal or server side uses the information sent by the network side (i.e. the output information of the network side model training) as one of the input contents for its own model training.

[0419] Offline model training is performed on the terminal side, network side, or server side, and then joint fine-tuning is performed on the actual network by the terminal side, network side, or server side.

[0420] In some embodiments of this application, when the training of the first AI model occurs or at least partially occurs on the terminal side or server side, there may be multiple trained AI models on the terminal side or server side. The terminal side or server side reports auxiliary information to the network side, and the network side determines and instructs the first device to use which AI model (such as the first AI model) for AI inference through a first instruction.

[0421] In some embodiments of this application, any two AI models among multiple AI models have at least one of the following differences:

[0422] Differences during the training phase, such as differences in the training dataset or the label set;

[0423] Differences in the reasoning stage, such as differences in input information and differences in reasoning output information.

[0424] In some embodiments of this application, when multiple trained AI models exist on the terminal side or server side, the aforementioned auxiliary information may include the identifiers of the multiple AI models and the classification identifiers of the training dataset.

[0425] In some embodiments of this application, the first instruction may indicate at least one of the following: the representation (Identity Document, ID) of the first AI model, the ID of the training dataset of the first AI model, or configuration information related to the acquisition of training data for the first AI model. Then, the terminal or server can determine which AI model to use for model inference based on the received AI model ID, training dataset ID, or configuration information related to training data acquisition. It is understood that, in this case, the first device can be a terminal or a server.

[0426] In some embodiments, the training of the AI ​​model occurs, or at least partially occurs, on the network side or server side. After the AI ​​model training is completed, the network side or server side can send the trained AI model to the terminal, and then the terminal can use the AI ​​model for inference.

[0427] In some embodiments of this application, the SDT determination method provided in this application may further include the following step 403.

[0428] Step 403: The first device trains the AI ​​model based on the fourth information to obtain the first AI model.

[0429] The fourth piece of information may include at least one of the following: training dataset related to SDT transmission, label set related to SDT transmission, optimization algorithm or index for AI model training, loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0430] It is understandable that the first device, when it has the capability to train an AI model, can train the AI ​​model based on the fourth information to obtain the first AI model.

[0431] In some embodiments of this application, the first device can first train a first AI model, and then use the first AI model to perform inference to obtain the aforementioned second information. It can be seen that the first device is the primary entity responsible for both the training and inference of the first AI model.

[0432] In some embodiments of this application, at least some information items in the training dataset match information items in the first information.

[0433] It is understood that matching at least some information items in the training dataset with information items in the first information can include: the training dataset and the first information including at least one identical information item.

[0434] For example, both the training dataset and the fourth information include signal strength information for at least one beam.

[0435] It is understandable that the information values ​​of information items that match the first information in the training dataset can be the same or different.

[0436] For example, the signal quality information on beam 1 in the first information is RSRP1, and the signal quality information on beam 1 in the first information is RSRP2.

[0437] In some embodiments of this application, the training dataset may also be referred to as training input information.

[0438] In some embodiments of this application, the tag set may include at least one of the following: an indication of whether to perform SDT transmission; an indication of whether to perform a first type of SDT transmission; the type of SDT transmission; a time point or time period for performing SDT transmission; a time point or time period for performing a first type of SDT transmission; a time point or time period for not performing SDT transmission; a time point or time period for not performing a first type of SDT transmission; and resources for SDT transmission.

[0439] In some embodiments of this application, the label set may be referred to as the truth set or the target of model training.

[0440] It is understood that a tag set may include multiple tag data or truth values. Each truth value can be any of the following: an indication of whether to perform SDT transmission; an indication of whether to perform Type I SDT transmission; the type of SDT transmission; the time point or time period for performing SDT transmission; the time point or time period for performing Type I SDT transmission; the time point or time period for not performing SDT transmission; the time point or time period for not performing Type I SDT transmission; and the resources for SDT transmission.

[0441] In some embodiments of this application, "the time point or time period for performing SDT transmission" may include at least one of the following: the time point in which at least one of RA-SDT, CG-SDT, Paging SDT, RACH-less DG-SDT, paging-based DL SDT (i.e. MDT), and Paging directly scheduled DL SDT occurs;

[0442] The time period occurring in RA-SDT, CG-SDT, Paging SDT, RACH-less DG-SDT, paging-based DL SDT (i.e. MDT), and DL SDT directly scheduled by paging.

[0443] In some embodiments of this application, "the time point or time period during which no first type of SDT transmission occurs" may include at least one of the following: the time point or time period during which at least one of RA-SDT, CG-SDT, Paging SDT, RACH-less DG-SDT, paging-based DL SDT (MDT), and DL SDT directly scheduled by paging does not occur. Specifically, it can be determined according to the first type.

[0444] In some embodiments of this application, at least some information items in the tag set match information items in the second information. For example, at least one information item is the same between the tag set and the second information. Similarly, the value of the at least one information item may be different in the tag set and the second information.

[0445] In some embodiments of this application, the training dataset may include information fed back by the terminal or information measured by the network-side device;

[0446] Alternatively, the tag set may include information fed back by the terminal or information measured by network-side devices.

[0447] In some embodiments of this application, the optimization algorithm for training the AI ​​model may include, but is not limited to, at least one of the following: error back propagation (BP) algorithm, stochastic gradient descent algorithm, etc.

[0448] In some embodiments of this application, the loss function for training the AI ​​model can be referred to as the objective function.

[0449] In some embodiments of this application, the reward information for AI model training can be referred to as the adjustment information or feedback information of the AI ​​model.

[0450] In some embodiments of this application, the reward information includes at least one of the following: the number of times the inference failed or succeeded, the difference between the inference result and the label data, and the number of times a rollback occurred.

[0451] In some embodiments of this application, the number of rollbacks can be the number of times a rollback occurs to a non-AI SDT transmission determination process.

[0452] In some embodiments of this application, the inference error can be any of the following: the error between the output information of the first AI model and the actual information, or the error between the output information of the first AI model and the label during model training. It is understood that the output information can also be referred to as the inference result.

[0453] Thus, since the first device can train the AI ​​model based on at least one of the following: the training dataset related to SDT transmission, the tag set related to SDT transmission, the optimization algorithm or index for AI model training, the loss function for AI model training, the reward information for AI model training, and the triggering condition for AI model training, and obtain the first AI model, the accuracy of model training can be improved, thereby improving the inference performance of the first AI model.

[0454] Furthermore, since the first AI model is used on the same device as the training device, the first device can combine its own reasoning needs to acquire fourth information and train the model based on the fourth information. This allows the reasoning performance of the trained first AI model to meet the reasoning needs of the first device, thereby improving the reasoning accuracy.

[0455] It should be noted that after the first AI model is trained, online training or model updates can be performed as needed, i.e., model supervision can be performed on the first AI model to ensure the reliability and accuracy of its inference.

[0456] In some embodiments of this application, the first device may also perform model supervision on the first AI model based on a supervision configuration for supervising the first AI model.

[0457] The aforementioned supervision configuration may include at least one of the following: the identifier of the first AI model; the model supervision cycle; the number of model supervisions; the model supervision duration; information related to the model supervision window; the model supervision triggering conditions; the model supervision label; and the model supervision metrics.

[0458] In some embodiments of this application, the identifier of the AI ​​model may include, but is not limited to, at least one of the following: AI structure identifier; AI algorithm identifier; identifier of a specific dataset associated with the first AI model; identifier of a specific scene, environment, channel characteristics, or device related to the first AI model; identifier of a function, characteristic, capability, or module related to the first AI model, etc. This application does not impose specific limitations on this.

[0459] In some embodiments of this application, the identifier of the AI ​​model may also be referred to as the AI ​​model identifier. The AI ​​model identifier can be described in various ways, such as functional ID, model ID, physical ID, logical ID, global ID, local ID, etc.

[0460] In some embodiments of this application, the model supervision period can be the interval between model supervision sessions; for example, model supervision can be performed once every 1 day.

[0461] In some embodiments of this application, the number of model supervision iterations includes at least one of the following:

[0462] The number of times the first AI model needs to be supervised within each model supervision cycle, for example, 10 supervision cycles for the model;

[0463] After receiving the supervision configuration, the total number of times the first AI model needs to be supervised is required. In some embodiments of this application, the model supervision duration may include at least one of the following:

[0464] The time required for model supervision of the first AI model each time within each model supervision cycle; that is, the time required for supervision each time the first AI model is supervised within each supervision cycle.

[0465] The duration of the supervision configuration; for example, the duration of the supervision configuration may include at least one model supervision cycle, or model supervision may only be performed within the duration of the configuration.

[0466] In some embodiments of this application, the model supervision window information may include at least one of the following:

[0467] Within each model supervision cycle, the actual duration of model supervision or the number of samples for which model supervision is performed (also known as the number of times);

[0468] Within each model supervision cycle, the duration of the window for performing model supervision, and the start or end time point;

[0469] In some embodiments of this application, the model supervision metrics may include at least one of the following: the error between the predicted value and the true value of the first AI model, and the performance metrics of the communication system.

[0470] In some embodiments of this application, the performance indicators of the communication system may include transmission latency, throughput, and the probability of collisions or conflicts occurring with resources such as RACH or PUSCH.

[0471] It can be understood that the predicted value of the first AI model refers to the inference result of the first AI model when inference is performed. The above-mentioned true value can be the actual situation at the time point or within the time period inferred by the first AI model, such as the actual SDT transmission situation.

[0472] It is understandable that the first device (such as a terminal, access network device, core network device, or server) can determine when to supervise the first AI model or how to supervise the first AI model based on the above configuration information.

[0473] In some embodiments of this application, the labels for model supervision described above can be labels used for model training or labels specifically used for model supervision.

[0474] In some embodiments of this application, the above-mentioned model supervision triggering conditions may include at least one of the following: when SDT transmission is performed based on the second information, there is no transmission configuration corresponding to the SDT; SDT transmission based on the second information fails; the inference error of the first AI model is greater than or equal to the sixth threshold; SDT resources are configured or reserved, but SDT transmission is not actually performed, for example, the terminal reserves SDT resources for a certain period of time, but there is no SDT transmission during that period of time; the model supervision index does not meet the requirements; the model supervision timer of the first AI model times out; the first AI model inference fails; the first AI model inference fails S times, where S is a positive integer; the number of inference failures of the first AI model reaches the seventh threshold; inference is performed using the first AI model; the probability of the first set of output information appearing is greater than or equal to the eighth threshold; the probability of the first set of output information appearing is less than the ninth threshold; the error of the second set of output information is greater than or equal to the tenth threshold, for example, based on historical information or traditional methods, the time point of beam failure is obtained, and if the first AI model gives a very different result, then the model supervision of the first AI model is triggered.

[0475] Thus, since the first device can perform model supervision on the first AI model based on the supervision configuration used for supervising the first AI model, the accuracy of model supervision performed by the first device can be improved, thereby enhancing the model supervision effect.

[0476] It is understood that the core network devices in the above embodiments can be dedicated to at least one of the core network devices for model training, inference, and supervision.

[0477] This application also provides a model training method. Figure 5 shows a flowchart of the model training method provided in this application. As shown in Figure 5, the model training method provided in this application may include the following step 501.

[0478] Step 501: The second device trains the AI ​​model based on the fourth information to obtain the first AI model.

[0479] The first AI model can be used for uplink transmission parameter inference;

[0480] The fourth piece of information includes at least one of the following: the training dataset related to SDT transmission, the label set related to SDT transmission, the optimization algorithm or index for AI model training, the loss function for AI model training, the reward information for AI model training, and the triggering conditions for AI model training.

[0481] In some embodiments of this application, the training dataset includes at least one of the following information: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device, or satellite device, panel orientation information of the terminal, network-side device, or satellite device, device information of the terminal, location-related information of the terminal, network-side device, or satellite device, motion-related information of the terminal, network-side device, or satellite device, perception information of the terminal, network-side device, or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information.

[0482] The first object includes any one of the following: beam, reference signal, TRP currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and BWP of the terminal.

[0483] In some embodiments of this application, the above-mentioned tag set includes at least one of the following information: an indication of whether to perform SDT transmission; an indication of whether to perform a first type of SDT transmission; the type of SDT transmission; a time point or time period for performing SDT transmission; a time point or time period for performing a first type of SDT transmission; a time point or time period for not performing SDT transmission; a time point or time period for not performing a first type of SDT transmission; and resources for SDT transmission.

[0484] In some embodiments of this application, the first type includes at least one of the following: RA-SDT, CG-SDT, paging SDT, RACH-less DG-SDT, paging-based downlink SDT, and paging-direct-scheduled downlink SDT.

[0485] In some embodiments of this application, the second device can be a terminal, a network-side device, or an AI-related server. The network-side device can be at least one of the following: an access network device, a core network device (such as a core network device specifically used for model training).

[0486] In some embodiments of this application, the second device may be the same as or different from the first device.

[0487] For further description of step 501, please refer to the relevant description of step 403 in the above embodiments. To avoid repetition, it will not be repeated here.

[0488] In the model training method provided in this application embodiment, since the second device can perform AI model training based on at least one of the training dataset related to SDT transmission, the label set related to SDT transmission, the optimization algorithm or index for AI model training, the loss function for AI model training, the reward information for AI model training, and the triggering condition for AI model training, and obtain the first AI model, the accuracy of model training can be improved, thereby improving the inference performance of the first AI model.

[0489] In some embodiments of this application, after step 501 above, the model training method provided in the embodiments of this application may further include the following step 502.

[0490] Step 502: The second device performs model supervision on the first AI model based on the supervision configuration used for supervising the first AI model.

[0491] The aforementioned supervision configuration may include at least one of the following: the identifier of the first AI model; the model supervision cycle; the number of model supervisions; the model supervision duration; information related to the model supervision window; the model supervision triggering conditions; the model supervision label; and the model supervision metrics.

[0492] In some embodiments of this application, the number of times the model supervision is requested may include at least one of the following: the number of times the first AI model needs to be supervised within each model supervision cycle; and the total number of times the first AI model needs to be supervised after receiving the supervision configuration.

[0493] In some embodiments of this application, the model supervision duration includes at least one of the following: the duration required for each model supervision of the first AI model within each model supervision cycle; and the effective duration of the supervision configuration.

[0494] In some embodiments of this application, the model supervision window information includes at least one of the following: the actual duration of model supervision or the number of samples for which model supervision is performed within each model supervision cycle; the duration of the window for which model supervision is required within each model supervision cycle, and the start or end time point.

[0495] In some embodiments of this application, the model supervision metrics include at least one of the following: the error between the predicted value and the true value of the first AI model, and the performance metrics of the communication system.

[0496] In some embodiments of this application, the above-mentioned model supervision triggering conditions may include at least one of the following: when SDT transmission is performed based on the inference result of the first AI model, there is no transmission configuration corresponding to the SDT; SDT transmission based on the inference result of the first AI model fails; the inference error of the first AI model is greater than or equal to the sixth threshold; SDT resources are configured or reserved, but SDT transmission is not actually performed; the model supervision index does not meet the requirements; the model supervision timer of the first AI model times out; the first AI model inference fails; the first AI model inference fails S times, where S is a positive integer; the number of inference failures of the first AI model reaches the seventh threshold; inference is performed using the first AI model; the probability of the first set of output information appearing is greater than or equal to the eighth threshold; the probability of the first set of output information appearing is less than the ninth threshold; the error of the second set of output information is greater than or equal to the tenth threshold.

[0497] In some embodiments of this application, the second device is any one of the following: a terminal, a network-side device, or a satellite device.

[0498] In some embodiments of this application, the second device has AI-related capabilities for determining SDT transmission with AI assistance;

[0499] The AI-related capabilities include at least one of the following: the ability to support the training of a first AI model; the ability to support inference using the first AI model; and the ability to send at least one type of auxiliary information used for the training or inference of the first AI model.

[0500] In some embodiments of this application, the aforementioned AI-related capabilities are determined by at least one of the following methods: the type of the first device; indication by one or more reference signals; capability information of AI-related capabilities carried by fifth information; wherein the fifth information includes at least one of the following: uplink control information, RRC signaling, and AI-related interface messages between the second device and the terminal, network-side device, or satellite device.

[0501] For further descriptions of step 502, please refer to the relevant descriptions in the above embodiments. To avoid repetition, they will not be repeated here.

[0502] In some embodiments of this application, the second device can perform model supervision on the first AI model during the inference process using the first AI model.

[0503] Thus, since the second device can perform model supervision on the first AI model based on the supervision configuration used for supervising the first AI model, the accuracy of model supervision performed by the first device can be improved, thereby enhancing the model supervision effect.

[0504] The SDT determination method provided in this application can be executed by an SDT determination device. This application uses an SDT determination device executing the SDT determination method as an example to illustrate the SDT determination device provided in this application.

[0505] This application provides an SDT determination device. As an example, the SDT determination device can be a communication device or a component within a communication device, such as a chip. The communication device can be a terminal, a network-side device, or a server, etc. Exemplarily, the terminal can be, but is not limited to, the type of terminal 11 listed above, and the network-side device can be, but is not limited to, the type of network-side device 12 listed above. This application does not impose specific limitations.

[0506] The SDT determination device includes a receiving module, a transmitting module, and a processing module. These modules can be implemented in software or hardware. When implemented in hardware, the processing module can be implemented by a processor. For example, the processor can include general-purpose processors, special-purpose processors, such as a Central Processing Unit (CPU), microprocessor, Digital Signal Processor (DSP), AI processor, Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC), Network Processor (NP), Field Programmable Gate Array (FPGA), or other programmable logic devices, gate circuits, transistors, discrete hardware components, etc. The receiving and transmitting modules can be implemented by a communication interface, which can include one or more of the following: transceiver, pins, circuits, bus, radio frequency unit, etc.

[0507] Specifically, referring to Figure 6, when the SDT determination device 600 is a terminal, a component in the terminal, a network-side device, or a server, the SDT determination device 600 includes a processing module 601, which is used to use a first AI model to obtain second information based on first information, wherein the first information is the input information required for the first AI model to perform model inference.

[0508] The second information is used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources.

[0509] In some embodiments of this application, the second information includes at least one of the following:

[0510] SDT transmission prediction results, connected state transmission indication, resource configuration of triggered SDT transmission, type of triggered SDT transmission, terminal state prediction results, uplink UL data volume prediction results, downlink measurement volume prediction results, candidate beams for SDT transmission and reference signals corresponding to the candidate beams.

[0511] The terminal state prediction result is used to indicate whether the terminal enters a first state or the probability of entering the first state within at least one time period.

[0512] In some embodiments of this application, the SDT transmission prediction result is used to indicate at least one of the following:

[0513] At least within a time period, whether SDT transmission is performed, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission.

[0514] Whether SDT transmission is performed at least at one time point, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission.

[0515] The time period or time point during which no SDT transmission or SDT transmission of the first type is performed, and the time period or time point during which SDT transmission or SDT transmission of the first type is performed, wherein the first type includes at least one of the following: Random Access RA-SDT, Configuration Authorization CG-SDT, Paging SDT, Dynamic Authorization DG-SDT without Random Access RACH-less, Paging-based Downlink SDT, and Paging Direct Scheduling Downlink SDT.

[0516] In some embodiments of this application, the first information includes at least one of the following: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device, or satellite device, panel orientation information of the terminal, network-side device, or satellite device, device information of the terminal, location-related information of the terminal, network-side device, or satellite device, motion-related information of the terminal, network-side device, or satellite device, perception information of the terminal, network-side device, or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information.

[0517] The first object includes any one of the following: beam, reference signal, transmit / receive point (TRP) currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and bandwidth portion (BWP) of the terminal.

[0518] In some embodiments of this application, the measurement-related information includes at least one of the following: current measurement value, measurement value change information over a time period, and historical measurement values;

[0519] The measured value change information includes at least one of the following: an indication of an increase or decrease in the measured value, the amount of change in the measured value, the percentage change in the measured value, an indication of whether the amount of change in the measured value is greater than a first threshold, and the number of times or duration that the amount of change in the measured value is greater than the first threshold.

[0520] The measured values ​​include at least one of the following: signal strength information, signal quality information, interference signal strength information, timing advance (TA) information, and link quality information.

[0521] In some embodiments of this application, the historical transmission-related information includes at least one of the following: historical service or load information, historical transmission performance parameters, historical modulation and coding information, historical link quality information, historical beam information, transmission time of the most recent downlink data, transmission time of the most recent uplink data, number of downlink data transmissions within a historical time period, number of uplink data transmissions within a historical time period, most recent transmission time of SDT data, most recent transmission time of SDT data in disconnected state, number of SDT data transmissions within a historical time period, and number of SDT data transmissions in disconnected state within a historical time period.

[0522] In some embodiments of this application, the beam configuration includes at least one of the following: the number of beams transmitted or received by the terminal, the beam direction, and the association between the beams or the beams and the reference signals corresponding to the beams; the number of beams transmitted or received by the network-side device, the beam direction, and the association between the beams or the beams and the reference signals corresponding to the beams; one or a set of reference signal indices; one or a set of beam indices; one or a set of beam direction information; and one or a set of Transmission Configuration Indicator (TCI) status information.

[0523] In some embodiments of this application, the location-related information includes at least one of the following: geographical location information of the terminal, network-side device or satellite device, geographical location change information of the terminal, network-side device or satellite device, cell reference location or movement trajectory of the terminal in a non-terrestrial network scenario, ephemeris information of the network-side device, and distribution information of the terminal.

[0524] In some embodiments of this application, the motion-related information includes at least one of the following: motion direction or motion direction change information of network-side devices; motion direction or motion direction change information of terminals; motion direction or motion direction change information of satellite devices; motion speed or motion speed change information of network-side devices; motion speed or motion speed change information of terminals; and motion speed or motion speed change information of satellite devices.

[0525] In some embodiments of this application, the model reasoning triggering conditions for using the first AI model for reasoning include at least one of the following: obtaining a first configuration; periodic triggering; receiving activation information for activating the first AI model reasoning;

[0526] The following events occur: inference-related timer timeout; before the terminal enters idle or inactive state; duration of no data transmission is greater than or equal to a first duration; cumulative data to be transmitted within a time period is greater than a second threshold; cumulative data to be transmitted within a time period is less than or equal to a second threshold; RRC state transition-related timer timeout; remaining duration of RRC state transition-related timer is less than or equal to a third threshold; after receiving an RRC release instruction from the upper layer; at least part of the information in the first information is received; within a time period, the measurement value or change in measurement value of the first object meets the condition; there is a service of the target type.

[0527] In some embodiments of this application, the inference stopping conditions for using the first AI model include at least one of the following: the terminal enters an idle state, an inactive state, a disconnected state, a power-saving state, or a connected state; the inference-related timer times out; the terminal receives an RRC release message; the random access channel (RACH) transmission is triggered; the SDT transmission is triggered; the predicted measurement value of the first object inferred by the first AI model is greater than or equal to a fourth threshold; or the predicted measurement value of the first object inferred by the first AI model is less than or equal to a fifth threshold.

[0528] In some embodiments of this application, the processing module is further configured to, after acquiring the second information, perform a first operation based on the second information, the first operation including at least one of the following: sending third information to a second device, the second device being a device for training the first AI model, the third information being used for fine-tuning or updating the first AI model; reverting to a non-AI SDT transmission determination process; triggering a switching or updating of the AI ​​model; triggering a change processing of the AI ​​model input information; triggering AI model training; triggering fine-tuning processing of the AI ​​model; triggering AI model supervision; not configuring or reserving the first SDT resource in advance; configuring or reserving only the second SDT resource; not configuring SDT-related fields in the system message block SIB1; not configuring SDT-related fields in the RRC release message; performing SDT transmission at the first resource location; stopping measurement for at least one time period; stopping measurement on the reference signal; stopping measurement on the reference signal for at least one time period; increasing the measurement period; performing continuous measurement for at least one time period; and performing measurement only on at least one beam or the reference signal corresponding to the at least one beam.

[0529] In some embodiments of this application, the processing module is specifically used to perform the first operation based on the second information when a first condition is met; wherein the first condition includes at least one of the following: the time for reasoning using the first AI model exceeds K time periods, and the second information still does not meet the performance requirements; reasoning using the first AI model fails; reasoning using the first AI model succeeds; the number of times reasoning using the first AI model reaches U times, where U is a positive integer.

[0530] In some embodiments of this application, the processing module is further configured to train an AI model based on the fourth information to obtain the first AI model;

[0531] The fourth piece of information includes at least one of the following: training dataset related to SDT transmission, label set related to SDT transmission, optimization algorithm or index for AI model training, loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0532] In some embodiments of this application, at least some information items in the training dataset match information items in the first information.

[0533] In some embodiments of this application, the tag set includes at least one of the following: an indication of whether to perform SDT transmission; an indication of whether to perform a first type of SDT transmission; the type of SDT transmission; a time point or time period for performing SDT transmission; a time point or time period for performing a first type of SDT transmission; a time point or time period for not performing SDT transmission; a time point or time period for not performing a first type of SDT transmission; and resources for SDT transmission.

[0534] In some embodiments of this application, the processing module is further configured to perform model supervision on the first AI model based on the supervision configuration for supervising the first AI model. The supervision configuration includes at least one of the following: the identifier of the first AI model; the model supervision period; the number of model supervisions; the model supervision duration; model supervision window related information; the model supervision triggering condition; the model supervision label; and the model supervision index.

[0535] In some embodiments of this application, the number of model supervision sessions includes at least one of the following: the number of times the first AI model needs to be supervised within each model supervision cycle; the total number of times the first AI model needs to be supervised after receiving the supervision configuration; wherein the model supervision duration includes at least one of the following: the duration required for each model supervision session of the first AI model within each model supervision cycle; the effective duration of the supervision configuration; wherein the model supervision window related information includes at least one of the following: the actual duration of model supervision or the number of samples for model supervision within each model supervision cycle; the duration of the window for model supervision within each model supervision cycle, the start time point or the end time point; wherein the model supervision metric includes at least one of the following: the error between the predicted value and the true value of the first AI model, and the communication system performance metric.

[0536] In some embodiments of this application, the model supervision triggering conditions include at least one of the following: no SDT transmission configuration is available when performing SDT transmission based on the inference result of the first AI model; SDT transmission fails based on the inference result of the first AI model; the inference error of the first AI model is greater than or equal to a sixth threshold; SDT resources are configured or reserved, but SDT transmission is not actually performed; the model supervision index does not meet the requirements; the model supervision timer of the first AI model times out; the first AI model fails to infer; the first AI model fails to infer S times (S is a positive integer); the number of inference failures of the first AI model reaches a seventh threshold; inference is performed using the first AI model; the probability of the first set of output information appearing is greater than or equal to an eighth threshold; the probability of the first set of output information appearing is less than a ninth threshold; the error of the second set of output information is greater than or equal to a tenth threshold.

[0537] In some embodiments of this application, the first device is any one of the following: a terminal, a network-side device.

[0538] In some embodiments of this application, the first device has AI-related capabilities for determining SDT transmission with AI assistance;

[0539] The AI-related capabilities include at least one of the following: the ability to support the training of the first AI model; the ability to support inference using the first AI model; and the ability to support sending at least one type of auxiliary information; wherein the auxiliary information is used for the training or inference of the first AI model.

[0540] In some embodiments of this application, the AI-related capabilities are determined by at least one of the following methods: the type of the first device; indication by one or more reference signals; capability information of the AI-related capabilities carried by fifth information; wherein the fifth information includes at least one of the following: uplink control information, RRC signaling, and AI-related interface messages between the first device and the terminal, network-side device, or satellite device.

[0541] In the SDT determination device provided in this application embodiment, since the second information can be obtained by reasoning using the first AI model, and the second information is used to determine whether SDT transmission will be triggered or whether SDT transmission resources need to be configured, SDT resources or SDT measurements are avoided when SDT transmission is not required, thereby avoiding waste of resources and configuration overhead.

[0542] The SDT determination device provided in this application embodiment can implement the various processes implemented in the method embodiment shown in 4 and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0543] This application provides a model training apparatus. As an example, the model training apparatus may be a communication device or a component within a communication device, such as a chip. The communication device may be a terminal, a network-side device, or a server, etc. Exemplarily, the terminal may include, but is not limited to, the type of terminal 11 listed above, and the network-side device may include, but is not limited to, the type of network-side device 12 listed above. This application does not impose specific limitations.

[0544] The model training device includes a receiving module, a transmitting module, and a processing module. These modules can be implemented in software or hardware. When implemented in hardware, the processing module can be implemented by a processor. For example, the processor can include general-purpose processors, special-purpose processors, such as a Central Processing Unit (CPU), microprocessor, Digital Signal Processor (DSP), Artificial Intelligence (AI) processor, Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC), Network Processor (NP), Field Programmable Gate Array (FPGA), or other programmable logic devices, gate circuits, transistors, discrete hardware components, etc. The receiving and transmitting modules can be implemented by a communication interface, which can include one or more of the following: transceiver, pins, circuits, bus, radio frequency unit, etc.

[0545] Specifically, referring to Figure 7, when the model training device 700 is a terminal, a component in a terminal, a network-side device, a server, or a component in a server, the model training device 700 includes a processing module 701, which is used to train an AI model based on the fourth information to obtain a first AI model. The first AI model is used for uplink transmission parameter inference.

[0546] The fourth piece of information includes at least one of the following: training dataset related to SDT transmission, label set related to SDT transmission, optimization algorithm or index for AI model training, loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0547] In some embodiments of this application, the training dataset includes at least one of the following information: the first information includes at least one of the following: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device or satellite device, panel orientation information of the terminal, network-side device or satellite device, device information of the terminal, location-related information of the terminal, network-side device or satellite device, motion-related information of the terminal, network-side device or satellite device, perception information of the terminal, network-side device or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information;

[0548] The first object includes any one of the following: beam, reference signal, TRP currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and BWP of the terminal.

[0549] In some embodiments of this application, the tag set includes at least one of the following information: an indication of whether SDT transmission is performed; an indication of whether a first type of SDT transmission is performed; the type of SDT transmission; a time point or time period for performing SDT transmission; a time point or time period for performing the first type of SDT transmission; a time point or time period for not performing SDT transmission; a time point or time period for not performing the first type of SDT transmission; and resources for SDT transmission.

[0550] In some embodiments of this application, the first type includes at least one of the following: RA-SDT, CG-SDT, paging SDT, RACH-less DG-SDT, paging-based downlink SDT, and paging-direct-scheduled downlink SDT.

[0551] In some embodiments of this application, after obtaining the first AI model, the method further includes:

[0552] Based on the supervision configuration used for supervising the first AI model, model supervision is performed on the first AI model. The supervision configuration includes at least one of the following: the identifier of the first AI model; the model supervision period; the number of model supervisions; the model supervision duration; model supervision window related information; model supervision trigger conditions; model supervision labels; and model supervision metrics.

[0553] In some embodiments of this application, the number of model supervision sessions includes at least one of the following: the number of times the first AI model needs to be supervised within each model supervision cycle; the total number of times the first AI model needs to be supervised after receiving the supervision configuration; wherein the model supervision duration includes at least one of the following: the duration required for each model supervision session of the first AI model within each model supervision cycle; the effective duration of the supervision configuration; wherein the model supervision window related information includes at least one of the following: the actual duration of model supervision or the number of samples for model supervision within each model supervision cycle; the duration of the window for model supervision within each model supervision cycle, the start time point or the end time point; wherein the model supervision metric includes at least one of the following: the error between the predicted value and the true value of the first AI model, and the communication system performance metric.

[0554] In some embodiments of this application, the model supervision triggering conditions include at least one of the following: no transmission configuration corresponding to the SDT is available during SDT transmission based on the inference result of the first AI model; SDT transmission based on the inference result of the first AI model fails; the inference error of the first AI model is greater than or equal to a sixth threshold; SDT resources are configured or reserved, but SDT transmission is not actually performed; the model supervision index does not meet the requirements; the model supervision timer of the first AI model times out; the first AI model inference fails; the first AI model inference fails S times, where S is a positive integer; the number of inference failures of the first AI model reaches a seventh threshold; inference is performed using the first AI model; the probability of the first set of output information appearing is greater than or equal to an eighth threshold; the probability of the first set of output information appearing is less than a ninth threshold; the error of the second set of output information is greater than or equal to a tenth threshold.

[0555] In some embodiments of this application, the second device is any one of the following: a terminal, a network-side device, or a satellite device.

[0556] In some embodiments of this application, the second device has AI-related capabilities for determining SDT transmission with AI assistance;

[0557] The AI-related capabilities include at least one of the following: the ability to support the training of the first AI model; the ability to support inference using the first AI model; and the ability to support sending at least one type of auxiliary information; wherein the auxiliary information is used for the training or inference of the first AI model.

[0558] In some embodiments of this application, the AI-related capabilities are determined by at least one of the following methods: the type of the first device; indication by one or more reference signals; capability information of the AI-related capabilities carried by fifth information; wherein the fifth information includes at least one of the following: uplink control information, RRC signaling, and AI-related interface messages between the second device and the terminal, network-side device, or AI-related server.

[0559] In the model training apparatus provided in this application embodiment, since AI model training can be performed based on at least one of the training dataset related to SDT transmission, the label set related to SDT transmission, the optimization algorithm or index for AI model training, the loss function for AI model training, the reward information for AI model training, and the triggering condition for AI model training, and a first AI model is obtained, the accuracy of model training can be improved, thereby improving the inference performance of the first AI model.

[0560] The model training device provided in this application embodiment can implement the various processes implemented in the method embodiment shown in 5 and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0561] As shown in Figure 8, this application embodiment also provides a communication device 800, including a processor 801 and a memory 802. The memory 802 stores programs or instructions that can run on the processor 801. For example, when the communication device 800 is a terminal, when the program or instructions are executed by the processor 801, they implement the various steps of the above-described SDT determination method embodiment and achieve the same technical effect; or they implement the various steps of the above-described model training method embodiment and achieve the same technical effect. To avoid repetition, further details are omitted here.

[0562] This application also provides a terminal, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps in the method embodiments shown in FIG4 or FIG5. This terminal embodiment corresponds to the above-described first device or second device-side method embodiments, and all implementation processes and methods of the above-described method embodiments can be applied to this terminal embodiment and can achieve the same technical effect. The terminal can be the SDT determination device shown in FIG6 or the model training device shown in FIG5. Specifically, FIG9 is a schematic diagram of the hardware structure of a terminal implementing an embodiment of this application.

[0563] The terminal 900 includes, but is not limited to, at least some of the following components: radio frequency unit 901, network module 902, audio output unit 903, input unit 904, sensor 905, display unit 906, user input unit 907, interface unit 908, memory 909, and processor 910.

[0564] Those skilled in the art will understand that the terminal 900 may also include a power supply (such as a battery) for powering various components. The power supply can be logically connected to the processor 910 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The terminal structure shown in Figure 9 does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0565] It should be understood that, in this embodiment, the input unit 904 may include a graphics processor 9041 and a microphone 9042. The graphics processor 9041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 906 may include a display panel 9061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit 907 includes at least one of a touch panel 9071 and other input devices 9072. The touch panel 9071 is also called a touch screen. The touch panel 9071 may include a touch detection device and a touch controller. Other input devices 9072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.

[0566] In this embodiment, after receiving downlink data from the network-side device, the radio frequency unit 901 can transmit it to the processor 910 for processing; in addition, the radio frequency unit 901 can send uplink data to the network-side device. Typically, the radio frequency unit 901 includes, but is not limited to, antennas, amplifiers, transceivers, couplers, low-noise amplifiers, duplexers, etc.

[0567] The memory 909 can be used to store software programs or instructions, as well as various data. The memory 909 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 909 may include volatile memory or non-volatile memory. The non-volatile memory may 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. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus RAM (DRRAM). The memory 909 in the embodiments of this application includes, but is not limited to, these and any other suitable types of memory.

[0568] Processor 910 may include one or more processing units; optionally, processor 910 integrates an application processor and a modem processor, wherein the application processor mainly handles operations involving the operating system, user interface, and applications, and the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into processor 910.

[0569] In one embodiment, terminal 900 can be the terminal described above:

[0570] The processor 910 is used to obtain second information based on first information using a first AI model, wherein the first information is the input information required for the first AI model to perform model inference.

[0571] The second information is used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources.

[0572] In some embodiments of this application, the second information includes at least one of the following:

[0573] SDT transmission prediction results, connected state transmission indication, resource configuration of triggered SDT transmission, type of triggered SDT transmission, terminal state prediction results, uplink UL data volume prediction results, downlink measurement volume prediction results, candidate beams for SDT transmission and reference signals corresponding to the candidate beams.

[0574] The terminal state prediction result is used to indicate whether the terminal enters a first state or the probability of entering the first state within at least one time period.

[0575] In some embodiments of this application, the SDT transmission prediction result is used to indicate at least one of the following:

[0576] At least within a time period, whether SDT transmission is performed, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission.

[0577] Whether SDT transmission is performed at least at one time point, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission.

[0578] The time period or time point during which no SDT transmission or SDT transmission of the first type is performed, and the time period or time point during which SDT transmission or SDT transmission of the first type is performed, wherein the first type includes at least one of the following: Random Access RA-SDT, Configuration Authorization CG-SDT, Paging SDT, Dynamic Authorization DG-SDT without Random Access RACH-less, Paging-based Downlink SDT, and Paging Direct Scheduling Downlink SDT.

[0579] In some embodiments of this application, the first information includes at least one of the following: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device or satellite device, panel orientation information of the terminal, network-side device or satellite device, device information of the terminal, location-related information of the terminal, network-side device or satellite device, motion-related information of the terminal, network-side device or satellite device, perception information of the terminal, network-side device or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information;

[0580] The first object includes any one of the following: beam, reference signal, transmit / receive point (TRP) currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and bandwidth portion (BWP) of the terminal.

[0581] In some embodiments of this application, the measurement-related information includes at least one of the following: current measurement value, measurement value change information over a time period, and historical measurement values;

[0582] The measured value change information includes at least one of the following: an indication of an increase or decrease in the measured value, the amount of change in the measured value, the percentage change in the measured value, an indication of whether the amount of change in the measured value is greater than a first threshold, and the number of times or duration that the amount of change in the measured value is greater than the first threshold.

[0583] The measured values ​​include at least one of the following: signal strength information, signal quality information, interference signal strength information, timing advance (TA) information, and link quality information.

[0584] In some embodiments of this application, the historical transmission-related information includes at least one of the following: historical service or load information, historical transmission performance parameters, historical modulation and coding information, historical link quality information, historical beam information, transmission time of the most recent downlink data, transmission time of the most recent uplink data, transmission time of the most recent SDT data, and the number of times SDT data was sent within a historical time period.

[0585] In some embodiments of this application, the beam configuration includes at least one of the following: the number of beams transmitted or received by the terminal, the beam direction, and the association between the beams or the beams and the reference signals corresponding to the beams; the number of beams transmitted or received by the network-side device, the beam direction, and the association between the beams or the beams and the reference signals corresponding to the beams; one or a set of reference signal indices; one or a set of beam indices; one or a set of beam direction information; and one or a set of Transmission Configuration Indicator (TCI) status information.

[0586] In some embodiments of this application, the location-related information includes at least one of the following: geographical location information of the terminal, network-side device or satellite device, geographical location change information of the terminal, network-side device or satellite device, cell reference location or movement trajectory of the terminal in a non-terrestrial network scenario, ephemeris information of the network-side device, and distribution information of the terminal.

[0587] In some embodiments of this application, the motion-related information includes at least one of the following: motion direction or motion direction change information of the network-side device; motion direction or motion direction change information of the terminal; motion speed or motion speed change information of the network-side device; and motion speed or motion speed change information of the terminal.

[0588] In some embodiments of this application, the model inference triggering conditions for using the first AI model for inference include at least one of the following: obtaining a first configuration; periodic triggering; receiving activation information for activating the first AI model inference; inference-related timer timeout; before the terminal enters an idle or inactive state; the duration of no data transmission is greater than or equal to a first duration; the cumulative amount of data to be transmitted within a time period is greater than a second threshold; the cumulative amount of data to be transmitted within a time period is less than or equal to the second threshold; the RRC state transition-related timer timeout; the remaining duration of the RRC state transition-related timer is less than or equal to a third threshold; after receiving an RRC release instruction from the upper layer; receiving at least a portion of the information in the first information; within a time period, the measurement value or change in measurement value of the first object meets the conditions; and the presence of a target type of service.

[0589] In some embodiments of this application, the inference stopping conditions for using the first AI model include at least one of the following: the terminal enters an idle state, an inactive state, a disconnected state, a power-saving state, or a connected state; the inference-related timer times out; the terminal receives an RRC release message; the random access channel (RACH) transmission is triggered; the SDT transmission is triggered; the predicted measurement value of the first object inferred by the first AI model is greater than or equal to a fourth threshold; or the predicted measurement value of the first object inferred by the first AI model is less than or equal to a fifth threshold.

[0590] In some embodiments of this application, the processing module is further configured to, after acquiring the second information, perform a first operation based on the second information, the first operation including at least one of the following: sending third information to a second device, the second device being a device for training the first AI model, the third information being used for fine-tuning or updating the first AI model; reverting to a non-AI SDT transmission determination process; triggering a switching or updating of the AI ​​model; triggering a change processing of the AI ​​model input information; triggering AI model training; triggering fine-tuning processing of the AI ​​model; triggering AI model supervision; not configuring or reserving the first SDT resource in advance; configuring or reserving only the second SDT resource; not configuring SDT-related fields in the system message block SIB1; not configuring SDT-related fields in the RRC release message; performing SDT transmission at the first resource location; stopping measurement for at least one time period; stopping measurement on the reference signal; stopping measurement on the reference signal for at least one time period; increasing the measurement period; performing continuous measurement for at least one time period; and performing measurement only on at least one beam or the reference signal corresponding to the at least one beam.

[0591] In some embodiments of this application, the processing module is specifically used to perform the first operation based on the second information when a first condition is met; wherein the first condition includes at least one of the following: the time for reasoning using the first AI model exceeds K time periods, and the second information still does not meet the performance requirements; reasoning using the first AI model fails; reasoning using the first AI model succeeds; the number of times reasoning using the first AI model reaches U times, where U is a positive integer.

[0592] In some embodiments of this application, the processing module is further configured to train an AI model based on the fourth information to obtain the first AI model;

[0593] The fourth piece of information includes at least one of the following: training dataset related to SDT transmission, label set related to SDT transmission, optimization algorithm or index for AI model training, loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0594] In some embodiments of this application, at least some information items in the training dataset match information items in the first information.

[0595] In some embodiments of this application, the tag set includes at least one of the following: an indication of whether to perform SDT transmission; an indication of whether to perform a first type of SDT transmission; the type of SDT transmission; a time point or time period for performing SDT transmission; a time point or time period for performing a first type of SDT transmission; a time point or time period for not performing SDT transmission; a time point or time period for not performing a first type of SDT transmission; and resources for SDT transmission.

[0596] In some embodiments of this application, the processing module is further configured to perform model supervision on the first AI model based on the supervision configuration for supervising the first AI model. The supervision configuration includes at least one of the following: the identifier of the first AI model; the model supervision period; the number of model supervisions; the model supervision duration; model supervision window related information; the model supervision triggering condition; the model supervision label; and the model supervision index.

[0597] In some embodiments of this application, the number of model supervision times includes at least one of the following: the number of times the first AI model needs to be supervised within each model supervision cycle; and the total number of times the first AI model needs to be supervised after receiving the supervision configuration.

[0598] The model supervision duration includes at least one of the following: the duration required for each model supervision of the first AI model within each model supervision cycle; and the effective duration of the supervision configuration.

[0599] The model supervision window information includes at least one of the following: the actual duration of model supervision or the number of samples for which model supervision is performed within each model supervision cycle; the duration of the window for which model supervision is required within each model supervision cycle, and the start or end time point.

[0600] The model supervision metrics include at least one of the following: the error between the predicted value and the true value of the first AI model, and the performance metrics of the communication system.

[0601] In some embodiments of this application, the model supervision triggering conditions include at least one of the following: no SDT transmission configuration is available when performing SDT transmission based on the inference result of the first AI model; SDT transmission fails based on the inference result of the first AI model; the inference error of the first AI model is greater than or equal to a sixth threshold; SDT resources are configured or reserved, but SDT transmission is not actually performed; the model supervision index does not meet the requirements; the model supervision timer of the first AI model times out; the first AI model fails to infer; the first AI model fails to infer S times (S is a positive integer); the number of inference failures of the first AI model reaches a seventh threshold; inference is performed using the first AI model; the probability of the first set of output information appearing is greater than or equal to an eighth threshold; the probability of the first set of output information appearing is less than a ninth threshold; the error of the second set of output information is greater than or equal to a tenth threshold.

[0602] In some embodiments of this application, the terminal has AI-related capabilities for determining SDT transmission based on AI assistance; wherein, the AI-related capabilities include at least one of the following: having the ability to support the training of the first AI model; having the ability to support inference using the first AI model; having the ability to support sending at least one type of auxiliary information; wherein, the auxiliary information is used for the training or inference of the first AI model.

[0603] In some embodiments of this application, the AI-related capabilities are determined by at least one of the following methods: the type of the terminal; indication by one or more reference signals; or capability information of the AI-related capabilities carried by fifth information.

[0604] The fifth piece of information includes at least one of the following: uplink control information, RRC signaling, and AI-related interface messages between the terminal and the terminal, network-side equipment, or satellite equipment.

[0605] In the terminal provided in this application embodiment, since a first AI model can be used, and the input information required for model reasoning based on the first AI model can be used to obtain second information, the second information can be used to determine whether SDT transmission needs to be triggered and to determine at least one of SDT transmission resources. Therefore, when the second information determines that SDT transmission needs to be triggered or SDT resources are determined, the terminal and the network-side device can configure SDT resources or perform SDT measurements, thereby avoiding resource saving.

[0606] In one embodiment, terminal 900 can be the second device described above:

[0607] Processor 910 is used to train an AI model based on fourth information to obtain a first AI model, and the first AI model is used for uplink transmission parameter inference.

[0608] The fourth piece of information includes at least one of the following: training dataset related to SDT transmission, label set related to SDT transmission, optimization algorithm or index for AI model training, loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

[0609] In some embodiments of this application, the training dataset includes at least one of the following information: the first information includes at least one of the following: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device or satellite device, panel orientation information of the terminal, network-side device or satellite device, device information of the terminal, location-related information of the terminal, network-side device or satellite device, motion-related information of the terminal, network-side device or satellite device, perception information of the terminal, network-side device or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information;

[0610] The first object includes any one of the following: beam, reference signal, TRP currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and BWP of the terminal.

[0611] In some embodiments of this application, the tag set includes at least one of the following information: an indication of whether SDT transmission is performed; an indication of whether a first type of SDT transmission is performed; the type of SDT transmission; a time point or time period for performing SDT transmission; a time point or time period for performing the first type of SDT transmission; a time point or time period for not performing SDT transmission; a time point or time period for not performing the first type of SDT transmission; and resources for SDT transmission.

[0612] In some embodiments of this application, the first type includes at least one of the following: RA-SDT, CG-SDT, paging SDT, RACH-less DG-SDT, paging-based downlink SDT, and paging-direct-scheduled downlink SDT.

[0613] In some embodiments of this application, after obtaining the first AI model, the method further includes:

[0614] Based on the supervision configuration used for supervising the first AI model, model supervision is performed on the first AI model. The supervision configuration includes at least one of the following: the identifier of the first AI model; the model supervision period; the number of model supervisions; the model supervision duration; model supervision window related information; model supervision trigger conditions; model supervision labels; and model supervision metrics.

[0615] In some embodiments of this application, the number of model supervision sessions includes at least one of the following: the number of times the first AI model needs to be supervised within each model supervision cycle; the total number of times the first AI model needs to be supervised after receiving the supervision configuration; wherein the model supervision duration includes at least one of the following:

[0616] The duration required for each model supervision of the first AI model within each model supervision cycle;

[0617] The duration of the monitoring configuration in effect;

[0618] The model supervision window related information includes at least one of the following:

[0619] Within each model supervision cycle, the actual duration of model supervision or the number of samples for which model supervision is performed;

[0620] Within each model supervision cycle, the duration of the window for performing model supervision, and the start or end time point;

[0621] The model supervision metrics include at least one of the following: the error between the predicted value and the true value of the first AI model, and the performance metrics of the communication system.

[0622] In some embodiments of this application, the model supervision triggering conditions include at least one of the following: no transmission configuration corresponding to the SDT is available during SDT transmission based on the inference result of the first AI model; SDT transmission based on the inference result of the first AI model fails; the inference error of the first AI model is greater than or equal to a sixth threshold; SDT resources are configured or reserved, but SDT transmission is not actually performed; the model supervision index does not meet the requirements; the model supervision timer of the first AI model times out; the first AI model inference fails; the first AI model inference fails S times, where S is a positive integer; the number of inference failures of the first AI model reaches a seventh threshold; inference is performed using the first AI model; the probability of the first set of output information appearing is greater than or equal to an eighth threshold; the probability of the first set of output information appearing is less than a ninth threshold; the error of the second set of output information is greater than or equal to a tenth threshold.

[0623] In some embodiments of this application, the terminal has AI-related capabilities for determining SDT transmission based on AI assistance;

[0624] The AI-related capabilities include at least one of the following: the ability to support the training of the first AI model; the ability to support inference using the first AI model; and the ability to support sending at least one type of auxiliary information; wherein the auxiliary information is used for the training or inference of the first AI model.

[0625] In some embodiments of this application, the AI-related capabilities are determined by at least one of the following methods: the type of the first device;

[0626] The AI-related capabilities are indicated by one or more reference signals; the AI-related capabilities are carried by the fifth information; wherein the fifth information includes at least one of the following: uplink control information, RRC signaling, and AI-related interface messages between the first device and the terminal, network-side device, or AI-related server.

[0627] In the terminal provided in this application embodiment, since the first AI model can be obtained by training the AI ​​model based on at least one of the training dataset related to SDT transmission, the tag set related to SDT transmission, the optimization algorithm or index for AI model training, the loss function for AI model training, the reward information for AI model training, and the triggering condition for AI model training, the accuracy of model training can be improved, thereby improving the inference performance of the first AI model.

[0628] It is understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description of the second device in the method embodiment and achieve the same or corresponding technical effects. To avoid repetition, it will not be described again here.

[0629] This application also provides a network-side device, including a processor and a communication interface. The communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps of the method embodiment shown in FIG4 or FIG5. This network-side device embodiment corresponds to the above-described network-side device method embodiment. All implementation processes and methods of the above-described method embodiments can be applied to this network-side device embodiment and can achieve the same technical effect.

[0630] Specifically, this application also provides a network-side device. As shown in FIG10, the network-side device 3000 includes a processor 3001, a network interface 3002, and a memory 3003. The network-side device may be the SDT determination device shown in FIG6 or the model training device shown in FIG7. The network interface 3002 is, for example, a common public radio interface (CPRI).

[0631] Specifically, the network-side device 3000 in this application embodiment further includes: instructions or programs stored in memory 3003 and executable on processor 3001. Processor 3001 calls the instructions or programs in memory 3003 to execute the methods executed by the modules shown in FIG6 or FIG7 and achieve the same technical effect. To avoid repetition, it will not be described in detail here.

[0632] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described SDT determination method embodiment or model training method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0633] The processor mentioned above is the processor in the terminal described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk. In some examples, the readable storage medium may be a non-transient readable storage medium.

[0634] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described SDT determination method embodiment or model training method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0635] 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.

[0636] This application also provides a computer program / program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the above-described SDT determination method embodiment or model training method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0637] This application also provides a communication system, including: a terminal and a network-side device, wherein the terminal can be used to perform the steps of the SDT determination method or the model determination method described above, and the network-side device can be used to perform the steps of the SDT determination method or the model determination method described above.

[0638] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0639] From the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of computer software products plus necessary general-purpose hardware platforms, and of course, they can also be implemented by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes several instructions to cause the terminal or network-side device to execute the methods described in the various embodiments of this application.

[0640] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other implementations under the guidance of this application without departing from the spirit and scope of the claims. All of these implementations are within the protection scope of this application.

Claims

1. A method for determining Small Data Transmission Time (SDT), comprising: The first device uses a first artificial intelligence (AI) model to obtain second information based on first information. The first information is the input information required for the first AI model to perform model inference. The second information is used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources.

2. The method according to claim 1, wherein, The second information includes at least one of the following: SDT transmission prediction results, connected state transmission indication, resource configuration of triggered SDT transmission, type of triggered SDT transmission, terminal state prediction results, uplink UL data volume prediction results, downlink measurement volume prediction results, candidate beams for SDT transmission and reference signals corresponding to the candidate beams. The terminal state prediction result is used to indicate whether the terminal enters a first state or the probability of entering the first state within at least one time period.

3. The method according to claim 2, wherein, The SDT transmission prediction result is used to indicate at least one of the following: At least within a time period, whether SDT transmission is performed, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission. Whether SDT transmission is performed at least at one time point, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission. The time period or time point during which no SDT transmission or SDT transmission of the first type is performed, and the time period or time point during which SDT transmission or SDT transmission of the first type is performed, wherein the first type includes at least one of the following: Random Access RA-SDT, Configuration Authorization CG-SDT, Paging SDT, Dynamic Authorization DG-SDT without Random Access RACH-less, Paging-based Downlink SDT, and Paging Direct Scheduling Downlink SDT.

4. The method according to any one of claims 1 to 3, wherein, The first information includes at least one of the following: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device or satellite device, panel orientation information of the terminal, network-side device or satellite device, device information of the terminal, location-related information of the terminal, network-side device or satellite device, motion-related information of the terminal, network-side device or satellite device, perception information of the terminal, network-side device or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information; The first object includes any one of the following: beam, reference signal, transmit / receive point (TRP) currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and bandwidth portion (BWP) of the terminal.

5. The method according to claim 4, wherein, The measurement-related information includes at least one of the following: current measurement value, measurement value change information within a time period, and historical measurement values; The measured value change information includes at least one of the following: an indication of an increase or decrease in the measured value, the amount of change in the measured value, the percentage change in the measured value, an indication of whether the amount of change in the measured value is greater than a first threshold, and the number of times or duration that the amount of change in the measured value is greater than the first threshold. The measured values ​​include at least one of the following: signal strength information, signal quality information, interference signal strength information, timing advance (TA) information, and link quality information.

6. The method according to claim 4 or 5, wherein, The historical transmission-related information includes at least one of the following: historical service or load information, historical transmission performance parameters, historical modulation and coding information, historical link quality information, historical beam information, transmission time of the most recent downlink data, transmission time of the most recent uplink data, number of downlink data transmissions within a historical time period, number of uplink data transmissions within a historical time period, most recent transmission time of SDT data, most recent transmission time of SDT data in disconnected state, number of SDT data transmissions within a historical time period, and number of SDT data transmissions in disconnected state within a historical time period.

7. The method according to any one of claims 4 to 6, wherein, The beam configuration includes at least one of the following: The number of beams transmitted or received by the terminal, the direction of the beams, and the correlation between the beams or the beams and the reference signals corresponding to the beams; The number of beams transmitted or received by network-side devices, beam direction, and the correlation between beams or beams and their corresponding reference signals; One or a set of reference signal indices; One or a group of beam indices; One or a group of beam direction information; One or a set of transport configuration indicators (TCI) status information.

8. The method according to any one of claims 4 to 7, wherein, The location-related information includes at least one of the following: geographical location information of the terminal, network-side equipment or satellite equipment; geographical location change information of the terminal, network-side equipment or satellite equipment; cell reference location or movement trajectory of the terminal in non-terrestrial network scenarios; ephemeris information of satellite equipment; and distribution information of the terminal.

9. The method according to any one of claims 4 to 8, wherein, The motion-related information includes at least one of the following: Information on the direction of movement or changes in the direction of movement of network-side devices; Information on the terminal's direction of motion or changes in direction of motion; Information on the direction of motion or changes in the direction of motion of satellite equipment; Information on the movement speed or changes in movement speed of network-side devices; Information on the terminal's movement speed or changes in movement speed; Information on the speed of satellite equipment or changes in its speed.

10. The method according to any one of claims 1 to 9, wherein, The model inference triggering conditions for using the first AI model include at least one of the following: Obtain the first configuration; Periodic triggering; Activation information for activating the inference of the first AI model is received; Inference-related timers timed out; Before the terminal enters the idle or inactive state; The duration of no data transmission is greater than or equal to the first duration; The cumulative amount of data to be transmitted within a certain time period exceeds the second threshold; The cumulative amount of data to be transmitted within a certain time period is less than or equal to the second threshold; RRC state transition related timer timeout; The remaining duration of the RRC state transition related timer is less than or equal to the third threshold; After receiving the RRC release instruction from the upper layer; Receive at least a portion of the information in the first message; Within a certain time period, the measured value or change in the measured value of the first object satisfies the condition; There are target types of business.

11. The method according to any one of claims 1 to 10, wherein, The reasoning stopping condition for using the first AI model includes at least one of the following: The terminal enters an idle state, an inactive state, a disconnected state, a power-saving state, or a connected state; Inference-related timers timed out; After the terminal receives the RRC release message; After triggering RACH transmission on the random access channel; After triggering SDT transmission; The predicted measurement value of the first object inferred by the first AI model is greater than or equal to the fourth threshold; The predicted measurement of the first object inferred by the first AI model is less than or equal to the fifth threshold.

12. The method according to any one of claims 1 to 11, wherein, After obtaining the second information, the method further includes: The first device performs a first operation based on the second information, the first operation including at least one of the following: Send a third message to a second device, which is the device used to train the first AI model, and the third message is used for fine-tuning or updating the first AI model; Revert to the non-AI SDT transmission determination process; Triggering the switching or updating of the AI ​​model; Trigger changes to the input information of the AI ​​model; Trigger AI model training; Triggering fine-tuning of the AI ​​model; Trigger AI model supervision; Without pre-configuring or reserving the first SDT resources; Configure or reserve only the second SDT resources; Do not configure SDT-related fields in system message block SIB1; Do not configure SDT-related fields in the RRC release message; SDT transmission is performed at the first resource location; Measurements must be stopped for at least one time period; Stop the measurement on the reference signal; The measurement on the reference signal is stopped for at least one time period; The measurement period becomes longer; Measurements were taken continuously over at least one time period; Measurements are performed only on at least one beam or on a reference signal corresponding to the at least one beam.

13. The method according to claim 12, wherein, The first operation based on the second information includes: If the first condition is met, the first device performs the first operation based on the second information; The first condition includes at least one of the following: The reasoning time using the first AI model exceeds K time intervals, and the second information still does not meet the performance requirements. Inference using the first AI model failed; Inference was successful using the first AI model; The number of times the first AI model is used for inference reaches U, where U is a positive integer.

14. The method according to any one of claims 1 to 13, wherein, The method further includes: The first device trains an AI model based on the fourth information to obtain the first AI model; The fourth piece of information includes at least one of the following: training dataset related to SDT transmission, label set related to SDT transmission, optimization algorithm or index for AI model training, loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

15. The method according to claim 14, wherein, At least some of the information items in the training dataset match the information items in the first information.

16. The method according to claim 14 or 15, wherein, The tag set includes at least one of the following: Indication regarding whether to perform SDT transmission; Indication regarding whether to perform a Type 1 SDT transmission; SDT transmission type; The time point or time period for SDT transmission; The time point or time period during which the first type of SDT transmission is performed; The time points or periods during which SDT transmission is not performed; The time points or time periods during which Type I SDT transmission is not performed; Resources transmitted via SDT.

17. The method according to any one of claims 1 to 16, wherein, The method further includes: The first device performs model supervision on the first AI model based on a supervision configuration for supervising the first AI model. The supervision configuration includes at least one of the following: the identifier of the first AI model; the model supervision period; the number of model supervisions; the model supervision duration; model supervision window related information; model supervision trigger conditions; model supervision labels; and model supervision metrics.

18. The method according to claim 17, wherein, The model supervision triggering condition includes at least one of the following: When performing SDT transmission based on the inference results of the first AI model, there is no transmission configuration corresponding to the SDT. SDT transmission based on the inference results of the first AI model failed; The inference error of the first AI model is greater than or equal to the sixth threshold; SDT resources were configured or reserved, but no SDT transmission was actually performed. If the model supervision index fails to meet the requirements; The model supervision timer for the first AI model timed out; After the first AI model's inference fails; After the first AI model fails inference S times, S is a positive integer; The number of times the first AI model failed inference reached the seventh threshold; Inference was performed using the first AI model; The probability of the first set of output information appearing is greater than or equal to the eighth threshold; The probability of the first set of output information appearing is less than the ninth threshold; The error of the second set of output information is greater than or equal to the tenth threshold.

19. The method according to any one of claims 1 to 18, wherein, The first device is any one of the following: a terminal, a network-side device, or an AI-related server.

20. The method according to any one of claims 1 to 19, wherein, The first device has AI-related capabilities for determining SDT transmission with AI assistance; The AI-related capabilities include at least one of the following: It has the capability to support the training of the first AI model; It has the ability to support reasoning using the first AI model; It has the ability to send at least one type of auxiliary information; The auxiliary information is used for training or inference of the first AI model.

21. The method according to claim 20, wherein, The AI-related capabilities are determined through at least one of the following methods: The type of the first device; Indicated by one or more reference signals; The fifth piece of information carries the capability information related to the AI; The fifth piece of information includes at least one of the following: uplink control information, RRC signaling, and AI-related interface messages between the first device and the terminal, network-side device, or satellite device.

22. A model training method, the method comprising: The second device trains an AI model based on the fourth information to obtain a first AI model, which is used for uplink transmission parameter inference. The fourth piece of information includes at least one of the following: training dataset related to SDT transmission, label set related to SDT transmission, optimization algorithm or index for AI model training, loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

23. The method according to claim 22, wherein, The training dataset includes at least one of the following: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device or satellite device, panel orientation information of the terminal, network-side device or satellite device, device information of the terminal, location-related information of the terminal, network-side device or satellite device, motion-related information of the terminal, network-side device or satellite device, perception information of the terminal, network-side device or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information; The first object includes any one of the following: beam, reference signal, TRP currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and BWP of the terminal.

24. The method according to claim 22 or 23, wherein, The tag set includes at least one of the following: Indication regarding whether to perform SDT transmission; Indication regarding whether to perform a Type 1 SDT transmission; SDT transmission type; The time point or time period for SDT transmission; The time point or time period during which the first type of SDT transmission is performed; The time points or periods during which SDT transmission is not performed; The time point or time period during which the first type of SDT transmission is not performed; Resources transmitted via SDT.

25. The method according to claim 24, wherein, The first type includes at least one of the following: RA-SDT, CG-SDT, paging SDT, RACH-less DG-SDT, paging-based downlink SDT, and paging-direct-scheduled downlink SDT.

26. The method according to any one of claims 22 to 25, wherein, The second device is any one of the following: a terminal, a network-side device, or an AI-related server.

27. The method according to any one of claims 22 to 26, wherein, The second device has AI-related capabilities for determining SDT transmission with AI assistance; The AI-related capabilities include at least one of the following: It has the capability to support the training of the first AI model; It has the ability to support reasoning using the first AI model; It has the ability to send at least one type of auxiliary information; The auxiliary information is used for training or inference of the first AI model.

28. The method according to claim 27, wherein, The AI-related capabilities are determined through at least one of the following methods: The type of the second device; Indicated by one or more reference signals; The fifth piece of information carries the capability information related to the AI; The fifth piece of information includes at least one of the following: uplink control information, RRC signaling, and AI-related interface messages between the first device and the terminal, network-side device, or AI-related server.

29. A small data transmission SDT determination apparatus, the apparatus comprising: Processing module; The processing module is used to use a first AI model to obtain second information based on first information, wherein the first information is the input information required for the first AI model to perform model inference; The second information is used to determine at least one of the following: whether SDT transmission is triggered, and SDT transmission resources.

30. The apparatus according to claim 29, wherein, The second information includes at least one of the following: SDT transmission prediction results, connected state transmission indication, resource configuration of triggered SDT transmission, type of triggered SDT transmission, terminal state prediction results, uplink UL data volume prediction results, downlink measurement volume prediction results, candidate beams for SDT transmission and reference signals corresponding to the candidate beams. The terminal state prediction result is used to indicate whether the terminal enters a first state or the probability of entering the first state within at least one time period.

31. The apparatus according to claim 30, wherein, The SDT transmission prediction result is used to indicate at least one of the following: At least within a time period, whether SDT transmission is performed, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission. Whether SDT transmission is performed at least at one time point, whether type 1 SDT transmission is performed, the probability of performing SDT transmission, the probability of performing type 1 SDT transmission, the probability of not performing SDT transmission, or the probability of not performing type 1 SDT transmission. The time period or time point during which SDT transmission or the first type of SDT transmission is not performed, and the time period or time point during which SDT transmission or the first type of SDT transmission is performed, wherein the first type includes at least one of the following: RA-SDT, Configuration Authorization CG-SDT, Paging SDT, Dynamic Authorization DG-SDT without Random Access RACH-less, Paging-based Downlink SDT, and Paging Direct Scheduling Downlink SDT.

32. The apparatus according to any one of claims 29 to 31, wherein, The first information includes at least one of the following: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device or satellite device, panel orientation information of the terminal, network-side device or satellite device, device information of the terminal, location-related information of the terminal, network-side device or satellite device, motion-related information of the terminal, network-side device or satellite device, perception information of the terminal, network-side device or satellite device, weather information or weather change information, security-related information of data or services, and historical transmission-related information; The first object includes any one of the following: beam, reference signal, transmit / receive point (TRP) currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and bandwidth portion (BWP) of the terminal.

33. The apparatus according to any one of claims 29 to 32, wherein, The processing module is further configured to, after acquiring the second information, perform a first operation based on the second information, wherein the first operation includes at least one of the following: Send a third message to a second device, which is the device used to train the first AI model, and the third message is used for fine-tuning or updating the first AI model; Revert to the non-AI SDT transmission determination process; Triggering the switching or updating of the AI ​​model; Trigger changes to the input information of the AI ​​model; Trigger AI model training; Triggering fine-tuning of the AI ​​model; Trigger AI model supervision; Without pre-configuring or reserving the first SDT resources; Configure or reserve only the second SDT resources; Do not configure SDT-related fields in system message block SIB1; Do not configure SDT-related fields in the RRC release message; SDT transmission is performed at the first resource location; Measurements must be stopped for at least one time period; Stop the measurement on the reference signal; The measurement on the reference signal is stopped for at least one time period; The measurement period becomes longer; Measurements were taken continuously over at least one time period; Measurements are performed only on at least one beam or on a reference signal corresponding to the at least one beam.

34. A model training apparatus, the apparatus comprising a processing module; The processing module is used to train an AI model based on the fourth information to obtain a first AI model, and the first AI model is used for uplink transmission parameter inference. in, The fourth piece of information includes at least one of the following: training dataset related to SDT transmission, label set related to SDT transmission, optimization algorithm or index for AI model training, loss function for AI model training, reward information for AI model training, and triggering conditions for AI model training.

35. The apparatus according to claim 34, wherein, The training dataset includes at least one of the following information: measurement-related information of at least one first object, distance or path loss information between the terminal and the network-side device, frequency domain resource information, reference signal configuration, beam configuration, network identifier, transmission power information of the terminal, network-side device or satellite device, panel orientation information of the terminal, network-side device or satellite device, device information of the terminal, location-related information of the terminal, network-side device or satellite device, motion-related information of the terminal, network-side device or satellite device, perception information of the terminal, network-side device or satellite device, weather information or weather change information, data or service security-related information, and historical transmission-related information. The first object includes any one of the following: beam, reference signal, TRP currently accessed by the terminal, TRP currently camped by the terminal, cell currently accessed by the terminal, cell currently camped by the terminal, frequency band of the terminal, carrier of the terminal, subband of the terminal, and BWP of the terminal.

36. The apparatus according to claim 34 or 35, wherein, The tag set includes at least one of the following: Indication regarding whether to perform SDT transmission; Indication regarding whether to perform a Type 1 SDT transmission; SDT transmission type; The time point or time period for SDT transmission; The time point or time period during which the first type of SDT transmission is performed; The time points or periods during which SDT transmission is not performed; The time point or time period during which the first type of SDT transmission is not performed; Resources transmitted via SDT.

37. A communication device comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the SDT determination method as claimed in any one of claims 1 to 21, or implementing the steps of the model training method as claimed in any one of claims 22 to 28.

38. A readable storage medium storing a program or instructions that, when executed by a processor, implement the SDT determination method as described in any one of claims 1 to 21, or implement the steps of the model training method as described in any one of claims 22 to 28.

39. A computer program product stored in a storage medium, the computer program product being executed by at least one processor to implement the SDT determination method as claimed in any one of claims 1 to 21, or to implement the model training method as claimed in any one of claims 22 to 28.

40. A chip comprising a processor and a communication interface coupled to the processor, the processor being configured to run a program or instructions to implement the SDT determination method as described in any one of claims 1 to 21, or to implement the model training method as described in any one of claims 22 to 28.

Images