Ai unit processing method and apparatus, and terminal and network-side device

Abstract

The present application belongs to the technical field of communications. Disclosed are an AI unit processing method and apparatus, and a terminal and a network-side device. The AI unit processing method in the embodiments of the present application comprises: a terminal executing, on the basis of first information, at least one of the following: disabling a first AI unit, and switching to a first mode; or, the terminal executing, on the basis of the first information, at least one of the following: enabling the first AI unit, and switching to a second mode, wherein in the first mode, the terminal executes a first operation on the basis of a non-AI unit, and in the second mode, the terminal executes the first operation on the basis of an AI unit, or executes the first operation on the basis of the non-AI unit and the AI unit.

Description

AI unit processing methods, devices, terminals, and network-side equipment

[0001] Cross-reference to related applications

[0002] This application claims priority to Chinese Patent Application No. 202411992093.0, filed on December 31, 2024, the disclosure of which is incorporated herein by reference in its entirety. Technical Field

[0003] This application belongs to the field of communication technology, specifically relating to an AI unit processing method, apparatus, terminal, and network-side equipment. Background Technology

[0004] In the field of communications, to achieve network and terminal energy conservation, functions such as Discontinuous Transmission (DTX) and Discontinuous Reception (DRX) have been defined. These functions allow the hardware of base stations or terminals to be temporarily shut down or switched to a more energy-efficient state when there is no data transmission or reception. Currently, with the introduction of Artificial Intelligence (AI) into communication systems, AI units also consume power, which may further increase the energy consumption of networks and terminals. Summary of the Invention

[0005] This application provides an AI unit processing method, apparatus, terminal, and network-side device, which can solve the problem of high power consumption of terminal and network-side devices in related technologies.

[0006] Firstly, an AI unit processing method is provided, executed by a terminal, the method comprising:

[0007] Based on the first information, the terminal performs at least one of the following: shuts down the first AI unit or switches to the first mode; or,

[0008] Based on the first information, the terminal performs at least one of the following: activates the first AI unit, or switches to the second mode;

[0009] In the first mode, the terminal performs the first operation based on a non-AI unit; in the second mode, the terminal performs the first operation based on an AI unit, or performs the first operation based on both a non-AI unit and an AI unit.

[0010] Secondly, an AI unit processing method is provided, executed by a network-side device, the method comprising:

[0011] The network-side device performs a second action, which includes at least one of the following:

[0012] Send first information to the terminal. The first information includes at least one of the following: first configuration, activation information of a first object, and deactivation information of the first object. The first object includes at least one of a first AI unit and the first configuration. The first configuration includes at least one of the following: first pattern configuration associated with the first AI unit, first DRX configuration, first DTX configuration, first cell DTX configuration, and first cell DRX configuration.

[0013] Based on the second information, perform at least one of the following: disable the third AI unit or switch to the third mode; or, based on the second information, perform at least one of the following: enable the third AI unit or switch to the fourth mode;

[0014] In the third mode, the network-side device performs the second operation based on the non-AI unit, and in the fourth mode, the network-side device performs the second operation based on the AI ​​unit, or based on both the non-AI unit and the AI ​​unit.

[0015] Thirdly, an AI unit processing device is provided for use in a terminal, the device comprising:

[0016] The first processing module is configured to perform at least one of the following based on the first information: shut down the first AI unit, switch to a first mode; or...

[0017] The first processing module is configured to perform at least one of the following based on the first information: activate the first AI unit, switch to the second mode;

[0018] In the first mode, the terminal performs the first operation based on a non-AI unit; in the second mode, the terminal performs the first operation based on an AI unit, or performs the first operation based on both a non-AI unit and an AI unit.

[0019] Fourthly, an AI unit processing device is provided, applied to network-side equipment, the device comprising:

[0020] The second processing module is configured to execute a second action, the second action including at least one of the following:

[0021] Send first information to the terminal. The first information includes at least one of the following: first configuration, activation information of a first object, and deactivation information of the first object. The first object includes at least one of a first AI unit and the first configuration. The first configuration includes at least one of the following: first pattern configuration associated with the first AI unit, first DRX configuration, first DTX configuration, first cell DTX configuration, and first cell DRX configuration.

[0022] Based on the second information, perform at least one of the following: disable the third AI unit or switch to the third mode; or, based on the second information, perform at least one of the following: enable the third AI unit or switch to the fourth mode;

[0023] In the third mode, the network-side device performs the second operation based on the non-AI unit, and in the fourth mode, the network-side device performs the second operation based on the AI ​​unit, or based on both the non-AI unit and the AI ​​unit.

[0024] Fifthly, an AI unit processing apparatus is provided, the apparatus being configured to perform the steps of the method described in the first aspect, or to implement the steps of the method described in the second aspect.

[0025] In a sixth aspect, a terminal is provided, the terminal including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the first aspect.

[0026] In a seventh aspect, a terminal is provided, including a processor and a communication interface, wherein the processor is configured to perform at least one of the following based on first information: turning off a first AI unit and switching to a first mode; or, based on the first information, performing at least one of the following: turning on the first AI unit and switching to a second mode;

[0027] In the first mode, the terminal performs the first operation based on a non-AI unit; in the second mode, the terminal performs the first operation based on an AI unit, or performs the first operation based on both a non-AI unit and an AI unit.

[0028] Eighthly, a network-side device is provided, the network-side device including a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the method as described in the second aspect.

[0029] A ninth aspect provides a network-side device, including a processor and a communication interface, wherein the processor is configured to perform a second action, the second action including at least one of the following:

[0030] Send first information to the terminal. The first information includes at least one of the following: first configuration, activation information of a first object, and deactivation information of the first object. The first object includes at least one of a first AI unit and the first configuration. The first configuration includes at least one of the following: first pattern configuration associated with the first AI unit, first DRX configuration, first DTX configuration, first cell DTX configuration, and first cell DRX configuration.

[0031] Based on the second information, perform at least one of the following: disable the third AI unit or switch to the third mode; or, based on the second information, perform at least one of the following: enable the third AI unit or switch to the fourth mode;

[0032] In the third mode, the network-side device performs the second operation based on the non-AI unit, and in the fourth mode, the network-side device performs the second operation based on the AI ​​unit, or based on both the non-AI unit and the AI ​​unit.

[0033] In a tenth aspect, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect, or implement the steps of the method described in the second aspect.

[0034] Eleventhly, a wireless communication system is provided, comprising: a terminal and a network-side device, wherein the terminal can be used to perform the steps of the method as described in the first aspect, and the network-side device can be used to perform the steps of the method as described in the second aspect.

[0035] In a twelfth aspect, a chip is provided, the chip including a processor and a communication interface coupled to the processor, the processor being configured to run programs or instructions to implement the method as described in the first aspect, or to implement the method as described in the second aspect.

[0036] In a thirteenth aspect, a computer program / program product is provided, the computer program / program product being stored in a storage medium, the computer program / program product being executed by at least one processor to implement the steps of the method as described in the first or second aspect.

[0037] In this embodiment, the terminal can disable the first AI unit and / or switch to a first mode based on non-AI units to perform the first operation according to the first information; alternatively, the terminal can enable the first AI unit and / or switch to a second mode according to the first information. This allows the terminal to control the first AI unit more flexibly based on the first information, preventing it from being constantly on and thus effectively saving terminal power. Furthermore, the terminal's ability to switch between the first and second modes based on the enabling or disabling of the first AI unit also prevents it from being constantly in the mode of performing the first operation based on the AI ​​unit. The terminal's ability to flexibly switch to the mode of performing the first operation based on non-AI units further contributes to energy conservation. Attached Figure Description

[0038] Figure 1a is a block diagram of a wireless communication system applicable to an embodiment of this application;

[0039] Figure 1b is a schematic diagram of AI-based CSI feedback and CSI prediction;

[0040] Figure 1c is a schematic diagram of NMSE for AI-based predicted CSI;

[0041] Figure 1d is one of the schematic diagrams of the DRX cycle;

[0042] Figure 1e is the second schematic diagram of the DRX cycle.

[0043] Figure 2 is a flowchart of an AI unit processing method provided in an embodiment of this application;

[0044] Figure 3 is a flowchart of another AI unit processing method provided in an embodiment of this application;

[0045] Figure 4 is a structural diagram of an AI unit processing device provided in an embodiment of this application;

[0046] Figure 5 is a structural diagram of another AI unit processing device provided in an embodiment of this application;

[0047] Figure 6 is a structural diagram of a communication device provided in an embodiment of this application;

[0048] Figure 7 is a structural diagram of a terminal provided in an embodiment of this application;

[0049] Figure 8 is a structural diagram of a network-side device provided in an embodiment of this application. Detailed Implementation

[0050] 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.

[0051] 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.

[0052] 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 the sender explicitly informing the receiver of specific information, the required operation, or the requested result in the instruction sent. An indirect instruction can be understood as the receiver determining the corresponding information based on the instruction sent by the sender, or making a judgment and determining the required operation or requested result based on the judgment result.

[0053] 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.

[0054] Figure 1a shows a block diagram of a wireless communication system applicable to an embodiment of this application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 can also be referred to as User Equipment (UE), and can be a mobile phone, tablet computer, laptop computer, notebook computer, personal digital assistant (PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile internet device (MID), augmented reality (AR), virtual reality (VR) device, robot, wearable device, flight vehicle, vehicle user equipment (VUE), shipboard equipment, pedestrian user equipment (PUE), smart home (home devices with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), game console, personal computer (PC), ATM, or self-service machine, etc. Wearable devices include: smartwatches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart chains, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among these, in-vehicle devices can also be referred to as in-vehicle terminals, in-vehicle controllers, in-vehicle modules, in-vehicle components, in-vehicle chips, or in-vehicle units, etc. It should be noted that the specific type of terminal 11 is not limited in this application embodiment. Network-side equipment 12 may include access network equipment or core network equipment, wherein access network equipment may also be referred to as Radio Access Network (RAN) equipment, radio access network function, or radio access network unit. Access network equipment may include base stations, Wireless Local Area Network (WLAN) access points (APs), or Wireless Fidelity (WiFi) nodes, etc.Among them, base stations 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), Non-Terrestrial Network (NTN) equipment (such as satellite or high altitude platform stations). The term "base station" can be any suitable term in the field, such as "station" or any other appropriate 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 term. It should be noted that the embodiments of this application only use the base station in the NR system as an example for introduction, and do not limit the specific type of base station.

[0055] 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 (L-NEF), and Binding Support. Functions include BSF, Application Function (AF), Location Management Function (LMF), Gateway Mobile Location Centre (GMLC), Network Data Analytics Function (NWDAF), and Non-Terrestrial Network (NTN) equipment (such as satellite or high altitude platform station).It should be noted that the embodiments of this application only use the core network equipment in the NR system as an example for introduction, and do not limit the specific type of core network equipment. If the name of the core network equipment mentioned in the embodiments of this application changes in subsequent protocol versions (e.g., 6G), it is also within the scope of protection of this application.

[0056] 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).

[0057] To better understand, the relevant concepts that may be involved in the embodiments of this application are explained below.

[0058] AI has been widely applied in various fields. AI modules can be implemented in various ways, such as neural networks, decision trees, support vector machines, and Bayesian classifiers. This application uses neural networks as an example, but does not limit the specific type of AI module.

[0059] Neural networks are composed of neurons. The parameters of a neural network are optimized using optimization algorithms. Optimization algorithms are a class of algorithms that help us minimize or maximize an objective function (sometimes called a loss function). The objective function is often a mathematical combination of model parameters and data. For example, given data X and its corresponding label Y, we construct a neural network model f(.). With the model, based on the input x, we can obtain the predicted output f(x), and we can calculate the difference between the predicted value and the true value (f(x) - Y), which is the loss function. Our goal is to find suitable weights and biases that minimize the value of the aforementioned loss function. The smaller the loss value, the closer our AI model is to reality.

[0060] Most common optimization algorithms are based on the error back propagation (BP) algorithm. The basic idea of ​​the BP algorithm is that the learning process consists of two parts: forward propagation of the signal and backward propagation of the error. During forward propagation, the input sample is introduced from the input layer, processed layer by layer by the hidden layers, and then propagated to the output layer. If the actual output of the output layer does not match the expected output, the process transitions to the error back propagation stage. Error back propagation involves propagating the output error back to the input layer layer by layer through the hidden layers, distributing the error to all units in each layer, thus obtaining the error signal of each unit. This error signal serves as the basis for adjusting the weights of each unit. This process of adjusting the weights through forward and backward propagation is repeated continuously. This continuous adjustment of weights is the learning and training process of the network. This process continues until the error of the network output is reduced to an acceptable level, or until the predetermined number of learning iterations is reached.

[0061] Generally, the AI ​​algorithms and models selected vary depending on the type of problem being solved. Among related technologies, the main method for improving 5G network performance using AI is to enhance or replace existing algorithms or processing modules through neural network-based algorithms and AI models. In specific scenarios, neural network-based algorithms and AI models can achieve better performance than deterministic algorithms. Commonly used neural networks include deep neural networks, convolutional neural networks, and recurrent neural networks. Existing AI tools can be used to build, train, and validate neural networks.

[0062] Replacing modules in existing systems with AI / machine learning (ML) methods can effectively improve system performance.

[0063] For example, Channel State Information (CSI) feedback and CSI prediction involve inputting historical CSI data into an AI model. The AI ​​model analyzes the temporal variation characteristics of the channel and outputs the future CSI. See Figure 1b for details.

[0064] The corresponding system performance is shown in Figure 1c. It can be seen that CSI feedback and CSI prediction offer a significant performance gain compared to the non-prediction approach. Furthermore, the achievable prediction accuracy varies depending on the predicted future time.

[0065] In addition to the aforementioned AI-based CSI feedback and CSI prediction use cases, AI can also be used to implement various other use cases such as CSI compression, positioning enhancement, beam management, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, and signal demodulation.

[0066] Network energy saving:

[0067] Network energy efficiency is crucial for environmental sustainability, reducing environmental impact (greenhouse gas emissions), and saving operating costs. As 5G becomes more widespread across industries and regions, and more advanced services and applications (such as XR) require very high data rates, networks are becoming denser, using more antennas, greater bandwidth, and more frequency bands. The environmental impact of 5G needs to be kept under control, necessitating the development of new solutions to improve network energy efficiency.

[0068] Discontinuous Transmission (DTX):

[0069] DTX is a technology used to reduce power consumption and improve spectrum efficiency. Its main goal is to save power and reduce interference by temporarily stopping data transmission when there is no useful information to send.

[0070] Cell Discontinuous Transmission (cell DTX) / Cell Discontinuous Reception (cell DRX):

[0071] The main energy-saving idea behind Cell DTX / DRX is to align the time-domain staggered DRX on-duration of different UEs to achieve DTX / DRX. This allows the cell to disable some or all base station signal transmission and reception, or shut down certain base station devices, during the aligned DTX / DRX off-duration, thus achieving energy savings. There are two potential implementation methods for Cell DTX / DRX: one is explicit Cell DTX / DRX configuration, considering how it works in conjunction with UE DRX; the other is for the base station to reconfigure the DRX of different UEs for time-domain alignment.

[0072] If Cell DTX / DRX is explicitly configured, it must include at least the following parameters: periodicity, start slot / offset, and on duration. Cell DTX / DRX can be activated dynamically via L1 / L2 signaling or Radio Resource Control (RRC) signaling.

[0073] Cell DTX configuration and Cell DRX configuration can be configured together or separately.

[0074] UE DRX (RRC Idle State DRX):

[0075] In idle mode, the device periodically wakes up to listen for paging occupancy (PO), an opportunity for the network to notify the device of incoming calls, messages, or notifications. The DRX cycle defines the frequency and duration of these wake-up cycles. A typical DRX cycle includes a DRX-on duration (during which the device listens for PO) and a DRX-off duration (during which the device turns off its radio to conserve power).

[0076] UE C-DRX (RRC Connected-State DRX):

[0077] Connected-mode DRX (C-DRX) was introduced in 5G / LTE to improve UE battery power consumption by allowing the UE to periodically enter a "sleep" state (off duration) during which monitoring of the Physical downlink control channel (PDCCH) is not required. To monitor the PDCCH for potential downlink / uplink data, the UE is allowed to periodically wake up and remain "awake" (on duration) for a certain amount of time before re-entering "sleep." Additionally, the UE may occasionally need to wake up to monitor the PDCCH, for example, to receive possible retransmissions.

[0078] A basic DRX cycle is shown in Figures 1d and 1e. Referring to Figure 1d, the DRX cycle consists of "On Duration" and "Opportunity for DRX": during the "On Duration" period, the UE listens to the PDCCH (not detected in the figure); during the "Opportunity for DRX" period, the UE does not listen to the PDCCH to save power.

[0079] In addition, as shown in Figure 1e, the network is configured with an inactivity timer. If a new PDCCH is received within the duration, the inactivity timer will be started or restarted to extend the duration for which the UE listens to the PDCCH.

[0080] It should be noted that the time-domain unit in the embodiments of this application may be one or more symbols, time slots, radio subframes, radio frames, ms, s, etc.

[0081] In the embodiments of this application, the frequency domain unit may be one or more resource elements (REs), resource blocks (RBs), RE groups, RB groups, etc.

[0082] The AI ​​unit described in this application embodiment may also be referred to as an AI model, AI structure, etc., or the AI ​​unit may refer to a processing unit capable of implementing specific algorithms, formulas, processing flows, capabilities, etc. related to AI, or the AI ​​unit may be a processing method, algorithm, function, module, or unit for a specific dataset, or the AI ​​unit may be a processing method, algorithm, function, module, or unit running on AI-related hardware such as a graphics processing unit (GPU), neural network processing unit (NPU), tensor processing unit (TPU), application-specific integrated circuit (ASIC), etc. This invention does not specifically limit these aspects. Optionally, the specific dataset includes the input and / or output of the AI ​​unit.

[0083] The identifier of the AI ​​unit may be an AI model identifier, AI structure identifier, AI algorithm identifier, function ID, physical identifier, logical identifier, global identifier, local identifier, or the identifier of a specific dataset associated with the AI ​​unit, or the identifier of a specific scenario, environment, channel characteristics, or device related to the AI, or the identifier of a function, characteristic, capability, or module related to the AI, etc.

[0084] The AI ​​unit can be a base station AI unit, such as AIDBA (discontinuous BS AI), which is only applied to the base station AI unit; the AI ​​unit can also be a terminal AI unit, such as AIDUA (discontinuous UE AI), which is only applied to the terminal AI unit; or AIDA can be applied to both the base station AI unit and the terminal AI unit.

[0085] The AI ​​unit processing method, apparatus, terminal, and network-side device provided in this application will be described in detail below with reference to the accompanying drawings and through some embodiments and application scenarios.

[0086] Please refer to Figure 2, which is a flowchart of an AI unit processing method provided in an embodiment of this application. The method is executed by a terminal. As shown in Figure 2, the method includes the following steps:

[0087] Step 201: The terminal performs at least one of the following actions based on the first information: turning off the first AI unit or switching to the first mode; or, the terminal performs at least one of the following actions based on the first information: turning on the first AI unit or switching to the second mode.

[0088] In the first mode, the terminal performs the first operation based on a non-AI unit; in the second mode, the terminal performs the first operation based on an AI unit, or performs the first operation based on both a non-AI unit and an AI unit. The terminal performing the first operation based on the AI ​​unit can be done by performing the first operation on top of the AI ​​unit, or with the assistance of the AI ​​unit, or by utilizing the AI ​​unit to perform the first operation.

[0089] It should be noted that the terminal obtains the first information before performing the above steps based on the first information.

[0090] Optionally, the method further includes:

[0091] The terminal receives the first information sent by the network-side device.

[0092] For example, the first information could be configuration information provided to the terminal by the network-side device, such as activation or deactivation information for the first AI unit. Thus, the terminal can enable or disable the first AI unit, and / or switch between the first and second modes, based on the first information sent by the network-side device, which helps to standardize the terminal's behavior.

[0093] Alternatively, the first information may be pre-configured or pre-agreed, or determined by the terminal itself based on information such as the service type.

[0094] In some implementations, the terminal shuts down the first AI unit based on first information. For example, if the first information includes deactivation information for the first AI unit, the terminal shuts down the first AI unit based on the deactivation information.

[0095] In some implementations, the terminal switches to a first mode based on the first information, that is, the terminal switches to a mode that performs the first operation based on a non-AI unit. In this case, the terminal may turn off the first AI unit, for example, turn off some less relevant AI units, so as to perform the first operation based on a non-AI unit, which can effectively improve the terminal's energy-saving effect. Alternatively, when the terminal performs the first operation based on a non-AI unit, the first AI unit can still be in an on state, for example, the first AI unit switches to a more energy-efficient state, in which case the terminal does not use the first AI unit and / or other AI units to perform the first operation.

[0096] In some implementations, the terminal shuts down the first AI unit and switches to a first mode based on the first information, that is, the terminal switches to a mode that performs the first operation based on a non-AI unit, and the first AI unit is in a shut-down state.

[0097] In the above implementation, the terminal shuts down the first AI unit according to the first information and / or switches to a first mode that performs the first operation based on the non-AI unit, effectively reducing the increase in energy consumption caused by turning on the AI ​​unit or performing related operations based on the AI ​​unit, thereby reducing the terminal's energy consumption.

[0098] In some implementations, the terminal activates the first AI unit based on first information. For example, if the first information includes activation information of the first AI unit, the terminal activates the first AI unit based on the activation information of the first AI unit.

[0099] In some implementations, the terminal switches to a second mode based on the first information, that is, the terminal switches to a mode in which the first operation is performed based on the AI ​​unit, or switches to a mode in which the first operation is performed based on both the non-AI unit and the AI ​​unit.

[0100] In some implementations, the terminal activates the first AI unit and switches to a second mode based on the first information, that is, switches to a mode that performs the first operation based on the AI ​​unit, or switches to a mode that performs the first operation based on both the non-AI unit and the AI ​​unit.

[0101] In the above embodiments, the terminal activates the first AI unit and / or switches to the second mode based on the first information. Thus, the terminal can activate the first AI unit and / or switch to the mode of performing the first operation based on the AI ​​unit based on the first information, so as to better manage the AI ​​unit on the terminal side and make the activation of the AI ​​unit more flexible.

[0102] Optionally, the first operation includes at least one of the following: beam management, mobility management, CSI feedback, CSI prediction, CSI compression, positioning, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, signal demodulation, sensing, synchronization, measurement, reporting, initial access, and random access.

[0103] In this embodiment, the terminal can turn off the first AI unit and / or switch to a first mode based on non-AI units to perform the first operation, thereby effectively saving terminal power; or, the terminal can turn on the first AI unit and / or switch to a second mode based on the first information. In this way, the terminal can turn the first AI unit on or off according to the first information, thus enabling more flexible control of the first AI unit and preventing it from always being on, thereby effectively saving terminal power. Furthermore, the terminal can switch between the first and second modes based on the turning the first AI unit on or off, which also prevents the terminal from always being in the mode of performing the first operation based on the AI ​​unit. The terminal can flexibly switch to the mode of performing the first operation based on non-AI units, which also helps save terminal power consumption.

[0104] Optionally, the first information includes at least one of the following:

[0105] First configuration;

[0106] Terminal connection status;

[0107] The energy-saving status of the terminal;

[0108] Activation information of the first object;

[0109] Deactivation information for the first object;

[0110] The first object includes at least one of the first AI unit and the first configuration, and the first configuration includes at least one of the following: the first pattern configuration, the first DRX configuration, the first DTX configuration, the first cell DTX configuration, and the first cell DRX configuration associated with the first AI unit.

[0111] For example, if the first information includes activation information for the first AI unit, the terminal activates the first AI unit based on the activation information; if the first information includes deactivation information for the first AI unit, the terminal deactivates the first AI unit based on the deactivation information. Alternatively, the terminal may activate the first AI unit upon receiving the activation information, and may choose not to activate or deactivate the first AI unit upon receiving the deactivation information.

[0112] For example, if the first information includes the first pattern configuration associated with the first AI unit, then the terminal can turn the first AI unit on or off according to the content of the first pattern configuration. For example, the first AI unit can be turned off when the first pattern configuration is inactive, and the first AI unit can be turned on when the first pattern configuration is active, etc., which will not be elaborated here.

[0113] For example, the first information includes the first DRX configuration. The terminal can enable the first AI unit when the terminal is in the active state of DRX, and disable the first AI unit and / or switch to the first mode when the terminal is in the inactive state of DRX, thereby saving terminal power consumption.

[0114] For example, the first information includes the first DTX configuration. The terminal can activate the first AI unit when the terminal is in the DTX active state, and deactivate the first AI unit and / or switch to the first mode when the terminal is in the DTX inactive state, thereby saving terminal power consumption.

[0115] For example, the first information includes the first cell DTX configuration or the first cell DRX configuration, so that the terminal can turn the first AI unit on or off according to the energy-saving mode of the network-side device. For example, the first AI unit is turned on in the active state of the first cell DRX and turned off in the inactive state. Or, the first AI unit is turned on in the active state of the first cell DTX and turned off in the inactive state, so as to save terminal power consumption. This also allows the terminal to combine the turning on or off of the first AI unit with the energy-saving mode of the network-side device.

[0116] It should be noted that the inactive state of the first configuration can also be described as the inactive period of the first configuration or the idle state of the first configuration; the active state of the first configuration can also be described as the active period of the first configuration.

[0117] For example, when the first information includes the terminal's connection status, the terminal performs different behaviors under different connection statuses. For instance, when the terminal is in an idle or inactive state, it may shut down the first AI unit or switch to a first mode to save terminal power consumption; or when the terminal is in a connected state, it may turn on the first AI unit or switch to a second mode.

[0118] In addition, the AI ​​units that can be enabled when the terminal is in a connected state, an inactive state, or an idle state can be different, or the AI ​​units that need to be disabled when the terminal is in a connected state, an inactive state, or an idle state can also be different. This defines which AI units the terminal is allowed to enable or need to disable in different states, which helps the terminal to have more flexible control over AI units.

[0119] Optionally, if the first information includes the connection status of the terminal, the first AI unit that is turned off by the terminal in the idle state includes at least one of the following:

[0120] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than a first preset power consumption, AI unit with computing power occupied greater than a first threshold, AI unit with priority lower than a first preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0121] Alternatively, the first AI unit activated by the terminal in the idle state includes at least one of the following:

[0122] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0123] Optionally, if the first information includes the connection status of the terminal, the first AI unit that is turned off by the terminal in the inactive state includes at least one of the following:

[0124] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than a first preset power consumption, AI unit with computing power occupied greater than a first threshold, AI unit with priority lower than a first preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0125] Alternatively, the first AI unit activated by the terminal in the inactive state includes at least one of the following:

[0126] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0127] Optionally, if the first information includes the connection status of the terminal, the first AI unit that is turned off by the terminal in the connected state includes at least one of the following:

[0128] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than a first preset power consumption, AI unit with computing power occupied greater than a first threshold, AI unit with priority lower than a first preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0129] Alternatively, the first AI unit activated by the terminal in the connected state includes at least one of the following:

[0130] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0131] For example, when the first information includes the terminal's energy-saving state, the terminal performs different behaviors in different energy-saving states. For instance, in the first energy-saving state, the terminal may turn the first AI unit on or off, or switch to a first mode, to save terminal power consumption; or the terminal may not turn on the first AI unit in the first energy-saving state, or switch to a second mode. The first energy-saving state is predefined by the protocol, preconfigured by the network side, configured by the network side, or determined by the terminal.

[0132] Optionally, if the first information includes the power-saving state of the terminal, the first AI unit that the terminal shuts down in the first power-saving state includes at least one of the following:

[0133] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than a first preset power consumption, AI unit with computing power occupied greater than a first threshold, AI unit with priority lower than a first preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0134] Alternatively, the first AI unit activated by the terminal in the first energy-saving state includes at least one of the following:

[0135] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0136] This application defines the content of the first information, so that the terminal can determine under what circumstances the first AI unit is turned on or off based on the content of the first information, effectively regulating the terminal behavior.

[0137] Optionally, the first pattern configuration includes at least one of the following:

[0138] The duration of the first AI active state timer;

[0139] The duration of the first AI inactive state timer;

[0140] The starting time position of the first pattern activation state within the cycle, for example, the starting time position of the first pattern activation state within the cycle of the first pattern;

[0141] The end time position of the first pattern activation state within the cycle;

[0142] The starting time position of the inactive state of the first pattern within the period, for example, the starting time position of the inactive state of the first pattern within the period of the first pattern;

[0143] The end time position of the inactive state of the first pattern within the cycle;

[0144] The effective start time and position of the first pattern;

[0145] The cycle of the first pattern;

[0146] The AI ​​unit or AI unit type that the terminal is allowed to activate when the first pattern is inactive;

[0147] The pattern configuration parameters applied when the terminal is in a connected state;

[0148] The pattern configuration parameters applied when the terminal is in an inactive state;

[0149] The pattern configuration parameters applied when the terminal is in an idle state.

[0150] For example, the first pattern configuration includes the duration of a first AI activation state timer within a first pattern cycle. Based on the duration of the first AI activation state timer and the start position of the first AI activation state within the first pattern, the terminal can determine the duration for which the first AI unit can be activated. This duration can be the duration of the first AI activation state timer, thus allowing the terminal to determine when to disable the first AI unit and / or switch to the first mode. In some embodiments, the start position of the first AI activation state within the first pattern cycle is not configured, indicating that the first AI activation state begins at the start position of the first pattern cycle or begins at the end position of the first AI inactive state. In some embodiments, the end position of the first AI activation state within the first pattern cycle is not configured, indicating that the first AI activation state ends at the end position of the first pattern cycle or ends at the start position of the first AI inactive state.

[0151] For example, the first pattern configuration includes the duration of a first AI inactive state timer within a first pattern period. Based on this first pattern configuration, the terminal can determine the duration the first AI unit is off or in the first mode, and thus know when to turn on the first AI unit and / or switch to the second mode. In some embodiments, the start position of the first AI inactive state within the first pattern period is not configured, indicating that the first AI inactive state begins at the start position of the first pattern period or begins at the end position of the first AI active state. In some embodiments, the end position of the first AI inactive state within the first pattern period is not configured, indicating that the first AI inactive state ends at the end position of the first pattern period or ends at the start position of the first AI active state.

[0152] It is understandable that a first pattern cycle includes a first AI active state and a first AI inactive state. The time occupied by the first AI active state plus the time occupied by the first AI inactive state constitutes a first pattern cycle.

[0153] For example, the first pattern configuration includes the AI ​​units or AI unit types that the terminal is allowed to enable when the first pattern is inactive. Based on this first pattern configuration, the terminal can determine which AI units or types of AI units can be enabled when the first pattern is inactive, and thus enable the corresponding AI units (these AI units can be the first AI unit). It should be noted that different pattern configurations may allow different AI units to be enabled or disabled, allowing for more flexible control over the terminal-side AI units by enabling or disabling different AI units according to different pattern configurations.

[0154] For example, depending on the different AI and energy-saving requirements of the terminal in different states (connected state, inactive state, idle state), the first pattern configuration may include the pattern configuration parameters applied when the terminal is in the connected state, or the pattern configuration parameters applied when the terminal is in the inactive state, or the pattern configuration parameters applied when the terminal is in the idle state. The pattern configuration parameters corresponding to different states can be different, so the terminal can enable or disable the AI ​​unit according to different pattern configurations in different states, making the control of the AI ​​unit more flexible.

[0155] Furthermore, the AI ​​units allowed to be enabled in the active state of the first pattern configuration can differ depending on whether the terminal is in a connected, inactive, or idle state. Similarly, the AI ​​units that need to be disabled in the active state of the first pattern configuration can also differ depending on whether the terminal is in a connected, inactive, or idle state. This defines which AI units the terminal is allowed to enable or disable in different states, facilitating more flexible control of the AI ​​units. Of course, the first pattern configuration can also include other content mentioned above, which will not be listed here.

[0156] In this embodiment, the terminal can enable or disable the first AI unit according to the specific content configured in the first pattern, and / or switch between the first mode and the second mode, which helps to reduce terminal power consumption and makes the terminal's control of the AI ​​unit more flexible.

[0157] Optionally, when the first information includes the first configuration, the first AI unit that the terminal disables in the first state includes at least one of the following:

[0158] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than a first preset power consumption, AI unit with computing power occupied greater than a first threshold, AI unit with priority lower than a first preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0159] Alternatively, the first AI unit activated by the terminal in the first state includes at least one of the following:

[0160] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0161] The first state includes at least one of the following: a first sub-state and a second sub-state. The first sub-state includes one of the following: the first configuration is inactive, the first configuration is active. The second sub-state includes one of the following: the terminal is in idle, the terminal is inactive, the terminal is connected, or the terminal is in a first power-saving state.

[0162] Optionally, the first energy-saving state is associated with at least one of the terminal's energy consumption status and the terminal's battery status. The first energy-saving state may be predefined by the protocol, preconfigured by the network-side device, or determined by the terminal. For example, when the terminal's battery is low, the terminal is in the first energy-saving state.

[0163] It should be noted that the AI ​​unit with the specific identifier can be predefined by the protocol, pre-configured or configured by the network-side device, or determined by the terminal.

[0164] For example, taking the first configuration including the first pattern configuration as an example, the first AI unit that needs to be turned off when the first pattern configuration is inactive includes at least one of the following: AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0165] The first AI unit, which can be enabled (or can also be described as allowed to be enabled, required to be enabled, desired to be enabled by the terminal, desired to be enabled by the network-side device, etc.) in the inactive state of the first pattern configuration, includes at least one of the following: AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, non-continuous base station AI unit, non-continuous terminal AI unit, non-continuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0166] For example, the activation or deactivation of the first AI unit is associated with the terminal's connection state (e.g., idle state, inactive state, connected state) or power-saving state (e.g., first power-saving state). For instance, when the terminal is in an idle state, the first AI unit that needs to be deactivated includes AI units related to CSI, which can be specifically identified AI units. That is, the first AI unit that can be activated when the terminal is in a connected state and in the inactive state of the first DRX is different from the first AI unit that can be activated when the terminal is in an idle state and in the inactive state of the first DRX. This is because the first operation that the terminal needs to perform may differ in different connection states, and the terminal's power-saving requirements also differ in different connection states.

[0167] It is understandable that the AI ​​units that can be enabled in the first configuration activation state can be different when the terminal is in the connected state, inactive state, and idle state, or the AI ​​units that need to be disabled in the first configuration activation state can also be different when the terminal is in the connected state, inactive state, and idle state. This defines which AI units the terminal is allowed to enable or need to disable in different states, which helps the terminal to have more flexible control over AI units.

[0168] Of course, the first configuration may also include the first DRX configuration, the first DTX configuration, the first cell DTX configuration, and the first cell DRX configuration. The first state may be the first configuration being in an inactive state or an active state, and / or the terminal being in an idle state / inactive state / connected state / first power-saving state. The first AI unit that the terminal needs to turn off or the first AI unit that needs to be turned on in these states can refer to the examples of the first pattern configuration above, which will not be specifically listed here.

[0169] In this embodiment of the application, when the first information includes the first configuration, the terminal can determine the first AI unit that needs to be turned off or the first AI unit that can be turned on based on the state of the terminal (e.g., idle state, inactive state, connected state, first energy-saving state) and the state of the first configuration (active state, inactive state). This allows the terminal to more flexibly control the first AI unit and also helps the terminal achieve energy saving.

[0170] Optionally, the first information includes the first configuration, and the terminal activates the first AI unit according to the first information, including:

[0171] When the terminal receives the activation information of the first configuration, the terminal activates the first AI unit in the activated state of the first configuration;

[0172] Alternatively, the method may further include:

[0173] When the terminal receives the activation information of the first configuration, the terminal shuts down the first AI unit and / or turns on the second AI unit in the inactive state of the first configuration.

[0174] For example, when the terminal receives activation information configured by the first pattern, the terminal activates the first AI unit in the activated state of the first pattern configuration, or deactivates the first AI unit in the deactivated state of the first pattern configuration. That is, whether the first pattern configuration is activated or deactivated requires activation and deactivation information. When the first pattern is not activated, it is not necessary to consider whether the first AI unit needs to be deactivated in the deactivated state of the first pattern or whether it needs to be activated in the activated state of the first pattern. Instead, it is only necessary to consider whether the first AI unit needs to be activated in the activated state of the first pattern or deactivated in the deactivated state after the first pattern is activated, thereby enabling the terminal to better control the first AI unit.

[0175] It is understandable that whether the first configuration is activated or not requires activation and deactivation information. When the first configuration is not activated, it is not necessary to consider whether the first AI unit needs to be turned off in the inactive state of the first configuration or whether it needs to be turned on in the active state of the first configuration. Instead, it is only necessary to consider whether the first AI unit needs to be turned on in the active state of the first configuration or whether it needs to be turned off in the inactive state after the first configuration is activated.

[0176] Optionally, the activation information configured in the first configuration is associated with N first cycles, or with a time length (also known as a time window), where N is a positive integer. Therefore, based on the N first cycles or time length, the terminal can determine the effective duration of the activation information configured in the first configuration, and the terminal can then perform actions such as activating the first AI unit or switching to the first mode within the effective duration of the first configuration activation information, effectively regulating terminal behavior. Optionally, the method further includes:

[0177] When the terminal activates the first AI unit, a first timer is activated; wherein, before the first timer expires, the terminal does not switch the first configuration to an inactive state and / or does not deactivate the first AI unit.

[0178] For example, taking the first configuration including the first pattern configuration as an example, the terminal starts the first AI unit and starts the first timer according to the first pattern configuration. Before the first timer expires, the terminal does not shut down the first AI unit, thereby avoiding the discontinuity of AI results or the reduction of the accuracy of AI unit results due to shutting down the first AI unit, thereby ensuring the accuracy, continuity and latency of the AI ​​results of the first AI unit.

[0179] Alternatively, the terminal can enable the first AI unit and start the first timer according to the first pattern configuration. Before the first timer expires, the terminal will not switch the first pattern configuration to the inactive state to avoid the first unit being shut down when the first pattern configuration is switched to the inactive state, thereby causing the AI ​​results of the first AI unit to be discontinuous or the accuracy to be reduced, or increasing the latency of the AI ​​results, so as to ensure the accuracy, continuity and latency of the AI ​​results of the first AI unit.

[0180] For example, the first configuration includes a first DRX configuration. The terminal starts the first AI unit and starts a first timer according to the first DRX configuration. Before the first timer expires, the terminal does not shut down the first AI unit and / or the terminal does not switch the first DRX configuration to an inactive state, so as to avoid the first AI unit being shut down due to the first DRX configuration being switched to an inactive state, so as to ensure the continuity and accuracy of the AI ​​results of the first AI unit.

[0181] Alternatively, the terminal can activate the first AI unit according to the first DRX configuration and start the first timer. Before the first timer expires, the terminal can still activate the first AI unit in the first DRX inactive state, thereby ensuring that the first AI unit remains activated in the first DRX inactive state and ensuring the continuity and accuracy of the AI ​​results of the first AI unit.

[0182] Of course, when the first configuration is the first DTX configuration, the first cell DRX configuration, or the first cell DTX configuration, the terminal can perform the same behavior as under the first pattern configuration and the first DRX configuration described above, which will not be elaborated here. In this embodiment of the application, the terminal activates the first AI unit according to the first information, including:

[0183] When the terminal receives the activation information of the first object, the terminal activates the first AI unit;

[0184] The terminal shuts down the first AI unit based on the first information, including:

[0185] Upon receiving deactivation information for the first object, the terminal shuts down the first AI unit.

[0186] For example, if the first object includes a first AI unit, the terminal will activate the first AI unit upon receiving activation information, or deactivate the first AI unit upon receiving deactivation information. Alternatively, if the first object includes a first configuration, the terminal will activate the first AI unit upon receiving activation information for the first configuration, or deactivate the first AI unit upon receiving deactivation information for the first configuration. Thus, the terminal can activate or deactivate the first AI unit based on the activation or deactivation information of the first object, facilitating flexible control of the first AI unit and enabling the terminal to deactivate the first AI unit based on deactivation information, preventing the first AI unit from being constantly activated and thus helping to save terminal energy.

[0187] It should be noted that after the terminal receives the activation information of the first object, the terminal activates the first AI unit when the first object is in an activated state, and / or deactivates the first AI unit when the first object is in an inactive state, thereby better standardizing the terminal's control behavior over the first AI unit.

[0188] Optionally, the method may further include:

[0189] When the terminal receives the deactivation information of the first object, the terminal does not turn on the first AI unit when the first object is in an active state, or does not turn off the first AI unit when the first object is in an inactive state.

[0190] For example, after the terminal activates the first AI unit according to the activation information of the first object, the terminal will not activate the first AI unit in the active state of the first object, or will not deactivate the first AI unit in the inactive state of the first object, until the terminal receives the deactivation information of the first object. Alternatively, after the terminal activates the first AI unit according to the activation information of the first configuration, the terminal activates the first AI unit in the active state of the first configuration before receiving the deactivation information of the first configuration. This can also better regulate the terminal's control behavior of the first AI unit.

[0191] It should be noted that when the terminal receives the activation information of the first object, the activation information of the first object indicates the content and / or configuration of the first AI unit. Therefore, based on the content and / or configuration of the first AI unit, the terminal can determine which first AI unit can be activated. For example, if the content of the first AI unit indicated in the activation information of the first object is related to CSI prediction, then the terminal can clearly identify the first AI unit that can be activated as the AI ​​unit used for CSI prediction.

[0192] In some implementations, the activation or deactivation information of the first object is sent by the network-side device and is displayed as an indication. In other implementations, the activation or deactivation information of the first object is sent from a higher layer of the terminal to a lower layer, for example, from the MAC layer to the physical layer, or from the RRC layer to the MAC / physical layer.

[0193] Optionally, if the first condition is met, the terminal expects to receive activation information of the first object sent by the network-side device, wherein the first condition includes at least one of the following:

[0194] (1) The network-side device instructs the terminal to report auxiliary information, which is used by the AI ​​unit of the network-side device;

[0195] (2) The network-side device triggers the terminal to perform layer 1 reporting or layer 3 reporting or location information reporting or beam management reporting;

[0196] (3) The network-side device receives the random access request from the terminal, or in other words, the network-side device receives the connection state establishment request from the terminal, so that the terminal expects to open the first AI unit for services in the connection state;

[0197] (4) The network-side device receives the WUS from the terminal;

[0198] (5) The network-side device enables the AI ​​unit, which is applicable to dual-end (e.g., terminal and base station) or multi-end (e.g., terminal, base station and core network equipment). In this case, when the network-side device enables the AI ​​unit, the terminal also needs to enable the AI ​​unit (e.g., the first AI unit).

[0199] (6) When the first object is related to the first AI unit, the battery power of the terminal is greater than the preset battery power.

[0200] (7) When the first object is related to the first AI unit, the energy consumption of the terminal is less than the preset energy consumption. The energy consumption of the terminal can be the current energy consumption or the estimated energy consumption after the first AI unit is turned on.

[0201] (8) The shutdown timer of the first AI unit fails. Optionally, if the second condition is met, the terminal expects to receive deactivation information of the first object sent by the network-side device, wherein the second condition includes at least one of the following:

[0202] (1) When the first object is related to the first AI unit, the battery power of the terminal is less than the preset battery power;

[0203] (2) When the first object is related to the first AI unit, the power consumption of the terminal is greater than the preset power consumption;

[0204] (3) The start timer of the first AI unit fails;

[0205] (4) The measurement result reported by the terminal is lower or higher than the second threshold;

[0206] (5) The network-side device shuts down the AI ​​unit. The AI ​​unit is applicable to dual-end (e.g., terminal and base station) or multi-end (e.g., terminal, base station and core network equipment). In this case, when the network-side device shuts down the AI ​​unit, the terminal also needs to shut down the AI ​​unit (e.g., the first AI unit).

[0207] (6) The network-side device monitors that the first parameter of the first AI unit of the terminal does not meet the preset requirements. The first parameter includes at least one of the following: accuracy, training result, model update, and model management.

[0208] In this embodiment, when the first condition is met, the network-side device can send activation information of the first object to the terminal, which helps the terminal to enable the first AI unit and / or switch to the second mode; and when the second condition is met, the network-side device can send deactivation information of the first object to the terminal, which helps the terminal to disable the first AI unit and / or switch to the first mode. In this way, by sending activation or deactivation information of the first object from the network side, the terminal's enabling or disabling behavior of the first AI unit can be better regulated.

[0209] Optionally, when the first object is associated with the first AI unit, the activation or deactivation information of the first AI unit is carried by at least one of the following: RRC signaling, Downlink Control Information (DCI), or Medium Access Control Element (MAC CE). For example, the activation or deactivation information of the first AI unit may be a 1-bit indication.

[0210] Optionally, when the first object is associated with the first AI unit, the activation or deactivation information of the first AI unit includes at least one of the following:

[0211] The terminal identifier can be used to determine which terminal needs to turn the first AI unit on or off.

[0212] The first AI unit identifier can be used to determine which AI unit needs to be turned on or off.

[0213] The activation time can be used to determine when the first AI unit is turned on (or for how long), and thus when it needs to be turned off.

[0214] The start activation time can be used to determine when the first AI unit is activated.

[0215] The activation end time can be used to determine when to shut down the first AI unit.

[0216] The activation period can be used to determine the activation duration of the first AI unit.

[0217] The deactivation time can be used to determine when to shut down the first AI unit.

[0218] Priority threshold, which can be the priority threshold of AI units, can determine which AI units are turned on or off based on the priority threshold.

[0219] In this embodiment of the application, the terminal determines the first AI unit based on target information, which includes at least one of the following: terminal type (e.g., power type), terminal capability, terminal energy consumption status, terminal battery status, terminal sleep level (e.g., whether the terminal can enter deep sleep, light sleep, or micro sleep), terminal service information, terminal service status, and network scheduling information.

[0220] It should be noted that the first information may be sent to the terminal by the network-side device, or it may be a protocol-defined provision, or it may be determined by the terminal itself. The terminal can determine the first AI unit based on the aforementioned first information, thereby enabling the terminal to better identify which first AI unit needs to be turned on or off. For example, the first information may include the terminal's power consumption status. If the terminal's power consumption is high, when the terminal receives the activation information of the first AI unit, it can determine, based on its own power consumption status, that the first AI unit to be turned on may be a lower-power AI unit, thereby helping to save terminal power consumption.

[0221] Optionally, the terminal activates the first AI unit based on the first information, including:

[0222] After receiving the first information at a first time interval at the reference time, the terminal activates the first AI unit;

[0223] Alternatively, the terminal may shut down the first AI unit based on the first information, including:

[0224] After receiving the first information at a second time interval at the reference time, the terminal shuts down the first AI unit.

[0225] For example, the first information includes the activation information of the first object. After receiving the activation information of the first object at a first time interval at a reference time, the terminal activates the first AI unit, so that the terminal has a certain amount of information processing time and preparation time.

[0226] Alternatively, the first information may include the deactivation information of the first object. After receiving the deactivation information of the first object at the second time interval of the reference time, the terminal shuts down the first AI unit. This allows the terminal to have sufficient information processing time and preparation time to shut down the first AI unit, while also ensuring the continuity and accuracy of the AI ​​results of the first AI unit.

[0227] Of course, the first information can also be other information, such as the first configuration, etc., which will not be specifically listed here.

[0228] Optionally, the first time interval or the second time interval includes one of the following:

[0229] The processing time for the activation or deactivation information of the first object;

[0230] Processing time of the first information;

[0231] The processing time when the first AI unit is activated;

[0232] The processing time for shutting down the first AI unit;

[0233] The switching time between the first and second modes;

[0234] The processing time of the first AI unit for data acquisition and / or model training is the processing time before the first AI unit can output AI results.

[0235] Optionally, the reference time is the time when the first information was received, or the time when the first information was received after a time offset value.

[0236] In this embodiment of the application, the meaning of the reference time, the first time interval, and the second time interval are clearly defined, which helps the terminal to clearly determine when to turn the first AI unit on or off, so that the terminal can better control the first AI unit.

[0237] It should be noted that the activation or deactivation information of the first object can also be an implicit indication.

[0238] Optionally, the terminal activates the first AI unit, including:

[0239] If the third condition is met, the terminal activates the first AI unit based on the first information, wherein the third condition includes at least one of the following:

[0240] When the first object is associated with the first AI unit, the terminal's battery level is greater than a preset battery level;

[0241] When the first object is associated with the first AI unit, the power consumption of the terminal is less than the preset power consumption;

[0242] The shutdown timer of the first AI unit has failed;

[0243] The terminal receives uplink data, and optionally, the priority of the uplink data is greater than a preset priority threshold;

[0244] The terminal receives scheduling from the network-side device;

[0245] The terminal performs the first operation;

[0246] The measurement result of the terminal is less than the third threshold.

[0247] It should be noted that the measurement results of the terminal include at least one of the following:

[0248] Signal strength, which includes Reference Signal Received Power (RSRP), Reference Signal Received Quality (RSRQ), Received Signal Strength Indication (RSSI), etc.

[0249] Signal quality, which includes signal-to-noise ratio (SNR), signal-to-noise and interference ratio (SINR), or latency information;

[0250] Path loss;

[0251] The distance between the terminal and network-side equipment or other terminals.

[0252] In this embodiment, the activation information of the first object is an implicit indication, meaning that the terminal activates the first AI unit only when the aforementioned third condition is met. This helps the terminal achieve energy saving by activating the first AI unit only when certain conditions are met. Alternatively, it can be understood that whether the third condition is met is also part of the first information, meaning that the first information may also include whether the third condition is met.

[0253] Optionally, the terminal shuts down the first AI unit, including:

[0254] If the fourth condition is met, the terminal shuts down the first AI unit based on the first information; wherein the fourth condition includes at least one of the following:

[0255] When the first object is associated with the first AI unit, the terminal's battery level is less than a preset battery level;

[0256] When the first object is associated with the first AI unit, the power consumption of the terminal is greater than the preset power consumption;

[0257] The start timer for the first AI unit has failed.

[0258] The terminal has not received any services and / or has not received scheduling from network-side devices during the first preset time period.

[0259] The terminal will not receive any services and / or will not receive scheduling from the network-side device during a second preset time period in the future.

[0260] The measurement result of the terminal is greater than the third threshold;

[0261] The terminal no longer performs the first operation.

[0262] In this embodiment, the deactivation information of the first object can be an implicit indication, that is, when the fourth condition is met, the terminal shuts down the first AI unit, thereby enabling the terminal to shut down the first AI unit under certain conditions, which helps the terminal achieve energy saving. Alternatively, it can be understood that whether the fourth condition is met is also part of the first information, that is, the first information can also include whether the fourth condition is met.

[0263] In this embodiment of the application, the first information includes a first configuration, the first configuration includes a first DRX configuration, wherein the first DRX in the first DRX configuration is a connected DRX or an idle DRX.

[0264] Understandably, the terminal can activate the first AI unit in the first DRX active state and deactivate the first AI unit in the first DRX inactive state. Thus, the activation or deactivation of the first AI unit is limited by the terminal-side DRX, which can reduce the terminal's power consumption in the inactive state.

[0265] Optionally, the first information includes a first configuration. When the terminal performs the first action, if the terminal switches to the inactive state of the first configuration, the terminal does not shut down the first AI unit, or starts the second timer.

[0266] The terminal performing the first action includes at least one of the following:

[0267] The terminal performs the first operation, for example, the terminal performs the first operation based on the first AI unit;

[0268] The terminal performs measurements prior to reporting triggered by the network-side device;

[0269] The terminal performs at least one of the following through the first AI unit: AI result processing, AI training, AI supervision, AI model management, and AI monitoring;

[0270] The terminal performs at least one of the following functions: AI training, AI computation, AI supervision, AI model management, AI monitoring, AI result output, and AI unit information input for dual-end AI units or multi-end AI units.

[0271] In this embodiment of the application, when the terminal performs the first action described above, if the terminal switches to the inactive state of the first configuration, the terminal does not shut down the first AI unit to avoid affecting other functions performed by the terminal; or a second timer is started. The second timer can also be called the delay timer of the active state. Before the second timer expires, the terminal will not switch to the inactive state of the first configuration or shut down the first AI unit. This allows the terminal to not switch to the inactive state of the first configuration or shut down the first AI unit for a certain period of time, thereby ensuring the continuity and accuracy of the AI ​​result output of the first AI unit.

[0272] It should be noted that the duration of the second timer is predefined by the protocol, preconfigured by the network-side device, or determined by the terminal based on the working behavior or working status of the first AI unit.

[0273] Optionally, the first configuration may include a first cell DTX configuration or a first cell DRX configuration, and the terminal activating the first AI unit includes at least one of the following:

[0274] If the first information includes the first cell DTX configuration, the terminal activates the first AI unit in the active state of the first cell DTX.

[0275] If the first information includes the DRX configuration of the first cell, the terminal activates the first AI unit in the active state of the DRX of the first cell.

[0276] Optionally, the terminal shuts down the first AI unit, including at least one of the following:

[0277] If the first information includes the first cell DTX configuration, the terminal disables the first AI unit and / or enables the second AI unit in the inactive state of the first cell DTX.

[0278] If the first information includes the DRX configuration of the first cell, the terminal disables the first AI unit and / or enables the second AI unit in the inactive state of the first cell's DRX.

[0279] In this embodiment, the terminal can activate the first AI unit in the active state of the first cell DTX and / or the active state of the first cell DRX. Alternatively, the terminal can deactivate the first AI unit and / or activate the second AI unit in the inactive state of the first cell DTX and / or the inactive state of the first cell DRX, thereby improving the terminal's control flexibility over activating or deactivating the first AI unit, and aligning with the network's sending and receiving behavior to achieve terminal energy saving.

[0280] Optionally, the terminal disables the first AI unit in the inactive state of the first cell DTX, including:

[0281] The terminal disables the first AI unit for receiving signals sent by network-side devices in the inactive state of the first cell DTX; or...

[0282] The terminal disables the first AI unit in the inactive state of the first cell DRX, including:

[0283] The terminal disables the first AI unit, which is used to send signals or report to network-side devices, in the inactive state of the first cell DRX.

[0284] In this embodiment, the type of the first AI unit that the terminal needs to shut down in the inactive state of the first cell DTX or the inactive state of the first cell DRX is further defined, thereby helping to improve the terminal's precise control over the AI ​​unit.

[0285] Please refer to Figure 3, which is a flowchart of an AI unit processing method provided in an embodiment of this application. The method is executed by a network-side device. As shown in Figure 3, the method includes the following steps:

[0286] Step 301: The network-side device performs a second action, which includes at least one of the following:

[0287] Send first information to the terminal, the first information including at least one of the following: first configuration, activation information of a first object, and deactivation information of a first object;

[0288] Based on the second information, perform at least one of the following: shut down the third AI unit or switch to the third mode; or, based on the second information, perform at least one of the following: turn on the third AI unit or switch to the fourth mode.

[0289] In the third mode, the network-side device performs the second operation based on the non-AI unit; in the fourth mode, the network-side device performs the second operation based on the AI ​​unit, or based on both the non-AI unit and the AI ​​unit.

[0290] The first object includes at least one of the first AI unit and the first configuration, wherein the first configuration includes at least one of the following: the first pattern configuration, the first DRX configuration, the first DTX configuration, the first cell DTX configuration, and the first cell DRX configuration associated with the first AI unit.

[0291] For example, the network-side device performing the second action includes: the network-side device sending first information to the terminal, and the terminal turning off the first AI unit and / or switching to a first mode according to the first information; or, the terminal turning on the first AI unit according to the first information and switching between the first mode and the second mode, thereby helping the terminal to achieve flexible control over the first AI unit and flexible control over the first mode and the second mode, which helps to save terminal power consumption. It should be noted that the specific implementation process of the terminal performing the above action according to the first information can be referred to the description in the method embodiment of Figure 2 above, and will not be repeated in this embodiment.

[0292] Alternatively, the network-side device may perform the second action by: disabling the third AI unit and / or switching to the third mode based on the second information; or by enabling the third AI unit and / or switching to the fourth mode based on the second information.

[0293] For example, the network-side device can disable the third AI unit based on the second information, such as the second information including deactivation information for the third AI unit. Alternatively, the network-side device can switch to a third mode based on the second information, that is, switch to a mode that performs the second operation based on a non-AI unit. Or, the network-side device can disable the third AI unit based on the second information and switch to the third mode. This effectively reduces the increased energy consumption caused by enabling the AI ​​unit or performing related operations based on the AI ​​unit, thereby reducing the energy consumption of the network-side device.

[0294] For example, the network-side device activates the third AI unit based on the second information; or, the network-side device switches to a fourth mode based on the second information, where the second operation is performed based on the AI ​​unit or on a combination of non-AI and AI units. Alternatively, the network-side device activates the third AI unit based on the second information and switches to the fourth mode. This allows the network-side device to activate the third AI unit or switch to the fourth mode based on the AI ​​unit to better manage the AI ​​unit and make its control more flexible.

[0295] Optionally, the second operation includes at least one of the following: beam management, motion management, CSI feedback, CSI prediction, positioning, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, signal demodulation, sensing, synchronization, and measurement.

[0296] Optionally, the second information can be determined by the network-side device itself, by the terminal, or by auxiliary information from the terminal.

[0297] In this embodiment, the network-side device can disable the third AI unit and / or switch to a mode that performs the second operation based on a non-AI unit according to the second information, thereby effectively saving terminal power consumption; or, the network-side device can enable the third AI unit and / or switch to a fourth mode according to the second information. In this way, the network-side device can also control the third AI unit more flexibly according to the second information, avoiding the third AI unit from always being on, thus effectively saving network-side device power consumption. Furthermore, the network-side device can switch between the third and fourth modes according to the enabling or disabling of the third AI unit, which also avoids the network-side device being constantly in the mode that performs the second operation based on the AI ​​unit. The network-side device can flexibly switch to a mode that performs the second operation based on a non-AI unit, which also helps save network-side device power consumption.

[0298] Optionally, the second information includes at least one of the following:

[0299] The energy-saving status of the network-side equipment;

[0300] Second configuration;

[0301] Activation information of the second object;

[0302] Deactivation information for the second object;

[0303] The second object includes at least one of the third AI unit and the second configuration, and the second configuration includes at least one of the following: second pattern configuration, second cell DTX configuration, second cell DRX configuration, second DRX configuration, and second DTX configuration.

[0304] For example, the second information includes a second DRX configuration. The network-side device can disable the third AI unit when the terminal is in DRX mode according to the second DRX configuration. Since the terminal does not perform reception at this time, the network-side device can disable the third AI unit, especially the AI ​​unit related to sending to the terminal, thereby saving network-side device power consumption.

[0305] For example, the second information includes a second DTX configuration. The network-side device can disable the third AI unit when the terminal is in DTX mode according to the second DTX configuration. Since the terminal does not perform transmission at this time, the network-side device can disable the third AI unit, such as the AI ​​unit related to receiving the content transmitted by the terminal, thereby saving the network-side device's power consumption.

[0306] For example, the second information includes the DTX configuration of the second cell. The network-side device can disable the third AI unit when the second cell DTX is inactive because the network does not perform transmission at this time. Therefore, the network-side device can disable the AI ​​unit related to transmission, thereby saving network-side device power consumption; or the network-side device can enable the third AI unit when the second cell DTX is active. In this way, the network-side device can flexibly control the third AI unit according to the second cell DTX configuration.

[0307] For example, the second information includes the DRX configuration of the second cell. The network-side device can disable the third AI unit when the second cell DRX is inactive, because the network does not perform reception at this time. Therefore, the network-side device can disable the AI ​​unit related to reception, thereby saving network-side device power consumption; or the network-side device can enable the third AI unit when the second cell DRX is active. In this way, the network-side device can flexibly control the third AI unit according to the second cell DRX configuration.

[0308] For example, the second information may include activation information for the third AI unit, and the network-side device may activate the third AI unit based on the activation information. Alternatively, the second information may include deactivation information for the third AI unit, and the network-side device may deactivate the third AI unit based on the deactivation information.

[0309] For example, when the second information includes the energy-saving status of the network-side device, the network-side device performs different behaviors under different energy-saving states. For instance, the network-side device may turn the third AI unit on or off, or switch to a third mode, in a specific energy-saving state to save energy. Alternatively, the network-side device may turn on the third AI unit or switch to a fourth mode when not in a specific energy-saving state. The specific energy-saving state is predefined by the protocol, pre-configured by the network side, configured by the network side, determined by the network side, or indicated by the terminal.

[0310] Optionally, if the second information includes the power-saving status of the network-side device, the third AI unit that is turned off by the network-side device in a specific power-saving state includes at least one of the following:

[0311] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than a first preset power consumption, AI unit with computing power occupied greater than a first threshold, AI unit with priority lower than a first preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0312] Alternatively, the third AI unit activated by the network-side device in the specific energy-saving state includes at least one of the following:

[0313] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0314] This application defines the content of the second information, so that the network-side device can determine under what circumstances the third AI unit is turned on or off, or switch between the third mode and the fourth mode, based on the content of the second information. This effectively regulates the behavior of the network-side device, helps the network-side device to flexibly control the third AI unit, and also helps the network-side device save energy.

[0315] Optionally, the second pattern configuration includes at least one of the following:

[0316] The duration of the second AI active state timer;

[0317] The duration of the second AI inactive state timer;

[0318] The starting time position of the second pattern activation state within the cycle;

[0319] The starting time position of the inactive state of the second pattern within the cycle;

[0320] The effective start time position of the second pattern;

[0321] The cycle of the second pattern.

[0322] For example, the second pattern configuration includes the duration of the second AI activation state timer. Based on the duration of the second AI activation state timer, the network-side device can determine the duration for which the third AI unit can be activated. This duration can be the duration of the second AI activation state timer, thus allowing the network-side device to determine when to disable the third AI unit and / or switch to the third mode. In some embodiments, the start position of the second AI activation state within the second pattern cycle is not configured, indicating that the second AI activation state begins at the start position of the second pattern cycle or begins at the end position of the second AI inactive state. In some embodiments, the end position of the second AI activation state within the second pattern cycle is not configured, indicating that the second AI activation state ends at the end position of the second pattern cycle or ends at the start position of the second AI inactive state.

[0323] For example, the second pattern configuration includes the duration of a second AI inactive state timer within a second pattern period. Based on this second pattern configuration, the network-side device can determine the duration the third AI unit is off or in the third mode, and thus, the network-side device can determine when to activate the third AI unit and / or switch to the fourth mode. In some embodiments, the start position of the second AI inactive state within the second pattern period is not configured, indicating that the second AI inactive state begins at the start position of the second pattern period or begins at the end position of the second AI active state. In some embodiments, the end position of the second AI inactive state within the second pattern period is not configured, indicating that the second AI inactive state ends at the end position of the second pattern period or ends at the start position of the second AI active state.

[0324] It can be understood that a second pattern cycle includes a second AI active state and a second AI inactive state. The time occupied by the second AI active state plus the time occupied by the second AI inactive state constitutes a second pattern cycle.

[0325] It should be noted that different second pattern configurations can have different pattern configurations, such as different periods, different durations of the second AI activation state, and different AI units that can be enabled or disabled. This allows for enabling or disabling different AI units based on different pattern configurations, making the control of AI units on the network side more flexible.

[0326] For example, the second pattern configuration includes the start time of the second pattern's effectiveness, so the network-side device can determine when the second pattern configuration takes effect, and thus determine when to perform actions such as enabling or disabling the third AI unit based on the second pattern configuration.

[0327] Of course, the second pattern configuration can also include other contents mentioned above, which will not be listed here.

[0328] In this embodiment, the network-side device can enable or disable the third AI unit according to the specific content configured in the second pattern, and / or switch between the third mode and the fourth mode, which helps to reduce the power consumption of the network-side device and makes the network-side device more flexible in controlling the AI ​​unit.

[0329] Optionally, the third AI unit that needs to be turned off when the second configuration is inactive includes at least one of the following:

[0330] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI unit monitoring other AI units, AI unit with power consumption greater than the second preset power consumption, AI unit occupying computing power greater than the fourth threshold, AI unit with priority lower than the second preset priority, AI unit with specific identification, and AI unit used to perform the second operation.

[0331] The third AI unit that can be activated when the second configuration is inactive includes at least one of the following:

[0332] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI unit that monitors other AI units, AI unit with power consumption less than the second preset power consumption, AI unit with computing power consumption less than the fourth threshold, AI unit with priority higher than the second preset priority, AI unit with a specific identifier, and AI unit used to perform the second operation.

[0333] In this embodiment, it is clarified which third AI units the network-side device needs to turn off and / or turn on in the second configuration inactive state, so that the network-side device can more flexibly control the third AI units.

[0334] Optionally, the network-side device disables the third AI unit, including:

[0335] The network-side device disables the third AI unit in the inactive state of the second configuration.

[0336] Alternatively, the network-side device may enable a third AI unit, including:

[0337] The network-side device activates the third AI unit in the active state of the second configuration.

[0338] In this embodiment, the network-side device can also enable the third AI unit in the second configured active state. Alternatively, the network-side device can disable the third AI unit in the second configured inactive state, thereby improving the network-side device's control flexibility in enabling or disabling the third AI unit.

[0339] In this embodiment of the application, the method further includes:

[0340] After the network-side device turns the third AI unit on or off, it sends a first indication message to the terminal. The first indication message is used to indicate that the network-side device has turned the third AI unit on or off.

[0341] For example, for dual-end AI units or multi-end AI units, after the network-side device turns on or off the third AI unit, it sends a first instruction message to the terminal. The terminal can then turn on or off its own AI unit according to the first instruction message, thereby enabling both the terminal and the network-side device to control the AI ​​unit simultaneously, which helps save energy for both the terminal and the network-side device.

[0342] Optionally, the method further includes:

[0343] If the first condition is met, the network-side device sends the activation information of the first object to the terminal;

[0344] The first condition includes at least one of the following:

[0345] The network-side device instructs the terminal to report auxiliary information, which is used by the third AI unit.

[0346] The network-side device triggers the terminal to perform layer 1 reporting, layer 3 reporting, location information reporting, or beam management reporting;

[0347] The network-side device receives a random access request from the terminal;

[0348] The network-side device receives the wake-up signal WUS from the terminal;

[0349] The network-side device activates the third AI unit;

[0350] When the first object is associated with the first AI unit, the terminal's battery level is greater than a preset battery level;

[0351] When the first object is associated with the first AI unit, the power consumption of the terminal is less than the preset power consumption;

[0352] The shutdown timer of the first AI unit has failed.

[0353] Optionally, the method further includes:

[0354] If the second condition is met, the network-side device sends the deactivation information of the first object to the terminal;

[0355] The second condition includes at least one of the following:

[0356] When the first object is associated with the first AI unit, the terminal's battery level is less than a preset battery level;

[0357] When the first object is associated with the first AI unit, the power consumption of the terminal is greater than the preset power consumption;

[0358] The start timer for the first AI unit has failed.

[0359] The measurement result reported by the terminal is lower or higher than the second threshold;

[0360] The network-side device shuts down the third AI unit;

[0361] The network-side device monitors that the first parameter of the first AI unit of the terminal does not meet the preset requirements. The first parameter includes at least one of the following: accuracy, training result, model update, and model management.

[0362] In this embodiment, when the first condition is met, the network-side device can send activation information of the first object to the terminal, which helps the terminal to enable the first AI unit and / or switch to the second mode; and when the second condition is met, the network-side device can send deactivation information of the first object to the terminal, which helps the terminal to disable the first AI unit and / or switch to the first mode. In this way, by sending activation or deactivation information of the first object from the network side, the terminal's enabling or disabling behavior of the first AI unit can be better regulated.

[0363] In this application, when determining whether the AI ​​unit is turned on or off, the terminal and network sides can, in addition to relying on the terminal type, support capabilities, power consumption and / or battery status, and sleep level, also rely on at least one of the following information:

[0364] (1) Business information; such as business type, business cycle, business changes, etc.

[0365] For example, the data arrival characteristics of different types of services, the arrival interval, period, priority, and size of uplink data.

[0366] (2) Network scheduling information; including scheduling level, quantity, time and frequency domain location, etc., and may also include the time domain location and time domain range of the uplink scheduling corresponding to the terminal's resource request.

[0367] (3) Signal strength information (such as RSRP, RSRQ, RSSI, etc.);

[0368] The signal strength information here can include the received uplink / downlink signal strength information between the terminal and network nodes of the source cell / target cell / currently camped cell / currently accessed cell / Pcell / Scell.

[0369] Specifically, for example:

[0370] a. Measured values ​​of RSRP, RSRQ, RSSI, etc., corresponding to one or a group of downlink broadcast signals / synchronization signals / reference signals (such as SSB, CSI-RS, TRS, etc.). If the measured values ​​are used for UE-side inference, they are obtained by the UE; if the measured values ​​are used for network-side inference, they are obtained by the UE measurement and reporting.

[0371] b. One or a set of uplink reference signals (such as SRS), and the corresponding RSRP, RSRQ, RSSI, and other measured values; if used for network-side inference of measured values, they are obtained from network measurements.

[0372] (4) Signal quality information (such as SNR, SINR, latency);

[0373] The signal quality information here may include the received SINR, SNR, and latency of uplink / downlink signals between the terminal and network nodes of the source cell / target cell / current camping cell / current access cell / Pcell / Scell.

[0374] (5) Path loss related information; including path loss between the terminal and network nodes of the source cell / target cell / currently camped cell / currently accessed cell / Pcell / Scell; it may also include path loss between network nodes of the source cell and target cell or Pcell and Scell.

[0375] (6) Distance information; including the distance between the terminal and the network nodes of the source cell / target cell / currently camped cell / currently accessed cell / Pcell / Scell; it may also include the distance between the source cell and the target cell or the network nodes of the Pcell and Scell.

[0376] (7) Frequency band / frequency point / carrier frequency / frequency layer / BWP information;

[0377] Specifically, this could include the source cell / target cell / currently camped cell / currently accessed cell / PCell / SCell's actual operating frequency band / frequency zone / frequency point / carrier frequency / frequency layer / BWP.

[0378] (8) One or a set of reference signal indices or beam indices or beam directions;

[0379] Specifically, for example, it may include the reference signal index, beam index, or beam direction actually transmitted between the terminal and the network nodes of the source cell / target cell / current camping cell / current access cell / Pcell / Scell.

[0380] (9) Propagation delay; the propagation delay can be the uplink / downlink or sidelink or signal propagation delay between network nodes;

[0381] Specifically, this could include, for example, the propagation delay of uplink / downlink signals between the terminal and the source / target / currently camped / currently accessed / Pcell / Scell.

[0382] (10) RTT (Round-Trip Time) round-trip delay;

[0383] Specifically, this could include the RTT latency between the terminal and network nodes of the source cell / target cell / currently camped cell / currently accessed cell / Pcell / Scell.

[0384] (11) Load conditions corresponding to SSB / beam / TCI / TRP, such as traffic volume and number of users;

[0385] (12) Interference conditions corresponding to SSB / beam / TCI / TRP, such as interference measurement values;

[0386] (13) Load conditions of one or more cells, carriers, and frequency points, such as traffic volume and number of users;

[0387] (14) Interference conditions of one or more cells, carriers, or frequency points, such as interference measurements;

[0388] (15) Transmission power related information of network devices; for example, transmission power related information of base stations or TRPs;

[0389] Specifically, this can include the actual transmission power of the base station in the source cell / target cell / currently camped cell / currently accessed cell / Pcell / Scell.

[0390] (16) Geographical location information of the base station or TRP;

[0391] Specifically, this can include the geographical location of the base station in the source cell, target cell, currently camped cell, currently accessed cell, Pcell, and Scell.

[0392] (17) The location information or distribution of the terminal; the location information may be specific geographical coordinates (such as GPS coordinates), or the approximate location range information of the terminal (such as the range information of which street it is located on), or the location information of the terminal relative to the station cell or access cell or a certain TRP or a group of TRPs (such as due east of the station cell); the distribution information may include the number of terminals in different areas (such as the station cell or access cell or a certain TRP or a group of TRPs);

[0393] (18) The direction of movement and / or speed of movement of the terminal, network node or satellite; the direction of movement may be an absolute direction, such as 40 degrees south of east; or it may be a relative direction, such as the direction relative to a certain base station or reference node.

[0394] (19) The terminal's energy consumption and / or power status;

[0395] (20) Terminal-supported operator information and network type information;

[0396] (21) Information on the orientation of the terminal, network node, or satellite panel;

[0397] (22) Network energy consumption and / or power consumption;

[0398] (23) Terminal type, such as power class;

[0399] (24) Sensing, synchronization, measurement, reporting, and initial access information;

[0400] Specifically, the sensing, synchronization, measurement, reporting, and initial access information include communication environment information, scenario information, channel state information (LOS / NLOS, obstruction, etc.), and the number of terminals.

[0401] For example, sensing, synchronization, measurement, reporting, and initial access to a specific beam to check for obstacles and / or the number of obstacles;

[0402] For example, sensing, synchronization, measurement, reporting, and the number of terminals or devices initially connected under the coverage of a specific beam;

[0403] (25) Network scene information (such as inH, Uma, RMa, etc., or homogeneous / heterogeneous networks (whether there is overlapping coverage));

[0404] (26) Environmental information (such as weather information);

[0405] (27) Ephemeris information;

[0406] (28) The reference position or movement trajectory of the cell in the NTN scenario; for example, in the LEO (Low-Earth Orbit) scenario, the cell on Earth moves as the satellite moves. In the GEO scenario, the cell on Earth is fixed as the satellite moves, and a reference position can be considered to exist.

[0407] (28) Multipath information of the channel; such as the first path or strongest path information of the terminal and different base stations / TRPs;

[0408] (29) Time information; the time can be a specific time, accurate to the second, such as 13:25:38. It can also be a time range, such as 13:00 to 14:00, morning / afternoon, day / night;

[0409] Alternatively, the time information can be timing information obtained through other RATs (Radio Access Technologies). The RAT can be, for example, Bluetooth, Wi-Fi, or 3G, 4G, or 5G.

[0410] (30) One or more of the following information currently used or within a certain time window, including at least one of the following: timing advance offset, timing pre-compensation, common timing advance offset, proprietary timing advance offset, frequency compensation, frequency offset, common frequency offset, proprietary frequency offset.

[0411] To better understand, the technical solutions provided in this application will be explained and illustrated below through specific embodiments.

[0412] In one implementation, an AI unit processing method is provided, comprising the following steps:

[0413] Step 1: The terminal receives the first DRX configuration, which can be sent to the terminal by the network-side device.

[0414] Step 2: The terminal determines whether to enable or disable the first AI unit based on the first DRX configuration.

[0415] The first DRX included in the first DRX configuration is either a connected DRX or an idle DRX.

[0416] For example, the terminal determines whether to enable or disable the first AI unit based on the first DRX configuration, including the following:

[0417] a) The terminal activates the first AI unit while in the first DRX active state;

[0418] b) The terminal disables the first AI unit and / or enables the second AI unit while in the first DRX inactive state;

[0419] c) When the terminal's first AI unit is executing the first action, if the terminal switches to the first DRX inactive state, the terminal does not shut down the first AI unit to avoid affecting other functions executed by the terminal; or the terminal starts a timer, thereby allowing the terminal to not switch to the DRX inactive state or shut down the first AI unit for a certain period of time through the timer's timing.

[0420] The first behavior includes at least one of the following:

[0421] 1. The terminal performs at least one of the following functions: beam management, mobility management, CSI feedback, CSI prediction, CSI compression, positioning enhancement, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, signal demodulation, sensing, synchronization, measurement, reporting, and initial access. For example, the terminal performs at least one of the following functions based on the first AI unit: beam management, mobility management, CSI feedback, CSI prediction, CSI compression, positioning enhancement, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, signal demodulation, sensing, synchronization, measurement, reporting, and initial access.

[0422] 2. Measurements taken by the terminal before reporting triggered by the network side;

[0423] 3. The terminal performs at least one of the following functions: AI result processing, AI training, AI supervision, AI model management, or monitoring other AI units.

[0424] 4. The terminal performs at least one of the following functions when executing a dual-end or multi-end model: training, computation, supervision, model management, monitoring, result output, or information input.

[0425] It should be noted that the duration of the timer is predefined by the protocol, preconfigured by the network side, or determined by the terminal based on the working behavior / state of the first AI unit.

[0426] e) The terminal activates the first AI unit and activates the first timer. Before the first timer expires, the terminal still activates the first AI unit in the first DRX inactive state.

[0427] f) The terminal activates the first AI unit and starts the first timer. Before the first timer expires, the terminal does not switch to the first DRX inactive state.

[0428] In this embodiment, the terminal receives the first cell DTX / DRX configuration and determines whether to enable or disable the first AI unit based on the first cell DTX / DRX configuration. In this way, the terminal's AI unit is limited by the base station's DTX / DRX, thereby maximizing the reduction of the terminal's power consumption in the inactive state.

[0429] Optionally, the terminal activates the first AI unit during the first cell DTX / DRX activity period, thereby adapting to the behavior of the base station. Activating the AI ​​unit when data needs to be sent and received with the base station helps save terminal power consumption.

[0430] Optionally, the terminal can turn off the first AI unit during the inactive period of the first cell DTX / DRX and / or turn on the second AI unit to keep synchronized with the base station. When it is necessary to send and receive data with the base station, the corresponding AI unit can be turned on, which helps to save terminal power consumption.

[0431] It should be noted that the first AI unit, used for receiving signals from network-side devices, is turned off during the inactive period of the first cell DTX, or the first AI unit, used for sending signals or reporting to the network side, is turned off during the inactive period of the first cell DRX. This allows for more targeted shutdown of the first AI unit, enabling the terminal to control it more flexibly and also contributing to energy saving.

[0432] In this embodiment, the on / off mechanism of the AI ​​unit on the terminal side under the existing DTX / DRX mechanism is defined, and it is clarified whether the AI ​​unit follows the existing energy-saving mechanism on the network side or the terminal side, thereby realizing energy saving on the network side or the terminal side, and also avoiding the effect of the existing energy-saving mechanism being affected by the operation of the AI ​​unit.

[0433] In one implementation, an AI unit processing method is provided, comprising the following steps:

[0434] Step 1: The terminal receives the first pattern configuration associated with the first AI unit, and determines whether to turn the first AI unit on or off based on the first pattern configuration or its indication information.

[0435] The first pattern configuration includes at least one of the following:

[0436] (1) AI on-duration timer or the duration of AI on-duration.

[0437] (2) AI inactivity timer or the duration of AI inactivity;

[0438] (3) The starting time position of the first pattern activation state within the cycle;

[0439] (4) The period of the first pattern (e.g., including long period / short period);

[0440] (4) AI units that the terminal is allowed to be enabled in the inactive state of the first pattern. For example, different pattern configurations correspond to different AI enable or disable units.

[0441] (5) The pattern configuration parameters applied when the terminal is in the connected state;

[0442] (6) The pattern configuration parameters applied when the terminal is in an inactive state;

[0443] (7) The pattern configuration parameters applied when the terminal is in an idle state.

[0444] The terminal determines whether to enable or disable the first AI unit based on the activation / deactivation information, including:

[0445] a) The terminal activates the first AI unit according to the activation information configured in the first pattern;

[0446] b) The terminal activates the information according to the first pattern configuration and shuts down the first AI unit.

[0447] Optionally, the activation or deactivation information configured in the first pattern is a display indication sent by the network side.

[0448] For example, when a first condition is met, the activation information configured in the first pattern is sent by the network side (e.g., the terminal expects the network side to send the activation information when the first condition is met), and the first condition includes at least one of the following:

[0449] 1. The network side instructs the terminal to report auxiliary information; optionally, the auxiliary information is used by the AI ​​unit on the network side;

[0450] 2. The network side triggers the terminal to perform L1 / L3 / location information / beam management reporting.

[0451] 3. The network side receives a random access request from the terminal (or in other words, the connection state is established, and the AI ​​function is enabled for services in the connection state);

[0452] 4. The network side receives the WUS from the terminal;

[0453] 5. Enable the AI ​​unit on the network side (the AI ​​unit is applicable to dual-end or multi-end AI units. When the network side is enabled, the terminal also needs to enable the AI ​​unit).

[0454] 6. The terminal's battery level exceeds the preset battery threshold;

[0455] 7. The terminal's energy consumption is lower than the preset energy consumption threshold;

[0456] 8. The first AI unit's shutdown timer failed;

[0457] For example, when the second condition is met, the deactivation information configured in the first pattern is sent by the network side, and the second condition includes at least one of the following:

[0458] 1. The terminal's battery level is lower than the preset battery threshold;

[0459] 2. The terminal's energy consumption exceeds the preset energy consumption threshold;

[0460] 3. The timer for the first AI unit failed to start;

[0461] 4. The measurement result reported by the terminal is below the threshold;

[0462] 5. The AI ​​unit is disabled on the network side;

[0463] 6. The accuracy, training results, model updates, and model management of the first AI unit monitored by the network side on the terminal side do not meet the requirements.

[0464] In this embodiment, the mechanism for the terminal to turn the first AI unit on or off according to the first pattern configuration is clarified, which improves the terminal's flexibility in controlling the first AI unit and also helps the terminal to save energy by turning off the first AI unit.

[0465] The AI ​​unit processing method provided in this application can be executed by an AI unit processing device. This application uses an AI unit processing device executing the AI ​​unit processing method as an example to illustrate the AI ​​unit processing device provided in this application.

[0466] This application provides an AI unit processing device. As an example, the AI ​​unit processing 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 type of terminal 11 listed above, and the network-side device may include, but is not limited to, the type of network-side device 12 listed above. This application does not impose specific limitations.

[0467] The AI ​​unit processing 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, etc., such as central processing units (CPUs), microprocessors, digital signal processors (DSPs), artificial intelligence (AI) processors, graphics processing units (GPUs), application-specific integrated circuits (ASICs), network processors (NPs), field-programmable gate arrays (FPGAs), 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: transceivers, pins, circuits, buses, radio frequency units, etc.

[0468] Specifically, referring to Figure 4, when the AI ​​unit processing device is a terminal or a component within a terminal, the AI ​​unit processing device 400 includes:

[0469] The first processing module 401 is configured to perform at least one of the following actions based on the first information: shutting down the first AI unit, switching to a first mode; or...

[0470] Based on the first information, perform at least one of the following: activate the first AI unit, switch to the second mode;

[0471] In the first mode, the terminal performs the first operation based on a non-AI unit; in the second mode, the terminal performs the first operation based on an AI unit, or performs the first operation based on both a non-AI unit and an AI unit.

[0472] Optionally, the first information includes at least one of the following:

[0473] First configuration;

[0474] Terminal connection status;

[0475] Activation information of the first object;

[0476] Deactivation information for the first object;

[0477] The first object includes at least one of the first AI unit and the first configuration; the first configuration includes at least one of the following: the first format pattern configuration, the first DRX configuration, the first DTX configuration, the first cell DTX configuration, and the first cell DRX configuration associated with the first AI unit.

[0478] Optionally, the first pattern configuration includes at least one of the following:

[0479] The duration of the first AI active state timer;

[0480] The duration of the first AI inactive state timer;

[0481] The starting time position of the first pattern activation state within the cycle;

[0482] The end time position of the first pattern activation state within the cycle;

[0483] The starting time position of the inactive state of the first pattern within the cycle;

[0484] The end time position of the inactive state of the first pattern within the cycle;

[0485] The effective start time and position of the first pattern;

[0486] The cycle of the first pattern;

[0487] The AI ​​unit or AI unit type that the terminal is allowed to activate when the first pattern is inactive;

[0488] The pattern configuration parameters applied when the terminal is in a connected state;

[0489] The pattern configuration parameters applied when the terminal is in an inactive state;

[0490] The pattern configuration parameters applied when the terminal is in an idle state.

[0491] Optionally, if the first information includes the first configuration or the connection state of the terminal, the first AI unit that is turned off by the terminal in the first state includes at least one of the following:

[0492] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than a first preset power consumption, AI unit with computing power occupied greater than a first threshold, AI unit with priority lower than a first preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0493] Alternatively, the first AI unit activated by the terminal in the first state includes at least one of the following:

[0494] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation.

[0495] The first state includes at least one of the following: a first sub-state and a second sub-state. The first sub-state includes one of the following: the first configuration is inactive, the first configuration is active. The second sub-state includes one of the following: the terminal is in idle, the terminal is inactive, the terminal is connected, or the terminal is in a first power-saving state.

[0496] Optionally, the first energy-saving state is associated with at least one of the terminal energy consumption status and the terminal power status.

[0497] Optionally, the first information includes the first configuration, and the first processing module 401 is further configured to:

[0498] Upon receiving the activation information of the first configuration, the first AI unit is activated in the activated state of the first configuration;

[0499] Alternatively, upon receiving activation information for the first configuration, the first AI unit may be turned off and / or the second AI unit may be turned on while the first configuration is inactive.

[0500] Optionally, the first processing module 401 is further configured to:

[0501] When the terminal activates the first AI unit, a first timer is activated; wherein, before the first timer expires, the terminal does not switch the first configuration to an inactive state and / or does not deactivate the first AI unit.

[0502] Optionally, the activation information of the first configuration is associated with N first cycles or with a single time length, where N is a positive integer.

[0503] Optionally, the first processing module 401 is further configured to:

[0504] Upon receiving the activation information from the first object, the first AI unit is activated;

[0505] Upon receiving the deactivation information for the first object, the first AI unit is shut down.

[0506] Optionally, when the terminal receives the deactivation information of the first object, the first processing module 401 is further configured to:

[0507] The first AI unit is not turned on when the first object is in an active state, or the first AI unit is not turned off when the first object is in an inactive state.

[0508] Optionally, the device further includes:

[0509] The determination module is used to determine the first AI unit based on target information, wherein the target information includes at least one of the following: terminal type, terminal capability, terminal energy consumption status, terminal battery status, terminal sleep level, terminal service information, terminal service status, and network scheduling information.

[0510] Optionally, when the first object is associated with the first AI unit, the activation or deactivation information of the first AI unit is carried by at least one of the following: Radio Resource Control (RRC) signaling, Downlink Control Information (DCI), and Media Access Control (MAC) Control Element (CE).

[0511] Optionally, when the first object is related to the first AI unit, the activation information or deactivation information of the first AI unit includes at least one of the following: terminal identifier, first AI unit identifier, activation time, start activation time, end activation time, activation period, deactivation time, and priority threshold.

[0512] Optionally, the first processing module 401 is further configured to:

[0513] After the terminal receives the first information at a first time interval at the reference time, the first AI unit is activated;

[0514] or,

[0515] After the terminal receives the first information at a second time interval at the reference time, the first AI unit is turned off.

[0516] Optionally, the first time interval or the second time interval includes one of the following:

[0517] The processing time for the activation or deactivation information of the first object;

[0518] Processing time of the first information;

[0519] The processing time when the first AI unit is activated;

[0520] The processing time for shutting down the first AI unit;

[0521] The switching time between the first and second modes;

[0522] The processing time for the first AI unit to collect data and / or train models.

[0523] Optionally, the reference time is the time when the first information was received, or the time when the first information was received after a time offset value.

[0524] Optionally, the first processing module 401 is further configured to:

[0525] If the third condition is met, the first AI unit is activated based on the first information; wherein the third condition includes at least one of the following:

[0526] When the first object is associated with the first AI unit, the terminal's battery level is greater than a preset battery level;

[0527] When the first object is associated with the first AI unit, the power consumption of the terminal is less than the preset power consumption;

[0528] The shutdown timer of the first AI unit has failed;

[0529] The terminal has received uplink data;

[0530] The terminal receives scheduling from the network-side device;

[0531] The measurement result of the terminal is less than the third threshold;

[0532] The terminal performs the first operation.

[0533] Optionally, the first processing module 401 is further configured to:

[0534] If the fourth condition is met, the first AI unit is shut down based on the first information; wherein the fourth condition includes at least one of the following:

[0535] When the first object is associated with the first AI unit, the terminal's battery level is less than a preset battery level;

[0536] When the first object is associated with the first AI unit, the power consumption of the terminal is greater than the preset power consumption;

[0537] The start timer for the first AI unit has failed.

[0538] The terminal has not received any services and / or has not received scheduling from network-side devices during the first preset time period.

[0539] The terminal will not receive any services and / or will not receive scheduling from the network-side device during a second preset time period in the future.

[0540] The measurement result of the terminal is greater than the third threshold;

[0541] The terminal no longer performs the first operation.

[0542] Optionally, the first information includes the first DRX configuration, wherein the first DRX in the first DRX configuration is a connected DRX or an idle DRX.

[0543] Optionally, the first information includes the first configuration. When the terminal performs the first action, if the terminal switches to the inactive state of the first configuration, the first processing module 401 does not shut down the first AI unit, or starts the second timer.

[0544] The terminal performing the first action includes at least one of the following:

[0545] The terminal performs the first operation;

[0546] The terminal performs measurements prior to reporting triggered by the network-side device;

[0547] The terminal performs at least one of the following through the first AI unit: AI result processing, AI training, AI supervision, AI model management, and AI monitoring;

[0548] The terminal performs at least one of the following functions: AI training, AI computation, AI supervision, AI model management, AI monitoring, AI result output, and AI unit information input for dual-end AI units or multi-end AI units.

[0549] Optionally, the first information includes the DTX configuration of the first cell, and the first processing module 401 is further configured to:

[0550] According to the DTX configuration of the first cell, the first AI unit used to receive signals sent by network-side devices is turned off in the inactive state of the first cell DTX.

[0551] Optionally, the first information includes the DRX configuration of the first cell, and the first processing module 401 is further configured to:

[0552] According to the DRX configuration of the first cell, the first AI unit used to send signals or report to network-side devices is turned off in the inactive state of the first cell DRX.

[0553] Optionally, the device further includes:

[0554] The receiving module is used to receive the first information sent by the network-side device.

[0555] Optionally, the first operation includes at least one of the following: beam management, mobility management, CSI feedback, CSI prediction, CSI compression, positioning, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, signal demodulation, sensing, synchronization, measurement, reporting, initial access, and random access.

[0556] The AI ​​unit processing device 400 provided in this application embodiment can implement the various processes implemented in the method embodiment described in FIG2 and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0557] Referring to Figure 5, when the AI ​​unit processing device is a network-side device or a component within a network-side device, the AI ​​unit processing device 500 includes:

[0558] The second processing module 501 is configured to execute a second action, the second action including at least one of the following:

[0559] Send first information to the terminal. The first information includes at least one of the following: first configuration, activation information of a first object, and deactivation information of the first object. The first object includes at least one of a first AI unit and the first configuration. The first configuration includes at least one of the following: first pattern configuration associated with the first AI unit, first DRX configuration, first DTX configuration, first cell DTX configuration, and first cell DRX configuration.

[0560] Based on the second information, perform at least one of the following: disable the third AI unit or switch to the third mode; or, based on the second information, perform at least one of the following: enable the third AI unit or switch to the fourth mode;

[0561] In the third mode, the network-side device performs the second operation based on the non-AI unit, and in the fourth mode, the network-side device performs the second operation based on the AI ​​unit, or based on both the non-AI unit and the AI ​​unit.

[0562] Optionally, the second information includes at least one of the following:

[0563] The energy-saving status of the network-side equipment;

[0564] Second configuration;

[0565] Activation information of the second object;

[0566] Deactivation information for the second object;

[0567] The second object includes at least one of the third AI unit and the second configuration, and the second configuration includes at least one of the following: second pattern configuration, second cell DTX configuration, second cell DRX configuration, second DRX configuration, and second DTX configuration.

[0568] Optionally, the second pattern configuration includes at least one of the following:

[0569] The duration of the second AI active state timer;

[0570] The duration of the second AI inactive state timer;

[0571] The starting time position of the second pattern activation state within the cycle;

[0572] The starting time position of the inactive state of the second pattern within the cycle;

[0573] The effective start time position of the second pattern;

[0574] The cycle of the second pattern;

[0575] The AI ​​units or AI unit types that are allowed to be enabled when the network-side device is in an inactive or idle state.

[0576] Optionally, the third AI unit that needs to be turned off when the second configuration is inactive includes at least one of the following:

[0577] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than the second preset power consumption, AI unit with computing power occupied greater than the fourth threshold, AI unit with priority lower than the second preset priority, AI unit with specific identification, and AI unit used to perform the second operation.

[0578] The third AI unit that can be activated when the second configuration is inactive includes at least one of the following:

[0579] AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than the second preset power consumption, AI unit with computing power occupied less than the fourth threshold, AI unit with priority higher than the second preset priority, AI unit with specific identification, and AI unit used to perform the second operation.

[0580] Optionally, the second processing module 501 is further configured to:

[0581] The third AI unit is disabled in the inactive state of the second configuration; or...

[0582] The third AI unit is activated in the active state of the second configuration.

[0583] Optionally, the device further includes:

[0584] The sending module is configured to send a first indication message to the terminal after the network-side device turns the third AI unit on or off. The first indication message is used to indicate that the network-side device has turned the third AI unit on or off.

[0585] Optionally, the sending module is further configured to:

[0586] If the first condition is met, the activation information of the first object is sent to the terminal;

[0587] The first condition includes at least one of the following:

[0588] The network-side device instructs the terminal to report auxiliary information, which is used by the third AI unit.

[0589] The network-side device triggers the terminal to perform layer 1 reporting, layer 3 reporting, location information reporting, or beam management reporting;

[0590] The network-side device receives a random access request from the terminal;

[0591] The network-side device receives the wake-up signal WUS from the terminal;

[0592] The network-side device activates the third AI unit;

[0593] When the first object is associated with the first AI unit, the terminal's battery level is greater than a preset battery level;

[0594] When the first object is associated with the first AI unit, the power consumption of the terminal is less than the preset power consumption;

[0595] The shutdown timer of the first AI unit has failed.

[0596] Optionally, the sending module is further configured to:

[0597] If the second condition is met, the deactivation information of the first object is sent to the terminal;

[0598] The second condition includes at least one of the following:

[0599] When the first object is associated with the first AI unit, the terminal's battery level is less than a preset battery level;

[0600] When the first object is associated with the first AI unit, the power consumption of the terminal is greater than the preset power consumption;

[0601] The start timer for the first AI unit has failed.

[0602] The measurement result reported by the terminal is lower or higher than the second threshold;

[0603] The network-side device shuts down the third AI unit;

[0604] The network-side device monitors that the first parameter of the first AI unit of the terminal does not meet the preset requirements. The first parameter includes at least one of the following: accuracy, training result, model update, and model management.

[0605] Optionally, the second operation includes at least one of the following: beam management, motion management, CSI feedback, CSI prediction, positioning, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, signal demodulation, sensing, synchronization, and measurement.

[0606] The AI ​​unit processing device 500 provided in this application embodiment can implement the various processes implemented in the method embodiment shown in FIG3 and achieve the same technical effect. To avoid repetition, it will not be described again here.

[0607] As shown in Figure 6, this application embodiment also provides a communication device 600, including a processor 601 and a memory 602. The memory 602 stores programs or instructions that can run on the processor 601. For example, when the communication device 600 is a terminal, the program or instructions executed by the processor 601 implement the various steps of the above-described AI unit processing method embodiment and achieve the same technical effect. When the communication device 600 is a network-side device, the program or instructions executed by the processor 601 implement the various steps of the above-described AI unit processing method embodiment and achieve the same technical effect. To avoid repetition, this will not be described again here.

[0608] This application also provides a terminal, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps in the method embodiment shown in FIG2. This terminal embodiment corresponds to the above-described terminal-side method embodiment, and all implementation processes and methods of the above-described method embodiments can be applied to this terminal embodiment and can achieve the same technical effect. The terminal may be the AI ​​unit processing device shown in FIG4. Specifically, FIG7 is a schematic diagram of the hardware structure of a terminal implementing an embodiment of this application.

[0609] 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.

[0610] Those skilled in the art will understand that the terminal 700 may also include a power supply (such as a battery) for powering various components. The power supply can be logically connected to the processor 710 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. The terminal structure shown in Figure 7 does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0611] 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.

[0612] 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.

[0613] 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.

[0614] 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.

[0615] The processor 710 is configured to perform at least one of the following actions based on the first information: shutting down the first AI unit and switching to a first mode; or, based on the first information, performing at least one of the following actions: turning on the first AI unit and switching to a second mode;

[0616] In the first mode, the terminal performs the first operation based on a non-AI unit; in the second mode, the terminal performs the first operation based on an AI unit, or performs the first operation based on both a non-AI unit and an AI unit.

[0617] In this embodiment, the terminal can disable the first AI unit and / or switch to a mode that performs the first operation based on a non-AI unit according to the first information, thereby effectively saving terminal power; or, the terminal can enable the first AI unit and / or switch to a second mode according to the first information. In this way, the terminal can control the first AI unit more flexibly based on the first information, avoiding the first AI unit from always being on, thus effectively saving terminal power. Furthermore, the terminal can switch between the first mode and the second mode based on the enabling or disabling of the first AI unit, preventing the terminal from always being in the mode that performs the first operation based on the AI ​​unit. The terminal can flexibly switch to a mode that performs the first operation based on a non-AI unit, which also helps save terminal power consumption.

[0618] It is understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description in the embodiment of the AI ​​unit processing method and achieve the same or corresponding technical effects. To avoid repetition, it will not be described again here.

[0619] This application also provides a network-side device, including a processor and a communication interface. The communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the steps of the method embodiment shown in FIG3. This network-side device embodiment corresponds to the above-described network-side device method embodiment. All implementation processes and methods of the above-described method embodiments can be applied to this network-side device embodiment and can achieve the same technical effect.

[0620] Specifically, this application embodiment also provides a network-side device, which may be the AI ​​unit processing device shown in FIG5. As shown in FIG8, the network-side device 800 includes: an antenna 81, a radio frequency device 82, a baseband device 83, a processor 84, and a memory 85. The antenna 81 is connected to the radio frequency device 82. In the uplink direction, the radio frequency device 82 receives information through the antenna 81 and sends the received information to the baseband device 83 for processing. In the downlink direction, the baseband device 83 processes the information to be transmitted and sends it to the radio frequency device 82. The radio frequency device 82 processes the received information and transmits it through the antenna 81.

[0621] The method executed by the network-side device in the above embodiments can be implemented in the baseband device 83, which includes a baseband processor.

[0622] The baseband device 83 may include at least one baseband board, on which multiple chips are disposed, as shown in FIG8. One of the chips is, for example, a baseband processor, which is connected to the memory 85 via a bus interface to call the program or instructions in the memory 85 to execute the network-side device operation shown in the above method embodiment.

[0623] The network-side device may also include a network interface 86, such as a Common Public Radio Interface (CPRI).

[0624] Wherein, processor 84 is used to execute the second action, the second action including at least one of the following:

[0625] Send first information to the terminal. The first information includes at least one of the following: first configuration, activation information of a first object, and deactivation information of the first object. The first object includes at least one of a first AI unit and the first configuration. The first configuration includes at least one of the following: first pattern configuration associated with the first AI unit, first DRX configuration, first DTX configuration, first cell DTX configuration, and first cell DRX configuration.

[0626] Based on the second information, perform at least one of the following: disable the third AI unit or switch to the third mode; or, based on the second information, perform at least one of the following: enable the third AI unit or switch to the fourth mode;

[0627] In the third mode, the network-side device performs the second operation based on the non-AI unit, and in the fourth mode, the network-side device performs the second operation based on the AI ​​unit, or based on both the non-AI unit and the AI ​​unit.

[0628] In addition, the network-side device 800 of this application embodiment also includes: a program or instructions stored in a memory 85 and executable on a processor 84. The processor 84 calls the program or instructions in the memory 85 to execute the methods executed by each module shown in FIG6 and achieve the same technical effect. To avoid repetition, it will not be described in detail here.

[0629] 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 AI ​​unit processing method embodiments described in FIG2 or FIG3 above, and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0630] The processor mentioned above is either the processor in the terminal described in the above embodiments or the processor in the network-side device. 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.

[0631] 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 AI ​​unit processing method embodiment described in FIG2 or FIG3 above, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0632] 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.

[0633] 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 AI ​​unit processing method embodiment described in FIG2 or FIG3 above, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0634] This application also provides a communication system, including: a terminal and a network-side device, wherein the terminal can be used to execute the steps of the AI ​​unit processing method described above, and the network-side device can be used to execute the steps of the AI ​​unit processing method described above.

[0635] 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.

[0636] 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 the computer software product includes several instructions to cause the terminal or network-side device to execute the methods described in the various embodiments of this application.

[0637] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other implementations under the guidance of this application without departing from the spirit and scope of the claims. All of these implementations are within the protection scope of this application.

Claims

1. A method for processing artificial intelligence (AI) units, comprising: Based on the first information, the terminal performs at least one of the following: shuts down the first AI unit or switches to the first mode; or, Based on the first information, the terminal performs at least one of the following: activates the first AI unit, or switches to the second mode; In the first mode, the terminal performs the first operation based on a non-AI unit; in the second mode, the terminal performs the first operation based on an AI unit, or performs the first operation based on both a non-AI unit and an AI unit.

2. The method according to claim 1, wherein, The first information includes at least one of the following: First configuration; Terminal connection status; The energy-saving status of the terminal; Activation information of the first object; Deactivation information for the first object; The first object includes at least one of the first AI unit and the first configuration; the first configuration includes at least one of the following: the first format pattern configuration, the first DRX configuration, the first DTX configuration, the first cell DTX configuration, and the first cell DRX configuration associated with the first AI unit.

3. The method according to claim 2, wherein, The first pattern configuration includes at least one of the following: The duration of the first AI active state timer; The duration of the first AI inactive state timer; The starting time position of the first pattern activation state within the cycle; The end time position of the first pattern activation state within the cycle; The starting time position of the inactive state of the first pattern within the cycle; The end time position of the inactive state of the first pattern within the cycle; The effective start time and position of the first pattern; The cycle of the first pattern; The AI ​​unit or AI unit type that the terminal is allowed to activate when the first pattern is inactive; The pattern configuration parameters applied when the terminal is in a connected state; The pattern configuration parameters applied when the terminal is in an inactive state; The pattern configuration parameters applied when the terminal is in an idle state.

4. The method according to claim 2, wherein, When the first information includes the first configuration or the connection status of the terminal, the first AI unit that is turned off by the terminal in the first state includes at least one of the following: AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than a first preset power consumption, AI unit with computing power occupied greater than a first threshold, AI unit with priority lower than a first preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation. Alternatively, the first AI unit activated by the terminal in the first state includes at least one of the following: AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than preset power consumption, AI unit with computing power occupied less than a first threshold, AI unit with priority higher than preset priority, discontinuous base station AI unit, discontinuous terminal AI unit, discontinuous AI unit, AI unit with specific identification, and AI unit for performing the first operation. The first state includes at least one of the following: a first sub-state and a second sub-state. The first sub-state includes one of the following: the first configuration is inactive, the first configuration is active. The second sub-state includes one of the following: the terminal is in idle, the terminal is inactive, the terminal is connected, or the terminal is in a first power-saving state.

5. The method according to claim 4, wherein, The first energy-saving state is associated with at least one of the terminal energy consumption status and the terminal power status.

6. The method according to any one of claims 2-5, wherein, The first information includes the first configuration, and the terminal activates the first AI unit according to the first information, including: When the terminal receives the activation information of the first configuration, the terminal activates the first AI unit in the activated state of the first configuration; Alternatively, the method may further include: When the terminal receives the activation information of the first configuration, the terminal shuts down the first AI unit and / or turns on the second AI unit in the inactive state of the first configuration.

7. The method according to any one of claims 2-6, wherein, The method further includes: When the terminal activates the first AI unit, a first timer is activated; wherein, before the first timer expires, the terminal does not switch the first configuration to an inactive state and / or does not deactivate the first AI unit.

8. The method according to any one of claims 2-7, wherein, The activation information of the first configuration is associated with N first cycles, or with a single time length, where N is a positive integer.

9. The method according to any one of claims 2-8, wherein, The terminal activates the first AI unit based on the first information, including: When the terminal receives the activation information of the first object, the terminal activates the first AI unit; The terminal shuts down the first AI unit based on the first information, including: Upon receiving deactivation information for the first object, the terminal shuts down the first AI unit.

10. The method according to any one of claims 2-9, wherein, The method further includes: When the terminal receives the deactivation information of the first object, the terminal does not turn on the first AI unit when the first object is in an active state, or does not turn off the first AI unit when the first object is in an inactive state.

11. The method according to any one of claims 2-10, wherein, The method further includes: The terminal determines the first AI unit based on the target information, which includes at least one of the following: terminal type, terminal capabilities, terminal energy consumption status, terminal battery status, terminal sleep level, terminal service information, terminal service status, and network scheduling information.

12. The method according to any one of claims 2-11, wherein, When the first object is associated with the first AI unit, the activation or deactivation information of the first AI unit is carried by at least one of the following: Radio Resource Control (RRC) signaling, Downlink Control Information (DCI), and Media Access Control (MAC) Control Element (CE).

13. The method according to any one of claims 2-12, wherein, When the first object is related to the first AI unit, the activation information or deactivation information of the first AI unit includes at least one of the following: terminal identifier, first AI unit identifier, activation time, start activation time, end activation time, activation cycle, deactivation time, and priority threshold.

14. The method according to any one of claims 2-13, wherein, The terminal activates the first AI unit based on the first information, including: After receiving the first information at a first time interval at the reference time, the terminal activates the first AI unit; Alternatively, the terminal may shut down the first AI unit based on the first information, including: After receiving the first information at a second time interval at the reference time, the terminal shuts down the first AI unit.

15. The method according to claim 14, wherein, The first time interval or the second time interval includes one of the following: The processing time for the activation or deactivation information of the first object; Processing time of the first information; The processing time when the first AI unit is activated; The processing time for shutting down the first AI unit; The switching time between the first and second modes; The processing time for the first AI unit to collect data and / or train models.

16. The method of claim 14, wherein, The reference time is the time when the first information was received, or the time after the time when the first information was received, after the time offset value.

17. The method according to any one of claims 2-16, wherein, The terminal activates the first AI unit, including: If the third condition is met, the terminal activates the first AI unit based on the first information; wherein the third condition includes at least one of the following: When the first object is associated with the first AI unit, the terminal's battery level is greater than a preset battery level; When the first object is associated with the first AI unit, the power consumption of the terminal is less than the preset power consumption; The shutdown timer of the first AI unit has failed; The terminal has received uplink data; The terminal receives scheduling from the network-side device; The measurement result of the terminal is less than the third threshold; The terminal performs the first operation.

18. The method according to any one of claims 2-17, wherein, The terminal shuts down the first AI unit, including: If the fourth condition is met, the terminal shuts down the first AI unit based on the first information; wherein the fourth condition includes at least one of the following: When the first object is associated with the first AI unit, the terminal's battery level is less than a preset battery level; When the first object is associated with the first AI unit, the power consumption of the terminal is greater than the preset power consumption; The start timer for the first AI unit has failed. The terminal has not received any services and / or has not received scheduling from network-side devices during the first preset time period. The terminal will not receive any services and / or will not receive scheduling from the network-side device during a second preset time period in the future. The measurement result of the terminal is greater than the third threshold; The terminal no longer performs the first operation.

19. The method according to any one of claims 2-18, wherein, The first information includes the first DRX configuration, wherein the first DRX in the first DRX configuration is a connected DRX or an idle DRX.

20. The method according to any one of claims 2-19, wherein, The first information includes the first configuration. When the terminal performs the first action, if the terminal switches to the inactive state of the first configuration, the terminal does not shut down the first AI unit or starts the second timer. The terminal performing the first action includes at least one of the following: The terminal performs the first operation; The terminal performs measurements prior to reporting triggered by the network-side device; The terminal performs at least one of the following through the first AI unit: AI result processing, AI training, AI supervision, AI model management, and AI monitoring; The terminal performs at least one of the following functions: AI training, AI computation, AI supervision, AI model management, AI monitoring, AI result output, and AI unit information input for dual-end AI units or multi-end AI units.

21. The method according to any one of claims 2-20, wherein, The first information includes the DTX configuration of the first cell, and the terminal disables the first AI unit according to the first information, including: According to the DTX configuration of the first cell, the terminal disables the first AI unit used to receive signals sent by the network-side device when the first cell DTX is inactive.

22. The method according to any one of claims 2-20, wherein, The first information includes the DRX configuration of the first cell, and the terminal disables the first AI unit according to the first information, including: According to the DRX configuration of the first cell, the terminal disables the first AI unit used to send signals or report to network-side devices when the first cell DRX is inactive.

23. The method according to any one of claims 1-22, wherein, The method further includes: The terminal receives the first information sent by the network-side device.

24. The method according to any one of claims 1-23, wherein, The first operation includes at least one of the following: beam management, mobility management, CSI feedback, CSI prediction, CSI compression, positioning, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, signal demodulation, sensing, synchronization, measurement, reporting, initial access, and random access.

25. An AI unit processing method, comprising: The network-side device performs a second action, which includes at least one of the following: Send first information to the terminal. The first information includes at least one of the following: first configuration, activation information of a first object, and deactivation information of the first object. The first object includes at least one of a first AI unit and the first configuration. The first configuration includes at least one of the following: first pattern configuration associated with the first AI unit, first DRX configuration, first DTX configuration, first cell DTX configuration, and first cell DRX configuration. Based on the second information, perform at least one of the following: disable the third AI unit or switch to the third mode; or, based on the second information, perform at least one of the following: enable the third AI unit or switch to the fourth mode; In the third mode, the network-side device performs the second operation based on the non-AI unit, and in the fourth mode, the network-side device performs the second operation based on the AI ​​unit, or based on both the non-AI unit and the AI ​​unit.

26. The method of claim 25, wherein, The second information includes at least one of the following: The energy-saving status of the network-side equipment; Second configuration; Activation information of the second object; Deactivation information for the second object; The second object includes at least one of the third AI unit and the second configuration, and the second configuration includes at least one of the following: second pattern configuration, second cell DTX configuration, second cell DRX configuration, second DRX configuration, and second DTX configuration.

27. The method according to claim 26, wherein, The second pattern configuration includes at least one of the following: The duration of the second AI active state timer; The duration of the second AI inactive state timer; The starting time position of the second pattern activation state within the cycle; The starting time position of the inactive state of the second pattern within the cycle; The effective start time position of the second pattern; The cycle of the second pattern; The AI ​​units or AI unit types that are allowed to be enabled when the network-side device is in an inactive or idle state.

28. The method according to claim 26 or 27, wherein, The third AI unit that needs to be turned off when the second configuration is inactive includes at least one of the following: AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption greater than the second preset power consumption, AI unit with computing power occupied greater than the fourth threshold, AI unit with priority lower than the second preset priority, AI unit with specific identification, and AI unit used to perform the second operation. The third AI unit that can be activated when the second configuration is inactive includes at least one of the following: AI receiving unit, AI training unit, AI supervision unit, AI model management unit, AI monitoring unit, AI unit with power consumption less than the second preset power consumption, AI unit with computing power occupied less than the fourth threshold, AI unit with priority higher than the second preset priority, AI unit with specific identification, and AI unit used to perform the second operation.

29. The method according to any one of claims 26-28, wherein, The network-side device shuts down the third AI unit, including one of the following: The network-side device disables the third AI unit in the inactive state of the second configuration. Alternatively, the network-side device may enable a third AI unit, including: The network-side device activates the third AI unit in the active state of the second configuration.

30. The method according to any one of claims 25-29, wherein, The method further includes: After the network-side device turns the third AI unit on or off, it sends a first indication message to the terminal. The first indication message is used to indicate that the network-side device has turned the third AI unit on or off.

31. The method according to claim 25, wherein, The method further includes: If the first condition is met, the network-side device sends the activation information of the first object to the terminal; The first condition includes at least one of the following: The network-side device instructs the terminal to report auxiliary information, which is used by the third AI unit. The network-side device triggers the terminal to perform layer 1 reporting, layer 3 reporting, location information reporting, or beam management reporting; The network-side device receives a random access request from the terminal; The network-side device receives the wake-up signal WUS from the terminal; The network-side device activates the third AI unit; When the first object is associated with the first AI unit, the terminal's battery level is greater than a preset battery level; When the first object is associated with the first AI unit, the power consumption of the terminal is less than the preset power consumption; The shutdown timer of the first AI unit has failed.

32. The method according to claim 25, wherein, The method further includes: If the second condition is met, the network-side device sends the deactivation information of the first object to the terminal; The second condition includes at least one of the following: When the first object is associated with the first AI unit, the terminal's battery level is less than a preset battery level; When the first object is associated with the first AI unit, the power consumption of the terminal is greater than the preset power consumption; The start timer for the first AI unit has failed. The measurement result reported by the terminal is lower or higher than the second threshold; The network-side device shuts down the third AI unit; The network-side device monitors that the first parameter of the first AI unit of the terminal does not meet the preset requirements. The first parameter includes at least one of the following: accuracy, training result, model update, and model management.

33. The method according to any one of claims 25-32, wherein, The second operation includes at least one of the following: beam management, motion management, CSI feedback, CSI prediction, positioning, load balancing, resource allocation, channel estimation, power amplifier nonlinearity suppression, signal demodulation, sensing, synchronization, and measurement.

34. An AI unit processing device, applied to a terminal, wherein, The device includes: The first processing module is configured to perform at least one of the following based on the first information: shut down the first AI unit, switch to a first mode; or... The first processing module is configured to perform at least one of the following based on the first information: activate the first AI unit, switch to the second mode; In the first mode, the terminal performs the first operation based on a non-AI unit; in the second mode, the terminal performs the first operation based on an AI unit, or performs the first operation based on both a non-AI unit and an AI unit.

35. The apparatus according to claim 34, wherein, The first information includes at least one of the following: First configuration; Terminal connection status; The energy-saving status of the terminal; The activation information of the first object or; Deactivation information for the first object; The first object includes at least one of the first AI unit and the first configuration; the first configuration includes at least one of the following: the first format pattern configuration, the first DRX configuration, the first DTX configuration, the first cell DTX configuration, and the first cell DRX configuration associated with the first AI unit.

36. The apparatus according to claim 35, wherein, The first processing module is also used for: Upon receiving the activation information from the first object, the first AI unit is activated; Upon receiving the deactivation information for the first object, the first AI unit is shut down.

37. The apparatus according to any one of claims 35 or 36, wherein, The first processing module is also used for: Upon receiving deactivation information for the first object, the first AI unit is not activated when the first object is in an activated state, or the first AI unit is not deactivated when the first object is in an inactive state.

38. The apparatus according to any one of claims 35-37, wherein, The device further includes: The determination module is used to determine the first AI unit based on target information, wherein the target information includes at least one of the following: terminal type, terminal capability, terminal energy consumption status, terminal battery status, terminal sleep level, terminal service information, terminal service status, and network scheduling information.

39. The apparatus according to any one of claims 35-38, wherein, The device further includes: The receiving module is used to receive the first information sent by the network-side device.

40. An AI unit processing device, applied to network-side equipment, wherein, The device includes: The second processing module is configured to execute a second action, the second action including at least one of the following: Send first information to the terminal. The first information includes at least one of the following: first configuration, activation information of a first object, and deactivation information of the first object. The first object includes at least one of a first AI unit and the first configuration. The first configuration includes at least one of the following: first pattern configuration associated with the first AI unit, first DRX configuration, first DTX configuration, first cell DTX configuration, and first cell DRX configuration. Based on the second information, perform at least one of the following: disable the third AI unit or switch to the third mode; or, based on the second information, perform at least one of the following: enable the third AI unit or switch to the fourth mode; In the third mode, the network-side device performs the second operation based on the non-AI unit, and in the fourth mode, the network-side device performs the second operation based on the AI ​​unit, or based on both the non-AI unit and the AI ​​unit.

41. The apparatus according to claim 40, wherein, The second information includes at least one of the following: The energy-saving status of the network-side equipment; Second configuration; Activation information of the second object; Deactivation information for the second object; The second object includes at least one of the third AI unit and the second configuration, and the second configuration includes at least one of the following: second pattern configuration, second cell DTX configuration, second cell DRX configuration, second DRX configuration, and second DTX configuration.

42. The apparatus according to claim 41, wherein, The second processing module is also used for: The third AI unit is disabled in the inactive state of the second configuration; or, The third AI unit is activated in the active state of the second configuration.

43. The apparatus according to any one of claims 40-42, wherein, The device further includes: The sending module is used to send a first indication message to the terminal after the network-side device turns the third AI unit on or off. The first indication message is used to indicate that the network-side device has turned the third AI unit on or off.

44. A terminal comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the AI ​​unit processing method as claimed in any one of claims 1-24.

45. A network-side device, comprising a processor and a memory, the memory storing a program or instructions executable on the processor, the program or instructions, when executed by the processor, implementing the steps of the AI ​​unit processing method as described in any one of claims 25-33.

46. ​​A readable storage medium, wherein, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the AI ​​unit processing method as described in any one of claims 1-24, or implement the steps of the AI ​​unit processing method as described in any one of claims 25-33.

47. A computer program product, wherein, The computer program product is stored in a storage medium and is executed by at least one processor to implement the steps of the AI ​​unit processing method as described in any one of claims 1-24, or to implement the steps of the AI ​​unit processing method as described in any one of claims 25-33.

Images