Communication method and apparatus, and device

Abstract

The present application relates to the field of communications, and discloses a communication method and apparatus, and a device. The communication method comprises: a first device acquires first information, wherein the first information is obtained on the basis of an artificial intelligence (AI) model, and the first information is used for a terminal to perform uplink transmission.

Description

Communication methods, devices and equipment

[0001] Cross-reference to related applications

[0002] This application claims priority to Chinese Patent Application No. 202411222417.2, filed on September 2, 2024, entitled "Communication Method, Apparatus and Device", 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 a communication method, apparatus, and device. Background Technology

[0004] Before performing uplink transmission, the user equipment (UE) needs to determine the uplink transmission parameters. For example, when performing random access channel (RACH), the UE needs to obtain the correct preamble, physical random access channel occasion (PRACH occasion, RO), and other relevant uplink transmission parameters.

[0005] Currently, network-side equipment pre-configures relevant uplink transmission parameters for terminals within the cell.

[0006] However, in real-world applications, terminals do not always need to communicate in every location within a cell, or the distribution of terminals within a cell is not necessarily uniform. This uneven distribution of users within a cell may lead to a higher probability of collisions in RACH resources under some beams, resulting in wasted RACH resources and low resource utilization. Summary of the Invention

[0007] This application provides a communication method, apparatus, and device that can solve the problem of low resource utilization that may result from existing uplink parameter determination schemes.

[0008] Firstly, a communication method is provided, the method comprising:

[0009] The first device acquires first information, wherein the first information is obtained based on an artificial intelligence (AI) model, and the first information is used by the terminal for uplink transmission.

[0010] Secondly, a communication device is provided, the device comprising:

[0011] The first acquisition module is used to acquire first information, wherein the first information is obtained based on an artificial intelligence (AI) model, and the first information is used by the terminal for uplink transmission.

[0012] Thirdly, a communication device is provided, the device being configured to perform the steps of the method described in the first aspect.

[0013] Fourthly, 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 aspect.

[0014] Fifthly, 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.

[0015] In a sixth aspect, a communication device is provided, including a processor and a communication interface, wherein the communication interface is used to acquire first information, wherein the first information is obtained based on an artificial intelligence (AI) model, and the first information is used by the terminal for uplink transmission.

[0016] In a seventh 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.

[0017] Eighthly, a chip is provided, the chip including a processor and a communication interface coupled to the processor, the processor being used to run programs or instructions to implement the method as described in the first aspect.

[0018] In a ninth aspect, a computer program / program product is provided, the computer program / program product being stored in a storage medium, the computer program / program product being executed by at least one processor to perform the steps of the method as described in the first aspect.

[0019] In this embodiment, since the first information for uplink transmission can be obtained from the artificial intelligence (AI) model, the uplink transmission parameters used by the terminal are matched with the actual application scenario, rather than being arbitrarily configured by the network-side device. Therefore, resource collisions and conflicts can be reduced, and resource waste can be avoided, thereby improving resource utilization. Attached Figure Description

[0020] Figure 1 is a block diagram of a wireless communication system applicable to an embodiment of this application.

[0021] Figure 2 is one of the schematic diagrams of the SSB to RO mapping results provided in the embodiments of this application.

[0022] Figure 3 is a second schematic diagram of the SSB to RO mapping result provided in the embodiments of this application.

[0023] Figure 4 is a flowchart illustrating a communication method proposed in an embodiment of this application.

[0024] Figure 5 is a schematic diagram of the structure of a communication device according to an embodiment of this application.

[0025] Figure 6 is a schematic diagram of the structure of a communication device proposed in an embodiment of this application.

[0026] Figure 7 is a schematic diagram of the structure of a terminal proposed in an embodiment of this application.

[0027] Figure 8 is a schematic diagram of the structure of a network-side device proposed in an embodiment of this application. Detailed Implementation

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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 New Radio (NR) systems 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.

[0032] 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 headphones, 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 (AS), or Wireless Fidelity (WiFi) nodes, etc.The term "base station" can be referred to as Node B (NB), Evolved Node B (eNB), Next Generation Node B (gNB), New Radio Node B (NR Node B), 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 Node B (HNB), Home Evolved Node B, Transmit / Receive Point (TRP), or any other suitable term in the relevant field, as long as the same technical effect is achieved. The term "base station" is not limited to any specific technical terminology. It should be noted that this application embodiment only uses a base station in an NR system as an example for description and does not limit the specific type of base station.

[0033] 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), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (or L-NEF), and Binding Support. The core network functions include: BSF (Block Network Function), 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 embodiment 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 embodiment changes in subsequent protocol versions (e.g., 6G), it will still be within the scope of protection of this application.

[0034] 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).

[0035] To better understand the technical solutions provided in this application, some technologies and concepts involved in this application will be explained or described below.

[0036] I. Random Access

[0037] Random Access Channel (RACH) refers to the process from when a terminal sends a random access preamble to attempt to access the network until a basic signaling connection is established with the network.

[0038] There are many purposes for random access. For example, random access triggered by the PDCCH order is mainly for the terminal to obtain uplink time synchronization. Another example is that when the terminal establishes the initial radio link, it can obtain the user identifier—the Cell-Radio Network Temporary Identifier (C-RNTI)—through the random access procedure.

[0039] Random access procedures include contention-based random access (CBRA) and contention-free random access (CFRA).

[0040] The CBRA process includes: 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 a Msg2 (Random Access Response (RAR)) message to the terminal. The RAR contains the number of the preamble detected by the network-side device and the uplink transmission resources allocated to the terminal for sending Msg3; after receiving Msg2, the terminal confirms that at least one of the preamble numbers carried in Msg2 matches the number or index of the preamble it sent, and then sends Msg3 containing contention resolution information according to the uplink transmission resources indicated by the RAR; after receiving Msg3, the network-side device sends Msg4 containing contention resolution information to the terminal; upon receiving Msg4, the terminal confirms that the contention resolution information is consistent with the one it sent in Msg3, thus completing the four-step random access process.

[0041] It should be further explained that in CBRA, the network-side device includes uplink grant (UL grant) information in the RAR, which indicates the scheduling information of the Physical Uplink Shared Channel (PUSCH) carrying Msg3, and includes information such as RAPID (RACH preamble ID), TC-RNTI, and uplink timing advance (TA). If the network-side device does not receive the Msg3 PUSCH, it can schedule a retransmission of the Msg3 PUSCH in the Physical Downlink Control Channel (PDCCH) scrambled with TC-RNTI. Since different terminals randomly select preambles for transmission in CBRA, different terminals may select the same preamble for transmission on the same time-frequency radio resources (RO resources), which can be understood as a preamble conflict between terminals.

[0042] In CFRA, the preamble is allocated by the network-side equipment. This preamble is called a dedicated random access preamble and is provided to the terminal by Radio Resource Control (RRC) signaling or PDCCH order signaling. Therefore, there is no preamble conflict. When dedicated random access preamble resources are insufficient, the network-side equipment notifies the terminal to initiate CBRA-based access.

[0043] 2. Preamble sequence

[0044] The PRACH signal consists of three parts: a cyclic prefix, one or more preamble sequences, and a guard time. The preamble supports four long preamble sequences of length 839 (formats 0 / 1 / 2 / 3) and nine short preamble sequences of length 139 (formats A1, A2, A3, B1, B2, B3, B4, C0, C1). The preamble sequence length is indicated by the higher-layer parameter prach-RootSequenceIndex. In frequency range 1 (FR1), long sequences and short sequences with subcarrier spacing of 15 kHz and 30 kHz are supported. In frequency range 2 (FR2), only short sequences with subcarrier spacing of 60 kHz and 120 kHz are supported.

[0045] III. Mapping Rules from SSB to RO

[0046] In related technologies, the configuration parameters of PRACH resources and SSB-RO are configured in System Information Block 1 (SIB1).

[0047] In NR, a cell can configure multiple Frequency Division Multiplexing (FDM) Physical Random Access Channel transmission occasions (PRACH Occasions, or ROs for short) at a single PRACH transmission time domain location. At any given time, the number of ROs that can be used for FDM can be {1, 2, 4, 8}, determined by the configuration of the higher-layer parameter msg1-FDM.

[0048] The random access preamble can only be transmitted on the time-domain resources configured by the parameter PRACHConfigurationIndex and the frequency-domain resources configured by the parameter msg1-FDM. PRACH frequency-domain resources n RA ∈{0,1,…,M-1}, where M equals the higher-layer parameter msg1-FDM. At initial access, the PRACH frequency domain resource n RA Starting with the lowest frequency RO resource within the initial active uplink bandwidth part, number them in ascending order; otherwise, use the PRACH frequency domain resource n. RAThe RO resources within the active uplink bandwidth part are numbered in ascending order, starting with the lowest frequency RO resource. For example, in Figure 2, the number of ROs in frequency division multiplexing at one time is 8 (msg1-FDM=8), and the RO resources are numbered sequentially from low to high frequency as RO#0 to RO#7.

[0049] In NR, there is an association between the Original Radio Signal (RO) and the actual transmitted synchronization signal / physical broadcast channel block (SS / PBCH block, SSB). ROs are associated with SSBs in the order of first frequency domain (from low to high frequency) and then time domain. One SSB may be associated with multiple consecutive ROs, or multiple SSBs may be associated with one RO (in which case, different SSBs correspond to different preambles). This is configured by the network-side equipment using the parameter ssb-perRACH-OccasionAndCB-PreamblesPerSSB.

[0050] After all SSBs have completed one round of association with RO, they constitute an SSB-RO mapping cycle. An SSB-RO association period may contain one or more SSB-RO mapping cycles. An SSB-RO association pattern period may contain one or more SSB-RO association periods. The SSB-RO mapping repeats with the association pattern period as the cycle, and the maximum association pattern period is 160ms.

[0051] Typically, network-side equipment can use different beams to transmit different SSBs. The number of SSBs is configured via the `ssb-PositionsInBurst` parameter; for FR2, the maximum number of SSBs is 64. The terminal selects the associated RO / "RO and preamble combination" of the SSB with the strongest signal based on the strength of the received downlink beam / SSB, and then transmits `Msg1`. In this way, the network-side equipment can determine the SSB selected by the terminal based on the received preamble RO / "RO and preamble combination". It then transmits `Msg2` on the corresponding downlink beam to ensure the quality of downlink signal reception.

[0052] Taking Figure 2 as an example, the number of ROs in frequency division multiplexing at a given time is 8, and the number of SSBs actually transmitted is 4, namely SSB#0, SSB#1, SSB#2, and SSB#3. Each SSB is associated with 2 ROs. If the UE determines to send PRACH / Msg1 on the RO corresponding to SSB#0, then the UE selects one RO from RO#0 and RO#1 to send PRACH.

[0053] Taking Figure 3 as an example, the number of ROs in frequency division multiplexing at a given time is 2, and the actual number of SSBs transmitted is 8, namely SSB#0, SSB#1, ..., SSB#7, with each pair of SSBs associated with one RO. When multiple SSBs share a single RO, the preamble sets associated with these multiple SSBs are different; that is, the same preamble cannot belong to different preamble sets associated with different SSBs simultaneously. Taking RO#0 in Figure 3 as an example, RO#0 has a total of 60 preambles, of which preambles with indices 0 to 29 are associated with SSB#0, and preambles with indices 30 to 59 are associated with SSB#1.

[0054] Before sending PRACH, the terminal first selects an SSB with an RSRP higher than a threshold based on the RSRP of the received beam. If multiple SSBs have RSRPs higher than the threshold, the terminal can select any SSB with an RSRP higher than the threshold. If there are no SSBs with RSRPs higher than the threshold, the terminal selects an SSB based on implementation.

[0055] Based on the network (NW) configuration, the terminal obtains the mapping between SSBs and ROs. After selecting an SSB, the RO corresponding to the selected SSB is used as the RO for transmitting PRACH / preamble / Msg1. If the selected SSB is associated with multiple ROs, the terminal can choose one of the ROs to transmit PRACH / preamble / Msg1.

[0056] For example, in the example shown in Figure 2, assuming the terminal selects SSB#1, the terminal can choose one of RO#2 and RO#3 for PRACH / Msg1 transmission. In the example shown in Figure 3, if the terminal selects SSB#1, it can choose the nearest available RO (RO#0 or 4) associated with SSB#1 for PRACH / Msg1 transmission. Within the selected RO, the terminal selects a preamble from the preamble set associated with the selected SSB for PRACH transmission. As shown in Figure 3, one RO is associated with two SSBs. Therefore, the available preamble set associated with each SSB in one RO is divided into two subsets, each corresponding to one SSB. The terminal will select a preamble sequence from the preamble subset corresponding to the selected SSB for PRACH / Msg1 transmission.

[0057] It should also be noted that the Synchronization Signal Block (SSB) or synchronization signal appearing in the embodiments of this application can also be any module that includes at least one of the following: synchronization signal, broadcast signal, Physical Broadcast Channel (PBCH), other system message downlink broadcast channel or its control channel.

[0058] In the embodiments of this application, the SSB to RO mapping can also refer to the generalized association between downlink signals and uplink signals / resources, such as the relationship between CSI-RS and RO.

[0059] In the embodiments of this application, the relevant uplink transmission parameters may include, but are not limited to, at least one of the following uplink transmission configuration parameters: Msg1, PRACH, preamble, MsgA, MsgA PRACH, MsgA PUSCH, Msg3 PUSCH, RACH based small data transmission (RA-SDT), configured grant PUSCH for SDT (CG PUSCH for SDT), configured grant PUSCH for RACH less handover, CG PUSCH for LTM (L1 / L2-triggered mobility) cell switching, PUCCH in RACH, and Sounding Reference Signal (SRS). For example, these relevant configuration parameters may include time-frequency domain resource-related parameters (such as the mapping ratio from SSB to RO, the mapping ratio from SSB to PUSCH resources, etc.) and power-related parameters (such as transmit power, target receive power value, maximum transmit power, power boost value, power back-off value, etc.).

[0060] IV. AI / ML

[0061] AI (Artificial Intelligence) 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.

[0062] In the embodiments of this application, the AI ​​model can be implemented in various ways, such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc., and the embodiments of this application do not specifically limit it.

[0063] In this application's embodiments, the AI ​​model may also be referred to as an AI unit, machine learning (ML) model, ML unit, AI structure, AI function, AI characteristic, machine learning model, neural network, neural network function, neural network functionality, etc. Alternatively, an AI model may refer to a processing unit capable of implementing specific algorithms, formulas, processing flows, capabilities, etc., related to AI. An AI model may also be a processing method, algorithm, function, module, or unit for a specific dataset. Furthermore, an AI model may be a processing method, algorithm, function, module, or unit running on AI / ML related hardware such as GPUs, NPUs, TPUs, and ASICs. This application does not impose specific limitations in these areas. The specific dataset may include the input and / or output of the AI ​​model.

[0064] In the embodiments of this application, the identifier of the AI ​​model can be an AI model identifier, an AI structure identifier, or an AI algorithm identifier. Alternatively, the identifier of the AI ​​model can be an identifier of a specific dataset associated with the AI ​​model. Alternatively, the identifier of the AI ​​model can be an identifier of a specific scenario, environment, channel characteristics, or device related to AI / MI. Alternatively, the identifier of the AI ​​model can be an identifier of a function, characteristic, capability, or module related to AI / ML. The embodiments of this application do not impose specific limitations on these aspects.

[0065] In this application embodiment, the server may refer to an entity used for training / prediction or providing AI-related information, or it may refer to services that bypass the operator or OTT (over the top).

[0066] In addition, during AI model training / machine learning, a label typically refers to the identification or annotation of the true category or target value of the sample data. Labels are used to represent the information that the AI ​​model should learn and predict.

[0067] In real-world deployment scenarios, terminals don't always need to communicate in every location within a cell, or their distribution within the cell isn't always uniform. For example, in office buildings or industrial parks, there are many users during the day, but only a few at night. Using the same RACH configuration at different times would result in unnecessary resource waste. Similarly, in Urban Macro (UMa) scenarios, the presence of buildings, roads, parks, etc., can lead to significant differences in user distribution across different areas / directions within the cell; there might be more users in one direction and fewer in another. Furthermore, on high-speed railways or highways, users typically switch from a specific beam direction in cell A to a specific beam direction in another cell, but traffic flow varies considerably at different times.

[0068] Uneven user distribution within a cell, whether in time or space, can increase the probability of RACH collisions in some beams, while wasting RACH resources in others. In existing methods, when a user initiates a RACH transmission, it randomly selects RACH resources that meet certain conditions. To ensure a low probability of RACH collisions, the network needs to allocate as many surplus RACH resources as possible. This requires reserving a significant amount of resources for RACH transmission, potentially leading to low resource utilization. Furthermore, if the terminal needs to maintain RACH transmission performance, it may need to use higher transmit power or select an appropriate beam through more accurate measurements, which could increase terminal power consumption and reduce energy efficiency.

[0069] To address at least one of the problems of low resource utilization and low terminal energy efficiency that may result from existing uplink parameter determination schemes, embodiments of this application propose a communication method that involves artificial intelligence (AI), specifically, uplink transmission parameters can be determined directly or indirectly based on AI.

[0070] The communication method provided in this application will be described in detail below with reference to the accompanying drawings and through some embodiments and application scenarios.

[0071] As shown in Figure 4, a communication method proposed in one embodiment of this application may include:

[0072] Step 401: The first device acquires first information, wherein the first information is obtained based on an artificial intelligence (AI) model, and the first information is used by the terminal for uplink transmission.

[0073] The first device can be at least one of a terminal, a network-side device, and a server.

[0074] The first information may include, but is not limited to, at least one of the following:

[0075] 1) User distribution information and / or load information of the community;

[0076] 2) Uplink transmission parameters.

[0077] The user distribution information and / or load information of the cell may include at least one of the following:

[0078] a. Third information;

[0079] b. The probability (or bias) of the occurrence of third information.

[0080] The third information includes at least one of the following:

[0081] The number of users under each beam direction, specifically, the number of users under each beam direction may include at least one of the following: the number of users stationed under each beam direction, the number of users accessing / switching over, and the number of users potentially triggering random access;

[0082] The number of users within the first area / coverage of the first beam, specifically, may include at least one of the following: the number of resident users within the first area / coverage of the first beam, the number of users accessing / switching over, and the number of users potentially triggering random access.

[0083] Traffic volume per beam;

[0084] Traffic volume within the first region / coverage area of ​​the first beam;

[0085] The probability of a terminal triggering random access transmission within the first area / coverage range of the first beam;

[0086] The probability of a terminal triggering random access transmission in each beam direction;

[0087] User types within the first area / first beam coverage area;

[0088] User power levels within the first area / first beam coverage area;

[0089] User type under each beam direction;

[0090] For each beam direction, the user power level may vary. For users with low power levels, it may be necessary to repetition access to a cell / transmission and receiving point (TRP) with a large coverage area.

[0091] In the embodiments of this application, the first region may be at least one specific region determined in advance, and the first beam may be at least one specific beam determined in advance.

[0092] The uplink transmission parameters may include at least one of the following:

[0093] a. At least one set of fourth information;

[0094] b. The probability (or bias) of the fourth piece of information appearing.

[0095] The fourth piece of information includes at least one of the following:

[0096] i. Configuration information for the Random Access Channel (RACH);

[0097] ii. Configuration information for the first Physical Uplink Shared Channel (PUSCH).

[0098] The configuration information of the RACH may include at least one of the following:

[0099] The number and / or index of at least one of the following associated with RACH: SSB, Channel State Information-Reference Signal (CSI-RS), and Transmission Configuration Indication (TCI);

[0100] The SSB-to-RO mapping ratio can optionally be predicted based on one or a set of SSB / CSI-RS / beam / TCI mapping ratios for RO / preamble mapping.

[0101] The number and / or indexes of SSBs associated with RO;

[0102] Number of ROs in the time domain;

[0103] The number of ROs in the frequency domain;

[0104] The number of ROs and / or indexes in the first SSB mapping;

[0105] Number and / or index of the preamble in the first SSB mapping;

[0106] The preamble format can optionally be predicted based on user distribution information and / or cell load information. For example, if the coverage area corresponding to SSB1 is large, a long format preamble is required, while if the coverage area corresponding to SSB2 is small, a short format preamble can be used.

[0107] The preamble-to-PUSCH mapping ratio;

[0108] The number and / or index of PUSCH resources in the first preamble mapping, for example, the number and / or index of PUSCH resources in a certain preamble mapping;

[0109] Configuration period of Physical Random Access Channel (PRACH);

[0110] The association cycle from SSB to PRACH resources;

[0111] SSB to PRACH resource association pattern cycle;

[0112] The mapping cycle from SSB to PRACH resources;

[0113] The determination period and / or time window used to determine the PRACH resource group for PRACH retransmission;

[0114] PRACH repeats the on state, wherein the on state includes being on or not being on;

[0115] The number of PRACH repetitions, for example, the number of PRACH repetitions predicted for each SSB / CSI-RS / beam / TCI;

[0116] PRACH repetition with the same beam or different beam;

[0117] The activation status of the PRACH resource, wherein the activation status includes activation or deactivation;

[0118] The trigger threshold for RACH associated with one or a group of SSBs / CSI-RS / beams / TCIs ​​can optionally be predicted based on the distribution and / or type of terminals. For example, if there are more Internet of Things (IoT) devices in the coverage area corresponding to SSB1, a lower RACH trigger threshold is required; if there are more mobile devices in the coverage area corresponding to SSB2, a normal RACH trigger threshold can be used.

[0119] Trigger thresholds for one or more RACHs associated with RO / preamble / preamble formats, wherein the trigger thresholds for RACHs that can be used for different RO / preamble / preamble formats can be set according to at least one of the following information: terminal distribution, terminal access probability, and terminal type.

[0120] Preamble received target power;

[0121] PRACH maximum transmission power (P CMAX ).

[0122] In the embodiments of this application, the first SSB may be at least one specific SSB that has been predetermined, and the first preamble may be at least one specific preamble that has been predetermined.

[0123] The configuration information of the first PUSCH includes at least one of the following:

[0124] The number and / or index of at least one of the SSB, CSI-RS, beam and TCI associated with the first PUSCH;

[0125] The SSB to PUSCH resource mapping ratio can optionally be predicted based on one or a group of SSBs / CSI-RS / beams / TCIs.

[0126] The number and / or indexes of SSBs associated with the first PUSCH resource;

[0127] The number of first PUSCH resources in the time and / or frequency domains;

[0128] The number and / or index of the first PUSCH resource mapped by the first SSB;

[0129] The cycle of CG PUSCH;

[0130] The association period from SSB to the first PUSCH resource;

[0131] The association pattern cycle from SSB to the first PUSCH resource;

[0132] The mapping cycle from SSB to the first PUSCH resource;

[0133] The first PUSCH repeats the on state, wherein the on state includes being on or not being on;

[0134] Number of times the first PUSCH is repeated;

[0135] The first PUSCH repetition uses the same or different beams.

[0136] The activation state of the first PUSCH resource, wherein the activation state includes activation or deactivation;

[0137] The target received power for the first PUSCH transmission;

[0138] Maximum transmission power of the first PUSCH;

[0139] Trigger threshold for one or more SSB / CSI-RS / beam / TCI associated message MsgA PUSCH;

[0140] The mapping ratio from MsgA PRACH / RO / preamble to MsgA PUSCH can optionally be predicted based on one or a set of RO / preamble / preamble formats.

[0141] The number and / or index of ROs / preambles associated with the first PUSCH resource;

[0142] The SSB or SSB group associated with the mapping from MsgA preamble to MsgA PUSCH;

[0143] The time-frequency domain resources of the MsgA PUSCH mapped by the first SSB or the first SSB group.

[0144] Wherein, the first PUSCH includes at least one of the following:

[0145] MsgA PUSCH;

[0146] Msg3 PUSCH;

[0147] PUSCH scheduled by downlink control information (DCI) scrambled with Temporary Cell Radio Network Temporary Identifier (TC-RNTI);

[0148] Configure grant PUSCH (configured grant PUSCH, CG PUSCH);

[0149] PUSCH in the RACH transmission process based on small data transmission;

[0150] PUSCH during RACH-free handover;

[0151] PUSCH during the LTM process;

[0152] CG PUSCH in SDT;

[0153] CG PUSCH during RACH-free handover;

[0154] CG PUSCH during LTM cell handover.

[0155] Optionally, the uplink transmission parameters are applicable to at least one of the following:

[0156] All beam directions / All synchronization signal blocks SSB / All channel status information reference signals CSI-RS;

[0157] A specific beam direction or group of beams / SSB / CSI-RS.

[0158] In other words, the aforementioned uplink transmission parameters can be parameters uniformly configured for all beam directions / SSB / CSI-RS, or they can be parameters individually configured for a specific beam direction / SSB / CSI-RS.

[0159] Optionally, the uplink transmission parameters can be understood as optimal or suitable uplink transmission parameters. Specifically, the uplink transmission parameters satisfy at least one of the following conditions:

[0160] 1) The usage rate of RACH resources is greater than the first threshold. For example, when uplink transmission parameters are used for uplink transmission, greater than or equal to x% of RACH resources are used.

[0161] 2) The RACH collision probability is lower than the second threshold. For example, when using the uplink transmission parameters for uplink transmission, the RACH collision probability is lower than y% for a period of time.

[0162] 3) The probability of access interruption is lower than the third threshold. For example, when using the aforementioned uplink transmission parameters for uplink transmission, the probability of access interruption is lower than z% for a period of time.

[0163] 4) Match the distribution of users or traffic within the cell. For example, if there are more interruptions in beam direction 1, more random access resources are allocated, and if there are fewer interruptions in beam direction 2, fewer random access resources are allocated accordingly.

[0164] 5) Match the probability of users initiating random access within the cell. For example, terminals in beam direction 1 have a higher probability of initiating random access and are allocated more random access resources, while terminals in beam direction 2 have a lower probability of initiating random access and are allocated fewer random access resources.

[0165] In some embodiments, the first information may be obtained by the first device itself based on an AI model. That is, step 401 may specifically include: the first device obtaining the first information based on the AI ​​model.

[0166] The first device (the execution entity that runs the AI ​​model for inference) may include at least one of the following:

[0167] 1) Terminal

[0168] The advantage of terminal inference is that the terminal can directly perform uplink transmission based on the inference results.

[0169] 2) Network-side equipment

[0170] The network-side equipment may include, but is not limited to, at least one of base stations, TRPs, and core network equipment (such as core network equipment specifically used for model training).

[0171] The base station can be at least one of the following: the base station where the terminal is currently camped, the base station is currently accessing, the base station is switching targets, and the base station is reselecting targets.

[0172] The advantage of network-side inference is that it can reduce the complexity and cost of the terminal, as well as reduce the power consumption of the terminal.

[0173] 3) Server

[0174] The server may be a device specifically used for training the AI ​​model, performing inference using the AI ​​model, and providing at least one of the following: training the AI ​​model, performing inference using the AI ​​model, and providing AI-related information; or, the server may be a third-party server; or, the server may be a device provided by an over-the-top (OTT) service provider, a third-party service provider, or the Internet.

[0175] It is understandable that using dedicated servers for AI inference can improve AI performance while reducing the complexity and cost of network-side devices and terminals.

[0176] Furthermore, if the first device includes network-side devices and / or servers, and the first information includes user distribution information and / or load information of the cell, then the method shown in Figure 4 may further include:

[0177] The first device determines the uplink transmission parameters based on the first information;

[0178] The first device sends the uplink transmission parameters to the terminal.

[0179] Furthermore, if the first device includes network-side devices and / or servers, and the first information includes uplink transmission parameters, the method shown in Figure 4 may further include:

[0180] The first device sends the uplink transmission parameters to the terminal.

[0181] In some embodiments, the first device may transmit the uplink transmission parameters to the terminal via at least one of the following signals / channels:

[0182] Paging messages, for example, network-side devices can use paging messages to indicate uplink transmission parameters inferred by the AI ​​model to the terminal. After receiving the paging message, the terminal can perform RACH transmission based on the uplink transmission parameters inferred by the AI ​​model.

[0183] Paging DCI;

[0184] Short message;

[0185] SSB;

[0186] System message;

[0187] Msg2;

[0188] Msg4;

[0189] MsgB;

[0190] Media Access Control Element (MAC CE) messages / RRC messages / Non-Access Stratum (NAS) messages.

[0191] In some detailed embodiments, the first device obtaining the first information based on the AI ​​model may include: the first device inputting second information into the AI ​​model to obtain the first information.

[0192] At least a portion of the second information is provided by the second device. Specifically:

[0193] In the first sub-implementation, if the first device is a network-side device, then the second device includes the terminal. That is, if the network-side device inputs the second information into the AI ​​model to obtain the first information, then at least a portion of the second information is reported by the terminal to the network-side device.

[0194] The signal by which the terminal reports at least a portion of the second information to the first device may include at least one of the following:

[0195] MAC CE message;

[0196] RRC message;

[0197] NAS message;

[0198] User face data;

[0199] MSG 1;

[0200] MSG A;

[0201] MSG 3;

[0202] Physical Uplink Control Channel (PUCCH);

[0203] PUSCH;

[0204] PRACH;

[0205] SRS;

[0206] Other upward reference signals.

[0207] Optionally, if the first device is a network-side device, before the first device obtains the first information based on the AI ​​model, the method further includes: the network-side device obtaining the AI ​​model from a terminal / other network-side device / server.

[0208] In the second sub-implementation, if the first device is a server, then the second device includes the terminal and / or network-side device. That is, if the server inputs the second information into the AI ​​model to obtain the first information, then at least a portion of the second information is indicated to the server by the terminal and / or network-side device. For example, this can be indicated via OTT messages.

[0209] In the third sub-implementation, if the first device is the terminal, then the second device includes a network-side device. That is, if the terminal inputs the second information into the AI ​​model to obtain the first information, then at least a portion of the second information is indicated to the terminal by the network-side device.

[0210] The signal by which the network-side device indicates at least a portion of the second information to the first device may include at least one of the following:

[0211] MAC CE message;

[0212] RRC message;

[0213] NAS message;

[0214] User face data;

[0215] DCI;

[0216] SIB;

[0217] Physical Downlink Control Channel (PDCCH);

[0218] Physical Downlink Shared Channel (PDSCH);

[0219] Msg2;

[0220] Msg4;

[0221] MsgB.

[0222] Optionally, if the first device is the terminal, before the first device obtains the first information based on the AI ​​model, the method further includes: the first device obtaining the AI ​​model from a second device, wherein the second device includes a network-side device and / or a server.

[0223] In other embodiments, the first information may be obtained by the second device based on the AI ​​model and then sent to the first device. That is, step 401 may specifically include: the first device receiving the first information sent by the second device, wherein the first information is obtained by the second device based on the AI ​​model.

[0224] In some detailed embodiments, the first information is obtained by the second device inputting the second information into the AI ​​model. Specifically:

[0225] If the first device includes network-side devices and / or servers, then the second device is the terminal. In other words, after the terminal uses an AI model to deduce the uplink transmission parameters, it can report the uplink transmission parameters to the network-side devices and / or servers.

[0226] If the first device is a terminal, then the second device includes a network-side device and / or a server. That is, after obtaining the first information based on the AI ​​model, the network-side device and / or the server send the first information to the terminal.

[0227] Optionally, at least part of the second information may be provided by the first device. The specific method of providing the information can be referred to in the three sub-implementations above, and will not be described again.

[0228] In some embodiments, the second information used as input to the AI ​​model during inference may include at least one of the following:

[0229] 1) Receive signal strength information.

[0230] The received signal strength information may include the received strength information of uplink and / or downlink signals between the terminal and the network node, such as at least one of Reference Signal Receiving Power (RSRP), Reference Signal Received Quality (RSRQ), and Received Signal Strength Indication (RSSI).

[0231] Specifically, the received signal strength information may include at least one of the following:

[0232] The measured values ​​of at least one of RSRP, RSRQ, and RSSI corresponding to one or a group of downlink broadcast signals / synchronization signals / reference signals (such as SSB, CSI-RS, TRS, etc.); Optionally, if these measured values ​​are input into the AI ​​model by a terminal for inference, then these measured values ​​are obtained by the terminal; if these measured values ​​are input into the AI ​​model by a network-side device for inference, then these measured values ​​are obtained by the network-side device.

[0233] Measurements of at least one of RSRP, RSRQ, and RSSI corresponding to one or a set of uplink reference signals (such as SRS); optionally, if these measurements are input into the AI ​​model for inference by a network-side device, then these measurements are obtained by the network-side device.

[0234] 2) Received signal quality information.

[0235] The received signal quality information may include received signal quality information of uplink and / or downlink signals between the terminal and the network node, such as at least one of signal-to-noise ratio (SNR), signal-to-interference-noise ratio (SINR), and latency.

[0236] 3) Path loss information.

[0237] The path loss information may include the path loss between the terminal and the network node. It may also be the path loss between network nodes of two cells, such as the path loss between the source cell and the target cell, or the path loss between the primary cell (Pcell) and the secondary cell (Scell).

[0238] 4) Reference signal related information.

[0239] The reference signal related information may include at least one of the following: reference signal pattern, reference signal time-domain position or index, reference signal frequency-domain position or index, and reference signal transmission power.

[0240] The reference signal may include at least one of the following for measuring received signal strength information such as RSRP: Primary Synchronization Signal (PSS), Secondary Synchronization Signal (SSS), CSI-RS, and Demodulation Reference Signal (DMRS).

[0241] 5) Communication distance information, such as the distance between the terminal and the network node.

[0242] 6) Absolutely timely advance (TA) information.

[0243] The absolute TA information may include one or more absolute TA values ​​used recently, for example, the absolute TA information may include the most recent absolute TA value between the terminal and the network node.

[0244] 7) Relative TA information.

[0245] The relative TA information may include one or more relative TA values ​​that are currently used, within a certain period of time, or recently used. For example, the relative TA information may include the most recent relative TA value between the terminal and the network node.

[0246] 8) TRP identifier or TRP group identifier.

[0247] 9) Community or community group identification.

[0248] 10) Tracking area marking.

[0249] Specifically, the tracking area identifier may include the identifier of the tracking area where the terminal is located.

[0250] 11) Radio Access Network (RAN) Notification Area Identifier, for example, the RAN notification area identifier can be the RAN identifier where the terminal is located.

[0251] 12) At least one of the following frequency domain information: frequency band information, frequency zone information, frequency point information, carrier frequency information, frequency layer information, and bandwidth part (BWP).

[0252] For example, the second information may specifically include at least one of the following: the frequency band, frequency point, carrier frequency, frequency layer, and BWP in which the cell actually operates.

[0253] 13) One / a set of reference signal indices.

[0254] 14) One or a group of beam indices.

[0255] 15) One or a group of beam directions.

[0256] For example, it may specifically include at least one of the reference signal index, beam index, and beam direction actually transmitted between the terminal and the network node of the cell.

[0257] 16) Transmission power related information, such as transmission power related information of the base station or TRP.

[0258] 17) Propagation delay.

[0259] 18) Round Trip Time (RTT).

[0260] For example, the first piece of information may specifically include the propagation delay or RTT between the terminal and the network node.

[0261] 19) Load information corresponding to the synchronization signal block SSB.

[0262] 20) Load information corresponding to the beam.

[0263] 21) Load information corresponding to TCI.

[0264] 22) Send the load information corresponding to the receiving point TRP.

[0265] 23) Interference information corresponding to the synchronization signal block SSB.

[0266] 24) Interference information corresponding to the beam.

[0267] 25) Interference information corresponding to TCI.

[0268] 26) Send the interference information corresponding to the receiving point TRP.

[0269] 27) Load information for one or more cells.

[0270] 28) Load information corresponding to one or more carriers.

[0271] 29) Load information corresponding to one or more frequency points.

[0272] In items 19) to 22) and 27) to 29), the load information may include at least one of the traffic volume and the number of users.

[0273] 30) Interference information corresponding to one or more cells.

[0274] 31) Interference information corresponding to one or more carriers.

[0275] 32) Interference information corresponding to at least one of the one or more frequency points.

[0276] In items 23) to 26) and 30) to 32), the interference information may include interference measurements.

[0277] 33) Geographic location information of the base station or TRP.

[0278] 34) Terminal location information.

[0279] 35) Terminal distribution information.

[0280] The location information may be specific geographic coordinates (such as GPS coordinates), or approximate location range information of the terminal (such as range information such as which street it is located on), or location information of the terminal relative to at least one of the following: the cell it is staying in, the cell it is accessing, a certain TRP, and a group of TRPs. For example, the location information may include the terminal being due east of the cell it is staying in.

[0281] 36) The direction and / or speed of movement of the terminal.

[0282] The direction of movement can be an absolute direction, such as 40 degrees east of south, or a relative direction, such as the direction relative to a certain base station or reference node.

[0283] 37) The direction and / or speed of movement of network-side devices.

[0284] The network-side device can be a network node or a satellite; the direction of movement can be an absolute direction, such as 40 degrees east of south, or a relative direction, such as the direction relative to a certain base station or reference node.

[0285] 38) Terminal energy consumption information.

[0286] 39) Terminal power information.

[0287] 40) Terminal type, such as power class.

[0288] 41) Operator information supported by the terminal.

[0289] 42) Information on network types supported by the terminal.

[0290] 43) Antenna orientation information of the terminal.

[0291] 44) Antenna orientation information of network-side devices.

[0292] The network-side device can be a network node or a satellite.

[0293] 45) Network structure information, or network scene information, for example, the network scene information may include inH, Uma, RMa, etc., or the network structure information may include homogeneous or heterogeneous networks (whether there is overlapping coverage).

[0294] 46) Historical movement trajectory information of terminals within the community.

[0295] 47) Multipath information of the channel, such as the first path or strongest path information of the terminal and different base stations / TRPs.

[0296] 48) Environmental information, such as weather information.

[0297] 49) Time information.

[0298] The time information can be specific and accurate to a particular time. For example, the time information can be accurate to a specific moment, such as 13:25:38. The time information can also be a time range, such as 13:00 to 14:00, morning / afternoon, daytime / nighttime.

[0299] The time information may be timing information obtained through other radio access technologies (RATs). The RAT may be Bluetooth, Wi-Fi, or 3G, 4G, or 5G, etc.

[0300] 50) Perceive information.

[0301] Specifically, the perceived information may include at least one of the following: communication environment information, scene information, channel state information, and the number of terminals. The channel state information here may include at least one of the following: line-of-sight (LOS), non-line-of-sight (NLOS), and obstruction information.

[0302] For example, the perceived information may specifically include:

[0303] To detect whether there are obstacles and / or the number of obstacles under a specific beam;

[0304] The number of terminals or devices covered by a specific beam can be sensed.

[0305] 51) Reference location or movement trajectory of a cell in a non-terrestrial network (NTN) scenario.

[0306] For example, in the case of a Low-Earth Orbit (LEO) satellite, the cell on Earth moves as the satellite moves.

[0307] For example, in the case of a Geosynchronous Earth Orbit (GEO) satellite, as the satellite moves, the cell on Earth is fixed, and a reference position can be considered to exist.

[0308] 52) Ephemeris information in NTN scenarios.

[0309] In some embodiments, the triggering conditions for the first device and / or the second device to perform inference using the AI ​​model may include at least one of the following:

[0310] 1) The fourth condition is met.

[0311] In other words, the AI ​​model's inference is supported by default if the fourth condition is met.

[0312] The fourth condition includes specific conditions such as having the ability to run the AI ​​model and / or being in the first usage scenario. For example, the first usage scenario can be a high-speed scenario, where the terminal is moving at high speed with a car on a highway, or the terminal is moving at high speed with a high-speed train.

[0313] 2) After the timer corresponding to the AI ​​model times out.

[0314] For example, a specific timer started for inference by the AI ​​model times out.

[0315] 3) After the terminal loses uplink synchronization.

[0316] 4) After the terminal loses downlink synchronization.

[0317] 5) After the terminal fails to access the random terminal for more than the first time period.

[0318] 6) After the terminal triggers random access for a second duration and still fails to successfully access the network.

[0319] 7) The number of times the terminal failed to access the system randomly exceeded the fourth threshold.

[0320] 8) After the terminal fails to access the network N times in a row.

[0321] 9) When the terminal meets the conditions / events for cell handover.

[0322] 10) The first device receives at least one of the second pieces of information, wherein the second pieces of information are inputs to the AI ​​model when it performs inference.

[0323] 11) Before the terminal triggers random access.

[0324] 12) Before the terminal triggers the initial access.

[0325] 13) The terminal detected a beam failure.

[0326] 14) The terminal detected a wireless link failure.

[0327] 15) The number of beam failure instances (BFI) exceeds the fifth threshold.

[0328] 16) BFI-related timers timed out.

[0329] 17) The terminal triggered a backoff indicator (BI).

[0330] 18) When the terminal receives a preamble sent by another terminal or that is not sent by itself.

[0331] For example, the preamble ID carried in the Msg2 received by the terminal is not the preamble ID it sent itself.

[0332] 19) The number of preambles that the terminal receives from other terminals or that are not sent by itself exceeds the sixth threshold.

[0333] 20) The terminal is configured with a 2-step RACH transmission.

[0334] Among them, 2-step RACH refers to 2-step RACH, also called 2-step PRACH or Type-2 RACH.

[0335] 21) The synchronization reference signal received power (SS-RSRP) of at least one SSB configured for configured grant Small Data Transmission (CG-SDT) is higher than the seventh threshold.

[0336] 22) TA invalid, that is, when TA is not valid.

[0337] 23) Used to determine if the change in RSRP of TA exceeds the eighth threshold.

[0338] 24) The terminal enters the connection state, that is, it is triggered only after entering the connection state.

[0339] 25) The current RACH configuration / RACH resources of the terminal do not match the terminal distribution under the first beam.

[0340] 26) The probability that a terminal initiating random access under the first beam where the current RACH configuration / RACH resources of the terminal do not match.

[0341] 27) The current RACH configuration / RACH resource usage ratio of the terminal is lower than the ninth threshold.

[0342] 28) The usage rate of RACH resources corresponding to one or a group of SSBs is higher than the tenth threshold.

[0343] 29) The probability of conflict of RACH resources corresponding to one or a group of SSBs is higher than the eleventh threshold.

[0344] 30) The probability of conflict of one or a group of RACH resources is higher than the twelfth threshold.

[0345] 31) The probability of access interruption of the current node or the current cell is higher than the thirteenth threshold.

[0346] 32) The type of terminal accessed / resided has changed.

[0347] 33) The power type of the terminal has changed.

[0348] It should be noted that the triggering condition for the above AI model to perform inference can mean that AI inference is triggered as long as at least one of the above conditions is met; or it can mean that after at least one of the above conditions is met, it is necessary to further combine other indication information from the network-side device (such as activation / deactivation information) to determine whether the AI ​​model is triggered to perform inference.

[0349] In this embodiment of the application, using the AI ​​model for reasoning can also be regarded as using the AI ​​model for prediction. Specifically, after inputting the second information mentioned above into the AI ​​model, the output of the AI ​​model is used as the first information.

[0350] In some embodiments, the performance metrics of the AI ​​model performing inference need to meet certain conditions to improve the reliability of the model's inference results. The performance metrics and the corresponding conditions they meet include at least one of the following:

[0351] 1) The complexity of the AI ​​model;

[0352] For example, different complexity index requirements for AI models can be defined for base stations or terminals of different types / capabilities. For instance, for ordinary terminals, the complexity of the AI ​​model used must not exceed a specific value.

[0353] 2) The inference latency of the AI ​​model;

[0354] For example, the duration of inferring optimal uplink transmission parameters using the AI ​​model cannot exceed a specific value.

[0355] 3) The success rate of the AI ​​model's inference;

[0356] For example, the success rate of using the AI ​​model to infer the optimal uplink transmission parameters must not be less than or must be greater than a specific value.

[0357] 4) The reliability of the reasoning results given by the AI ​​model.

[0358] Specifically, the reliability can be defined as the probability of collisions during random access using uplink transmission parameters inferred from an AI model being less than a threshold value.

[0359] Optionally, before step 401, the communication method proposed in this application embodiment may further include:

[0360] Train the AI ​​model.

[0361] In other words, before using the AI ​​model for specific reasoning, it needs to be trained to generate an AI model with the corresponding functions. Furthermore, online training or model updates are necessary as needed to ensure the reliability and accuracy of the AI ​​model's reasoning.

[0362] Specifically, the AI ​​model can be trained based on the training dataset.

[0363] In some embodiments, the training dataset for the AI ​​model includes at least one of the following:

[0364] 1) Sample data, wherein the sample data includes second information, which is the input when training the AI ​​model;

[0365] 2) Labels (also called truth values) of the sample data;

[0366] 3) Basic algorithms and / or indexes of basic algorithms, such as the error back propagation (BP) algorithm, stochastic gradient descent algorithm, etc.

[0367] 4) Loss function (also called objective function);

[0368] 5) Reward information (also known as adjustment / feedback information).

[0369] The labels of the sample data include at least one of the following:

[0370] An indicator of whether the uplink transmission parameters are reasonable. For example, if the uplink transmission parameters corresponding to the sample data are reasonable, the indicator is 1, indicating that the sample data is a positive sample data; if the uplink transmission parameters corresponding to the sample data are unreasonable, the indicator is 0, indicating that the sample data is a negative sample data.

[0371] The number of uplink transmission parameters obtained;

[0372] Number of users within the first region / coverage of the first beam;

[0373] Traffic volume within the first region / coverage area of ​​the first beam;

[0374] The probability of a terminal initiating random access in the first area / first beam;

[0375] RACH usage ratio in the first region / first beam;

[0376] RACH collision probability in the first region / first beam;

[0377] The number of at least one of the following in the first region / first beam: RACH resources, transmission opportunities (RO), and preamble;

[0378] The number of PUSCH resources in the first region / first beam;

[0379] Access interruption probability in the first region / first beam;

[0380] Obtain at least one set of true values ​​for uplink transmission parameters;

[0381] At least one of the first pieces of information.

[0382] The first region can be a specific region determined in advance, and the first beam can be a specific beam determined in advance.

[0383] The reward information includes at least one of the following:

[0384] The number of times the system reverts to the first mode, which is a mode that does not use an AI model to determine uplink transmission parameters;

[0385] The number of times the AI ​​model inference failed and / or succeeded;

[0386] The difference between the inference results of the AI ​​model and the labels;

[0387] The collision probability of RACH.

[0388] In some embodiments, the AI ​​model is trained by at least one of the terminal, network-side device, and server. That is, the entity that trains the AI ​​model includes at least one of the following:

[0389] 1) Terminal

[0390] 2) Network-side equipment

[0391] The network-side equipment may include at least one of a base station, a TRP, and a core network device (such as a core network device specifically used for model training).

[0392] The base station can be one of the base stations where the current terminal is camped, accessing, switching the terminal target, or reselecting the terminal target.

[0393] 3) Server

[0394] The server can be a device specifically used for training or inference or providing AI-related information, or it can be a third-party server, or a device provided by an OTT service provider, a third-party service provider, or the Internet.

[0395] In some embodiments, the AI ​​model may be jointly trained by at least two of the terminal, network-side device, and server. Specifically, the training process of the AI ​​model includes at least one of the following:

[0396] The third device sends at least a portion of the output information during the training of the AI ​​model to the fourth device, and the fourth device uses the received output information as one of the input information for training the AI ​​model. The third device is one of the terminal, the network-side device, and the server, and the fourth device is the other of the terminal, the network-side device, and the server.

[0397] The third device and / or the fourth device deploy the AI ​​model online after training it offline, and then fine-tune the AI ​​model online.

[0398] For example:

[0399] 1) The terminal reports at least a portion of the output information of the AI ​​model during the training process to the network-side device and / or server, and the network-side device and / or server uses the information reported by the terminal as one of the input contents for training the AI ​​model.

[0400] 2) The network-side device sends at least a portion of the output information of the AI ​​model during training to the terminal and / or server, and the terminal and / or server uses the information sent by the network-side device as one of the input contents for training the AI ​​model.

[0401] 3) The terminal, network-side device or server performs online model training, and then the terminal, network-side device or server performs fine-tuning in the actual network.

[0402] In some embodiments, where the training device for the AI ​​model includes the terminal and / or the server, the method shown in FIG4 may further include:

[0403] The terminal and / or the server report auxiliary information to the network-side device, the auxiliary information being used by the network-side device to determine the AI ​​model for inferring the first information from multiple candidate AI models;

[0404] The terminal and / or the server receive instruction information sent by the network-side device, the instruction information being used to instruct the AI ​​model for reasoning the first information.

[0405] For example, when AI model training occurs, or at least partially occurs, on a terminal or server, the terminal or server may have multiple pre-trained AI models. The terminal or server can report auxiliary information to the network-side device, which then determines and instructs which AI model to use for AI inference. Accordingly, the terminal or server determines which AI model to use for model inference based on at least one of the following: the received AI model ID, the dataset ID, and data acquisition-related configuration information.

[0406] Among them, the multiple candidate AI models have at least one of the following differences:

[0407] Different training datasets;

[0408] Different sample data;

[0409] The labels on the sample data are different;

[0410] The output information is different.

[0411] The auxiliary information includes at least one of the following:

[0412] The identifiers of the multiple candidate AI models;

[0413] The classification labels of the training datasets of the multiple candidate AI models.

[0414] The indication information includes at least one of the following:

[0415] Identification of AI models;

[0416] Classification labels for the training dataset;

[0417] Configuration information related to data acquisition.

[0418] In other embodiments, where the training device for the AI ​​model includes the network-side device and / or the server, the method shown in FIG4 may further include:

[0419] The network-side device and / or the server send the trained AI model to the terminal so that the terminal can use the AI ​​model to perform reasoning on the first information.

[0420] In other words, AI model training occurs, or at least partially occurs, on network-side devices or servers. After the AI ​​model training is completed, the network-side devices or servers distribute the trained AI model to the terminal, which then uses the model for inference.

[0421] In some embodiments, the training triggering conditions for the AI ​​model include at least one of the following:

[0422] The first condition and / or the first event is triggered;

[0423] Triggered periodically;

[0424] Semi-static triggering.

[0425] In the first sub-implementation, the training of the AI ​​model is triggered by a first condition and / or a first event, wherein the first condition and / or the first event includes at least one of the following:

[0426] 1) The relevant timer of the AI ​​model times out, for example, the timer set for training the AI ​​model times out.

[0427] 2) Instructions from network-side devices, such as base stations sending model training instructions to terminals via TA MAC CE commands.

[0428] 3) The terminal accesses the first cell, switches to the first frequency point / first frequency band, and switches to the first operator / first public land mobile network (PLMN), wherein the first cell, the first frequency point / first frequency band, and the first operator / first public land mobile network can be a specific cell, a specific frequency point / specific frequency band, and a specific operator / specific public land mobile network that are determined in advance.

[0429] 4) After the AI ​​model inference fails.

[0430] 5) After the AI ​​model fails to reason N times consecutively.

[0431] 6) The number of inference failures of the AI ​​model reached the fourteenth threshold.

[0432] 7) After each time the AI ​​model is used for inference.

[0433] 8) The variation in the interference level of the first link exceeds the fifteenth threshold, where the first link can be a specific link that has been predetermined.

[0434] 9) The measured value of the signal reception quality of the terminal / network side device changes or the amount of change exceeds the sixteenth threshold, wherein the signal reception quality includes at least one of RSRP, RSRQ and RSSI.

[0435] 10) The timer related to model training timeout has expired for the third duration.

[0436] 11) There are N consecutive instances of model supervision or N instances of model supervision have occurred.

[0437] 12) Changes in the external environment, such as obtaining information about environmental changes through sensors.

[0438] 13) The network-side device is configured with uplink resources to fall back to the first mode, and the network-side device is configured to enable the second mode. The first mode is a mode that does not use the AI ​​model to determine the uplink transmission parameters, and the second mode is a mode that uses the AI ​​model to determine the uplink transmission parameters. For example, the second mode can be the AI ​​for RACH mode.

[0439] 14) In the primary usage scenario, such as in a high-speed scenario.

[0440] 15) Reselect the cell or TRP.

[0441] 16) Cell or TRP handover.

[0442] 17) The conditions or events for cell handover are met.

[0443] 18) The terminal triggers access to the secondary cell.

[0444] 19) Receive at least one of the second information, the second information being used to train the AI ​​model, the specific content of the second information being referred to above.

[0445] 20) The number of times BFI is triggered exceeds the seventeenth threshold;

[0446] 21) Whether BFI is triggered;

[0447] 22) When the terminal receives a preamble sent by another terminal or not sent by itself, for example, the terminal receives a preamble ID in Msg2 that is not sent by itself.

[0448] 23) The number of preambles received by the terminal from other terminals or that are not sent by itself exceeds the eighteenth threshold.

[0449] It should be noted that the triggering conditions for the above-mentioned AI model training can mean that AI training is triggered as long as at least one of the above-mentioned triggering conditions is met; or it can mean that after at least one of the above-mentioned triggering conditions is met, further consideration should be given to AI-related capabilities and other indications from network-side devices to determine whether to trigger AI model training.

[0450] In the second sub-implementation, the training of the AI ​​model is triggered periodically, wherein the configuration information for periodic triggering includes at least one of the following:

[0451] The starting point for periodic training, such as when the terminal receives the network configuration information and begins periodic model training;

[0452] Period length, for example, triggering at least one model training session at each preset interval;

[0453] The number of model training sessions and / or the training duration within a single period.

[0454] In the third sub-implementation, the training of the AI ​​model is semi-statically triggered, wherein the semi-static triggering includes at least one of the following:

[0455] The issuance and activation of semi-static configuration information are triggered by a second condition / second event, wherein the semi-static configuration information includes at least one of the following: the start time of model training, the cycle length, the number of model training sessions within a cycle, and / or the model training duration.

[0456] Semi-static configuration information is configured via RRC messages, and the AI ​​model is activated / deactivated via DCI or MAC-CE for semi-static training.

[0457] In some embodiments, the criteria for determining whether the AI ​​model training is complete include at least one of the following:

[0458] 1) The loss calculated by the loss function satisfies the third condition, such as the third condition being: the loss is less than a predefined threshold value;

[0459] Specifically, the loss function can be used to calculate at least one of the following:

[0460] The mean square error or normalized mean square error between the predicted and actual values;

[0461] The mean absolute error between the predicted and actual values.

[0462] 2) The model training iterations have reached the nineteenth threshold;

[0463] 3) The number of fine-tuning iterations reaches the twentieth threshold;

[0464] 4) At least one item in the label of the sample data is greater than or equal to or less than or equal to the threshold value;

[0465] 5) If the training time of the model is greater than or equal to the fourth training time, for example, if the training time of the AI ​​model is greater than or equal to the target training time, then the training is considered complete.

[0466] 6) If the energy consumption for model training is greater than or equal to the twenty-first threshold, for example, if the energy consumption for AI model training is greater than or equal to the target value, then training is considered complete.

[0467] In some embodiments, the behavior after obtaining the first information based on the AI ​​model includes at least one of the following:

[0468] 1) Collect at least one of the input and output information of this AI inference and provide it to the training device of the AI ​​model.

[0469] 2) Fall back to the first mode, which is a mode in which the uplink transmission parameters are not determined using the AI ​​model;

[0470] For example, reverting to the first mode may include at least one of the following:

[0471] Revert to the configuration information calculation and determination process of RACH without using an AI model (non-AI);

[0472] Revert to the configuration information calculation and determination process of MsgA PUSCH without using the AI ​​model;

[0473] The process of calculating and determining configuration information for at least one of the following is reverted to without using an AI model: CG PUSCH for SDT, RACH without handover, and cell handover under LTM mechanism.

[0474] 3) Trigger the switching of the AI ​​model, such as switching to another AI model.

[0475] 4) Trigger changes to the input information of the AI ​​model.

[0476] 5) Trigger at least one of the following: training, retraining, fine-tuning, and updating of the AI ​​model.

[0477] 6) Trigger supervision of the inference performance of the AI ​​model.

[0478] In some embodiments, the behavior after obtaining the first information based on the AI ​​model includes triggering supervision of the inference performance of the AI ​​model, and supervising the inference performance of the AI ​​model based on configuration information related to model supervision.

[0479] For example, when the inference environment differs significantly from the training environment, the TA inference performance based on the AI ​​model will become very poor, meaning the accuracy of the inferred TA-related information may be insufficient. Therefore, it is necessary to supervise the actual inference performance of AI-assisted TA-related information and trigger a series of adjustment measures based on the supervision results to ensure the reliability of the AI ​​model.

[0480] The configuration information related to model supervision includes at least one of the following:

[0481] 1) Identification of models that require inference performance supervision.

[0482] 2) The cycle of supervising model inference performance.

[0483] The model supervision period can be the interval between model supervision sessions, for example, performing model supervision once every day.

[0484] 3) The number of times the model inference performance is supervised.

[0485] The number of times model inference performance supervision is performed refers to the number of times the AI ​​model needs to be supervised within an inference performance supervision cycle. For example, the model needs to be supervised 10 times within one inference performance supervision cycle. Alternatively, the number of times model inference performance supervision is the total number of times the AI ​​model needs to be supervised after receiving the inference performance supervision configuration information.

[0486] 4) Duration of supervision for model inference performance.

[0487] The duration of model inference performance supervision refers to the duration of supervision required each time model inference performance supervision is performed within the inference performance supervision cycle. Alternatively, the duration of model inference performance supervision refers to the effective duration of the inference performance supervision configuration information, such as N model supervision cycles within the duration of model inference performance supervision, or model supervision only being performed within the duration of model inference performance supervision.

[0488] 5) Information related to the time window for supervising model inference performance.

[0489] The time window information related to model inference performance supervision refers to the actual duration of model inference performance supervision or the number of samples (times) of model inference performance supervision performed within a model inference performance supervision period. Alternatively, the time window information related to model inference performance supervision refers to at least one of the following: the duration, start time, and end time of the window in which model inference performance supervision needs to be performed within a model inference performance supervision period.

[0490] 6) Triggering conditions for model inference performance supervision.

[0491] 7) Labels for supervising model inference performance.

[0492] The labels for supervising model inference performance are the same as those for model training, or the labels for supervising model inference performance are different from those for model training.

[0493] 8) Supervised inference performance metrics, such as the error between AI model predictions and actual values, or communication system performance metrics; among which, communication system performance metrics may include at least one of transmission latency, throughput, and RACH / PUSCH resource collision probability.

[0494] It is understandable that the terminal / network testing equipment / server can determine when and / or how to supervise the model inference performance based on the above configuration information.

[0495] In some embodiments, the triggering conditions for the model inference performance supervision include at least one of the following:

[0496] 1) The probability of a collision when using the first information obtained by the AI ​​for random access exceeds a threshold;

[0497] 2) At least one of the inference performance metrics of the AI ​​model fails to meet the requirements;

[0498] 3) The timer related to the inference performance supervision of the AI ​​model timed out;

[0499] 4) The terminal switches to at least one of a new cell, a new TRP, and a new beam;

[0500] 5) The number of terminals switching to at least one of the following: a new cell, a new TRP, and a new beam exceeds the twenty-second threshold;

[0501] 6) The AI ​​model inference failed;

[0502] 7) The AI ​​model fails to infer N times consecutively;

[0503] 8) The number of inference failures of the AI ​​model reaches the twenty-third threshold;

[0504] 9) Each time the AI ​​model is used;

[0505] 10) The number of times BFI is triggered exceeds the twenty-fourth threshold;

[0506] 11) Whether BFI is triggered;

[0507] 12) The terminal receives a preamble sent by another terminal or that is not sent by itself. For example, in Msg2, it receives a preamble ID that is not sent by itself.

[0508] 13) The number of preambles received by the terminal from other terminals or that are not sent by itself exceeds the twenty-fifth threshold;

[0509] 14) The probability of the third and / or fourth information appearing is greater than or equal to or less than or equal to the twenty-sixth threshold;

[0510] 15) The deviation value of the third and / or fourth information exceeds the twenty-seventh threshold.

[0511] The third information includes at least one of the following:

[0512] Number of users in each beam direction;

[0513] Number of users within the first region / coverage of the first beam;

[0514] Traffic volume per beam;

[0515] Traffic volume within the first region / coverage area of ​​the first beam;

[0516] The probability of a terminal triggering random access transmission within the first area / coverage range of the first beam;

[0517] The probability of a terminal triggering random access transmission in each beam direction;

[0518] User types within the first area / first beam coverage area;

[0519] User power levels within the first area / first beam coverage area;

[0520] User type under each beam direction;

[0521] User power level in each beam direction.

[0522] The fourth piece of information includes at least one of the following (see above for details):

[0523] Configuration information for the Random Access Channel (RACH);

[0524] Configuration information for the first physical uplink shared channel (PUSCH).

[0525] In some embodiments, the execution device performing the relevant operations on the AI ​​model has relevant capabilities. These relevant operations include at least one of the following:

[0526] It has the ability to train the AI ​​model;

[0527] It has the ability to perform reasoning using the AI ​​model;

[0528] It has the ability to supervise the inference performance of the AI ​​model;

[0529] It has the ability to provide one or more auxiliary information, which is used for at least one of training the AI ​​model and performing inference using the AI ​​model.

[0530] In other words, before terminal / network-side devices / servers perform AI model training / inference / inference performance supervision, it is necessary to define the relevant capabilities of the terminal / network-side devices / servers and provide methods for determining these capabilities. Introducing a definition and determination mechanism for relevant capabilities helps each node flexibly implement corresponding features and facilitates feature implementation between nodes.

[0531] For example, one or more AI-related capabilities can be defined for AI-assisted determination of uplink transmission parameters:

[0532] 1) The terminal has the capability to support the training of the AI ​​model, including at least one of the following:

[0533] a. The ability to train AI models to determine the distribution of terminals or traffic within a cell;

[0534] b. The ability to train an AI model to determine the optimal uplink transmission parameters, such as training the optimal uplink transmission parameters for each beam direction.

[0535] 2) The terminal has the capability to support AI inference using the AI ​​model, including at least one of the following:

[0536] a. Use the AI ​​model to infer the UE distribution or traffic distribution within the cell;

[0537] b. Use the AI ​​model to infer the optimal uplink transmission parameters, such as training the optimal uplink transmission parameters for each beam direction.

[0538] 3) The terminal has the ability to report one or more auxiliary information for the training / inference of the AI ​​model.

[0539] 4) Network-side equipment (such as base stations, TRPs, or core network equipment) has the capability to support the training of the AI ​​model.

[0540] Force, including at least one of the following:

[0541] a. The ability to train AI models to determine UE distribution or traffic distribution within a cell;

[0542] b. The ability to train an AI model to determine the optimal uplink transmission parameters, such as training the optimal uplink transmission parameters for each beam direction.

[0543] 5) Network-side equipment (such as base stations, TRPs, or core network equipment) has the capability to support AI inference using the AI ​​model, including at least one of the following:

[0544] a. Use the AI ​​model to infer the UE distribution or traffic distribution within the cell;

[0545] b. Use the AI ​​model to infer the optimal uplink transmission parameters, such as training the optimal uplink transmission parameters for each beam direction.

[0546] 6) Network-side equipment (such as base stations, TRPs, or core network equipment) has the capability to support the indication of one or more auxiliary information for the training / inference of the AI ​​model.

[0547] 7) The server has the capability to support the training of the AI ​​model, including at least one of the following:

[0548] a. The ability to train AI models to determine UE distribution or traffic distribution within a cell;

[0549] b. The ability to train an AI model to determine the optimal uplink transmission parameters, such as training the optimal uplink transmission parameters for each beam direction.

[0550] a) The server has the capability to support AI inference using the AI ​​model, including at least one of the following:

[0551] a. Use the AI ​​model to infer the UE distribution or traffic distribution within the cell;

[0552] b. Use the AI ​​model to infer the optimal uplink transmission parameters, such as training the optimal uplink transmission parameters for each beam direction.

[0553] 8) The server has the ability to report one or more auxiliary information for the training / inference of the AI ​​model.

[0554] In some embodiments, the relevant capability is obtained based on at least one of the following:

[0555] 1) Terminal type, for example, different AI-related capabilities are introduced for different terminal types (such as Redcap, IoT), or the related capabilities are related to the terminal type, and the related capabilities are different for different terminal types;

[0556] The correspondence between the relevant capabilities and terminal types includes at least one of the following:

[0557] The input information used for model training / inference varies depending on the type of terminal. For example, less input information should be used for model training on terminal devices with weaker capabilities.

[0558] Different types of terminals use different labels for model training;

[0559] The execution devices for model training vary depending on the type of terminal. For example, for terminals with weaker capabilities, model training can be performed only on the network side device, or the terminal can only perform a small part of the joint model training (for example, model training involving user privacy data can be performed on the terminal).

[0560] The underlying algorithms used for model training differ depending on the type of terminal. For example, overly complex underlying algorithms may not be applicable to devices with weaker capabilities.

[0561] 2) Network type, for example, introducing different AI-related capabilities for different network types (such as NTN, TN);

[0562] 3) One or more reference signals indicate, for example, specific resources of PRACH (such as specific ROs or preambles) that indicate that the terminal has the ability to train the AI ​​model or to use the AI ​​model for inference;

[0563] 4) Uplink Control Information (UCI) is carried, such as physical layer control information (e.g., UCI information reported to the network for the terminal).

[0564] 5) RRC message carrying;

[0565] 6) The first interface message carried between the terminal and the server;

[0566] 7) The first interface message carried between the terminal and the network-side equipment;

[0567] 8) The first interface message carried between the server and the network-side device.

[0568] The first interface message may include messages related to the first AI model or messages related to all AI models.

[0569] The communication method proposed in this application can obtain the first information for uplink transmission based on the AI ​​model. This makes the uplink transmission parameters used by the terminal match the actual application scenario, rather than being arbitrarily configured by the network-side device. Therefore, it can reduce resource collisions and conflicts, avoid resource waste, improve resource utilization, enhance the reliability of uplink transmission, and reduce terminal power consumption.

[0570] More specifically, the communication method proposed in this application uses an AI model to predict user distribution or optimal uplink transmission parameters, ensuring reasonable uplink transmission parameter configuration in each beam direction, reducing resource collisions and conflicts, improving transmission reliability, reducing latency, and saving network or terminal energy consumption. For example, it ensures a reasonable SSB-to-RO mapping ratio in each beam direction, reducing RACH conflicts and improving the reliability of random access.

[0571] The following describes a communication method proposed in this application through two specific embodiments.

[0572] Example 1: A solution for terminal-based AI model inference (prediction)

[0573] 1) The terminal uses an AI model to predict RACH transmission resources or RACH transmission power information;

[0574] For example, the terminal can use AI models to predict the PRACH format, PRACH repetition count, and PRACH transmission power of this RACH transmission.

[0575] 2) The training of the AI ​​model can be performed on a terminal or network-side device;

[0576] If training is performed on the terminal side, the network-side device can notify the terminal of relevant secondary or auxiliary information, such as network load and terminal distribution information under each beam or SSB. The advantage of training the AI ​​model on the terminal side is that it can acquire more real-time information.

[0577] If training is performed on the network-side device, the terminal reports relevant secondary or auxiliary information to the network-side device. The advantage of training on the network-side device is that it can save terminal power consumption and reduce the complexity of terminal training.

[0578] Example 2: A solution for network-side devices to use AI models for inference (prediction)

[0579] Network-side equipment uses AI models to predict terminal distribution information / traffic volume information within the cell, and determines appropriate RACH resource configuration or RACH transmission power information based on the terminal distribution information / traffic volume information within the cell.

[0580] For example, network-side equipment can further determine the number of ROs, the PRACH format used, target-Rx power, and other uplink transmission parameters for each beam / SSB based on terminal distribution information / traffic volume information within the cell.

[0581] Alternatively, network-side devices can use AI models to predict RACH transmission resources for a group of terminals:

[0582] One group of terminals can be those that meet at least one of the following conditions: same beam, same associated RS, geographical proximity, and similar RSRP.

[0583] The advantage of training the model on the network side is that it can save the terminal's energy consumption and reduce the complexity of the terminal's prediction.

[0584] This application provides a communication method in which the executing entity can be a virtual device. This application uses a virtual device executing the communication method as an example to illustrate the communication device provided in this application.

[0585] As shown in Figure 5, one embodiment of this application proposes a communication device 500, which can be used in a first device. The device 500 may include: a first acquisition module 501, used to acquire first information, wherein the first information is obtained according to an artificial intelligence (AI) model, and the first information is used by the terminal for uplink transmission.

[0586] The first device can be at least one of a terminal, a network-side device, and a server.

[0587] The first information may include, but is not limited to, at least one of the following:

[0588] 1) User distribution information and / or load information of the community;

[0589] 2) Uplink transmission parameters.

[0590] The user distribution information and / or load information of the cell may include at least one of the following:

[0591] a. Third information;

[0592] b. The probability (or bias) of the occurrence of third information.

[0593] The third information includes at least one of the following:

[0594] The number of users under each beam direction, specifically, the number of users under each beam direction may include at least one of the following: the number of users stationed under each beam direction, the number of users accessing / switching over, and the number of users potentially triggering random access;

[0595] The number of users within the first area / coverage of the first beam, specifically, may include at least one of the following: the number of resident users within the first area / coverage of the first beam, the number of users accessing / switching over, and the number of users potentially triggering random access.

[0596] Traffic volume per beam;

[0597] Traffic volume within the first region / coverage area of ​​the first beam;

[0598] The probability of a terminal triggering random access transmission within the first area / coverage range of the first beam;

[0599] The probability of a terminal triggering random access transmission in each beam direction;

[0600] User types within the first area / first beam coverage area;

[0601] User power levels within the first area / first beam coverage area;

[0602] User type under each beam direction;

[0603] For each beam direction, the user power level may vary. For users with low power levels, it may be necessary to repetition access to a cell / transmission and receiving point (TRP) with a large coverage area.

[0604] In the embodiments of this application, the first region may be at least one specific region determined in advance, and the first beam may be at least one specific beam determined in advance.

[0605] The uplink transmission parameters may include at least one of the following:

[0606] a. At least one set of fourth information;

[0607] b. The probability (or bias) of the fourth piece of information appearing.

[0608] The fourth piece of information includes at least one of the following:

[0609] i. Configuration information for the Random Access Channel (RACH);

[0610] ii. Configuration information for the first Physical Uplink Shared Channel (PUSCH).

[0611] The configuration information of the RACH may include at least one of the following:

[0612] The number and / or index of at least one of the following associated with RACH: SSB, Channel State Information-Reference Signal (CSI-RS), and Transmission Configuration Indication (TCI);

[0613] The SSB-to-RO mapping ratio can optionally be predicted based on one or a set of SSB / CSI-RS / beam / TCI mapping ratios for RO / preamble mapping.

[0614] The number and / or indexes of SSBs associated with RO;

[0615] Number of ROs in the time domain;

[0616] The number of ROs in the frequency domain;

[0617] The number of ROs and / or indexes in the first SSB mapping;

[0618] Number and / or index of the preamble in the first SSB mapping;

[0619] The preamble format can optionally be predicted based on user distribution information and / or cell load information. For example, if the coverage area corresponding to SSB1 is large, a long format preamble is required, while if the coverage area corresponding to SSB2 is small, a short format preamble can be used.

[0620] The preamble-to-PUSCH mapping ratio;

[0621] The number and / or index of PUSCH resources in the first preamble mapping, for example, the number and / or index of PUSCH resources in a certain preamble mapping;

[0622] Configuration period of Physical Random Access Channel (PRACH);

[0623] The association cycle from SSB to PRACH resources;

[0624] SSB to PRACH resource association pattern cycle;

[0625] The mapping cycle from SSB to PRACH resources;

[0626] The determination period and / or time window used to determine the PRACH resource group for PRACH retransmission;

[0627] PRACH repeats the on state, wherein the on state includes being on or not being on;

[0628] The number of PRACH repetitions, for example, the number of PRACH repetitions predicted for each SSB / CSI-RS / beam / TCI;

[0629] PRACH repetition with the same beam or different beam;

[0630] The activation status of the PRACH resource, wherein the activation status includes activation or deactivation;

[0631] The trigger threshold for RACH associated with one or a group of SSBs / CSI-RS / beams / TCIs ​​can optionally be predicted based on the distribution and / or type of terminals. For example, if there are more Internet of Things (IoT) devices in the coverage area corresponding to SSB1, a lower RACH trigger threshold is required; if there are more mobile devices in the coverage area corresponding to SSB2, a normal RACH trigger threshold can be used.

[0632] Trigger thresholds for one or more RACHs associated with RO / preamble / preamble formats, wherein the trigger thresholds for RACHs that can be used for different RO / preamble / preamble formats can be set according to at least one of the following information: terminal distribution, terminal access probability, and terminal type.

[0633] Preamble received target power;

[0634] PRACH maximum transmission power (P CMAX ).

[0635] In the embodiments of this application, the first SSB may be at least one specific SSB that has been predetermined, and the first preamble may be at least one specific preamble that has been predetermined.

[0636] The configuration information of the first PUSCH includes at least one of the following:

[0637] The number and / or index of at least one of the SSB, CSI-RS, beam and TCI associated with the first PUSCH;

[0638] The SSB to PUSCH resource mapping ratio can optionally be predicted based on one or a group of SSBs / CSI-RS / beams / TCIs.

[0639] The number and / or indexes of SSBs associated with the first PUSCH resource;

[0640] The number of first PUSCH resources in the time and / or frequency domains;

[0641] The number and / or index of the first PUSCH resource mapped by the first SSB;

[0642] The cycle of CG PUSCH;

[0643] The association period from SSB to the first PUSCH resource;

[0644] The association pattern cycle from SSB to the first PUSCH resource;

[0645] The mapping cycle from SSB to the first PUSCH resource;

[0646] The first PUSCH repeats the on state, wherein the on state includes being on or not being on;

[0647] Number of times the first PUSCH is repeated;

[0648] The first PUSCH repetition uses the same or different beams.

[0649] The activation state of the first PUSCH resource, wherein the activation state includes activation or deactivation;

[0650] The target received power for the first PUSCH transmission;

[0651] Maximum transmission power of the first PUSCH;

[0652] Trigger threshold for one or more SSB / CSI-RS / beam / TCI associated message MsgA PUSCH;

[0653] The mapping ratio from MsgA PRACH / RO / preamble to MsgA PUSCH can optionally be predicted based on one or a set of RO / preamble / preamble formats.

[0654] The number and / or index of ROs / preambles associated with the first PUSCH resource;

[0655] The SSB or SSB group associated with the mapping from MsgA preamble to MsgA PUSCH;

[0656] The time-frequency domain resources of the MsgA PUSCH mapped by the first SSB or the first SSB group.

[0657] Wherein, the first PUSCH includes at least one of the following:

[0658] MsgA PUSCH;

[0659] Msg3 PUSCH;

[0660] PUSCH scheduled by downlink control information (DCI) scrambled with Temporary Cell Radio Network Temporary Identifier (TC-RNTI);

[0661] Configure grant PUSCH (configured grant PUSCH, CG PUSCH);

[0662] PUSCH in the RACH transmission process based on small data transmission;

[0663] PUSCH during RACH-free handover;

[0664] PUSCH during the LTM process;

[0665] CG PUSCH in SDT;

[0666] CG PUSCH during RACH-free handover;

[0667] CG PUSCH during LTM cell handover.

[0668] Optionally, the uplink transmission parameters are applicable to at least one of the following:

[0669] All beam directions / All synchronization signal blocks SSB / All channel status information reference signals CSI-RS;

[0670] A specific beam direction or group of beams / SSB / CSI-RS.

[0671] In other words, the aforementioned uplink transmission parameters can be parameters uniformly configured for all beam directions / SSB / CSI-RS, or they can be parameters individually configured for a specific beam direction / SSB / CSI-RS.

[0672] Optionally, the uplink transmission parameters can be understood as optimal or suitable uplink transmission parameters. Specifically, the uplink transmission parameters satisfy at least one of the following conditions:

[0673] 1) The usage rate of RACH resources is greater than the first threshold. For example, when uplink transmission parameters are used for uplink transmission, greater than or equal to x% of RACH resources are used.

[0674] 2) The RACH collision probability is lower than the second threshold. For example, when using the uplink transmission parameters for uplink transmission, the RACH collision probability is lower than y% for a period of time.

[0675] 3) The probability of access interruption is lower than the third threshold. For example, when using the aforementioned uplink transmission parameters for uplink transmission, the probability of access interruption is lower than z% for a period of time.

[0676] 4) Match the distribution of users or traffic within the cell. For example, if there are more interruptions in beam direction 1, more random access resources are allocated, and if there are fewer interruptions in beam direction 2, fewer random access resources are allocated accordingly.

[0677] 5) Match the probability of users initiating random access within the cell. For example, terminals in beam direction 1 have a higher probability of initiating random access and are allocated more random access resources, while terminals in beam direction 2 have a lower probability of initiating random access and are allocated fewer random access resources.

[0678] In some embodiments, the first information may be obtained by the first device itself based on an AI model. That is, the first acquisition module 501 may be specifically used to obtain the first information based on the AI ​​model.

[0679] Furthermore, if the first device includes network-side equipment and / or a server, and the first information includes user distribution information and / or load information of the cell, then the device 500 shown in FIG5 may further include:

[0680] The first determining module is used to determine the uplink transmission parameters based on the first information;

[0681] The first sending module is used to send the uplink transmission parameters to the terminal.

[0682] Furthermore, if the first device includes network-side devices and / or servers, and the first information includes uplink transmission parameters, then the apparatus 500 shown in FIG5 may further include:

[0683] The first sending module is used to send the uplink transmission parameters to the terminal.

[0684] In some embodiments, the first transmitting module may transmit the uplink transmission parameters to the terminal via at least one of the following signals / channels: paging message, paging DCI, short message, SSB, system message, Msg2, Msg4, MsgB, MAC CE message / RRC message / NAS message.

[0685] In some detailed embodiments, the first acquisition module 501 is specifically used to: input second information into the AI ​​model to obtain the first information. Wherein, at least a portion of the second information is provided by the second device. Specifically:

[0686] In the first sub-implementation, if the first device is a network-side device, then the second device includes the terminal. That is, if the network-side device inputs the second information into the AI ​​model to obtain the first information, then at least a portion of the second information is reported by the terminal to the network-side device.

[0687] The signal by which the terminal reports at least a portion of the second information to the first device may include at least one of the following: MAC CE, RRC message, NAS message, user plane data, MSG 1, MSG A, MSG 3, PUCCH, PUSCH, PRACH, SRS, and other uplink reference signals.

[0688] Optionally, if the first device is a network-side device, the communication device 500 may further include: an AI model acquisition module, used to obtain the AI ​​model from the terminal / other network-side device / server before obtaining the first information based on the AI ​​model.

[0689] In the second sub-implementation, if the first device is a server, then the second device includes the terminal and / or network-side device. That is, if the server inputs the second information into the AI ​​model to obtain the first information, then at least a portion of the second information is indicated to the server by the terminal and / or network-side device. For example, this can be indicated via OTT messages.

[0690] In the third sub-implementation, if the first device is the terminal, then the second device includes a network-side device. That is, if the terminal inputs the second information into the AI ​​model to obtain the first information, then at least a portion of the second information is indicated to the terminal by the network-side device. The signal from which the network-side device indicates at least a portion of the second information to the first device may include at least one of the following: MAC CE message, RRC message, NAS message, user plane data, DCI, SIB, PDCCH, PDSCH, Msg2, Msg4, and MsgB.

[0691] Optionally, if the first device is the terminal, the communication device 500 further includes: an AI model acquisition module, used to acquire the AI ​​model from a second device before obtaining the first information based on the AI ​​model, wherein the second device includes a network-side device and / or a server.

[0692] In other embodiments, the first information may be obtained by the second device based on the AI ​​model and then sent to the first device. That is, the first acquisition module 501 may be specifically used to: receive the first information sent by the second device, wherein the first information is obtained by the second device based on the AI ​​model.

[0693] In some detailed embodiments, the first information is obtained by the second device inputting the second information into the AI ​​model. Specifically:

[0694] If the first device includes network-side devices and / or servers, then the second device is the terminal. In other words, after the terminal uses an AI model to deduce the uplink transmission parameters, it can report the uplink transmission parameters to the network-side devices and / or servers.

[0695] If the first device is a terminal, then the second device includes a network-side device and / or a server. That is, after obtaining the first information based on the AI ​​model, the network-side device and / or the server send the first information to the terminal.

[0696] In some embodiments, the second information used as input to the AI ​​model during inference may include at least one of the following:

[0697] 1) Receive signal strength information;

[0698] 2) Received signal quality information;

[0699] 3) Path loss information;

[0700] 4) Reference signal related information;

[0701] 5) Communication distance information;

[0702] 6) Absolutely timely advance (TA) information;

[0703] 7) Relative TA information;

[0704] 8) TRP identifier or TRP group identifier;

[0705] 9) Community or community group signage;

[0706] 10) Tracking area markings;

[0707] 11) Radio Access Network (RAN) Notification Area Identifier, for example, the RAN notification area identifier can be the RAN identifier where the terminal is located;

[0708] 12) At least one of the following frequency domain information: frequency band information, frequency zone information, frequency point information, carrier frequency information, frequency layer information, and bandwidth part (BWP);

[0709] 13) One / a set of reference signal indices;

[0710] 14) One / a group of beam indices;

[0711] 15) One / a group of beam directions;

[0712] 16) Transmission power related information, such as transmission power related information of the base station or TRP;

[0713] 17) Propagation delay;

[0714] 18) Round Trip Time (RTT);

[0715] 19) Load information corresponding to the synchronization signal block SSB;

[0716] 20) Load information corresponding to the beam;

[0717] 21) Load information corresponding to TCI;

[0718] 22) Send the load information corresponding to the receiving point TRP;

[0719] 23) Interference information corresponding to the synchronization signal block SSB;

[0720] 24) Interference information corresponding to the beam;

[0721] 25) Interference information corresponding to TCI;

[0722] 26) Send the interference information corresponding to the receiving point TRP;

[0723] 27) Load information for one or more cells;

[0724] 28) Load information corresponding to one or more carriers;

[0725] 29) Load information corresponding to one or more frequency points;

[0726] 30) Interference information corresponding to one or more cells;

[0727] 31) Interference information corresponding to one or more carriers;

[0728] 32) Interference information corresponding to at least one of one or more frequency points;

[0729] 33) Geographic location information of the base station or TRP;

[0730] 34) Terminal location information;

[0731] 35) Terminal distribution information;

[0732] 36) The direction and / or speed of movement of the terminal;

[0733] 37) The direction and / or speed of movement of the network-side devices;

[0734] 38) Terminal energy consumption information;

[0735] 39) Terminal battery level information;

[0736] 40) Terminal type;

[0737] 41) Terminal supported operator information;

[0738] 42) Information on network types supported by the terminal;

[0739] 43) Antenna orientation information of the terminal;

[0740] 44) Antenna orientation information of network-side devices;

[0741] The network-side device may be a network node or a satellite;

[0742] 45) Network structure information, or network scenario information;

[0743] 46) Historical movement trajectory information of terminals within the community;

[0744] 47) Multipath information of the channel, such as the first path or strongest path information of the terminal and different base stations / TRPs;

[0745] 48) Environmental information, such as weather information;

[0746] 49) Time information;

[0747] 50) Perceiving information;

[0748] 51) Reference location or movement trajectory of a cell in a non-terrestrial network (NTN) scenario;

[0749] 52) Ephemeris information in NTN scenarios.

[0750] In some embodiments, the triggering conditions for the first device and / or the second device to perform inference using the AI ​​model may include at least one of the following:

[0751] 1) The fourth condition is met.

[0752] In other words, triggering the AI ​​model's inference is supported by default if the fourth condition is met. This fourth condition includes specific conditions such as having the ability to run the AI ​​model and / or being in the first use case.

[0753] 2) After the timer corresponding to the AI ​​model times out.

[0754] 3) After the terminal loses uplink synchronization.

[0755] 4) After the terminal loses downlink synchronization.

[0756] 5) After the terminal fails to access the random terminal for more than the first time period.

[0757] 6) After the terminal triggers random access for a second duration and still fails to successfully access the network.

[0758] 7) The number of times the terminal failed to access the system randomly exceeded the fourth threshold.

[0759] 8) After the terminal fails to access the network N times in a row.

[0760] 9) When the terminal meets the conditions / events for cell handover.

[0761] 10) The first device receives at least one of the second pieces of information, wherein the second pieces of information are inputs to the AI ​​model when it performs inference.

[0762] 11) Before the terminal triggers random access.

[0763] 12) Before the terminal triggers the initial access.

[0764] 13) The terminal detected a beam failure.

[0765] 14) The terminal detected a wireless link failure.

[0766] 15) The number of beam failure instances (BFI) exceeds the fifth threshold.

[0767] 16) BFI-related timers timed out.

[0768] 17) The terminal triggered a backoff indicator (BI).

[0769] 18) When the terminal receives a preamble sent by another terminal or that is not sent by itself.

[0770] 19) The number of preambles that the terminal receives from other terminals or that are not sent by itself exceeds the sixth threshold.

[0771] 20) The terminal is configured with a 2-step RACH transmission.

[0772] 21) The synchronization reference signal received power (SS-RSRP) of at least one SSB configured for configured grant Small Data Transmission (CG-SDT) is higher than the seventh threshold.

[0773] 22) TA invalid, that is, when TA is not valid.

[0774] 23) Used to determine if the change in RSRP of TA exceeds the eighth threshold.

[0775] 24) The terminal enters the connection state, that is, it is triggered only after entering the connection state.

[0776] 25) The current RACH configuration / RACH resources of the terminal do not match the terminal distribution under the first beam.

[0777] 26) The probability that a terminal initiating random access under the first beam where the current RACH configuration / RACH resources of the terminal do not match.

[0778] 27) The current RACH configuration / RACH resource usage ratio of the terminal is lower than the ninth threshold.

[0779] 28) The usage rate of RACH resources corresponding to one or a group of SSBs is higher than the tenth threshold.

[0780] 29) The probability of conflict of RACH resources corresponding to one or a group of SSBs is higher than the eleventh threshold.

[0781] 30) The probability of conflict of one or a group of RACH resources is higher than the twelfth threshold.

[0782] 31) The probability of access interruption of the current node or the current cell is higher than the thirteenth threshold.

[0783] 32) The type of terminal accessed / resided has changed.

[0784] 33) The power type of the terminal has changed.

[0785] In some embodiments, the performance metrics of the AI ​​model performing inference need to meet certain conditions to improve the reliability of the model's inference results. The performance metrics and the corresponding conditions they meet include at least one of the following:

[0786] 1) The complexity of the AI ​​model;

[0787] For example, different complexity index requirements for AI models can be defined for base stations or terminals of different types / capabilities. For instance, for ordinary terminals, the complexity of the AI ​​model used must not exceed a specific value.

[0788] 2) The inference latency of the AI ​​model;

[0789] For example, the duration of inferring optimal uplink transmission parameters using the AI ​​model cannot exceed a specific value.

[0790] 3) The success rate of the AI ​​model's inference;

[0791] For example, the success rate of using the AI ​​model to infer the optimal uplink transmission parameters must not be less than or must be greater than a specific value.

[0792] 4) The reliability of the reasoning results given by the AI ​​model.

[0793] Specifically, the reliability can be defined as the probability of collisions during random access using uplink transmission parameters inferred from an AI model being less than a threshold value.

[0794] Optionally, the communication device 500 proposed in this application embodiment may further include: a model training module for training the AI ​​model.

[0795] Specifically, the model training module can train the AI ​​model based on the training dataset.

[0796] In some embodiments, the training dataset for the AI ​​model includes at least one of the following:

[0797] 1) Sample data, wherein the sample data includes second information, which is the input when training the AI ​​model;

[0798] 2) Labels (also called truth values) of the sample data;

[0799] 3) Basic algorithms and / or indexes of basic algorithms, such as the error back propagation (BP) algorithm, stochastic gradient descent algorithm, etc.

[0800] 4) Loss function (also called objective function);

[0801] 5) Reward information (also known as adjustment / feedback information).

[0802] The labels of the sample data include at least one of the following:

[0803] An indicator of whether the uplink transmission parameters are reasonable. For example, if the uplink transmission parameters corresponding to the sample data are reasonable, the indicator is 1, indicating that the sample data is a positive sample data; if the uplink transmission parameters corresponding to the sample data are unreasonable, the indicator is 0, indicating that the sample data is a negative sample data.

[0804] The number of uplink transmission parameters obtained;

[0805] Number of users within the first region / coverage of the first beam;

[0806] Traffic volume within the first region / coverage area of ​​the first beam;

[0807] The probability of a terminal initiating random access in the first area / first beam;

[0808] RACH usage ratio in the first region / first beam;

[0809] RACH collision probability in the first region / first beam;

[0810] The number of at least one of the following in the first region / first beam: RACH resources, transmission opportunities (RO), and preamble;

[0811] The number of PUSCH resources in the first region / first beam;

[0812] Access interruption probability in the first region / first beam;

[0813] Obtain at least one set of true values ​​for uplink transmission parameters;

[0814] At least one of the first pieces of information.

[0815] The first region can be a specific region determined in advance, and the first beam can be a specific beam determined in advance.

[0816] The reward information includes at least one of the following:

[0817] The number of times the system reverts to the first mode, which is a mode that does not use an AI model to determine uplink transmission parameters;

[0818] The number of times the AI ​​model inference failed and / or succeeded;

[0819] The difference between the inference results of the AI ​​model and the labels;

[0820] The collision probability of RACH.

[0821] In some embodiments, the AI ​​model is trained by at least one of the terminal, network-side device, and server.

[0822] In some embodiments, the AI ​​model may be jointly trained by at least two of the terminal, network-side device, and server. Specifically, the training process of the AI ​​model includes at least one of the following:

[0823] The third device sends at least a portion of the output information during the training of the AI ​​model to the fourth device, and the fourth device uses the received output information as one of the input information for training the AI ​​model. The third device is one of the terminal, the network-side device, and the server, and the fourth device is the other of the terminal, the network-side device, and the server.

[0824] The third device and / or the fourth device deploy the AI ​​model online after training it offline, and then fine-tune the AI ​​model online.

[0825] In some embodiments, where the training device (first device) for the AI ​​model includes the terminal and / or the server, the apparatus shown in FIG5 may further include:

[0826] An information reporting module is used to report auxiliary information to the network-side device. The auxiliary information is used by the network-side device to determine the AI ​​model for inferring the first information from multiple candidate AI models.

[0827] The information receiving module is used to receive indication information sent by the network-side device, the indication information being used to instruct the AI ​​model for reasoning the first information.

[0828] Among them, the multiple candidate AI models have at least one of the following differences:

[0829] Different training datasets;

[0830] Different sample data;

[0831] The labels on the sample data are different;

[0832] The output information is different.

[0833] The auxiliary information includes at least one of the following:

[0834] The identifiers of the multiple candidate AI models;

[0835] The classification labels of the training datasets of the multiple candidate AI models.

[0836] The indication information includes at least one of the following:

[0837] Identification of AI models;

[0838] Classification labels for the training dataset;

[0839] Configuration information related to data acquisition.

[0840] In other embodiments, where the training device (first device) of the AI ​​model includes the network-side device and / or the server, the communication device 500 shown in FIG5 may further include: a model distribution module for distributing the trained AI model to the terminal so that the terminal can use the AI ​​model to perform reasoning on the first information.

[0841] In some embodiments, the training triggering conditions for the AI ​​model include at least one of the following:

[0842] The first condition and / or the first event is triggered;

[0843] Triggered periodically;

[0844] Semi-static triggering.

[0845] In the first sub-implementation, the training of the AI ​​model is triggered by a first condition and / or a first event, wherein the first condition and / or the first event includes at least one of the following:

[0846] 1) The relevant timer of the AI ​​model times out.

[0847] 2) Instructions from network-side devices, such as base stations sending model training instructions to terminals via TA MAC CE commands.

[0848] 3) The terminal accesses the first cell, switches to the first frequency point / first frequency band, and switches to the first operator / first public land mobile network (PLMN), wherein the first cell, the first frequency point / first frequency band, and the first operator / first public land mobile network can be a specific cell, a specific frequency point / specific frequency band, and a specific operator / specific public land mobile network that are determined in advance.

[0849] 4) After the AI ​​model inference fails.

[0850] 5) After the AI ​​model fails to reason N times consecutively.

[0851] 6) The number of inference failures of the AI ​​model reached the fourteenth threshold.

[0852] 7) After each time the AI ​​model is used for inference.

[0853] 8) The variation in the interference level of the first link exceeds the fifteenth threshold, where the first link can be a specific link that has been predetermined.

[0854] 9) The measured value of the signal reception quality of the terminal / network side device changes or the amount of change exceeds the sixteenth threshold, wherein the signal reception quality includes at least one of RSRP, RSRQ and RSSI.

[0855] 10) The timer related to model training timeout has expired for the third duration.

[0856] 11) There are N consecutive instances of model supervision or N instances of model supervision have occurred.

[0857] 12) Changes in the external environment, such as obtaining information about environmental changes through sensors.

[0858] 13) The network-side device is configured with uplink resources to fall back to the first mode, and the network-side device is configured to enable the second mode, wherein the first mode is a mode that does not use the AI ​​model to determine the uplink transmission parameters, and the second mode is a mode that uses the AI ​​model to determine the uplink transmission parameters.

[0859] 14) In the primary usage scenario, such as in a high-speed scenario.

[0860] 15) Reselect the cell or TRP.

[0861] 16) Cell or TRP handover.

[0862] 17) The conditions or events for cell handover are met.

[0863] 18) The terminal triggers access to the secondary cell.

[0864] 19) Receive at least one of the second information, the second information being used to train the AI ​​model, the specific content of the second information being referred to above.

[0865] 20) The number of times BFI is triggered exceeds the seventeenth threshold;

[0866] 21) Whether BFI is triggered;

[0867] 22) When the terminal receives a preamble sent by another terminal or not sent by itself, for example, the terminal receives a preamble ID in Msg2 that is not sent by itself.

[0868] 23) The number of preambles received by the terminal from other terminals or that are not sent by itself exceeds the eighteenth threshold.

[0869] In the second sub-implementation, the training of the AI ​​model is triggered periodically, wherein the configuration information for periodic triggering includes at least one of the following:

[0870] The starting point for periodic training, such as when the terminal receives the network configuration information and begins periodic model training;

[0871] Period length, for example, triggering at least one model training session at each preset interval;

[0872] The number of model training sessions and / or the training duration within a single period.

[0873] In the third sub-implementation, the training of the AI ​​model is semi-statically triggered, wherein the semi-static triggering includes at least one of the following:

[0874] The issuance and activation of semi-static configuration information are triggered by a second condition / second event, wherein the semi-static configuration information includes at least one of the following: the start time of model training, the cycle length, the number of model training sessions within a cycle, and / or the model training duration.

[0875] Semi-static configuration information is configured via RRC messages, and the AI ​​model is activated / deactivated via DCI or MAC-CE for semi-static training.

[0876] In some embodiments, the criteria for determining whether the AI ​​model training is complete include at least one of the following:

[0877] 1) The loss calculated by the loss function satisfies the third condition, such as the loss being less than a predefined threshold value.

[0878] 2) The model training iterations have reached the nineteenth threshold;

[0879] 3) The number of fine-tuning iterations reaches the twentieth threshold;

[0880] 4) At least one item in the label of the sample data is greater than or equal to or less than or equal to the threshold value;

[0881] 5) If the training time of the model is greater than or equal to the fourth training time, for example, if the training time of the AI ​​model is greater than or equal to the target training time, then the training is considered complete.

[0882] 6) If the energy consumption for model training is greater than or equal to the twenty-first threshold, for example, if the energy consumption for AI model training is greater than or equal to the target value, then training is considered complete.

[0883] In some embodiments, the behavior after obtaining the first information based on the AI ​​model includes at least one of the following:

[0884] 1) Collect at least one of the input and output information of this AI inference and provide it to the training device of the AI ​​model.

[0885] 2) Fall back to the first mode, which is a mode in which the uplink transmission parameters are not determined using the AI ​​model;

[0886] For example, reverting to the first mode may include at least one of the following:

[0887] Revert to the configuration information calculation and determination process of RACH without using an AI model (non-AI);

[0888] Revert to the configuration information calculation and determination process of MsgA PUSCH without using the AI ​​model;

[0889] The process of calculating and determining configuration information for at least one of the following is reverted to without using an AI model: CG PUSCH for SDT, RACH without handover, and cell handover under LTM mechanism.

[0890] 3) Trigger the switching of the AI ​​model, such as switching to another AI model.

[0891] 4) Trigger changes to the input information of the AI ​​model.

[0892] 5) Trigger at least one of the following: training, retraining, fine-tuning, and updating of the AI ​​model.

[0893] 6) Trigger supervision of the inference performance of the AI ​​model.

[0894] In some embodiments, the behavior after obtaining the first information based on the AI ​​model includes triggering supervision of the inference performance of the AI ​​model, and supervising the inference performance of the AI ​​model based on configuration information related to model supervision.

[0895] The configuration information related to model supervision includes at least one of the following:

[0896] 1) Identification of models that require inference performance supervision.

[0897] 2) The cycle of supervising model inference performance.

[0898] The model supervision period can be the interval between model supervision sessions, for example, performing model supervision once every day.

[0899] 3) The number of times the model inference performance is supervised.

[0900] 4) Duration of supervision for model inference performance.

[0901] 5) Information related to the time window for supervising model inference performance.

[0902] 6) Triggering conditions for model inference performance supervision.

[0903] 7) Labels for supervising model inference performance.

[0904] 8) Supervised inference performance metrics, such as the error between AI model predictions and actual values, or communication system performance metrics; among which, communication system performance metrics may include at least one of transmission latency, throughput, and RACH / PUSCH resource collision probability.

[0905] It is understandable that the terminal / network testing equipment / server can determine when and / or how to supervise the model inference performance based on the above configuration information.

[0906] In some embodiments, the triggering conditions for the model inference performance supervision include at least one of the following:

[0907] 1) The probability of a collision when using the first information obtained by the AI ​​for random access exceeds a threshold;

[0908] 2) At least one of the inference performance metrics of the AI ​​model fails to meet the requirements;

[0909] 3) The timer related to the inference performance supervision of the AI ​​model timed out;

[0910] 4) The terminal switches to at least one of a new cell, a new TRP, and a new beam;

[0911] 5) The number of terminals switching to at least one of the following: a new cell, a new TRP, and a new beam exceeds the twenty-second threshold;

[0912] 6) The AI ​​model inference failed;

[0913] 7) The AI ​​model fails to infer N times consecutively;

[0914] 8) The number of inference failures of the AI ​​model reaches the twenty-third threshold;

[0915] 9) Each time the AI ​​model is used;

[0916] 10) The number of times BFI is triggered exceeds the twenty-fourth threshold;

[0917] 11) Whether BFI is triggered;

[0918] 12) The terminal receives a preamble sent by another terminal or that is not sent by itself. For example, in Msg2, it receives a preamble ID that is not sent by itself.

[0919] 13) The number of preambles received by the terminal from other terminals or that are not sent by itself exceeds the twenty-fifth threshold;

[0920] 14) The probability of the third and / or fourth information appearing is greater than or equal to or less than or equal to the twenty-sixth threshold;

[0921] 15) The deviation value of the third and / or fourth information exceeds the twenty-seventh threshold.

[0922] In some embodiments, the execution device performing the relevant operations on the AI ​​model has relevant capabilities. These relevant operations include at least one of the following:

[0923] It has the ability to train the AI ​​model;

[0924] It has the ability to perform reasoning using the AI ​​model;

[0925] It has the ability to supervise the inference performance of the AI ​​model;

[0926] It has the ability to provide one or more auxiliary information, which is used for at least one of training the AI ​​model and performing inference using the AI ​​model.

[0927] In other words, before terminal / network-side devices / servers perform AI model training / inference / inference performance supervision, it is necessary to define the relevant capabilities of the terminal / network-side devices / servers and provide methods for determining these capabilities. Introducing a definition and determination mechanism for relevant capabilities helps each node flexibly implement corresponding features and facilitates feature implementation between nodes.

[0928] In some embodiments, the relevant capability is obtained based on at least one of the following:

[0929] 1) Terminal type, for example, different AI-related capabilities are introduced for different terminal types (such as Redcap, IoT), or the related capabilities are related to the terminal type, and the related capabilities are different for different terminal types;

[0930] The correspondence between the relevant capabilities and terminal types includes at least one of the following:

[0931] The input information used for model training / inference varies depending on the type of terminal. For example, less input information should be used for model training on terminal devices with weaker capabilities.

[0932] Different types of terminals use different labels for model training;

[0933] The execution devices for model training vary depending on the type of terminal. For example, for terminals with weaker capabilities, model training can be performed only on the network side device, or the terminal can only perform a small part of the joint model training (for example, model training involving user privacy data can be performed on the terminal).

[0934] The underlying algorithms used for model training differ depending on the type of terminal. For example, overly complex underlying algorithms may not be applicable to devices with weaker capabilities.

[0935] 2) Network type, for example, introducing different AI-related capabilities for different network types (such as NTN, TN);

[0936] 3) One or more reference signals indicate, for example, specific resources of PRACH (such as specific ROs or preambles) that indicate that the terminal has the ability to train the AI ​​model or to use the AI ​​model for inference;

[0937] 4) Uplink Control Information (UCI) is carried, such as physical layer control information (e.g., UCI information reported to the network for the terminal).

[0938] 5) RRC message carrying;

[0939] 6) The first interface message carried between the terminal and the server;

[0940] 7) The first interface message carried between the terminal and the network-side equipment;

[0941] 8) The first interface message carried between the server and the network-side device.

[0942] The first interface message may include messages related to the first AI model or messages related to all AI models.

[0943] The communication device 500 provided in this application embodiment can implement the various processes implemented in the method embodiment of FIG4 and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0944] As shown in Figure 6, this application embodiment also provides a communication device 600, including a processor 601 and a memory 602. The memory 602 stores programs or instructions that can run on the processor 601. For example, when the communication device 600 is a terminal, the program or instructions executed by the processor 601 implement the various steps of the above-described communication method embodiment and achieve the same technical effect. When the communication device 600 is a network-side device, the program or instructions executed by the processor 601 implement the various steps of the above-described communication method embodiment and achieve the same technical effect. To avoid repetition, further details are omitted here.

[0945] 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 embodiment shown in FIG4. This terminal embodiment corresponds to the above-described terminal-side method embodiment, 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 may be the communication device shown in FIG5. Specifically, FIG7 is a schematic diagram of the hardware structure of a terminal implementing an embodiment of this application.

[0946] The terminal 700 includes, but is not limited to, at least some of the following components: radio frequency unit 701, network module 702, audio output unit 703, input unit 704, sensor 705, display unit 706, user input unit 707, interface unit 708, memory 709, and processor 710.

[0947] Those skilled in the art will understand that the terminal 700 may also include a power supply (such as a battery) for powering various components. The power supply can be logically connected to the processor 710 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. The terminal structure shown in Figure 7 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.

[0948] It should be understood that, in this embodiment, the input unit 704 may include a graphics processor 7041 and a microphone 7042. The graphics processor 7041 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 706 may include a display panel 7061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit 707 includes at least one of a touch panel 7071 and other input devices 7072. The touch panel 7071 is also called a touch screen. The touch panel 7071 may include a touch detection device and a touch controller. Other input devices 7072 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.

[0949] In this embodiment, after receiving downlink data from the network-side device, the radio frequency unit 701 can transmit it to the processor 710 for processing; in addition, the radio frequency unit 701 can send uplink data to the network-side device. Typically, the radio frequency unit 701 includes, but is not limited to, antennas, amplifiers, transceivers, couplers, low-noise amplifiers, duplexers, etc.

[0950] The memory 709 can be used to store software programs or instructions, as well as various data. The memory 709 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 709 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 link dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 709 in the embodiments of this application includes, but is not limited to, these and any other suitable types of memory.

[0951] Processor 710 may include one or more processing units; optionally, processor 710 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 710.

[0952] The radio frequency unit 701 is used to acquire first information, which is obtained based on an artificial intelligence (AI) model and is used by the terminal for uplink transmission.

[0953] Alternatively, the processor 710 is configured to acquire first information, wherein the first information is obtained based on an artificial intelligence (AI) model, and the first information is used by the terminal for uplink transmission.

[0954] The terminal proposed in this application can obtain the first information for uplink transmission based on the artificial intelligence (AI) model. This makes the uplink transmission parameters used by the terminal match the actual application scenario, rather than being arbitrarily configured by the network-side device. Therefore, it can reduce resource collisions and conflicts, avoid resource waste, improve resource utilization, enhance the reliability of uplink transmission, and reduce terminal power consumption.

[0955] It is understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description of the method embodiment and achieve the same or corresponding technical effect. To avoid repetition, it will not be described again here.

[0956] 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. 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.

[0957] Specifically, this application embodiment also provides a network-side device, which can be the communication device shown in FIG5. As shown in FIG8, the network-side device 800 includes: an antenna 81, a radio frequency device 82, a baseband device 83, a processor 84, and a memory 85. The antenna 81 is connected to the radio frequency device 82. In the uplink direction, the radio frequency device 82 receives information through the antenna 81 and sends the received information to the baseband device 83 for processing. In the downlink direction, the baseband device 83 processes the information to be transmitted and sends it to the radio frequency device 82. The radio frequency device 82 processes the received information and transmits it through the antenna 81.

[0958] The method executed by the network-side device in the above embodiments can be implemented in the baseband device 83, which includes a baseband processor.

[0959] The baseband device 83 may include at least one baseband board, on which multiple chips are disposed, as shown in FIG8. One of the chips is, for example, a baseband processor, which is connected to the memory 85 via a bus interface to call the program in the memory 85 and execute the network-side device operation shown in the above method embodiment.

[0960] The network-side device may also include a network interface 86, such as a Common Public Radio Interface (CPRI).

[0961] Specifically, the network-side device 800 in this application embodiment further includes: instructions or programs stored in memory 85 and executable on processor 84. Processor 84 calls the instructions or programs in memory 85 to execute the methods executed by each module shown in FIG5 and achieve the same technical effect. To avoid repetition, it will not be described in detail here.

[0962] 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 communication method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.

[0963] 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.

[0964] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described communication method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0965] 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.

[0966] 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 communication method embodiments, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0967] This application also provides a communication system, including a terminal and a network-side device, wherein the terminal or network-side device can be used to execute the steps of the communication method described in FIG4.

[0968] 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.

[0969] 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.

[0970] 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 communication method, the method comprising: The first device acquires first information, wherein the first information is obtained based on an artificial intelligence (AI) model, and the first information is used by the terminal for uplink transmission.

2. The method according to claim 1, wherein, The first device acquires first information, including: The first device obtains the first information based on the AI ​​model.

3. The method according to claim 2, wherein, The first device includes network-side equipment and / or a server, the first information includes user distribution information of the cell and / or load information of the cell, and the method further includes: The first device determines the uplink transmission parameters based on the first information; The first device sends the uplink transmission parameters to the terminal.

4. The method according to claim 2, wherein, The first device includes network-side devices and / or servers, the first information includes uplink transmission parameters, and the method further includes: The first device sends the uplink transmission parameters to the terminal.

5. The method according to claim 3 or 4, wherein, The first device sends the uplink transmission parameters to the terminal via at least one of the following signals: Paging messages; Paging downlink control information (DCI); SMS; Synchronization signal block (SSB); System message; Msg2; Msg4; MsgB; Media Access Control Unit (MAC) CE messages / Radio Resource Control Protocol (RRC) messages / Non-Access Stratum (NAS) messages.

6. The method according to claim 2, wherein, The first device is the terminal. Before the first device obtains the first information based on the AI ​​model, the method further includes: The first device obtains the AI ​​model from the second device, wherein the second device includes a network-side device and / or a server.

7. The method according to any one of claims 2-6, wherein, The first device obtains the first information based on the AI ​​model, including: The first device inputs the second information into the AI ​​model to obtain the first information.

8. The method according to claim 7, wherein, At least part of the second information is provided by the second device.

9. The method according to claim 8, wherein, If the first device is a network-side device, then the second device includes the terminal; If the first device is a server, then the second device includes the terminal and / or network-side device; If the first device is the terminal, then the second device includes network-side devices.

10. The method according to claim 1, wherein, The first device acquires first information, including: The first device receives the first information sent by the second device, wherein the first information is obtained by the second device based on the AI ​​model.

11. The method according to claim 10, wherein, If the first device includes network-side devices and / or servers, then the second device is the terminal; If the first device is a terminal, then the second device includes network-side devices and / or servers.

12. The method according to claim 3, wherein, The user distribution information and / or load information of the cell include at least one of the following: Third information; The probability of the third piece of information appearing; The third information includes at least one of the following: Number of users in each beam direction; Number of users within the first region / coverage of the first beam; Traffic volume per beam; Traffic volume within the first region / coverage area of ​​the first beam; The probability of a terminal triggering random access transmission within the first area / coverage range of the first beam; The probability of a terminal triggering random access transmission in each beam direction; User types within the first area / first beam coverage area; User power levels within the first area / first beam coverage area; User type under each beam direction; User power level in each beam direction.

13. The method according to any one of claims 3-5, 11, wherein, The uplink transmission parameters include at least one of the following: At least one set of fourth information; The probability of the fourth piece of information appearing; The fourth piece of information includes at least one of the following: Configuration information for the Random Access Channel (RACH); Configuration information for the first physical uplink shared channel (PUSCH).

14. The method according to claim 13, wherein, The RACH configuration information includes at least one of the following: The quantity and / or index of at least one of the following associated with RACH: Synchronization Signal Block (SSB), Channel State Information Reference Signal (CSI-RS), and Beam and Transmission Configuration Indicator (TCI); The mapping ratio from SSB to uplink transmission opportunity (RO) of random access channel; The number and / or indexes of SSBs associated with RO; The number of ROs in the time domain; The number of ROs in the frequency domain; The number of ROs and / or indexes in the first SSB mapping; Number and / or index of the preamble in the first SSB mapping; Preamble format; The mapping ratio from preamble to PUSCH; The number and / or index of PUSCH resources mapped in the first preamble; The configuration period of the Physical Random Access Channel (PRACH); The association cycle from SSB to PRACH resources; SSB to PRACH resource association pattern cycle; The mapping cycle from SSB to PRACH resources; The determination period and / or time window used to determine the PRACH resource group for PRACH retransmission; PRACH repeatedly turns on; Number of PRACH repetitions; The PRACH may reuse the same or different beams; The activation status of the PRACH resource; Trigger thresholds for one or more SSB / CSI-RS / beam / TCI associated RACHs; Trigger thresholds for one or more RACHs associated with RO / preamble / preamble formats; The preamble receives the target power; PRACH maximum transmission power.

15. The method according to claim 13, wherein, The configuration information of the first PUSCH includes at least one of the following: The number and / or index of at least one of the following associated with the first PUSCH: the synchronization signal block SSB, the channel state information reference signal CSI-RS, the beam, and the TCI; The mapping ratio from SSB to the first PUSCH resource; The number and / or indexes of SSBs associated with the first PUSCH resource; The number of first PUSCH resources in the time and / or frequency domains; The number and / or index of the first PUSCH resource mapped by the first SSB; Configure the authorization cycle for CG PUSCH; The association period from SSB to the first PUSCH resource; The association pattern cycle from SSB to the first PUSCH resource; The mapping cycle from SSB to the first PUSCH resource; The first PUSCH is repeatedly in the on state; Number of times the first PUSCH is repeated; The first PUSCH may reuse the same or different beams; The activation status of the first PUSCH resource; The target received power for the first PUSCH transmission; Maximum transmission power of the first PUSCH; Trigger threshold for one or more SSB / CSI-RS / beam / TCI associated message MsgA PUSCH; The mapping ratio from MsgA PRACH / RO / preamble to MsgA PUSCH; The number and / or index of ROs / preambles associated with the first PUSCH resource; The SSB or SSB group associated with the mapping from MsgA preamble to MsgA PUSCH; The time-frequency domain resources of the MsgA PUSCH mapped by the first SSB or the first SSB group.

16. The method according to claim 15, wherein, The first PUSCH includes at least one of the following: MsgA PUSCH; Msg3 PUSCH; PUSCH scheduled by downlink control information (DCI) scrambled with the temporary cell radio network identifier (TC-RNTI); Configure and authorize CG PUSCH; PUSCH in the RACH transmission process based on small data transmission; PUSCH during RACH-free handover; PUSCH during the LTM process; CG PUSCH in Small Data Transmission DT; CG PUSCH during RACH-free handover; CG PUSCH during LTM cell handover.

17. The method according to any one of claims 3-5, 11, and 13-16, wherein, The uplink transmission parameters satisfy at least one of the following conditions: The usage rate of RACH resources exceeds the first threshold; The collision probability of RACH is below the second threshold; The probability of access interruption is below the third threshold; Matching the user distribution or business volume distribution within the community; It matches the probability of users initiating random access within the community.

18. The method according to any one of claims 1-17, wherein, The triggering conditions for using the AI ​​model for reasoning include at least one of the following: The fourth condition is met, wherein the fourth condition includes having the ability to run the AI ​​model and / or being in the first use case; After the timer corresponding to the AI ​​model times out; After the terminal loses uplink synchronization; After the terminal loses downlink synchronization; After the terminal's random access failure exceeds the first time period; After the terminal triggers random access for a second duration and still fails to successfully access; The number of times the terminal failed to access the network randomly exceeded the fourth threshold. After the terminal randomly fails to access the network N times consecutively; If the terminal meets the conditions / events for cell handover; The first device receives at least one of the second pieces of information, where the second pieces of information are inputs to the AI ​​model during inference. Before the terminal triggers random access; Before the terminal triggers initial access; The terminal detected a beam failure. The terminal detected a wireless link failure. The number of beam failure instances (BFI) exceeds the fifth threshold; BFI-related timers timed out; The terminal triggered a rollback instruction (BI); When the terminal receives a preamble sent by another terminal or that was not sent by itself; The number of preambles received by the terminal from other terminals or that are not sent by itself exceeds the sixth threshold. The terminal is configured with 2-step RACH transmission; The synchronization reference signal received power (SS-RSRP) of at least one SSB configured for authorized small data transmission CG-SDT is higher than the seventh threshold. Pre-delay expiration; The change in RSRP used to determine if the TA exceeds the eighth threshold; The terminal enters the connected state; The current RACH configuration / RACH resources of the terminal do not match the terminal distribution under the first beam; The probability that a terminal initiates random access under the first beam when the terminal's current RACH configuration / RACH resources do not match. The current RACH configuration / RACH resource usage ratio of the terminal is lower than the ninth threshold; The usage rate of RACH resources corresponding to one or a group of SSBs is higher than the tenth threshold; The probability of conflict of RACH resources corresponding to one or a group of SSBs is higher than the eleventh threshold; The probability of conflict for one or a group of RACH resources is higher than the twelfth threshold; The probability of access interruption for the current node or the current cell is higher than the thirteenth threshold; The type of terminal accessed / resided has changed; The power type of the terminal has changed.

19. The method according to any one of claims 1-18, wherein, The AI ​​model is trained by at least one of the terminal, network-side device, and server.

20. The method according to claim 19, wherein, When the training device for the AI ​​model includes the terminal and / or the server, the method further includes: The terminal and / or the server report auxiliary information to the network-side device, the auxiliary information being used by the network-side device to determine the AI ​​model for inferring the first information from multiple candidate AI models; The terminal and / or the server receive instruction information sent by the network-side device, the instruction information being used to instruct the AI ​​model for reasoning the first information.

21. The method according to claim 19 or 20, wherein, The training dataset for the AI ​​model includes at least one of the following: Sample data, the sample data including second information; Labels for sample data; Basic algorithm and / or index of the basic algorithm; Loss function; Reward information.

22. The method according to claim 21, wherein, The labels of the sample data include at least one of the following: An indicator of whether the uplink transmission parameters are reasonable; The number of uplink transmission parameters obtained; Number of users within the first region / coverage of the first beam; Traffic volume within the first region / coverage area of ​​the first beam; The probability of a terminal initiating random access in the first area / first beam; RACH usage ratio in the first region / first beam; RACH collision probability in the first region / first beam; The number of at least one of the following in the first region / first beam: RACH resources, transmission opportunities (RO), and preamble; The number of PUSCH resources in the first region / first beam; Access interruption probability in the first region / first beam; Obtain at least one set of true values ​​for uplink transmission parameters; At least one of the first pieces of information.

23. The method according to claim 21, wherein, The reward information includes at least one of the following: The number of times the system reverts to the first mode, which is a mode that does not use an AI model to determine uplink transmission parameters; The number of times the AI ​​model inference failed and / or succeeded; The difference between the inference results of the AI ​​model and the labels; Collision probability of the Random Access Channel (RACH).

24. The method according to any one of claims 7-9, 18, and 21-23, wherein, The second information includes at least one of the following: Receive signal strength information; Received signal quality information; Path loss information; Reference signal related information; Communication distance information; Absolutely timed advance notice of TA information; Relative TA information; Send / receive point TRP identifier or TRP group identifier; Community or community group signage; Tracking area markings; Radio Access Network (RAN) Notification Area Identifier; Frequency band information; Frequency band information; Frequency information; Carrier frequency information; Frequency layer information; Partial bandwidth BWP; One / a set of reference signal indices; One / a group of beam indices; One or a group of beam directions; Transmission power related information; Propagation delay; Round-trip time (RTT) Load information corresponding to the synchronization signal block SSB; Beam-related load information; Transmission configuration indicates the load information corresponding to TCI; Send the load information corresponding to the receiving point TRP; Interference information corresponding to the synchronization signal block SSB; Interference information corresponding to the beam; Interference information corresponding to TCI; Transmit interference information corresponding to the receiving point TRP; Load information for one or more cells; Load information corresponding to one or more carriers; Load information corresponding to one or more frequency points; Interference information for one or more cells; Interference information corresponding to one or more carriers; Interference information corresponding to one or more frequency points; Geographic location information of the base station or TRP; Terminal location information; Terminal distribution information; The direction and / or speed of movement of the terminal; The direction and / or speed of movement of network-side devices Terminal energy consumption information; Battery information; Terminal type; Information on the carriers supported by the terminal; Information on network types supported by the terminal; Information on the orientation of the terminal's antenna; Information on the antenna orientation of network-side devices; Network structure information; Historical movement trajectory information of terminals within the community; Multipath information of the channel; Environmental information; Time information; Perceive information; Reference location or movement trajectory of a cell in a non-terrestrial network (NTN) scenario; Ephemeris information in NTN scenarios.

25. The method according to any one of claims 19-24, wherein, The training trigger conditions for the AI ​​model include at least one of the following: The first condition and / or the first event is triggered; Triggered periodically; Semi-static triggering.

26. The method of claim 25, wherein, The first condition and / or the first event includes at least one of the following: After the relevant timer of the AI ​​model times out; Indication information from network-side devices; The terminal accesses the first cell, switches to the first frequency point / first frequency band, and switches to the first operator / first public terrestrial mobile network (PLMN). After the AI ​​model inference fails; After the AI ​​model fails inference N times consecutively; The number of inference failures of the AI ​​model reached the fourteenth threshold. After each time the AI ​​model is used for inference; The variation in the interference level of the first link exceeds the fifteenth threshold; The measured value of the signal reception quality of the terminal / network side device changes or the amount of change exceeds the sixteenth threshold, wherein the signal reception quality includes at least one of the following: Reference Signal Received Power (RSRP), Reference Signal Received Quality (RSRQ), and Received Signal Strength Indication (RSSI). The third timeout duration related to model training; N consecutive model supervision events or N model supervision events have occurred; The external environment has changed; The network-side device is configured with uplink resources to fall back to the first mode, and the network-side device is configured to enable the second mode. The first mode is a mode that does not use the AI ​​model to determine the uplink transmission parameters, and the second mode is a mode that uses the AI ​​model to determine the uplink transmission parameters. In the primary use case; Cell or TRP reselection; Cell or TRP handover; The conditions or events for cell handover are met; The terminal triggers access to the secondary cell; Receive at least one of the second pieces of information, the second pieces of information being used to train the AI ​​model; The number of times the Beam Failure Instance (BFI) is triggered exceeds the seventeenth threshold; Does it trigger a beam failure instance (BFI)? When the terminal receives a preamble sent by another terminal or that was not sent by itself; The number of preambles received by the terminal from other terminals or that are not sent by itself exceeds the eighteenth threshold.

27. The method according to claim 25, wherein, The training of the AI ​​model is semi-statically triggered, wherein the semi-static triggering includes at least one of the following: The issuance and activation of semi-static configuration information are triggered by a second condition / second event, wherein the semi-static configuration information includes at least one of the following: the starting time of model training, the cycle length, the number of model training sessions within a cycle, and / or the model training duration. Semi-static training of the AI ​​model is performed by configuring semi-static configuration information through Radio Resource Control Protocol (RRC) and activating / deactivating it through Downlink Control Information (DCI) or Media Access Control Unit (MAC-CE).

28. The method according to any one of claims 1-27, wherein, The behavior after obtaining the first information based on the AI ​​model includes at least one of the following: Collect at least one of the input and output information for this AI inference and provide it to the training device of the AI ​​model; Fall back to the first mode, which is a mode in which the uplink transmission parameters are not determined using the AI ​​model; Trigger the switching of the AI ​​model; This triggers a change in the input information of the AI ​​model; Trigger at least one of the following: training, retraining, fine-tuning, and updating of the AI ​​model; This triggers the supervision of the inference performance of the AI ​​model.

29. The method according to claim 28, wherein, The actions taken after obtaining the first information from the AI ​​model include triggering supervision of the AI ​​model's inference performance, and supervising the AI ​​model's inference performance based on configuration information related to model supervision.

30. The method according to claim 29, wherein, The configuration information related to model supervision includes at least one of the following: Identifiers of models that require inference performance supervision; The cycle of supervising model inference performance; The number of times the model inference performance is supervised; Duration of supervision for model inference performance; Information related to the time window for supervising model inference performance; Triggering conditions for model inference performance supervision; Labels for supervising model inference performance; Inference performance metrics that require supervision.

31. The method according to claim 29, wherein, The triggering conditions for the model inference performance supervision include at least one of the following: The probability of a collision when using the first information obtained by the AI ​​for random access exceeds a threshold. At least one of the inference performance metrics of the AI ​​model fails to meet the requirements; The timer related to the inference performance supervision of the AI ​​model timed out; The terminal switches to at least one of a new cell, a new TRP, and a new beam; The number of terminals switching to at least one of the following: a new cell, a new TRP, and a new beam exceeds the twenty-second threshold; The AI ​​model inference failed. The AI ​​model failed inference N times consecutively. The number of inference failures of the AI ​​model reached the twenty-third threshold. Each time the AI ​​model is used; The number of times the Beam Failure Instance (BFI) is triggered exceeds the twenty-fourth threshold; Does BFI need to be triggered? The terminal receives a preamble sent by another terminal or a preamble that it did not send itself; The number of preambles received by the terminal from other terminals or that are not sent by itself exceeds the twenty-fifth threshold. The probability of the third and / or fourth information appearing is greater than or equal to or less than or equal to the twenty-sixth threshold. The deviation values ​​of the third and / or fourth information exceed the twenty-seventh threshold; The third information includes at least one of the following: Number of users in each beam direction; Number of users within the first region / coverage of the first beam; Traffic volume per beam; Traffic volume within the first region / coverage area of ​​the first beam; The probability of a terminal triggering random access transmission within the first area / coverage range of the first beam; The probability of a terminal triggering random access transmission in each beam direction; User types within the first area / first beam coverage area; User power levels within the first area / first beam coverage area; User type under each beam direction; User power level in each beam direction; The fourth piece of information includes at least one of the following: Configuration information for the Random Access Channel (RACH); Configuration information for the first physical uplink shared channel (PUSCH).

32. The method according to any one of claims 1-31, wherein, The execution device that performs the relevant operations on the AI ​​model has the relevant capabilities, wherein the relevant operations include at least one of the following: Train the AI ​​model; Use the AI ​​model to perform inference; The inference performance of the AI ​​model is supervised; Provide auxiliary information for the inference or training of the AI ​​model.

33. The method according to claim 32, wherein, The relevant capabilities include at least one of the following: It has the ability to train the AI ​​model; It has the ability to perform reasoning using the AI ​​model; It has the ability to supervise the inference performance of the AI ​​model; It has the ability to provide one or more auxiliary information, which is used for at least one of training the AI ​​model and performing inference using the AI ​​model.

34. The method according to claim 32, wherein, The relevant capabilities are obtained based on at least one of the following: Terminal type; Network type; One or more reference signals indicate; Uplink control information (UCI) is carried; Radio Resource Control (RRC) message carries; The first interface message between the terminal and the server carries; The first interface message between the terminal and the network-side equipment carries; The first interface message between the server and the network-side device carries; The first interface message may include messages related to the first AI model or messages related to all AI models.

35. A communication device, comprising: The first acquisition module is used to acquire first information, wherein the first information is obtained based on an artificial intelligence (AI) model, and the first information is used by the terminal for uplink transmission.

36. The apparatus according to claim 35, wherein, The first acquisition module is specifically used for: The first information is obtained based on the AI ​​model.

37. The apparatus according to claim 25, wherein, The first acquisition module is specifically used for: The first information is received from the second device, wherein the first information is obtained by the second device based on the AI ​​model.

38. The apparatus according to any one of claims 35-37, wherein, The first information includes at least one of the following: The first information includes user distribution information and / or load information of the cell; Uplink transmission parameters.

39. The apparatus according to claim 38, wherein, The user distribution information and / or load information of the cell include at least one of the following: Third information; The probability of the third piece of information appearing; The third information includes at least one of the following: Number of users in each beam direction; Number of users within the first region / coverage of the first beam; Traffic volume per beam; Traffic volume within the first region / coverage area of ​​the first beam; The probability of a terminal triggering random access transmission within the first area / coverage range of the first beam; The probability of a terminal triggering random access transmission in each beam direction; User types within the first area / first beam coverage area; User power levels within the first area / first beam coverage area; User type under each beam direction; User power level in each beam direction.

40. The apparatus according to claim 38, wherein, The uplink transmission parameters include at least one of the following: At least one set of fourth information; The probability of the fourth piece of information appearing; The fourth piece of information includes at least one of the following: Configuration information for the Random Access Channel (RACH); Configuration information for the first physical uplink shared channel (PUSCH).

41. The apparatus according to claim 40, wherein, The RACH configuration information includes at least one of the following: The quantity and / or index of at least one of the following associated with RACH: Synchronization Signal Block (SSB), Channel State Information Reference Signal (CSI-RS), and Beam and Transmission Configuration Indicator (TCI); The mapping ratio from SSB to uplink transmission opportunity (RO) of random access channel; The number and / or indexes of SSBs associated with RO; The number of ROs in the time domain; The number of ROs in the frequency domain; The number of ROs and / or indexes in the first SSB mapping; Number and / or index of the preamble in the first SSB mapping; Preamble format; The mapping ratio from preamble to PUSCH; The number and / or index of PUSCH resources mapped in the first preamble; The configuration period of the Physical Random Access Channel (PRACH); The association cycle from SSB to PRACH resources; SSB to PRACH resource association pattern cycle; The mapping cycle from SSB to PRACH resources; The determination period and / or time window used to determine the PRACH resource group for PRACH retransmission; PRACH repeatedly turns on; Number of PRACH repetitions; The PRACH may reuse the same or different beams; The activation status of the PRACH resource; Trigger thresholds for one or more SSB / CSI-RS / beam / TCI associated RACHs; Trigger thresholds for one or more RACHs associated with RO / preamble / preamble formats; The preamble receives the target power; PRACH maximum transmission power.

42. The apparatus according to claim 40, wherein, The configuration information of the first PUSCH includes at least one of the following: The number and / or index of at least one of the following associated with the first PUSCH: the synchronization signal block SSB, the channel state information reference signal CSI-RS, the beam, and the TCI; The mapping ratio from SSB to the first PUSCH resource; The number and / or indexes of SSBs associated with the first PUSCH resource; The number of first PUSCH resources in the time and / or frequency domains; The number and / or index of the first PUSCH resource mapped by the first SSB; Configure the authorization cycle for CG PUSCH; The association period from SSB to the first PUSCH resource; The association pattern cycle from SSB to the first PUSCH resource; The mapping cycle from SSB to the first PUSCH resource; The first PUSCH is repeatedly in the on state; Number of times the first PUSCH is repeated; The first PUSCH may reuse the same or different beams; The activation status of the first PUSCH resource; The target received power for the first PUSCH transmission; Maximum transmission power of the first PUSCH; Trigger threshold for one or more SSB / CSI-RS / beam / TCI associated message MsgA PUSCH; The mapping ratio from MsgA PRACH / RO / preamble to MsgA PUSCH; The number and / or index of ROs / preambles associated with the first PUSCH resource; The SSB or SSB group associated with the mapping from MsgA preamble to MsgA PUSCH; Time-frequency domain resources of MsgA PUSCH mapped by the first SSB or the first SSB group; Wherein, the first PUSCH includes at least one of the following: MsgA PUSCH; Msg3 PUSCH; PUSCH scheduled by downlink control information (DCI) scrambled with the temporary cell radio network identifier (TC-RNTI); Configure and authorize CG PUSCH; PUSCH in the RACH transmission process based on small data transmission; PUSCH during RACH-free handover; PUSCH during the LTM process; CG PUSCH in Small Data Transmission DT; CG PUSCH during RACH-free handover; CG PUSCH during LTM cell handover.

43. The method according to any one of claims 35-42, wherein, The first information is obtained by inputting the second information into the AI ​​model; The second information includes at least one of the following: Receive signal strength information; Received signal quality information; Path loss information; Reference signal related information; Communication distance information; Absolutely timed advance notice of TA information; Relative TA information; Send / receive point TRP identifier or TRP group identifier; Community or community group signage; Tracking area markings; Radio Access Network (RAN) Notification Area Identifier; Frequency band information; Frequency band information; Frequency information; Carrier frequency information; Frequency layer information; Partial bandwidth BWP; One / a set of reference signal indices; One / a group of beam indices; One or a group of beam directions; Transmission power related information; Propagation delay; Round-trip time (RTT) Load information corresponding to the synchronization signal block SSB; Beam-related load information; Transmission configuration indicates the load information corresponding to TCI; Send the load information corresponding to the receiving point TRP; Interference information corresponding to the synchronization signal block SSB; Interference information corresponding to the beam; Interference information corresponding to TCI; Transmit interference information corresponding to the receiving point TRP; Load information for one or more cells; Load information corresponding to one or more carriers; Load information corresponding to one or more frequency points; Interference information for one or more cells; Interference information corresponding to one or more carriers; Interference information corresponding to one or more frequency points; Geographic location information of the base station or TRP; Terminal location information; Terminal distribution information; The direction and / or speed of movement of the terminal; The direction and / or speed of movement of network-side devices Terminal energy consumption information; Battery information; Terminal type; Information on the carriers supported by the terminal; Information on network types supported by the terminal; Information on the orientation of the terminal's antenna; Information on the antenna orientation of network-side devices; Network structure information; Historical movement trajectory information of terminals within the community; Multipath information of the channel; Environmental information; Time information; Perceive information; Reference location or movement trajectory of a cell in a non-terrestrial network (NTN) scenario; Ephemeris information in NTN scenarios.

44. 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 steps of the communication method as claimed in any one of claims 1 to 34.

45. A readable storage medium storing a program or instructions that, when executed by a processor, implement the communication method as described in any one of claims 1-34.

Images