Random access transmission method, terminal and network side equipment

Abstract

The invention discloses a random access transmission method, a terminal and network side equipment, and belongs to the technical field of communication, and the random access transmission method comprises the steps that first equipment obtains first information based on an artificial intelligence (AI) model; performing random access channel (RACH) transmission based on the first information; wherein the first information comprises at least one of the following items: beam information, RACH collision probability, RACH resource information and a transmitting power parameter.

Description

Technical Field

[0001] This application belongs to the field of communication technology, specifically relating to a random access transmission method, a terminal, and network-side equipment. Background Technology

[0002] In some scenarios, users within a cell are often unevenly distributed, such as a high density of users in one direction and a sparser distribution in another. This uneven distribution can increase the probability of collisions on the Random Access Channel (RACH) under certain beams.

[0003] Currently, users typically initiate RACH transmissions by randomly selecting RACH resources that meet certain criteria. This selection method leads to lower resource utilization and increases terminal power consumption, thus reducing terminal energy efficiency. Summary of the Invention

[0004] This application provides a random access transmission method, a terminal, and a network-side device that can solve the problems of low resource utilization and low energy efficiency.

[0005] In a first aspect, a random access transmission method is provided, comprising: a first device obtaining first information based on an artificial intelligence (AI) model; and performing random access channel (RACH) transmission based on the first information; wherein the first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

[0006] Secondly, a random access transmission device is provided, comprising: an acquisition module for obtaining first information based on an AI model; and a transmission module for performing RACH transmission based on the first information; wherein the first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

[0007] Thirdly, a random access transmission apparatus is provided, the apparatus being configured to perform the steps of the method described in the first aspect.

[0008] 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.

[0009] Fifthly, a terminal is provided, including a processor and a communication interface, wherein the processor is used to implement the steps of the method described in the first aspect, and the communication interface is used to couple with the processor.

[0010] In a sixth aspect, 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.

[0011] In a seventh aspect, a network-side device is provided, including a processor and a communication interface, wherein the processor is used to implement the steps of the method described in the first aspect, and the communication interface is used to couple with the processor.

[0012] In an eighth 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.

[0013] A ninth aspect provides a wireless communication system, comprising: a terminal and a network-side device, wherein the terminal is configured to perform the steps of the method described in the first aspect, and the network-side device is configured to perform the steps of the method described in the first aspect.

[0014] In a tenth aspect, a chip is provided, the chip including a processor and a communication interface coupled to the processor, the processor being configured to run a program or instructions to implement the steps of the method described in the first aspect.

[0015] Eleventhly, 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.

[0016] In this embodiment, the first device obtains first information based on an artificial intelligence (AI) model and performs RACH transmission based on the first information. The first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters. The above process can determine the resources for RACH transmission based on the AI ​​model, thereby improving resource utilization. Furthermore, performing RACH transmission based on the resources can avoid unnecessary RACH retransmissions, saving resources and reducing overhead, thereby reducing device energy consumption and improving device performance. Attached Figure Description

[0017] Figure 1 This diagram illustrates a block diagram of a wireless communication system to which embodiments of this application may be applied;

[0018] Figure 2 This illustration shows a flowchart of a random access transmission method provided in an embodiment of this application.

[0019] Figure 3This illustration shows another flowchart of the random access transmission method provided in an embodiment of this application;

[0020] Figure 4 This illustration shows another flowchart of the random access transmission method provided in an embodiment of this application;

[0021] Figure 5 This illustration shows a structural schematic diagram of a random access transmission apparatus provided in an embodiment of this application;

[0022] Figure 6 This illustration shows a structural diagram of a communication device provided in an embodiment of this application;

[0023] Figure 7 This illustration shows a hardware structure diagram of a terminal provided in an embodiment of this application;

[0024] Figure 8 This diagram illustrates the hardware structure of a network-side device according to an embodiment of this application. Detailed Implementation

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] Figure 1This diagram illustrates a block diagram of a wireless communication system applicable to embodiments 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 devices (home appliances with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), game consoles, personal computers (PCs), ATMs, or self-service machines, 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.

[0030] 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 Function. Support Functions (BSF), Application Functions (AF), Location Management Functions (LMF), Gateway Mobile Location Centres (GMLC), and Network Data Analytics Functions (NWDAF), etc. 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.

[0031] 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).

[0032] The random access transmission method, terminal, and network-side equipment provided in this application will be described in detail below with reference to the accompanying drawings and through some embodiments and application scenarios.

[0033] This application relates to artificial intelligence (AI) models, including but not limited to various implementation methods such as neural networks, decision trees, support vector machines, or Bayesian classifiers. Generally, the selected AI algorithm and AI model vary depending on the type of problem. In some scenarios, neural network-based algorithms and AI models can achieve better performance than deterministic algorithms. Commonly used neural networks include deep neural networks, convolutional neural networks, and recurrent neural networks. AI models enable the construction, training, and validation of neural networks.

[0034] In practice, due to the insufficient size of real-time acquired datasets, directly training a neural network often fails to achieve convergence. A common approach is to pre-train the network using a large amount of offline collected data until it converges. Then, the parameters of the pre-trained neural network are fine-tuned using real-time acquired data to adapt the network to the real-world environment. Fine-tuning can be considered a training process that uses the parameters of the pre-trained neural network as initialization.

[0035] In machine learning and deep learning, a label typically refers to the identification or annotation of the true class or target value of a data sample. Labels are used to represent the information that the model should learn and predict, including but not limited to: labels in classification tasks, labels in object detection, labels in regression tasks, and labels in sequence labeling. Labels are a key component in supervised learning tasks, used to train machine learning models; the quality and accuracy of labels are crucial to the model's performance.

[0036] AI model lifecycle management comprises several AI functional modules: model training, model deployment, model inference, model monitoring, and model updates. Performing AI model training, validation, and testing generates model performance metrics that can be used as part of the model testing process. If a model storage function is available, it's used to transfer trained, validated, and tested AI models to the storage function, or to transfer updated versions of the model. Monitoring AI model operations or the deployment of AI functions (such as model selection / activation / deactivation / switching / rollback) provides feedback on model monitoring performance and makes decisions based on data received from the data collection and inference modules to ensure correct inference operations. Model inference can use inference data provided by the data collection function as input to provide the output of the applied AI model.

[0037] Figure 2 This diagram illustrates a flowchart of a random access transmission method provided in an embodiment of this application. Figure 2 As shown, the method 200 may include the following steps.

[0038] S202: The first device obtains the first information based on the AI ​​model.

[0039] The first piece of information may include at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

[0040] S204: RACH transmission based on the first information.

[0041] In this embodiment, the first device can be a terminal or a network-side device. The network-side device includes, but is not limited to, one of the following: a base station, a TRP, a core network device, or a server. The base station includes, but is not limited to, one of the following: the base station of the cell the terminal is currently camped on, the base station of the cell the terminal is currently accessing, the base station of the target cell after the terminal's handover, the base station of the target cell after the terminal's reselection, the base station of the Pcell the terminal is currently accessing, the base station of the Scell ​​the terminal is currently accessing, etc.

[0042] The aforementioned server refers to network-side devices used for training, inference, supervision, or providing AI-related information. This server includes, but is not limited to, one of the following: servers operating across different carriers (Over-The-Top, OTT), servers from third-party service providers, or internet servers.

[0043] It should be noted that in the scenario where the server obtains the first information based on the AI ​​model, the above step S204 can be performed by other devices, such as terminals, base stations, TRPs, or core network equipment. In other words, in this scenario, the server does not participate in the RACH transmission process; its function is to obtain the first information based on the AI ​​model.

[0044] In this embodiment of the application, RACH transmission based on the first information may include one of the following:

[0045] The first method: Directly use the first information for RACH transmission;

[0046] The second method is to select a RACH resource that meets the requirements for RACH transmission based on the first information. The RACH resource that meets the requirements refers to the RACH resource that can avoid conflicts.

[0047] For example, based on the beam information and / or RACH collision probability in the first information, select the required SSB, CSI-RS, preamble index, PRACH resources for transmitting the preamble, or PRACH power, etc., for RACH transmission.

[0048] The first method described above simplifies the AI-assisted random access transmission process. The second method flexibly obtains suitable RACH resources based on the output information of the AI ​​model, avoiding access failures due to resource conflicts, improving the accuracy and efficiency of RACH transmission, ensuring RACH transmission performance, and reducing the energy consumption of the first device.

[0049] In this embodiment of the application, the RACH transmission described above may include at least one of the following:

[0050] 1) RACH transmission corresponding to the first RACH resource, wherein the first RACH resource may include at least one of the following: random access timing (RO) resource, preamble format, preamble index, and preamble sequence;

[0051] 2) RACH transmission corresponding to the first reference signal, wherein the first reference signal may include at least one of the following: synchronization signal physical broadcast channel block (SSB) and channel state information reference signal (CSI-RS);

[0052] 3) RACH transmission corresponding to the first Transmission Configuration Indicator (TCI);

[0053] 4) RACH transmission corresponding to the first TRP.

[0054] In this embodiment, RO resources may include at least one of the following: time-domain resources, frequency-domain resources, number, number of transmissions, number of repetitions, etc. of the RO. The time-domain resources of the RO may include: the time unit in which the RO is located, such as a slot, symbol, or subframe. The frequency-domain resources of the RO may include: the frequency-domain unit in which the RO is located, such as a bandwidth part (BWP), subband, RB set, resource block (RB), carrier, etc.

[0055] In this embodiment, beam information can also be referred to as spatial characteristics, such as TCI or QCL. The beam information in the first information may include at least one of the following:

[0056] 1) Beam information of the terminal;

[0057] 2) Beam information between the terminal and the first cell, wherein the first cell may include one of the following: source cell, target cell, currently camped cell, currently accessed cell, candidate cell, primary cell, secondary cell;

[0058] 3) Beam information between the terminal and the first transmit / receive point (TRP), wherein the first TRP may include one of the following: source TRP, target TRP, currently camped TRP, currently accessing TRP, candidate TRP, primary TRP, secondary TRP;

[0059] 4) Beam information within the preset area. It should be noted that the preset in the embodiments of this application may be pre-configured, given, or specific, and is not limited in specific terms. The presets mentioned below all have the same meaning and will not be declared one by one.

[0060] 5) Beam information for preset path loss range;

[0061] 6) Beam information for preset signal quality strength range;

[0062] 7) Beam information for preset RACH configuration parameters.

[0063] In addition, regardless of which of the seven types of beam information is mentioned above, it can include at least one of the following: reference signal index, beam index, and beam direction. The reference signal corresponding to the reference signal index can include at least one of the following: SSB, CSI-RS.

[0064] The aforementioned first device obtains beam information based on an AI model. When the first information includes beam information, it can select appropriate RACH resources for RACH transmission based on the beam information, thereby improving the success rate of RACH transmission and reducing retransmissions after RACH transmission failures due to collisions, thus reducing the power consumption of the device.

[0065] In this embodiment of the application, the RACH collision probability in the first information may include at least one of the following:

[0066] 1) The RACH collision probability of the terminal;

[0067] 2) RACH collision probability between the terminal and the first cell, wherein the first cell may include one of the following: source cell, target cell, currently camped cell, currently accessed cell, candidate cell, primary cell, secondary cell;

[0068] 3) RACH collision probability between the terminal and the first TRP, wherein the first TRP may include one of the following: source TRP, target TRP, currently residing TRP, currently accessing TRP, candidate TRP, primary TRP, secondary TRP;

[0069] 4) Preset RACH collision probability for the beam direction;

[0070] 5) RACH collision probability within the preset area;

[0071] 6) RACH collision probability within the preset road loss range;

[0072] 7) Preset RACH collision probability within the signal quality strength range;

[0073] 8) RACH collision probability of preset RACH configuration parameters.

[0074] The first device described above obtains the RACH collision probability based on an AI model. That is, when the first information includes the RACH collision probability, it can select appropriate RACH resources for RACH transmission based on the RACH collision probability, thereby improving the success rate of RACH transmission and reducing retransmissions after RACH transmission failures due to collisions, thus reducing the power consumption of the device.

[0075] In this embodiment, the RACH resource information in the first information may include one of the following: a set of candidate RACH resource information, a set of candidate RACH resource information, and probability values ​​associated with each candidate RACH resource information. A set of candidate RACH resource information refers to two or more candidate RACH resource information. The probability value represents the probability that the corresponding candidate RACH resource information will ultimately be used for RACH transmission. The first device can select the corresponding RACH resource information for RACH transmission based on the probability values ​​associated with each candidate RACH resource information, such as selecting the candidate RACH resource information with the highest probability value as the RACH resource information used for RACH transmission.

[0076] In this embodiment of the application, the RACH resource information in the first information, whether it is one or a group, can include at least one of the following: RO resource, preamble resource, and associated reference signal resource.

[0077] In one implementation, the RACH resource information in the first information may include at least one of the following:

[0078] 1) Terminal RACH resource information;

[0079] 2) RACH resource information between the terminal and the first cell, wherein the first cell may include one of the following: source cell, target cell, currently camped cell, currently accessed cell, candidate cell, primary cell, secondary cell;

[0080] 3) RACH resource information between the terminal and the first TRP, wherein the first TRP may include one of the following: source TRP, target TRP, currently residing TRP, currently accessing TRP, candidate TRP, primary TRP, secondary TRP;

[0081] 4) RACH resource information for preset beam directions;

[0082] 5) RACH resource information within the preset area;

[0083] 6) RACH resource information for preset road loss range;

[0084] 7) RACH resource information for preset signal quality strength range;

[0085] 8) RACH resource information with preset RACH configuration parameters, wherein the RACH configuration parameters may include at least one of the following: preset preamble sequence, preamble format, RO, and RO group.

[0086] The first device described above obtains RACH resource information based on an AI model. That is, when the first information includes RACH resource information, it can obtain RACH resource information such as RACH RO resources, preamble resources, or associated reference signal resources. RACH transmission can be performed directly based on the RACH resources obtained from the AI ​​model, thereby simplifying the RACH resource selection process and improving the efficiency of RACH transmission.

[0087] In this embodiment, the transmit power parameter in the first information may include one of the following: a set of candidate RACH transmission transmit power parameters, a set of candidate RACH transmission transmit power parameters, and probability values ​​associated with each candidate RACH transmission transmit power parameter. A set of candidate RACH transmission transmit power parameters refers to two or more candidate RACH transmission transmit power parameters. The probability value represents the probability that the corresponding candidate RACH transmission transmit power parameter will ultimately be used for RACH transmission. The first device can select the corresponding RACH transmission transmit power parameter for RACH transmission based on the probability values ​​associated with each candidate RACH transmission transmit power parameter, such as selecting the candidate RACH transmission transmit power parameter with the highest probability value as the RACH transmission transmit power parameter used for RACH transmission.

[0088] In this embodiment, the transmit power parameter in the first information, whether singular or a set, can include at least one of the following: transmit power, target receive power value, maximum transmit power, power boost value, and power fallback value. The power boost value refers to how much the power has increased, such as 3dB or 6dB, and this value is related to factors such as device load, interference, power consumption, and type. It should be noted that the power values ​​involved in this embodiment can refer to absolute probability values ​​or relative power values; there is no specific limitation.

[0089] For example, the transmit power parameters in the first information can be Tx power, PCmax, and Power ramping values, or Tx power, PCmax, and Power ramping values ​​of RACH-related PUSCH, such as MsgAPUSCH, msg3PUSCH, etc.

[0090] In one implementation, the transmit power parameter in the first information may include at least one of the following:

[0091] 1) RACH transmission transmit power parameters of the terminal;

[0092] 2) RACH transmission transmit power parameters between the terminal and the first cell, wherein the first cell may include one of the following: source cell, target cell, currently camped cell, currently accessed cell, candidate cell, primary cell, secondary cell;

[0093] 3) RACH transmission transmit power parameters between the terminal and the first TRP, wherein the first TRP may include one of the following: source TRP, target TRP, currently camped TRP, currently accessing TRP, candidate TRP, primary TRP, secondary TRP;

[0094] 4) Preset RACH transmission transmit power parameters for the beam direction;

[0095] 5) RACH transmission transmit power parameters within the preset area;

[0096] 6) RACH transmission transmit power parameters within the preset path loss range;

[0097] 7) Preset RACH transmission transmit power parameters within the preset signal quality strength range;

[0098] 8) RACH transmission transmit power parameters of preset RACH configuration parameters.

[0099] The aforementioned method of obtaining the transmit power parameters based on an AI model, i.e., when the first information includes the transmit power parameters, can effectively improve the performance of RACH transmission and also avoid the problems of increased interference and increased device power consumption caused by using excessively high transmit power.

[0100] In this embodiment of the application, the AI ​​model can obtain output information after reasoning through the input information. The input information of the AI ​​model can be one or more pieces, and the output information can also be one or more pieces; there is no specific limitation.

[0101] In this embodiment of the application, the input information of the AI ​​model may include at least one of the following:

[0102] 1) Signal strength information between the first and second devices;

[0103] The aforementioned signal strength information may include one of the following:

[0104] 1a) Uplink signal reception strength information between the first device and the second device, wherein the uplink signal reception strength information may include at least one of the following: reference signal received power RSRP, reference signal received quality RSRQ, and received signal strength indication RSSI obtained based on uplink signal measurements; if the first device is a terminal, the reception strength information is obtained by the terminal measurement; if the first device is a network-side device, the reception strength information is obtained by the terminal measurement and reporting.

[0105] 1b) Downlink signal reception strength information between the first device and the second device, wherein the downlink signal reception strength information may include at least one of the following: RSRP, RSRQ, RSSI obtained based on downlink signal measurements; if the first device is a terminal, the reception strength information is obtained by the terminal measurement; if the first device is a network-side device, the reception strength information is obtained by the terminal measurement and reporting.

[0106] 2) Signal quality information between the first and second devices;

[0107] The aforementioned signal quality information may include one of the following:

[0108] 2a) Signal quality information of the uplink signal between the first device and the second device;

[0109] 2b) Signal quality information of the downlink signal between the first and second devices;

[0110] The uplink signals involved in the above input information 1)-2) may include at least one of the following: uplink detection signal, uplink reference signal, and uplink synchronization signal. The downlink signals involved in the above input information 1)-2) may include at least one of the following: downlink broadcast signal, synchronization signal, and reference signal.

[0111] The signal quality information involved in the above input information 2a)-2b) may include at least one of the following: received signal-to-interference-plus-noise ratio (SINR), signal-to-noise ratio (SNR), and latency.

[0112] 3) Path loss between the first and second devices;

[0113] 4) Distance information between the first and second devices;

[0114] 5) The number of random access failures in RACH transmission between the first and second devices;

[0115] The number of random access failures in the aforementioned RACH transmission may include at least one of the following:

[0116] 5a) The number of random access failures in the RACH transmission corresponding to the first RACH resource, wherein the first RACH resource may include at least one of the following: RO resource, preamble format, preamble index, and preamble sequence;

[0117] 5b) The number of random access failures in the RACH transmission corresponding to the first reference signal, wherein the first reference signal may include at least one of the following: SSB, CSI-RS;

[0118] 5c) The number of random access failures in the RACH transmission corresponding to the first TCI;

[0119] 5d) The number of random access failures in the RACH transmission corresponding to the first TRP.

[0120] 6) The number of times the random access response message reception failed during RACH transmission between the first and second devices;

[0121] The number of times the random access response message reception failed during the aforementioned RACH transmission may include at least one of the following:

[0122] 6a) The number of times the random access response message reception failed for the RACH transmission corresponding to the first RACH resource, wherein the first RACH resource may include at least one of the following: RO resource, preamble format, preamble index, and preamble sequence;

[0123] 6b) The number of times the random access response message received by the RACH transmission corresponding to the first reference signal failed, wherein the first reference signal may include at least one of the following: SSB, CSI-RS;

[0124] 6c) The number of times the random access response message was received during the RACH transmission corresponding to the first TCI;

[0125] 6d) The number of times the random access response message was received failed during the RACH transmission corresponding to the first TRP.

[0126] 7) The maximum number of RACH transmissions between the first and second devices;

[0127] The maximum number of RACH transmissions mentioned above may include at least one of the following:

[0128] 7a) The maximum number of RACH transmissions corresponding to the first RACH resource, wherein the first RACH resource may include at least one of the following: RO resource, preamble format, preamble index, and preamble sequence;

[0129] 7b) The maximum number of RACH transmissions corresponding to the first reference signal, wherein the first reference signal may include at least one of the following: SSB, CSI-RS;

[0130] 7c) The maximum number of RACH transmissions corresponding to the first TCI;

[0131] 7d) The maximum number of RACH transmissions corresponding to the first TRP.

[0132] 8) The number of retransmissions of RACH transmission between the first and second devices;

[0133] The number of retransmissions in the aforementioned RACH transmission may include at least one of the following:

[0134] 8a) The number of retransmissions of the RACH transmission corresponding to the first RACH resource, wherein the first RACH resource may include at least one of the following: RO resource, preamble format, preamble index, and preamble sequence;

[0135] 8b) The number of retransmissions of the RACH transmission corresponding to the first reference signal, wherein the first reference signal may include at least one of the following: SSB, CSI-RS;

[0136] 8c) The number of retransmissions of the RACH transmission corresponding to the first TCI;

[0137] 8d) The number of retransmissions of the RACH transmission corresponding to the first TRP.

[0138] 9) The number of failures when switching between different RACH transmission types between the first and second devices;

[0139] The RACH transmission types include: two-step random access procedure (also known as Type-2 random access procedure) and four-step random access procedure (also known as Type-1 random access procedure). The aforementioned handover includes: switching from a two-step random access procedure to a four-step random access procedure, and switching from a four-step random access procedure to a two-step random access procedure, which is not specifically limited.

[0140] 10) The absolute timing advance (TA) value between the first and second devices;

[0141] The TA value mentioned above can be an absolute TA value used between the first device and the second device, currently, within a specified time period, or recently, such as the RAR TA value, and is not specifically limited.

[0142] 11) The relative TA value between the first and second devices;

[0143] The aforementioned relative TA value can be a relative TA value between the first device and the second device, currently, within a specified time period, or recently used, such as the MAC CE TA value, and is not specifically limited.

[0144] 12) Beam information between the first and second devices;

[0145] 13) Propagation delay between the first device and the second device, wherein the propagation delay can be the signal propagation delay between uplink, downlink, sidelink, or network nodes;

[0146] 14) Round-trip time (RTT) between the first and second devices;

[0147] In this embodiment of the application, the second device refers to a device that performs RACH transmission with the first device, including one of the following scenarios: the first device is a terminal and the second device is a network-side device; the first device is a network-side device and the second device is a terminal; the specific scenario is not limited.

[0148] In one implementation, the first device involved in the above input information 1)-14) can be a terminal, and the second device involved can be one of the following: source cell, target cell, currently camped cell, currently accessed cell, network node of PCell, network node of SCell.

[0149] 15) Maximum transmit power for RACH transmission;

[0150] The maximum transmit power of the aforementioned RACH transmission may include at least one of the following:

[0151] 15a) The maximum transmit power of RACH transmission corresponding to the first RACH resource, wherein the first RACH resource may include at least one of the following: RO resource, preamble format, preamble index, and preamble sequence;

[0152] 15b) The maximum transmit power of the RACH transmission corresponding to the first reference signal, wherein the first reference signal may include at least one of the following: SSB, CSI-RS;

[0153] 15c) The maximum transmit power of the RACH transmission corresponding to the first TCI;

[0154] 15d) The maximum transmit power of the RACH transmission corresponding to the first TRP.

[0155] Furthermore, regardless of which of the above-mentioned maximum transmit power for RACH transmission, it can be the maximum transmit power of the first device for RACH transmission in at least one of the following: cell, cell group, frequency layer, TA area, TRP, network node, carrier, band, subband, and partial bandwidth.

[0156] 16) TRP identifier, wherein the TRP identifier corresponds to at least one of the following TRPs: source TRP, target TRP, currently selected TRP, residing TRP, currently accessed TRP, primary TRP, secondary TRP;

[0157] 17) TRP group identifier, wherein the TRP group identifier corresponds to at least one of the following TRP groups: source TRP group, target TRP group, currently selected TRP group, residing TRP group, currently accessed TRP group, primary TRP group, secondary TRP group;

[0158] 18) Cell identifier, wherein the cell identifier can be the physical layer cell identifier of the first cell;

[0159] 19) Cell group identifier, wherein the cell group identifier can be the identifier of the first cell group, and the first cell group can include one of the following: source cell group, target cell group, currently camped cell group, currently accessing cell group, primary cell group, secondary cell group;

[0160] 20) Time Advance Group (TAG) identifier, where the TAG identifier can be the TAG identifier of the first cell;

[0161] 21) Tracking area identifier, wherein the tracking area identifier can be the tracking area identifier of the first cell;

[0162] 22) Radio Access Network Notification Area (RNA) identifier, wherein the RNA identifier can be the RNA identifier of the first cell;

[0163] In this embodiment of the application, the first cell involved in the above input information 20)-24) may include one of the following: source cell, target cell, currently camped cell, currently accessed cell, primary cell, and secondary cell.

[0164] 23) Frequency domain resource information, which includes at least one of the following: frequency band, frequency zone, frequency point, carrier frequency, frequency layer, and BWP;

[0165] 24) Load information, wherein the load information can be the load information corresponding to at least one of SSB, beam, TCI, TRP, cell, carrier and frequency point. For example, the load information can be the traffic volume or the number of users.

[0166] 25) Interference information, wherein the interference information may be interference information corresponding to at least one of SSB, beam, TCI, TRP, cell, carrier and frequency point, and for example, the interference information may be interference measurement value;

[0167] 26) Information of the first device; wherein the information of the first device may include at least one of the following: transmission power information, location information, distribution information, direction of movement, speed of movement, energy consumption information, power consumption information, operator information, network type information, panel orientation information, device type, sensing information, network scene information, environmental information (such as weather information), ephemeris information, reference position of the cell in the Non-Terrestrial Network (NTN) scene, movement trajectory of the cell in the NTN scene, multipath information of the channel (such as first path or strongest path information), time information, compensation information, and offset information.

[0168] In this embodiment of the application, the location information of the first device may be specific geographical coordinates (such as GPS coordinates), or approximate location range information (such as which street it is located on), or location information relative to the host cell, access cell, a certain TRP or a group of TRPs (such as due east of the host cell), and there is no specific limitation.

[0169] In this embodiment of the application, the distribution information of the first device may include: the number of the first device in different areas (such as the stationed cell, the access cell, a certain TRP or a group of TRPs), etc., and is not specifically limited.

[0170] In this embodiment, the direction of movement of the first device can be an absolute direction, such as 40 degrees east of south; or it can be a relative direction, such as the direction relative to a certain base station or reference node, and the specific direction is not limited.

[0171] In this embodiment, the sensing information of the first device may include: communication environment information, scene information, channel state information (such as LOS, NLOS, obstruction), or the number of terminals. For example, sensing whether there are obstacles under a specific beam and the number of obstacles, or sensing the number of terminals or network-side devices covered by a specific beam.

[0172] In this embodiment of the application, network scene information may include: inH, Uma, RMa, homogeneous network, or heterogeneous network (with or without overlapping coverage), etc., and is not specifically limited.

[0173] In this embodiment, the reference position of a cell in an NTN scenario, such as in a GEO scenario, is fixed on Earth as the satellite moves, and can be considered to have a reference position. In an NTN scenario, such as in a LEO (Low-Earth Orbit) scenario, the cell's trajectory moves on Earth as the satellite moves.

[0174] In this embodiment, the time information of the first device can be a precise moment, such as 13:25:38, accurate to the second; or it can be a time range, such as 13:00 to 14:00, morning, afternoon, daytime, or nighttime. Alternatively, this time information can be timing information obtained through other radio access technologies (RATs), such as timing information obtained through Bluetooth, Wi-Fi, 3G, 4G, or 5G.

[0175] In this embodiment, the compensation information of the first device may include time-domain compensation information and / or frequency-domain compensation information. The time-domain compensation information may be timing pre-compensation, and the frequency-domain compensation information may be frequency compensation.

[0176] In one implementation, the compensation information of the first device may include at least one of the following:

[0177] 1) Timing pre-compensation, wherein the offset information includes at least one of the following: timing advance offset, common timing advance offset, and dedicated timing advance offset;

[0178] 2) Frequency compensation, wherein the offset information includes at least one of the following: frequency offset, common frequency offset, and proprietary frequency offset.

[0179] In this embodiment, the process of obtaining the first information based on the AI ​​model is the AI ​​model's reasoning process. The AI ​​model's reasoning can be triggered by a triggering condition, i.e., the AI ​​model is triggered to perform reasoning when the triggering condition is met; alternatively, no triggering condition may be set, i.e., the AI ​​model is triggered to perform reasoning at any time or randomly, without specific limitations. Furthermore, the aforementioned triggering conditions can be used to trigger AI model reasoning or to set the AI ​​model to an enabled state, without specific limitations.

[0180] In this embodiment of the application, the triggering condition for AI model inference may include at least one of the following:

[0181] 1) AI timer timed out;

[0182] 2) Use RACH configuration with network settings;

[0183] 3) Use the RACH type configured in the network;

[0184] 4) The collision probability of RACH transmission is greater than or equal to the first threshold;

[0185] 5) The number of random access failures in the RACH transmission corresponding to the first information is greater than or equal to the second threshold;

[0186] 6) The number of failed receptions of the random access response message transmitted in the RACH corresponding to the first information is greater than or equal to the third threshold;

[0187] 7) The number of RACH transmissions corresponding to the first information that reach the maximum number of transmissions is greater than or equal to the fourth threshold;

[0188] 8) The number of repeated transmissions used in the RACH transmission corresponding to the first information is greater than or equal to the fifth threshold;

[0189] 9) The number of failed attempts to switch between different RACH transmission types is greater than or equal to the sixth threshold;

[0190] The RACH transmission types include: two-step random access procedure (also known as Type-2 random access procedure) and four-step random access procedure (also known as Type-1 random access procedure). The aforementioned handover includes: switching from a two-step random access procedure to a four-step random access procedure, and switching from a four-step random access procedure to a two-step random access procedure, which is not specifically limited.

[0191] 10) The RACH transmission corresponding to the first message reaches the maximum transmit power;

[0192] 11) The transmit power of the RACH transmission corresponding to the first information is greater than or equal to the seventh threshold. For example, the difference between the transmit power of the RACH transmission corresponding to the first information and the maximum transmit power is greater than or equal to a specified threshold.

[0193] 12) The number of power boosts for the RACH transmission corresponding to the first information is greater than or equal to the eighth threshold;

[0194] 13) The number of failed RACH transmissions corresponding to different first information is greater than or equal to the ninth threshold;

[0195] The first information involved in the triggering conditions 5)-8) and 10)-13) above may include at least one of the following: RO resources, preamble format, preamble index, preamble sequence, SSB, beam, TCI, TRP.

[0196] 14) Community re-selection;

[0197] 15) TRP reselection;

[0198] 16) Cell handover;

[0199] 17) TRP switching;

[0200] 18) The cell handover conditions are met, which may include at least one of the following, for example: the measurement results of the source cell (RSRP, RSRQ or RSSI of L1 or L3, etc.) are greater than or less than a specified threshold; the measurement results of the target cell or neighboring cell are better than those of the source cell; the measurement results of the target cell or neighboring cell are greater than a specified threshold; the measurement results of the source cell are less than a specified threshold and the measurement results of the target cell or neighboring cell are greater than another specified threshold; the load of the source cell is greater than a specified threshold.

[0201] 19) The first device triggers access to the secondary cell;

[0202] 20) The first device moves to a preset position (e.g., the edge of the cell);

[0203] 21) Upstream business arrives;

[0204] 22) Downlink traffic arrives;

[0205] 23) Preset type of service arrival (such as services with high latency requirements or high reliability requirements).

[0206] 24) Network indication triggered;

[0207] 25) The speed of the first device is greater than or equal to the tenth threshold;

[0208] 26) The acceleration of the first device is greater than or equal to the eleventh threshold;

[0209] 27) The position of the first piece of equipment has changed;

[0210] 28) The network location has changed;

[0211] 29) The position change of the first device is greater than or equal to the twelfth threshold;

[0212] 30) The change in network location is greater than or equal to the thirteenth threshold;

[0213] 31) The beam selected by the first device changes;

[0214] 32) The spatial transmission filter selected by the first device changes;

[0215] In this embodiment of the application, the first to thirteenth thresholds can be preset, and the specific values ​​are not limited.

[0216] It should be noted that the AI ​​model can be triggered to perform inference after at least one of the above 32 triggering conditions is met. Alternatively, the AI ​​model can be enabled after at least one of the conditions is met (i.e., the user has permission to use the AI ​​model). However, whether to use the AI ​​model for inference and when to use it can be determined by considering the relevant capabilities of the AI ​​model, its application scope, or instructions from the network side, and there are no specific limitations.

[0217] In this embodiment, step S202 is the AI ​​model inference step. The inference metrics of the AI ​​model may include at least one of the following: AI model complexity, AI model inference latency (e.g., not exceeding a specified value), AI model inference success rate (e.g., not less than one specified value or greater than another specified value), and AI model inference result reliability (e.g., the probability that the inference result meets performance requirements is greater than or equal to a specified value). These inference metrics can be used to measure the accuracy and precision of the AI ​​model, and can also guide AI model training. By training to achieve preset inference metrics, the accuracy of AI model inference can be improved.

[0218] The complexity of the AI ​​model can be set based on different devices, such as setting the complexity of the AI ​​model on the terminal or the complexity of the AI ​​model on the base station.

[0219] In this embodiment of the application, the method may further include: a first device acquiring first indication information, and activating or deactivating the AI ​​model based on the first indication information. The first indication information is activation or deactivation information for the AI ​​model. Activating the AI ​​model indicates that it can enter an enabled state and can be used for inference. Deactivating the AI ​​model indicates that it enters a disabled state, in which case inference cannot be performed and the model must wait for reactivation before inference can be executed.

[0220] In this embodiment of the application, before using the AI ​​model for inference, the above method may further include: obtaining first configuration information of the AI ​​model. The first configuration information may include at least one of the following:

[0221] 1) AI model;

[0222] 2) Identification of AI models;

[0223] 3) The application scope of AI model inference, wherein the application scope of AI model inference may include at least one of the following: the frequency domain range of AI model inference application, the cell (which may be one or more) of AI model inference application, the range of the terminal's location, and the range of the distance between the terminal and the base station.

[0224] 4) The inference cycle of the AI ​​model;

[0225] 5) The effective duration of AI model inference;

[0226] 6) Triggering conditions for AI model inference;

[0227] 7) Configuration parameters of the AI ​​model, wherein the configuration parameters of the AI ​​model may include at least one item: input information of the AI ​​model, output information of the AI ​​model, type of input information of the AI ​​model, and type of output information of the AI ​​model;

[0228] 8) Does it support joint inference across multiple devices?

[0229] In this embodiment of the application, before step S202, the following may be included: the first device sends third information to the network side, receives an instruction sent by the network side, and determines the AI ​​model to be used among multiple AI models based on the instruction.

[0230] Among them, any two of the multiple AI models have differences, which include at least one of the following: differences in training datasets, differences in label information, differences in input information, and differences in output information.

[0231] The third piece of information may include at least one of the following: identifiers of multiple AI models and classification identifiers of datasets. The aforementioned instructions may include at least one of the following: AI model IDs, dataset identifiers, and data collection configuration information.

[0232] In this embodiment of the application, the above method may further include: performing at least one of the following first operations:

[0233] 1) Send one or more of the input information, output information, and current state information of the first device to the device that trained the AI ​​model;

[0234] 2) If the first information obtained based on the AI ​​model is determined to be paused or abandoned, the result of the AI ​​model inference does not meet expectations. The result can be abandoned, or the AI ​​model inference can be abandoned during the next RACH transmission (i.e., paused once), or the AI ​​model inference can be abandoned in all subsequent RACH transmissions (i.e., abandoned and no longer used for AI model inference).

[0235] 3) Triggering adjustments to the AI ​​model, where adjustments to the AI ​​model may include at least one of the following: model update, model change, model switching, model fine tuning, changes to model input information, changes to model output information, etc.

[0236] The first operation described above can be triggered when at least one of the following conditions is met:

[0237] 1) The RACH transmission information that does not meet the performance requirements is not obtained after the AI ​​model inference exceeds M time, where M can be preset and the specific value is not limited;

[0238] 2) The AI ​​model failed to complete the AI ​​inference.

[0239] 3) Successfully completed AI inference using the AI ​​model;

[0240] 4) The AI ​​model was used to perform AI inference N times, where N is an integer that can be preset and the specific value is not limited.

[0241] In this embodiment of the application, before training, inference, or supervising the AI ​​model, the method may further include: the first device sending indication information of its AI capability information to the second device. The AI ​​capability information may include at least one of the following:

[0242] 1) The first device may or may not have the ability to train AI models;

[0243] 2) The first device may or may not have the ability to use AI models for AI reasoning;

[0244] 3) The first device may or may not have the ability to send auxiliary information, which is used for AI model inference.

[0245] The first device that sends the instruction information on AI capabilities may include one of the following: a terminal, a base station, a TRP, a core network device, or a server, without specific limitations.

[0246] In this embodiment of the application, AI capability information can be determined through one or more of the following methods:

[0247] 1) It depends on the type of terminal, such as introducing different AI-related capabilities for different types of terminals (e.g., Redcap, IoT);

[0248] 2) It depends on the network type, such as introducing different AI-related capabilities for different network types (e.g., NTN, TN);

[0249] 3) Indicated by one or more reference signals, such as specific resources of PRACH (e.g., specific RO or preamble), indicating that the first device has the ability to train or infer AI models;

[0250] 4) Carried by uplink control information, such as physical layer control information (e.g., UCI information reported to the network for the terminal);

[0251] 5) Carried by RRC signaling;

[0252] 6) Carried by specific interface messages between the terminal and the server, which can be: messages related to a specific AI model or messages related to all AI models;

[0253] 7) Carried by a specific interface message between the terminal and the network, which can be: a message related to a specific AI model or a message related to all AI models;

[0254] 8) Carried by specific interface messages between the server and the network, which can be: messages related to a specific AI model or messages related to all AI models.

[0255] The method provided in this application embodiment involves a first device obtaining first information based on an AI model and performing RACH transmission based on the first information. The first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters. The above process can determine the resources for RACH transmission based on the AI ​​model, thereby improving resource utilization. Furthermore, performing RACH transmission based on the resources can avoid unnecessary RACH retransmissions, saving resources and reducing overhead, thus reducing device energy consumption and improving device performance.

[0256] Figure 3 This illustration shows another flowchart of the random access transmission method provided in an embodiment of this application. Figure 3 As shown, the method 300 may include the following steps.

[0257] S302: Train the AI ​​model.

[0258] In this embodiment, the training of the AI ​​model can be performed before inference to obtain the AI ​​model. Alternatively, the AI ​​model can be trained online to update it, thereby improving its accuracy and reliability. The training of the AI ​​model can be performed by a terminal, base station, TRP, core network equipment, or server, and is not specifically limited to any particular device. The server includes, but is not limited to, one of the following: an OTT server, a third-party service provider's server, or an internet server.

[0259] S304: Obtain the first information based on the AI ​​model.

[0260] The first piece of information may include at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

[0261] S306: RACH transmission based on the first information.

[0262] In this embodiment, step S302 is the step of training the AI ​​model, and step S304 is the step of using the AI ​​model for inference. These two steps can be performed by the same device, or they can be performed by different devices; there is no specific limitation.

[0263] For example, in the first scenario: Device A first trains an AI model, and then uses the AI ​​model to perform inference to obtain the first information. In this scenario, the same device can complete both training and inference, which is highly efficient and saves transmission resources as there is no need to obtain the AI ​​model from other devices.

[0264] For example, in the second scenario: Device B trains an AI model, and then Device C uses the AI ​​model to perform inference and obtain the first information. In this scenario, different devices cooperate, and Device C needs to obtain the AI ​​model from Device B before performing inference. That is to say, in the second scenario, step S304 specifically includes: first obtaining the AI ​​model from the device that trained the AI ​​model, and then obtaining the first information based on the AI ​​model. The advantage of this scenario is that it eliminates the need to train the AI ​​model locally on the inference device, thus saving the processing resources of the inference device.

[0265] In this embodiment of the application, if the step of a first device acquiring the AI ​​model is included before performing AI model inference, it can include the following two scenarios: First scenario: The first device is a terminal, and the AI ​​model is acquired by the terminal. Second scenario: The first device is a network-side device, and the AI ​​model is acquired by the network-side device.

[0266] In the first scenario mentioned above, the steps for the terminal to acquire the AI ​​model can specifically include:

[0267] The terminal obtains the input information of the AI ​​model through the first message sent by the network side. The first message may include at least one of the following: Media Access Control Element (MAC CE), Radio Resource Control (RRC) message, Non-Access Stratum (NAS) message, User Plane Data, Downlink Control Information (DCI) information, System Information Block (SIB), Physical Downlink Control Channel (PDCCH), Physical Downlink Shared Channel (PDSCH), MSG 2 message, MSG 4 message, and MSG B message.

[0268] In the second scenario mentioned above, the steps for the network-side device to acquire the AI ​​model can specifically include:

[0269] Network-side devices obtain input information for the AI ​​model through a second message reported by the terminal. This second message may include at least one of the following: MAC CE, RRC message, NAS message, user plane data, MSG 1 message, MSG A message, MSG3 message, Physical Uplink Control Channel (PUCCH), Physical Uplink Shared Channel (PUSCH), Physical Random Access Channel (PRACH), SRS, and other uplink reference signals (such as WUS).

[0270] In this embodiment, the AI ​​model inference steps described above can be executed on the terminal, which has the advantage of enabling inference based on information obtained by the terminal in a timely manner, thus improving efficiency. The AI ​​model inference steps can also be executed on network-side devices, which has the advantage of reducing terminal complexity and cost, and lowering terminal power consumption. Alternatively, the AI ​​model inference steps can be executed on a server, such as an OTT server or a third-party server, which has the advantage of improving AI performance, reducing the complexity and cost of network-side devices and terminals, and lowering power consumption of network-side devices and terminals.

[0271] In one implementation, step S302 may specifically include: a third device acquiring training samples, the training samples including at least one label and input information used by the AI ​​model for inference; a fourth device using the training samples to train the model to obtain an AI model, and the AI ​​model outputting second information. This approach belongs to a scenario where the third and fourth devices jointly perform training, wherein either the third or fourth device can be a device performing inference, including one of the following: a terminal, a base station, a TRP, a core network device, or a server, without specific limitations.

[0272] The second information is used for RACH transmission and includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

[0273] In one implementation, the first device and the second device jointly train an AI model. In this case, the first device uses the trained AI model to perform the inference process, while the second device only participates in the training process.

[0274] In this embodiment of the application, in a scenario where AI model inference is performed by a network-side device, the input information of the AI ​​model can be indicated by a terminal or other network-side device through cross-carrier OTT messages.

[0275] For example, in a scenario where AI model inference is performed by a server, the input information of the AI ​​model can be indicated by a terminal, base station, TRP, core network equipment or other server via OTT messages.

[0276] In this embodiment of the application, step S302 may include one of the following:

[0277] The first scenario: Under the condition that the preset conditions are met, the training of the AI ​​model is triggered;

[0278] The second scenario: periodically triggering AI model training, such as triggering AI model training every K intervals, where K can be preset and the specific value is not limited;

[0279] The third scenario involves semi-static triggering of AI model training. Semi-static triggering refers to initiating AI model training only when semi-static configuration information is met. This semi-static configuration information can include at least one of the following: the start point of model training, the training period, and the duration within the period. Furthermore, the semi-static configuration information can be configured via RRC, and semi-static triggering of AI model training can be activated or deactivated via DCI or MAC-CE, with no specific limitations.

[0280] In the first scenario mentioned above, the preset conditions used to trigger AI model training may include at least one of the following:

[0281] 1) AI timer timed out;

[0282] 2) Use RACH configuration with network settings;

[0283] 3) Use the RACH type configured in the network;

[0284] 4) The collision probability corresponding to RACH transmission is greater than or equal to the fourteenth threshold;

[0285] 5) The number of random access failures in the RACH transmission corresponding to the first information is greater than or equal to the fifteenth threshold;

[0286] 6) The number of failed receptions of the access response message transmitted in the RACH corresponding to the first information is greater than or equal to the sixteenth threshold;

[0287] 7) The number of RACH transmissions corresponding to the first message that reach the maximum number of transmissions is greater than or equal to the seventeenth threshold;

[0288] 8) The number of repeated transmissions used in the RACH transmission corresponding to the first information is greater than or equal to the eighteenth threshold;

[0289] 9) The number of failed attempts to switch between different RACH transmission types is greater than or equal to the nineteenth threshold;

[0290] The RACH transmission types include: two-step random access procedure (also known as Type-2 random access procedure) and four-step random access procedure (also known as Type-1 random access procedure). The aforementioned handover includes: switching from a two-step random access procedure to a four-step random access procedure, and switching from a four-step random access procedure to a two-step random access procedure, which is not specifically limited.

[0291] 10) The RACH transmission corresponding to the first message reaches the maximum transmit power;

[0292] 11) The transmit power of the RACH transmission corresponding to the first information is greater than or equal to the twentieth threshold. For example, the difference between the transmit power of the RACH transmission corresponding to the first information and the maximum transmit power is greater than or equal to a specified threshold.

[0293] 12) The number of power boosts for the RACH transmission corresponding to the first information is greater than or equal to the twenty-first threshold;

[0294] 13) The number of failed RACH transmissions corresponding to different first information is greater than or equal to the twenty-second threshold;

[0295] The first information involved in the triggering conditions 5)-8) and 10)-13) above may include at least one of the following: RO resources, preamble format, preamble index, preamble sequence, SSB, beam, TCI, TRP.

[0296] 14) Community re-selection;

[0297] 15) TRP reselection;

[0298] 16) Cell handover;

[0299] 17) TRP switching;

[0300] 18) The cell handover conditions are met, which may include at least one of the following, for example: the measurement results of the source cell (RSRP, RSRQ or RSSI of L1 or L3, etc.) are greater than or less than a specified threshold; the measurement results of the target cell or neighboring cell are better than those of the source cell; the measurement results of the target cell or neighboring cell are greater than a specified threshold; the measurement results of the source cell are less than a specified threshold and the measurement results of the target cell or neighboring cell are greater than another specified threshold; the load of the source cell is greater than a specified threshold.

[0301] 19) The first device triggers access to the secondary cell;

[0302] 20) The first device moves to a preset position (e.g., the edge of the cell);

[0303] 21) Upstream business arrives;

[0304] 22) Downlink traffic arrives;

[0305] 23) Preset type of service arrival (such as services with high latency requirements or high reliability requirements).

[0306] 24) Network indication trigger, such as the base station sending an indication to the terminal via MAC CE, triggering AI model training;

[0307] 25) The speed of the first device is greater than or equal to the twenty-third threshold;

[0308] 26) The acceleration of the first device is greater than or equal to the twenty-fourth threshold;

[0309] 27) The position of the first piece of equipment has changed;

[0310] 28) The network location has changed;

[0311] 29) The change in position of the first device is greater than or equal to the twenty-fifth threshold;

[0312] 30) The change in network location is greater than or equal to the 26th threshold;

[0313] 31) The external environment of the first device changes, such as by obtaining information about the environmental changes through sensors;

[0314] 32) The spatial transmission filter selected by the first device changes;

[0315] 33) The beam selected by the first device changes;

[0316] 34) The TCI selected by the first device changes;

[0317] 35) The SSB selected by the first device changes;

[0318] 36) TAG timer timed out;

[0319] 37) Instructions from the network side;

[0320] 38) The first device connects to the new cell;

[0321] 39) The first device switches frequencies;

[0322] 40) The first device switches frequency bands;

[0323] 41) The first device switches operators;

[0324] 42) The first device switches to the Public Land Mobile Network (PLMN);

[0325] 43) AI model inference failed;

[0326] 44) The AI ​​model fails to infer N times consecutively;

[0327] 45) The number of inference failures of the AI ​​model is greater than or equal to the 27th threshold;

[0328] 46) Use AI models for reasoning;

[0329] 47) The first piece of equipment was moved to the new residential area;

[0330] 48) The first device moves to the tracking area;

[0331] 49) The first device moves to the preset geographical location;

[0332] 50) The change in the moving speed of the first device exceeds the specified level, that is, a significant change occurs, such as a sudden drop in a short period of time, from 250km / h to 3km / h, etc.

[0333] 51) The RSRP measured by the first device changes;

[0334] 52) The change in RSRP measured by the first device is greater than or equal to the 28th threshold;

[0335] 53) AI model training timer timeout preset time;

[0336] 54) N ​​consecutive model supervisions occur;

[0337] 55) N instances of model supervision occur;

[0338] 56) The number of random access failures in the RACH transmission corresponding to the first information in the AI ​​model inference is greater than or equal to the 29th threshold;

[0339] 57) The number of failed receptions of the RACH transmission corresponding to the first information in the AI ​​model inference is greater than or equal to the thirtieth threshold;

[0340] 58) The number of RACH transmissions corresponding to the first information in the AI ​​model inference that reach the maximum number of transmissions is greater than or equal to the thirty-first threshold;

[0341] 59) The RACH transmission corresponding to the first piece of information inferred by the AI ​​model reaches its maximum transmit power;

[0342] 60) The transmit power of the RACH transmission corresponding to the first information inferred by the AI ​​model is greater than or equal to the thirty-second threshold;

[0343] 61) The number of power boosts in the RACH transmission corresponding to the first information in the AI ​​model inference is greater than or equal to the thirty-third threshold;

[0344] 62) The number of failed RACH transmissions corresponding to different first information needs to be switched during AI model inference, and the number of failures is greater than or equal to the thirty-fourth threshold.

[0345] In this embodiment of the application, the fourteenth to thirty-fourth thresholds can be preset, and the specific values ​​are not limited.

[0346] It should be noted that AI model training can be triggered after meeting at least one of the 62 triggering conditions mentioned above. Alternatively, AI model training can be enabled after meeting at least one of the conditions (i.e., having the permission to train the AI ​​model). However, whether to train the AI ​​model and when to train it can be determined by considering the AI ​​model's capabilities, application scope, or network-side instructions, and is not specifically limited.

[0347] In this embodiment of the application, the method may further include: obtaining training-related information of the AI ​​model. The training-related information may include at least one of the following:

[0348] 1) Input information for AI model training;

[0349] 2) Labels for AI model training;

[0350] 3) Loss function for AI model training;

[0351] 4) The AI ​​algorithm used to train the AI ​​model or an index of the AI ​​algorithm;

[0352] 5) Reward information for AI model training;

[0353] 6) Adjustment information or feedback information for AI model training, wherein the adjustment information or feedback information may include at least one of the following: whether a rollback occurred (i.e., RACH resources were acquired using the traditional method), the number of rollbacks, the difference between the collision probability of the RACH resources in AI inference and the collision probability of the actual RACH transmission, and the number of AI inference failures.

[0354] In this embodiment, the aforementioned tag can be the output information of an AI model, and the tag can be obtained through terminal feedback, terminal measurement, or measurement by a network-side device; the specific method is not limited. In one implementation, the aforementioned tag may include at least one of the following:

[0355] 1) Whether RACH transmission information that meets performance requirements is obtained;

[0356] 2) The success rate of RACH is greater than or equal to the 35th threshold;

[0357] 3) The RACH failure rate is less than the 36th threshold;

[0358] 4) The transmission power required for a successful RACH is less than the thirty-seventh threshold;

[0359] 5) The number of transmissions required for a successful RACH is less than the thirty-eighth threshold;

[0360] 6) The number of RACH transmission messages obtained;

[0361] 7) The training time of the AI ​​model. If the training time of the AI ​​model is greater than or equal to the target time, the training is considered complete.

[0362] 8) Energy consumption for AI model training: If the energy consumption for AI model training is greater than or equal to the target value, then training is considered complete.

[0363] 9) The RACH transmission information required for transmission between the terminal and the cell network node is transmitted. If the performance of RACH transmission between the terminal and the target cell (or secondary cell) is greater than or equal to the target value, the training is considered complete.

[0364] In this embodiment of the application, the fourteenth to thirty-eighth thresholds can be preset, and the specific values ​​are not limited.

[0365] In this embodiment of the application, whether the AI ​​model training is complete can be determined by at least one of the following criteria:

[0366] 1) The loss function meets the predefined requirements or values, such as the training error being less than a predefined threshold value; where the loss function may include: the mean squared error or normalized mean squared error between the predicted value and the true value, and / or the mean absolute error between the predicted value and the true value.

[0367] 2) The model has been trained a predetermined number of times;

[0368] 3) The number of iterations for fine tuning has reached the predetermined value;

[0369] 4) At least one of the above labels meets the corresponding threshold value.

[0370] It is worth mentioning that, in this embodiment of the application, if the AI ​​model training is performed by a terminal, different types of terminals can be configured to have different AI model training capabilities based on the type of terminal. This configuration can include at least one of the following:

[0371] 1) The input information used for model training varies depending on the type of terminal. For example, less input information should be used for model training on terminal devices with weaker capabilities.

[0372] 2) Different labels are used for model training depending on the type of terminal;

[0373] 3) The execution method of model training varies depending on the type of terminal. For example, for terminal devices with weaker capabilities, it is possible to consider performing model training only on the network side, or to perform only a small part of the joint model training on the terminal side (such as model training involving user privacy data can be performed on the terminal side).

[0374] 4) Different types of terminals require different AI models for training. For example, overly complex AI models may not be applicable to terminals with weaker capabilities.

[0375] The method provided in this application training method trains an AI model, obtains first information based on the AI ​​model, and performs RACH transmission based on the first information. The first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters. The above process can determine the RACH transmission resources based on the AI ​​model, improving resource utilization. Furthermore, performing RACH transmission based on the RACH transmission resources can avoid unnecessary RACH transmission retransmissions, saving resources and reducing overhead, thereby reducing equipment energy consumption and improving equipment performance.

[0376] Figure 4 This illustration shows another flowchart of the random access transmission method provided in an embodiment of this application. Figure 4 As shown, the method 400 may include the following steps.

[0377] S402: Obtain the first information based on the AI ​​model.

[0378] The first piece of information may include at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

[0379] S404: RACH transmission based on the first information.

[0380] S406: Obtain the supervision configuration information of the AI ​​model.

[0381] In this embodiment of the application, the supervision configuration information may include at least one of the following:

[0382] 1) Identification of AI models to be supervised;

[0383] 2) The cycle of model supervision;

[0384] 3) Duration of model supervision;

[0385] 4) Window-related information for model supervision, including: the duration of the supervision window and / or the number of samples supervised by the model, etc.

[0386] 5) Duration of the supervision window;

[0387] 6) The number of samples used for model supervision;

[0388] 7) Triggering conditions for model supervision;

[0389] 8) Metrics for model supervision;

[0390] 9) Labels for model supervision.

[0391] S408: Supervise the AI ​​model based on the supervision configuration information.

[0392] In this embodiment, when the inference environment differs significantly from the training environment, the performance of RACH transmission based on the inference results of the AI ​​model becomes very poor; that is, the inferred first information or the RACH resources determined based on the first information are not accurate enough. Therefore, supervising the AI ​​model and triggering adjustment measures based on the supervision results can improve the accuracy and reliability of the AI ​​model.

[0393] In this embodiment of the application, the labels used during the training of the AI ​​model can also be used in part or in whole for the supervision of the AI ​​model, which will not be elaborated here.

[0394] In this embodiment, step S408 is an AI model supervision step. AI model supervision can be performed when triggering conditions are met, or it can be performed at any time or randomly without setting triggering conditions; the specific execution is not limited. The triggering conditions for AI model supervision may include at least one of the following:

[0395] 1) The results of AI inference do not meet the accuracy requirements;

[0396] 2) At least one of the inference metrics of the AI ​​model fails to meet the requirements;

[0397] 3) The AI ​​model supervision metrics do not meet the requirements. The AI ​​model supervision metrics include: the error between the AI ​​model's predicted value and the actual value, and / or, the network performance metrics; where network performance metrics may include: transmission latency and / or throughput, etc.

[0398] 4) The timer for AI model supervision timed out;

[0399] 5) The terminal switches to a new cell or TRP or switches beams;

[0400] 6) AI model inference failed;

[0401] 7) The AI ​​model fails to infer N times consecutively;

[0402] 8) The number of inference failures by the AI ​​model reached the thirty-ninth threshold;

[0403] 9) Use AI models for reasoning;

[0404] 10) The terminal has moved to a new cell, tracking area, or geographical location;

[0405] 11) The terminal’s moving speed changes more than the 40th threshold within a specified time period, i.e., a significant change occurs, such as a sudden drop in speed within a short period of time, from 250 km / h to 3 km / h, etc.

[0406] 12) The external environment in which the AI ​​model's inference device is located changes, such as obtaining information about environmental changes through sensors.

[0407] In this embodiment of the application, the aforementioned thirty-ninth and fortieth thresholds can be preset, and their specific values ​​are not limited.

[0408] It is worth mentioning that, in this embodiment of the application, if the AI ​​model inference is performed by a terminal, different types of terminals can be configured to have different AI model inference capabilities based on the type of terminal. This configuration can include at least one of the following:

[0409] 1) The input information used for model inference varies depending on the type of terminal. For example, less input information should be used for model inference on terminal devices with weaker capabilities.

[0410] 2) Different labels are used for model inference depending on the type of terminal;

[0411] 3) The execution method of model inference is different for different types of terminals. For example, for terminal devices with weaker capabilities, it is possible to consider performing model inference only on the network side, or performing only part of the inference on the terminal side (such as model inference involving user privacy data can be performed on the terminal side).

[0412] 4) Different types of terminals require different AI models for model inference. For example, overly complex AI models may not be applicable to terminals with weaker capabilities.

[0413] The method provided in this application embodiment obtains first information based on an AI model, performs RACH transmission based on the first information, acquires supervisory configuration information of the AI ​​model, and supervises the AI ​​model according to the supervisory configuration information. The first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters. This process can determine RACH transmission resources based on the AI ​​model, improving resource utilization. Furthermore, performing RACH transmission based on these resources avoids unnecessary RACH retransmissions, saving resources, reducing overhead, and thus lowering equipment energy consumption and improving equipment performance.

[0414] The following describes the application scenarios of the above method embodiments.

[0415] The first scenario: The terminal uses an AI model for prediction. In this scenario, the terminal predicts RACH transmission resources (including collision probability), which can predict RACH transmission resources for one or more Cells, TRPs, Beams, SSBs, RSRP ranges, geographical locations, etc. AI model training can be performed on the terminal side or the network side. If trained on the terminal side, the network side can notify the terminal of relevant parameters or information, such as network load, per-beam distribution, and SSB distribution. The advantage of terminal-side training is that it can obtain more real-time information. If trained on the network side, the terminal specifically reports relevant parameters or information to the network side. The advantage of network-side training is that it can save terminal energy and reduce the complexity of terminal training.

[0416] The second scenario involves using AI models for prediction on the network side. In this scenario, network-side devices can predict RACH transmission resources (including collision probabilities) for one or more TRPs, Beams, SSBs, RSRP ranges, geographical locations, etc., or they can predict the RACH transmission resources for a group of terminals. A group of terminals can be terminals with the same beam or the same associated RS, or terminals geographically close, or terminals with similar RSRPs, etc. Model training is performed on the network side, which has the advantage of saving terminal power consumption and reducing the complexity of prediction on the terminal.

[0417] Taking the cell-free scenario as an example, the prediction scheme for the first scenario mentioned above can be as follows:

[0418] 1) The UE measures and obtains multiple SSBs / Beams / TRPs that meet the conditions, and different SSBs are associated with different RACH resources;

[0419] 2) Based on the target AI model, the UE predicts RACH transmission related information corresponding to multiple SSBs / Beams / TRPs. Specifically, for example, the AI ​​model inference obtains the collision probability of the RACH resources associated with each SSB that meets the conditions, the AI ​​model inference obtains the transmit power of the PRACH transmission corresponding to each SSB that meets the conditions, and the AI ​​model inference obtains the RACH resources (time and frequency resources, preamble sequence) selected by each SSB that meets the conditions.

[0420] Based on the above prediction scheme, the corresponding RACH transmission method can be described as follows:

[0421] Step 1: The UE trains the target AI model based on multiple TRPs in a cell-free cell;

[0422] Step 2: When the service arrives, the network paging the UE or the UE has an uplink service that needs to be transmitted, the UE needs to select to establish an RRC connection with one or more TRPs to transmit fast data, and the UE needs to initiate RACH transmission.

[0423] Step 3: The UE uses the target AI model to perform AI inference to obtain RACH transmission related information between the UE and multiple TRPs within the coverage area of ​​the macro base station;

[0424] Step 4: The RACH transmission information obtained by the AI ​​model inference is used to initiate RACH transmission.

[0425] Taking the cell-free scenario as an example again, the prediction scheme for the second scenario mentioned above can be as follows:

[0426] 1) Macro base stations predict RACH transmission-related information between multiple TRPs and a group of terminals based on information such as UE distribution or service distribution;

[0427] For example, different SSBs / Beams / TRPs are associated with different RACH resources; the AI ​​model inference yields the collision probability of the RACH resources associated with the SSB corresponding to one or more TRPs; the AI ​​model inference yields the transmit power of the PRACH transmission corresponding to the SSB corresponding to one or more TRPs; the AI ​​model inference yields the RACH resources (time-frequency resources, preamble sequences) selected by the SSB corresponding to one or more TRPs.

[0428] 2) The macro base station performs AI inference to obtain RACH transmission related information between multiple TRPs and a group of terminals, and in the configuration information, it indicates at least one specific TRP and terminal RACH transmission related information to the terminal.

[0429] Specifically, the terminal is notified via system messages from the macro base station, or via broadcast messages or RRC messages from the TRP through interaction between the macro base station and the TRP.

[0430] Based on the above prediction scheme, the corresponding RACH transmission method can be described as follows:

[0431] Step 1: The macro base station of the cell-free cell trains the target AI model based on multiple TRPs;

[0432] Step 2: When the service arrives, the network paging the UE or the UE has an uplink service that needs to be transmitted, the UE needs to select to establish an RRC connection with one or more TRPs to transmit fast data, and the UE needs to initiate RACH transmission.

[0433] Step 3: The macro base station of the cell-free cell uses the target AI model to perform AI inference to obtain RACH transmission information between multiple TRPs and a group of terminals within the coverage area of ​​the macro base station;

[0434] Step 4: The terminal is notified via system messages from the macro base station, or via broadcast messages or RRC messages from the TRP through interaction between the macro base station and the TRP.

[0435] Step 5: The UE uses the RACH transmission information obtained from the network indication to select the appropriate TRP and initiate RACH transmission.

[0436] The relevant terms used in the above embodiments are explained below.

[0437] In this embodiment of the application, random access resources may include: random access resources based on contention-based random access (CBRA), and / or random access resources based on contention-free random access (CFRA). Alternatively, they may include: random access resources corresponding to 4-step RACH, or random access resources corresponding to 2-step RACH.

[0438] In this embodiment of the application, the random access process can be performed by the terminal in either a connected state or an unconnected state (idle / inactive).

[0439] In this embodiment of the application, SSB to RO mapping can also refer to the association between downlink signals, uplink signals, and resources in a broader sense, such as the relationship between CSI-RS and RO.

[0440] In this embodiment, RACH Occasion (RO) and PRACH Occasion both refer to the time-frequency resources required to transmit a PRACH sequence. RA and Random Access both refer to the random access procedure. When the RACH transmission is MsgA, MsgAPRACH, MsgAPUSCH, Msg3 PUSCH, the corresponding transmission of RA-SDT (RACH based small data transmission), the PUSCH scrambled by TC-RNTI and scheduled by DCI, the CG PUSCH transmission, the PUSCH in the RACH less handover process, the PUSCH in the LTM process, the PUCCH in the RACH process, SRS, etc., RO and PRACH Occasion represent the time-frequency resources required for the corresponding transmission, and the PRACH sequence is the corresponding transport channel or the corresponding DMRS resource.

[0441] In this application embodiment, SSB and SS / PBCH block can be used interchangeably or with other names. They can refer to any module containing at least some synchronization signals, broadcast signals or other downlink broadcast signals or their control channels.

[0442] In this embodiment of the application, RACH or PRACH transmission includes msg1, PRACH, preamble, MsgA, MsgAPRACH, MsgA PUSCH, Msg3 PUSCH, RA-SDT (RACH based small data transmission) corresponding transmission, PUSCH scrambled by TC-RNTI and scheduled by DCI, CG PUSCH transmission, PUSCH during RACH less handover, PUSCH during LTM process, PUCCH during random access process, SRS transmission, etc.

[0443] In this embodiment of the application, RACH resources include RO, msg1, PRACH, preamble, MsgA, MsgA PRACH, MsgAPUSCH, Msg3 PUSCH, RA-SDT (RACH based small data transmission) corresponding transmissions, PUSCH scrambled by TC-RNTI and scheduled by DCI, CG PUSCH resources, PUSCH during RACH less handover, PUSCH during LTM, PUCCH during random access, SRS and other transmissions, and corresponding time domain and / or frequency domain resources.

[0444] In this embodiment of the application, the RACH-related PDCCH includes at least one of the following: the PDCCH corresponding to Msg2, RAR, MsgB, Msg4, and Msg3.

[0445] In this application embodiment, the AI ​​model may also be referred to as an AI unit, AI structure, etc., or the AI ​​model may refer to a processing unit that can implement specific algorithms, formulas, processing flows, capabilities, etc. related to AI, or the AI ​​model may be a processing method, algorithm, function, module or unit for a specific dataset, or the AI ​​model may be a processing method, algorithm, function, module or unit running on AI-related hardware such as GPU, NPU, TPU, ASIC, etc., without specific limitations.

[0446] In this application embodiment, the identifier (i.e. ID) of the AI ​​model can be an AI model identifier, an AI structure identifier, an AI algorithm identifier, or an identifier of a specific dataset associated with the AI ​​model, or an identifier of a specific scenario, environment, channel characteristics, or device related to AI, or an identifier of a function, characteristic, capability, or module related to AI, and there is no specific limitation on these.

[0447] In this embodiment of the application, the index of the AI ​​unit can be described in various ways, such as functional ID, model ID, physical model ID, logical model ID, global model ID, local model ID, etc.

[0448] In this embodiment of the application, AI can also be represented as machine learning (ML), which has various implementation methods, such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc., and no specific limitation is made thereto.

[0449] In this application embodiment, "server" may specifically refer to an entity used for training, predicting, or providing AI-related information, or it may refer to services that bypass operators or OTT services.

[0450] In this application embodiment, RACH or preamble can also be called any module that includes at least one of synchronization signal, random access signal / channel, uplink control signal / channel, and other control channels for access.

[0451] In this application embodiment, TA-related information includes, but is not limited to, at least one of: timing advance, timing advance offset, timing pre-compensation, common timing advance offset, and proprietary timing advance offset. TA-related information can also be applied to other related information, such as frequency error-related information, including, but not limited to, frequency compensation, frequency offset, common frequency offset, and proprietary frequency offset.

[0452] In this application embodiment, the TA-related information or first input information between a certain device (such as a terminal) and another device (such as a network node) may also refer to the timing advance information required for transmission from one device to another, or the first input information required to estimate the timing advance information.

[0453] The TA-related information or first input information between a certain device (such as a terminal) and a certain cell can also refer to the timing advance information required for signal transmission from the device to a certain cell, or the first input information required to estimate the timing advance information.

[0454] The TA-related information or first input information between a certain cell and another cell can also refer to the TA-related information required for signal transmission from a cell network node to another cell network node, or the first input information required to estimate the timing advance information, or the TA difference information required for signal transmission between a certain device in a cell and another cell, or the first input information required to estimate the TA difference information.

[0455] Figure 5 This illustration shows a structural diagram of a random access transmission device provided in an embodiment of this application, such as... Figure 5 As shown, the device may include an acquisition module 501 and a transmission module 502.

[0456] The acquisition module 501 is used to obtain the first information based on the AI ​​model.

[0457] The transmission module 502 is used for RACH transmission based on the first information.

[0458] The first piece of information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

[0459] In this embodiment of the application, the beam information includes at least one of the following:

[0460] Beam information of the terminal;

[0461] Beam information between the terminal and the first cell;

[0462] Beam information between the terminal and the first transmitting / receiving point (TRP);

[0463] Beam information within the preset area;

[0464] Beam information for preset path loss range;

[0465] Beam information within the preset signal quality strength range;

[0466] Beam information for preset RACH configuration parameters.

[0467] In this embodiment of the application, the RACH collision probability includes at least one of the following:

[0468] The RACH collision probability of the terminal;

[0469] RACH collision probability between the terminal and the first cell;

[0470] The probability of RACH collision between the terminal and the first TRP;

[0471] RACH collision probability in the preset beam direction;

[0472] The probability of RACH collision within the preset area;

[0473] RACH collision probability within the preset road loss range;

[0474] RACH collision probability within a preset signal quality strength range;

[0475] The RACH collision probability of the preset RACH configuration parameters.

[0476] In this embodiment of the application, the RACH resource information includes at least one of the following:

[0477] Terminal RACH resource information;

[0478] RACH resource information between the terminal and the first cell;

[0479] RACH resource information between the terminal and the first TRP;

[0480] RACH resource information for preset beam directions;

[0481] RACH resource information within the preset area;

[0482] RACH resource information with preset road loss range;

[0483] RACH resource information for preset signal quality strength range;

[0484] RACH resource information with preset RACH configuration parameters.

[0485] In this embodiment of the application, the transmit power parameters include at least one of the following:

[0486] The terminal's RACH transmission transmit power parameters;

[0487] RACH transmission transmit power parameters between the terminal and the first cell;

[0488] RACH transmission transmit power parameters between the terminal and the first TRP;

[0489] RACH transmission transmit power parameters with preset beam direction;

[0490] RACH transmission transmit power parameters within the preset area;

[0491] RACH transmission transmit power parameters within the preset path loss range;

[0492] RACH transmission transmit power parameters within a preset signal quality strength range;

[0493] The preset RACH configuration parameters include the RACH transmission transmit power parameters.

[0494] In this embodiment of the application, the RACH resource information includes one of the following:

[0495] A set of candidate RACH resource information;

[0496] A set of candidate RACH resource information and the probability value associated with each candidate RACH resource information.

[0497] In this embodiment of the application, the transmit power parameters include one of the following:

[0498] A set of candidate RACH transmit power parameters;

[0499] A set of candidate RACH transmit power parameters and the probability values ​​associated with each candidate RACH transmit power parameter.

[0500] In this embodiment of the application, the input information of the AI ​​model includes at least one of the following:

[0501] Uplink signal reception strength information between the first and second devices;

[0502] Downlink signal reception strength information between the first device and the second device;

[0503] Signal quality information of the uplink signal between the first and second devices;

[0504] Signal quality information of the downlink signal between the first device and the second device;

[0505] Path loss between the first and second devices;

[0506] Distance information between the first and second devices;

[0507] The number of random access failures in RACH transmission between the first and second devices;

[0508] The number of times the random access response message reception failed during RACH transmission between the first and second devices;

[0509] The maximum number of RACH transmissions between the first and second devices;

[0510] The number of retransmissions of RACH transmission between the first and second devices;

[0511] The number of failed attempts to switch between different RACH transmission types between the first and second devices;

[0512] The absolute timing advance (TA) value between the first and second devices;

[0513] The relative TA value between the first and second devices;

[0514] Beam information between the first and second devices;

[0515] Propagation delay between the first and second devices;

[0516] Round-trip time (RTT) between the first and second devices;

[0517] The maximum transmit power of RACH transmission, TRP identifier, TRP group identifier, cell identifier, cell group identifier, timing advance group TAG identifier, tracking area identifier, radio access network notification area RNA identifier, frequency domain resource information, load information, interference information, and information of the first device.

[0518] In this embodiment of the application, the information of the first device includes at least one of the following:

[0519] Transmission power information, location information, distribution information, direction of movement, speed of movement, energy consumption information, power consumption information, operator information, network type information, panel orientation information, device type, sensing information, network scene information, environmental information, ephemeris information, reference position of the cell in the non-terrestrial network NTN scene, movement trajectory of the cell in the NTN scene, multipath information of the channel, time information, compensation information, and offset information.

[0520] In this embodiment of the application, the compensation information includes at least one of the following:

[0521] Timing pre-compensation, wherein the offset information includes at least one of the following: timing advance offset, common timing advance offset, and dedicated time advance offset;

[0522] Frequency compensation, wherein the offset information includes at least one of the following: frequency offset, common frequency offset, and proprietary frequency offset.

[0523] In this embodiment of the application, RACH transmission includes at least one of the following:

[0524] RACH transmission corresponding to the first RACH resource;

[0525] RACH transmission corresponding to the first reference signal;

[0526] The first transmission configuration indicates the RACH transmission corresponding to TCI;

[0527] RACH transmission corresponding to the first TRP;

[0528] The first RACH resource includes at least one of the following: RO resource, preamble format, preamble index, and preamble sequence; the first reference signal includes at least one of the following: synchronization signal physical broadcast channel block (SSB) and channel state information reference signal (CSI-RS).

[0529] In this embodiment of the application, the maximum transmit power is the maximum transmit power of the first device when performing RACH transmission in at least one of the cell, cell group, frequency layer, TA area, TRP, network node, carrier, band, subband and partial bandwidth.

[0530] In this embodiment of the application, the TRP identifier corresponds to at least one of the following TRPs: source TRP, target TRP, currently selected TRP, residing TRP, currently accessed TRP, primary TRP, and secondary TRP;

[0531] The TRP group identifier corresponds to at least one of the following TRP groups: source TRP group, target TRP group, currently selected TRP group, residing TRP group, currently accessed TRP group, primary TRP group, and secondary TRP group;

[0532] The community identifier is the physical layer community identifier of the first community;

[0533] The community group identifier is the identifier for the first community group;

[0534] The TAG identifier is the TAG identifier of the first community;

[0535] The tracking zone identifier is the tracking zone identifier of the first cell;

[0536] The RNA identifier is the RNA identifier of the first cell;

[0537] The first cell includes one of the following: source cell, target cell, currently stationed cell, currently accessing cell, primary cell, and secondary cell. The first cell group includes one of the following: source cell group, target cell group, currently stationed cell group, currently accessing cell group, primary cell group, and secondary cell group.

[0538] In this embodiment of the application, the beam information includes at least one of the following: reference signal index, beam index, and beam direction;

[0539] The load information is the load information corresponding to at least one of SSB, beam, TCI, TRP, cell, carrier, and frequency point;

[0540] The interference information is the interference information corresponding to at least one of SSB, beam, TCI, TRP, cell, carrier and frequency point.

[0541] In this embodiment of the application, the triggering condition for AI model inference includes at least one of the following:

[0542] AI timer timed out;

[0543] RACH configuration using network configuration;

[0544] Use the RACH type configured in the network;

[0545] The collision probability of RACH transmission is greater than or equal to the first threshold;

[0546] The number of random access failures in the RACH transmission corresponding to the first information is greater than or equal to the second threshold;

[0547] The number of failed receptions of the random access response message for the RACH transmission corresponding to the first information is greater than or equal to the third threshold;

[0548] The number of RACH transmissions corresponding to the first information that reach the maximum number of transmissions is greater than or equal to the fourth threshold;

[0549] The number of repeated transmissions used in the RACH transmission corresponding to the first information is greater than or equal to the fifth threshold;

[0550] The number of failed attempts to switch between different RACH transmission types is greater than or equal to the sixth threshold;

[0551] The RACH transmission corresponding to the first message reaches maximum transmit power;

[0552] The transmit power of the RACH transmission corresponding to the first information is greater than or equal to the seventh threshold;

[0553] The number of power boosts in the RACH transmission corresponding to the first information is greater than or equal to the eighth threshold;

[0554] The number of failed RACH transmissions corresponding to switching different first information is greater than or equal to the ninth threshold;

[0555] Community re-selection;

[0556] TRP reselection;

[0557] Cell handover;

[0558] TRP switching;

[0559] The conditions for cell handover are met;

[0560] The first device triggers access to the secondary cell;

[0561] The first device moves to the preset position;

[0562] Upstream business has arrived;

[0563] Downlink traffic has arrived;

[0564] Preset type of service arrives;

[0565] Network indication triggered;

[0566] The speed of the first device is greater than or equal to the tenth threshold;

[0567] The acceleration of the first device is greater than or equal to the eleventh threshold;

[0568] The location of the first device has changed;

[0569] The network location has changed;

[0570] The position change of the first device is greater than or equal to the twelfth threshold;

[0571] The network location change is greater than or equal to the thirteenth threshold;

[0572] The beam selected by the first device changes;

[0573] The spatial transmission filter selected by the first device has changed;

[0574] The first piece of information includes at least one of the following: RO resources, preamble format, preamble index, preamble sequence, SSB, beam, TCI, and TRP.

[0575] In this embodiment of the application, the above-described apparatus can also be used for:

[0576] Obtain training samples, which include at least one label and input information used by the AI ​​model for inference; and / or,

[0577] The AI ​​model is obtained by training the model using training samples. The AI ​​model outputs second information, which is used for RACH transmission and includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

[0578] In this embodiment of the application, the first device is a network-side device, and the input information of the AI ​​model is indicated by the terminal or other network-side devices through cross-carrier OTT messages.

[0579] In this embodiment of the application, the above-mentioned device is applied to a first device, and the device is also used to: perform joint training with a second device to obtain an AI model.

[0580] In this embodiment of the application, the above-described apparatus is further configured to perform one of the following:

[0581] Under preset conditions, training of the AI ​​model is triggered;

[0582] Periodically trigger the training of the AI ​​model;

[0583] Semi-static triggering is used to train the AI ​​model.

[0584] In this embodiment of the application, the above-mentioned preset conditions include at least one of the following:

[0585] AI timer timed out;

[0586] RACH configuration using network configuration;

[0587] Use the RACH type configured in the network;

[0588] The collision probability corresponding to RACH transmission is greater than or equal to the fourteenth threshold;

[0589] The number of random access failures in the RACH transmission corresponding to the first information is greater than or equal to the fifteenth threshold;

[0590] The number of failed RACH transmissions for the first information is greater than or equal to the sixteenth threshold.

[0591] The number of RACH transmissions corresponding to the first information that reach the maximum number of transmissions is greater than or equal to the seventeenth threshold;

[0592] The number of repeated transmissions used in the RACH transmission corresponding to the first information is greater than or equal to the eighteenth threshold;

[0593] The number of failed attempts to switch between different RACH transmission types is greater than or equal to the nineteenth threshold;

[0594] The RACH transmission corresponding to the first message reaches maximum transmit power;

[0595] The transmit power of the RACH transmission corresponding to the first information is greater than or equal to the twentieth threshold;

[0596] The number of power boosts in the RACH transmission corresponding to the first information is greater than or equal to the twenty-first threshold;

[0597] The number of failed RACH transmissions corresponding to switching different first information is greater than or equal to the twenty-second threshold;

[0598] Community re-selection;

[0599] TRP reselection;

[0600] Cell handover;

[0601] TRP switching;

[0602] The conditions for cell handover are met;

[0603] The first device triggers access to the secondary cell;

[0604] The first device moves to the preset position;

[0605] Upstream business has arrived;

[0606] Downlink traffic has arrived;

[0607] Preset type of service arrives;

[0608] Network indication triggered;

[0609] The speed of the first device is greater than or equal to the twenty-third threshold;

[0610] The acceleration of the first device is greater than or equal to the twenty-fourth threshold;

[0611] The location of the first device has changed;

[0612] The network location has changed;

[0613] The change in the position of the first device is greater than or equal to the twenty-fifth threshold;

[0614] The network location change is greater than or equal to the 26th threshold;

[0615] The external environment of the first device changes;

[0616] The spatial transmission filter selected by the first device has changed;

[0617] The beam selected by the first device changes;

[0618] The TCI selected by the first device has changed;

[0619] The SSB selected by the first device has changed;

[0620] TAG timer timed out;

[0621] Instructions from the network side;

[0622] The first device is connected to the new community;

[0623] The first device switches frequency points;

[0624] The first device switches to the frequency band;

[0625] First device switches carriers;

[0626] The first device switches to the Public Land Mobile Network (PLMN);

[0627] AI model inference failed;

[0628] The AI ​​model failed inference N times in a row;

[0629] The number of AI model inference failures is greater than or equal to the 27th threshold;

[0630] Use AI models for reasoning;

[0631] The first piece of equipment was moved to the new residential area;

[0632] The first device moves to the tracking area;

[0633] The first device moves to the preset geographical location;

[0634] The moving speed of the first device changes;

[0635] The RSRP measured by the first device changed;

[0636] The RSRP change measured by the first device is greater than or equal to the 28th threshold;

[0637] AI model training timer timeout preset time;

[0638] N consecutive model supervisions occur;

[0639] N instances of model supervision occurred;

[0640] The number of random access failures in the RACH transmission corresponding to the first information in the AI ​​model inference is greater than or equal to the twenty-ninth threshold.

[0641] The number of failed RACH transmission access response messages corresponding to the first information in the AI ​​model inference is greater than or equal to the thirtieth threshold.

[0642] The number of RACH transmissions corresponding to the first information in the AI ​​model inference that reach the maximum number of transmissions is greater than or equal to the thirty-first threshold.

[0643] The RACH transmission corresponding to the first information inferred by the AI ​​model reaches the maximum transmission power.

[0644] The transmit power of the RACH transmission corresponding to the first information inferred by the AI ​​model is greater than or equal to the thirty-second threshold;

[0645] The number of RACH transmission power boosts corresponding to the first information inferred by the AI ​​model is greater than or equal to the thirty-third threshold.

[0646] The AI ​​model inference requires switching between different first information corresponding to RACH transmission failures with a number greater than or equal to the thirty-fourth threshold.

[0647] In this embodiment of the application, the above-mentioned device is further used for:

[0648] Obtain training-related information for the AI ​​model, which includes at least one of the following:

[0649] Input information for AI model training;

[0650] Labels for AI model training;

[0651] Loss function for AI model training;

[0652] AI algorithms for training AI models;

[0653] Reward information for AI model training;

[0654] Adjustment information for AI model training.

[0655] In this embodiment of the application, the label includes at least one of the following:

[0656] Whether RACH transmission information that meets performance requirements is obtained;

[0657] The success rate of RACH is greater than or equal to the 35th threshold;

[0658] The RACH failure rate is less than the 36th threshold;

[0659] The transmission power required for a successful RACH is less than the thirty-seventh threshold;

[0660] The number of transmissions required for a successful RACH is less than the thirty-eighth threshold;

[0661] The number of RACH transmission messages obtained;

[0662] The duration of AI model training;

[0663] Energy consumption for AI model training;

[0664] The RACH transmission information required for transmission between the terminal and the cell network node.

[0665] In this embodiment of the application, the inference metrics of the AI ​​model include at least one of the following:

[0666] The complexity of AI models;

[0667] Latency of AI model inference;

[0668] Success rate of AI model inference;

[0669] The reliability of AI model inference results.

[0670] In this embodiment of the application, the above-described apparatus can also be used for:

[0671] Obtain first instruction information, which is used to indicate the activation or deactivation information of the AI ​​model;

[0672] The AI ​​model is activated or deactivated based on the first instruction information.

[0673] In this embodiment of the application, the above-described apparatus can also be used for:

[0674] Obtain the first configuration information of the AI ​​model, which includes at least one of the following:

[0675] AI models;

[0676] Identification of AI models;

[0677] The application scope of AI model reasoning;

[0678] The inference cycle of AI models;

[0679] The effective duration of AI model inference;

[0680] Triggering conditions for AI model inference;

[0681] The configuration parameters of an AI model include at least one of the following: the input information of the AI ​​model, the output information of the AI ​​model, the type of the input information of the AI ​​model, and the type of the output information of the AI ​​model.

[0682] Does it support joint inference across multiple devices?

[0683] In this embodiment of the application, the above-described apparatus can also be used for:

[0684] Send third-party information to the network side;

[0685] Receive instructions sent from the network side;

[0686] Based on the instructions, determine the AI ​​model to use from among multiple AI models;

[0687] The third information includes at least one of the following: the identifiers of multiple AI models and the classification identifiers of the datasets. The indications include at least one of the following: the identifier ID of the AI ​​model, the identifier of the dataset, and the configuration information for data collection.

[0688] In this embodiment of the application, the above-described apparatus is further configured to perform at least one of the following:

[0689] Send one or more of the input information, output information, and current state information of the first device to the device that trained the AI ​​model;

[0690] The decision to pause or abandon is based on the initial information obtained from the AI ​​model;

[0691] This triggers adjustments to the AI ​​model.

[0692] In this embodiment of the application, the above-mentioned device is applied to the first device, and the device is also used to: send indication information of the AI ​​capability information of the first device to the second device;

[0693] The AI ​​capability information includes at least one of the following:

[0694] The first device may or may not have the ability to train AI models;

[0695] The first device may or may not have the ability to use AI models for AI reasoning;

[0696] The first device may or may not have the ability to send auxiliary information, which is used for AI model inference.

[0697] In this embodiment of the application, the above-mentioned device is further used for:

[0698] Obtain the supervised configuration information of the AI ​​model;

[0699] The AI ​​model is supervised based on the supervision configuration information.

[0700] In this embodiment of the application, the monitoring configuration information includes at least one of the following:

[0701] Identification of AI models to be supervised;

[0702] The cycle of model supervision;

[0703] The duration of model supervision;

[0704] Window-related information for model supervision;

[0705] Duration of the supervision window;

[0706] The number of samples used for model supervision;

[0707] Triggering conditions for model supervision;

[0708] Metrics for model supervision;

[0709] Labels for model supervision.

[0710] In this embodiment of the application, the triggering conditions for AI model supervision include at least one of the following:

[0711] The results of AI inference do not meet the accuracy requirements;

[0712] At least one of the inference metrics of the AI ​​model fails to meet the requirements;

[0713] The metrics for AI model supervision do not meet the requirements. The metrics for AI model supervision include: the error between the AI ​​model's predicted values ​​and the actual values, and / or, the network's performance metrics.

[0714] AI model supervision timer timed out;

[0715] The terminal switches to a new cell or TRP or switches beams;

[0716] AI model inference failed;

[0717] The AI ​​model failed inference N times in a row;

[0718] The number of inference failures by the AI ​​model reached the 39th threshold.

[0719] Using AI model inference;

[0720] The terminal has moved to a new cell, tracking area, or geographical location;

[0721] The terminal's moving speed changes beyond the fortieth threshold within a specified time period;

[0722] The external environment in which the AI ​​model's inference device is located changes.

[0723] The apparatus provided in this application embodiment can execute the methods in any of the above method embodiments. For detailed processes, please refer to the description in the method embodiments, which will not be repeated here.

[0724] The apparatus provided in this application embodiment obtains first information based on an AI model and performs RACH transmission based on the first information. The first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters. The above process can determine the RACH transmission resources based on the AI ​​model, thereby improving resource utilization. Furthermore, performing RACH transmission based on the RACH transmission resources can avoid unnecessary RACH transmission retransmissions, saving resources and reducing overhead, thereby reducing equipment energy consumption and improving equipment performance.

[0725] This application provides a random access transmission device. As an example, the device may be a communication device or a component within a communication device, such as a chip. The communication device may be a terminal, a network-side device, or a server, etc. Exemplarily, the terminal may include, but is not limited to, the types of terminals listed above, and the network-side device may include, but is not limited to, the types of network-side devices listed above. This application does not impose specific limitations on these types.

[0726] The random access transmission device includes a receiving module, a transmitting module, and a processing module. These modules can be implemented in software or hardware. When implemented in hardware, the processing module can be implemented by a processor. For example, the processor can include general-purpose processors, special-purpose processors, such as a Central Processing Unit (CPU), microprocessor, Digital Signal Processor (DSP), Artificial Intelligence (AI) processor, Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC), Network Processor (NP), Field Programmable Gate Array (FPGA), or other programmable logic devices, gate circuits, transistors, discrete hardware components, etc. The receiving and transmitting modules can be implemented by a communication interface, which can include one or more of the following: transceiver, pins, circuits, bus, radio frequency unit, etc.

[0727] The random access transmission device provided in this application embodiment can achieve... Figures 2 to 4 The various processes implemented in the method embodiments achieve the same technical effect, and will not be described again here to avoid repetition.

[0728] like Figure 6 As shown in the illustration, this application 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 method embodiments 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 method embodiments and achieve the same technical effect. To avoid repetition, further details are omitted here.

[0729] This application embodiment also provides a terminal, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps in the method embodiments shown above. This terminal embodiment corresponds to the above-described terminal-side method embodiments, and all implementation processes and methods of the above-described method embodiments can be applied to this terminal embodiment and achieve the same technical effect. The terminal can be... Figure 5 The apparatus shown. Specifically, Figure 7 A schematic diagram of the hardware structure of a terminal to implement an embodiment of this application.

[0730] 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.

[0731] Those skilled in the art will understand that the terminal 700 may also include a power supply (such as a battery) for supplying power to various components. The power supply may 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. Figure 7 The terminal structure shown 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.

[0732] 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.

[0733] 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.

[0734] 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.

[0735] 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.

[0736] The processor 710 is used to obtain first information based on the AI ​​model and to perform RACH transmission based on the first information.

[0737] The terminal provided in this application embodiment obtains first information based on an AI model and performs RACH transmission based on the first information. The first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters. The above process can determine the RACH transmission resources based on the AI ​​model, thereby improving resource utilization. Furthermore, performing RACH transmission based on the RACH transmission resources can avoid unnecessary RACH transmission retransmissions, saving resources and reducing overhead, thereby reducing device energy consumption and improving device performance.

[0738] 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.

[0739] 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 above. 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 achieve the same technical effects.

[0740] Specifically, embodiments of this application also provide a network-side device. For example... Figure 8 As shown, the network-side device 800 includes: a processor 801, a network interface 802, and a memory 803. This network-side device can be... Figure 5 The device shown. The network interface 802 is, for example, a common public radio interface (CPRI).

[0741] Specifically, the network-side device 800 in this application embodiment further includes: instructions or programs stored in memory 803 and executable on processor 801, wherein processor 801 calls the instructions or programs in memory 803 to execute. Figure 5 The methods executed by each module shown achieve the same technical effect, and to avoid repetition, they will not be described in detail here.

[0742] 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 method embodiments and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0743] 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.

[0744] 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 method embodiments and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0745] 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.

[0746] 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 method embodiments and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0747] This application also provides a random access transmission system, including: a terminal and a network-side device, wherein the terminal can be used to perform the steps of the random access transmission method described above, and the network-side device can be used to perform the steps of the random access transmission method described above.

[0748] 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.

[0749] 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.

[0750] 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 random access transmission method, characterized in that, include: The first device obtains the first information based on an artificial intelligence (AI) model. Based on the first information, random access channel (RACH) transmission is performed; The first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

2. The method according to claim 1, characterized in that, The beam information includes at least one of the following: Beam information of the terminal; Beam information between the terminal and the first cell; Beam information between the terminal and the first transmitting / receiving point (TRP); Beam information within the preset area; Beam information for preset path loss range; Beam information within the preset signal quality strength range; Beam information for preset RACH configuration parameters.

3. The method according to claim 1, characterized in that, The RACH collision probability includes at least one of the following: The RACH collision probability of the terminal; RACH collision probability between the terminal and the first cell; The probability of RACH collision between the terminal and the first TRP; RACH collision probability in the preset beam direction; The probability of RACH collision within the preset area; RACH collision probability within the preset road loss range; RACH collision probability within a preset signal quality strength range; The RACH collision probability of the preset RACH configuration parameters.

4. The method according to claim 1, characterized in that, The RACH resource information includes at least one of the following: Terminal RACH resource information; RACH resource information between the terminal and the first cell; RACH resource information between the terminal and the first TRP; RACH resource information for preset beam directions; RACH resource information within the preset area; RACH resource information with preset road loss range; RACH resource information for preset signal quality strength range; RACH resource information with preset RACH configuration parameters.

5. The method according to claim 1, characterized in that, The transmit power parameter includes at least one of the following: The terminal's RACH transmission transmit power parameters; RACH transmission transmit power parameters between the terminal and the first cell; RACH transmission transmit power parameters between the terminal and the first TRP; RACH transmission transmit power parameters with preset beam direction; RACH transmission transmit power parameters within the preset area; RACH transmission transmit power parameters within the preset path loss range; RACH transmission transmit power parameters within a preset signal quality strength range; The preset RACH configuration parameters include the RACH transmission transmit power parameters.

6. The method according to any one of claims 1-5, characterized in that, The RACH resource information includes one of the following: A set of candidate RACH resource information; A set of candidate RACH resource information and the probability value associated with each candidate RACH resource information.

7. The method according to any one of claims 1-5, characterized in that, The transmit power parameter includes one of the following: A set of candidate RACH transmit power parameters; A set of candidate RACH transmit power parameters and the probability values ​​associated with each candidate RACH transmit power parameter.

8. The method according to claim 1, characterized in that, The input information of the AI ​​model includes at least one of the following: Uplink signal reception strength information between the first and second devices; Downlink signal reception strength information between the first device and the second device; Signal quality information of the uplink signal between the first and second devices; Signal quality information of the downlink signal between the first device and the second device; Path loss between the first and second devices; Distance information between the first and second devices; The number of random access failures in RACH transmission between the first and second devices; The number of times the random access response message reception failed during RACH transmission between the first and second devices; The maximum number of RACH transmissions between the first and second devices; The number of retransmissions of RACH transmission between the first and second devices; The number of failed attempts to switch between different RACH transmission types between the first and second devices; The absolute timing advance (TA) value between the first and second devices; The relative TA value between the first and second devices; Beam information between the first and second devices; Propagation delay between the first and second devices; Round-trip time (RTT) between the first and second devices; The maximum transmit power of RACH transmission, TRP identifier, TRP group identifier, cell identifier, cell group identifier, timing advance group TAG identifier, tracking area identifier, radio access network notification area RNA identifier, frequency domain resource information, load information, interference information, and information of the first device.

9. The method according to claim 8, characterized in that, The information of the first device includes at least one of the following: Transmission power information, location information, distribution information, direction of movement, speed of movement, energy consumption information, power consumption information, operator information, network type information, panel orientation information, device type, sensing information, network scene information, environmental information, ephemeris information, reference position of the cell in the non-terrestrial network NTN scene, movement trajectory of the cell in the NTN scene, multipath information of the channel, time information, compensation information, and offset information.

10. The method according to claim 9, characterized in that, The compensation information includes at least one of the following: Timing pre-compensation, wherein the offset information includes at least one of the following: timing advance offset, common timing advance offset, and dedicated time advance offset; Frequency compensation, wherein the offset information includes at least one of the following: frequency offset, common frequency offset, and proprietary frequency offset.

11. The method according to claim 8, characterized in that, The RACH transmission includes at least one of the following: RACH transmission corresponding to the first RACH resource; RACH transmission corresponding to the first reference signal; The first transmission configuration indicates the RACH transmission corresponding to TCI; RACH transmission corresponding to the first TRP; The first RACH resource includes at least one of the following: RO resource, preamble format, preamble index, and preamble sequence; the first reference signal includes at least one of the following: synchronization signal physical broadcast channel block (SSB) and channel state information reference signal (CSI-RS).

12. The method according to claim 8, characterized in that, The maximum transmit power is the maximum transmit power of the first device for RACH transmission in at least one of the following: cell, cell group, frequency layer, TA area, TRP, network node, carrier, band, subband, and partial bandwidth.

13. The method according to claim 8, characterized in that, The TRP identifier corresponds to at least one of the following TRPs: source TRP, target TRP, currently selected TRP, residing TRP, currently accessed TRP, primary TRP, and secondary TRP; The TRP group identifier corresponds to at least one of the following TRP groups: source TRP group, target TRP group, currently selected TRP group, residing TRP group, currently accessed TRP group, primary TRP group, and secondary TRP group; The cell identifier is the physical layer cell identifier of the first cell; The cell group identifier is the identifier of the first cell group; The TAG identifier is the TAG identifier of the first cell; The tracking area identifier is the tracking area identifier of the first cell; The RNA identifier is the RNA identifier of the first cell; The first cell includes one of the following: source cell, target cell, currently stationed cell, currently accessing cell, primary cell, and secondary cell; the first cell group includes one of the following: source cell group, target cell group, currently stationed cell group, currently accessing cell group, primary cell group, and secondary cell group.

14. The method according to claim 8, characterized in that, The beam information includes at least one of the following: reference signal index, beam index, and beam direction; The load information is the load information corresponding to at least one of SSB, beam, TCI, TRP, cell, carrier and frequency point; The interference information is interference information corresponding to at least one of SSB, beam, TCI, TRP, cell, carrier, and frequency point.

15. The method according to claim 1, characterized in that, The triggering conditions for the AI ​​model inference include at least one of the following: AI timer timed out; RACH configuration using network configuration; Use the RACH type configured in the network; The collision probability of RACH transmission is greater than or equal to the first threshold; The number of random access failures in the RACH transmission corresponding to the first information is greater than or equal to the second threshold; The number of failed receptions of the random access response message for the RACH transmission corresponding to the first information is greater than or equal to the third threshold; The number of RACH transmissions corresponding to the first information that reach the maximum number of transmissions is greater than or equal to the fourth threshold; The number of repeated transmissions used in the RACH transmission corresponding to the first information is greater than or equal to the fifth threshold; The number of failed attempts to switch between different RACH transmission types is greater than or equal to the sixth threshold; The RACH transmission corresponding to the first message reaches maximum transmit power; The transmit power of the RACH transmission corresponding to the first information is greater than or equal to the seventh threshold; The number of power boosts in the RACH transmission corresponding to the first information is greater than or equal to the eighth threshold; The number of failed RACH transmissions corresponding to switching different first information is greater than or equal to the ninth threshold; Community re-selection; TRP reselection; Cell handover; TRP switching; The conditions for cell handover are met; The first device triggers access to the secondary cell; The first device moves to the preset position; Upstream business has arrived; Downlink traffic has arrived; Preset type of service arrives; Network indication triggered; The speed of the first device is greater than or equal to the tenth threshold; The acceleration of the first device is greater than or equal to the eleventh threshold; The location of the first device has changed; The network location has changed; The position change of the first device is greater than or equal to the twelfth threshold; The network location change is greater than or equal to the thirteenth threshold; The beam selected by the first device changes; The spatial transmission filter selected by the first device has changed; The first information includes at least one of the following: RO resources, preamble format, preamble index, preamble sequence, SSB, beam, TCI, and TRP.

16. The method according to claim 1, characterized in that, Also includes: A third device acquires training samples, which include at least one label and input information used by the AI ​​model for inference. The fourth device uses the training samples to train the model to obtain the AI ​​model. The AI ​​model outputs second information, which is used for RACH transmission and includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

17. The method according to claim 1, characterized in that, The first device is a network-side device, and the input information for the AI ​​model is indicated by the terminal or other network-side devices through cross-carrier OTT messages.

18. The method according to claim 1, characterized in that, Also includes: The AI ​​model is obtained by jointly training the first device and the second device.

19. The method according to claim 1, characterized in that, It also includes one of the following: Under preset conditions, training of the AI ​​model is triggered; Periodically trigger the training of the AI ​​model; Semi-static triggering is used to train the AI ​​model.

20. The method according to claim 19, characterized in that, The preset condition includes at least one of the following: AI timer timeout; RACH configuration using network configuration; Use the RACH type configured in the network; The collision probability corresponding to RACH transmission is greater than or equal to the fourteenth threshold; The number of random access failures in the RACH transmission corresponding to the first information is greater than or equal to the fifteenth threshold; The number of failed access response messages received by the RACH transmission corresponding to the first information is greater than or equal to the sixteenth threshold; the number of RACH transmissions corresponding to the first information that reach the maximum number of transmissions is greater than or equal to the seventeenth threshold. The number of repeated transmissions used in the RACH transmission corresponding to the first information is greater than or equal to the eighteenth threshold; The number of failed attempts to switch between different RACH transmission types is greater than or equal to the nineteenth threshold; The RACH transmission corresponding to the first message reaches maximum transmit power; The transmit power of the RACH transmission corresponding to the first information is greater than or equal to the twentieth threshold; The number of power boosts in the RACH transmission corresponding to the first information is greater than or equal to the twenty-first threshold; The number of failed RACH transmissions corresponding to switching different first information is greater than or equal to the twenty-second threshold; Community re-selection; TRP reselection; Cell handover; TRP switching; The conditions for cell handover are met; The first device triggers access to the secondary cell; The first device moves to the preset position; Upstream business has arrived; Downlink traffic has arrived; Preset type of service arrives; Network indication triggered; The speed of the first device is greater than or equal to the twenty-third threshold; The acceleration of the first device is greater than or equal to the twenty-fourth threshold; The location of the first device has changed; The network location has changed; The change in the position of the first device is greater than or equal to the twenty-fifth threshold; The network location change is greater than or equal to the 26th threshold; The external environment of the first device changes; The spatial transmission filter selected by the first device has changed; The beam selected by the first device changes; The TCI selected by the first device has changed; The SSB selected by the first device has changed; TAG timer timed out; Instructions from the network side; The first device is connected to the new community; The first device switches frequency points; The first device switches to the frequency band; First device switches carriers; The first device switches to the Public Land Mobile Network (PLMN); AI model inference failed; The AI ​​model failed inference N times in a row; The number of AI model inference failures is greater than or equal to the 27th threshold; Use AI models for reasoning; The first piece of equipment was moved to the new residential area; The first device moves to the tracking area; The first device moves to the preset geographical location; The moving speed of the first device changes; The RSRP measured by the first device changed; The RSRP change measured by the first device is greater than or equal to the 28th threshold; AI model training timer timeout preset time; N consecutive model supervisions occur; N instances of model supervision occurred; The number of random access failures in the RACH transmission corresponding to the first information in the AI ​​model inference is greater than or equal to the twenty-ninth threshold. The number of failed RACH transmission access response messages corresponding to the first information in the AI ​​model inference is greater than or equal to the thirtieth threshold. The number of RACH transmissions corresponding to the first information in the AI ​​model inference that reach the maximum number of transmissions is greater than or equal to the thirty-first threshold. The RACH transmission corresponding to the first information inferred by the AI ​​model reaches the maximum transmission power. The transmit power of the RACH transmission corresponding to the first information inferred by the AI ​​model is greater than or equal to the thirty-second threshold; The number of RACH transmission power boosts corresponding to the first information inferred by the AI ​​model is greater than or equal to the thirty-third threshold. The AI ​​model inference requires switching between different first information corresponding to RACH transmission failures with a number greater than or equal to the thirty-fourth threshold.

21. The method according to claim 1, characterized in that, Also includes: Obtain training-related information of the AI ​​model, wherein the training-related information includes at least one of the following: Input information for AI model training; Labels for AI model training; Loss function for AI model training; AI algorithms for training AI models; Reward information for AI model training; Adjustment information for AI model training.

22. The method according to claim 21, characterized in that, The label includes at least one of the following: Whether RACH transmission information that meets performance requirements is obtained; The success rate of RACH is greater than or equal to the 35th threshold; The RACH failure rate is less than the 36th threshold; The transmission power required for a successful RACH is less than the thirty-seventh threshold; The number of transmissions required for a successful RACH is less than the thirty-eighth threshold; The number of RACH transmission messages obtained; The duration of AI model training; Energy consumption for AI model training; The RACH transmission information required for transmission between the terminal and the cell network node.

23. The method according to any one of claims 1-22, characterized in that, The inference metrics of the AI ​​model include at least one of the following: The complexity of AI models; Latency of AI model inference; Success rate of AI model inference; The reliability of AI model inference results.

24. The method according to claim 1, characterized in that, Also includes: Obtain first indication information, which is used to indicate the activation or deactivation information of the AI ​​model; The AI ​​model is activated or deactivated based on the first indication information.

25. The method according to claim 1, characterized in that, Also includes: Obtain the first configuration information of the AI ​​model, wherein the first configuration information includes at least one of the following: AI models; Identification of AI models; The application scope of AI model reasoning; The inference cycle of AI models; The effective duration of AI model inference; Triggering conditions for AI model inference; The configuration parameters of the AI ​​model include at least one of the following: the input information of the AI ​​model, the output information of the AI ​​model, the type of the input information of the AI ​​model, and the type of the output information of the AI ​​model. Does it support joint inference across multiple devices? 26. The method according to any one of claims 1-25, characterized in that, Also includes: The first device sends third information to the network side; Receive instructions sent from the network side; Based on the aforementioned instructions, determine the AI ​​model to be used from among multiple AI models; The third information includes at least one of the following: the identifiers of multiple AI models and the classification identifiers of datasets; the indication includes at least one of the following: the identifier ID of the AI ​​model, the identifier of the dataset, and the configuration information for data collection.

27. The method according to any one of claims 1-25, characterized in that, It also includes at least one of the following: Send one or more of the input information, output information, and current state information of the first device to the device that trained the AI ​​model; The decision to pause or abandon is based on the initial information obtained from the AI ​​model; This triggers adjustments to the AI ​​model.

28. The method according to any one of claims 1-25, characterized in that, Also includes: The first device sends an instruction to the second device to provide information about the first device's AI capabilities. The AI ​​capability information includes at least one of the following: The first device may or may not have the ability to train AI models; The first device may or may not have the ability to use AI models for AI reasoning; The first device may or may not have the ability to send auxiliary information, which is used for AI model inference.

29. The method according to claim 1, characterized in that, Also includes: Obtain the supervised configuration information of the AI ​​model; The AI ​​model is supervised based on the supervision configuration information.

30. The method according to claim 29, characterized in that, The supervision configuration information includes at least one of the following: Identification of AI models to be supervised; The cycle of model supervision; The duration of model supervision; Window-related information for model supervision; Duration of the supervision window; The number of samples used for model supervision; Triggering conditions for model supervision; Metrics for model supervision; Labels for model supervision.

31. The method according to claim 29 or 30, characterized in that, The triggering conditions for AI model supervision include at least one of the following: The results of AI inference do not meet the accuracy requirements; At least one of the inference metrics of the AI ​​model fails to meet the requirements; The AI ​​model supervision metrics do not meet the requirements. The AI ​​model supervision metrics include: the error between the AI ​​model's predicted value and the actual value, and / or, the network's performance metrics. AI model supervision timer timed out; The terminal switches to a new cell or TRP or switches beams; AI model inference failed; The AI ​​model failed inference N times in a row; The number of inference failures by the AI ​​model reached the 39th threshold. Using AI model inference; The terminal has moved to a new cell, tracking area, or geographical location; The terminal's moving speed changes beyond the fortieth threshold within a specified time period; The external environment in which the AI ​​model's inference device is located changes.

32. A random access transmission device, characterized in that, include: The acquisition module is used to obtain initial information based on the AI ​​model. The transmission module is used to perform RACH transmission based on the first information; The first information includes at least one of the following: beam information, RACH collision probability, RACH resource information, and transmit power parameters.

33. A terminal, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the random access transmission method as described in any one of claims 1-31.

34. A network-side device, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the random access transmission method as described in any one of claims 1-31.

35. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the random access transmission method as described in any one of claims 1-31.

Images