Method for determining state of artificial intelligence ai unit, terminal and network side device

Abstract

The application discloses a state determination method of an artificial intelligence (AI) unit, and belongs to the technical field of communication. The method of the application embodiment comprises the following steps: a terminal acquires first information from a network side device, wherein the first information is used for indicating monitoring related information of an AI unit in a first radio resource control (RRC) state; and the terminal determines the state of the AI unit in the first RRC state according to the first information; wherein the first RRC state comprises at least one of a connected state, a non-connected state, an idle state and an inactive state, and the state of the AI unit comprises at least one of an available state, an unavailable state, an activated state and a deactivated state.

Description

Technical Field

[0001] This application belongs to the field of communication technology, specifically relating to a method for determining the state of an artificial intelligence (AI) unit, a terminal, and a network-side device. Background Technology

[0002] Model monitoring refers to the process of monitoring the model inference of an AI unit. For example, in related technologies, when the model inference performance of an AI unit deteriorates due to a shift in the distribution of inference data or other reasons, model monitoring can promptly detect this performance degradation, thus preventing excessive negative impact on the performance of the communication system.

[0003] However, the model monitoring solutions for AI units provided in related technologies still have problems such as poor accuracy in determining the state of AI units during the model monitoring process and large monitoring latency, which affect the performance of the communication system. Summary of the Invention

[0004] This application provides a method for determining the state of an AI unit, a terminal, and a network-side device, which can improve the accuracy of the state determination results of the AI ​​unit, reduce monitoring latency, and improve the performance of the communication system.

[0005] In a first aspect, a method for determining the state of an AI unit is provided, comprising: a terminal acquiring first information from a network-side device, wherein the first information is used to indicate monitoring-related information of the AI ​​unit in a first Radio Resource Control (RRC) state; the terminal determining the state of the AI ​​unit in the first RRC state based on the first information; wherein the first RRC state includes at least one of a connected state, a disconnected state, an idle state, and an inactive state, and the state of the AI ​​unit includes at least one of an available state, an unavailable state, an active state, and a deactivated state.

[0006] Secondly, a method for determining the state of an AI unit is provided, comprising: a network-side device sending first information to a terminal; wherein the first information is used to indicate monitoring-related information of the AI ​​unit in a first RRC state, wherein the first RRC state includes at least one of a connected state, a disconnected state, an idle state, and an inactive state.

[0007] Thirdly, a state determination device for an AI unit is provided, comprising: a transmission module for acquiring first information from a network-side device, wherein the first information is used to indicate monitoring-related information of the AI ​​unit in a first RRC state; and a processing module for determining the state of the AI ​​unit in the first RRC state based on the first information; wherein the first RRC state includes at least one of a connected state, a disconnected state, an idle state, and an inactive state, and the state of the AI ​​unit includes at least one of an available state, an unavailable state, an active state, and a deactivated state.

[0008] Fourthly, a state determination device for an AI unit is provided, comprising: a transmission module for sending first information to a terminal; wherein the first information is used to indicate monitoring-related information of the AI ​​unit in a first Radio Resource Control (RRC) state, wherein the first RRC state includes at least one of a connected state, a disconnected state, an idle state, and an inactive state.

[0009] Fifthly, an apparatus for determining the state of an AI unit 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.

[0010] 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.

[0011] In a seventh aspect, a terminal is provided, including a processor and a communication interface, wherein the communication interface is used to acquire first information from a network-side device, wherein the first information is used to indicate monitoring-related information of an AI unit in a first RRC state; the processor is used to determine the state of the AI ​​unit in the first RRC state based on the first information; wherein the first RRC state includes at least one of a connected state, a disconnected state, an idle state, and an inactive state, and the state of the AI ​​unit includes at least one of an available state, an unavailable state, an active state, and a deactivated state.

[0012] 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.

[0013] In a ninth aspect, a network-side device is provided, including a processor and a communication interface, wherein the communication interface is used to send first information to a terminal; wherein the first information is used to indicate monitoring-related information of an AI unit in a first RRC state, wherein the first RRC state includes at least one of a connected state, a disconnected state, an idle state, and an inactive state.

[0014] 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.

[0015] 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.

[0016] 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 a program or instructions to implement the steps of the method described in the first aspect, or to implement the steps of the method described in the second aspect.

[0017] In a thirteenth aspect, a computer program / program product is provided, which is stored in a storage medium and is executed by at least one processor to implement the steps of the method as described in the first aspect, or to implement the steps of the method as described in the second aspect.

[0018] In this embodiment, the terminal determines the state of the AI ​​unit in the first RRC state, such as at least one of an available state, an unavailable state, an active state, or a deactivated state, by utilizing monitoring-related information of the AI ​​unit indicated by the network-side device in the first RRC state. This improves the accuracy of the AI ​​unit state determination during model monitoring, reduces monitoring latency, and ensures the performance of the communication system. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of a wireless communication system provided in an exemplary embodiment of this application.

[0020] Figure 2 This is one of the flowcharts illustrating the state determination method for an AI unit provided in an exemplary embodiment of this application.

[0021] Figure 3 This is a second flowchart illustrating the state determination method for an AI unit provided in an exemplary embodiment of this application.

[0022] Figure 4a This is one of the interactive flow diagrams of the AI ​​unit state determination method provided in an exemplary embodiment of this application.

[0023] Figure 4b This is the second interactive flow diagram of the AI ​​unit state determination method provided in an exemplary embodiment of this application.

[0024] Figure 4c This is the third interactive flowchart of the AI ​​unit state determination method provided in an exemplary embodiment of this application.

[0025] Figure 5 This is the third flowchart illustrating the state determination method for an AI unit provided in an exemplary embodiment of this application.

[0026] Figure 6 This is one of the structural schematic diagrams of the state determination device of the AI ​​unit provided in an exemplary embodiment of this application.

[0027] Figure 7 This is a second schematic diagram of the structure of the state determination device of the AI ​​unit provided in an exemplary embodiment of this application.

[0028] Figure 8 This is a schematic diagram of the structure of a communication device provided in an exemplary embodiment of this application.

[0029] Figure 9 This is a schematic diagram of the structure of a terminal provided in an exemplary embodiment of this application.

[0030] Figure 10 This is a schematic diagram of the structure of a network-side device provided in an exemplary embodiment of this application. Detailed Implementation

[0031] 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.

[0032] 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.

[0033] The term "instruction" in this application can be either a direct instruction (or explicit instruction) or an indirect instruction (or implicit instruction). A direct instruction can be understood as one in which the sender explicitly informs the receiver of specific information, the operation to be performed, or the requested result, etc.; an indirect instruction can be understood as one in which the receiver determines the corresponding information based on the instruction sent by the sender, or makes a judgment and determines the operation to be performed or the requested result, etc., based on the judgment result.

[0034] 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.

[0035] Figure 1This diagram illustrates a block diagram of a wireless communication system applicable to embodiments of this application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 can be a mobile phone, tablet computer, laptop computer, notebook computer, personal digital assistant (PDA), handheld computer, netbook, ultra-mobile personal computer (UMPC), mobile internet device (MID), augmented reality (AR), virtual reality (VR) device, robot, wearable device, flight vehicle, vehicle user equipment (VUE), shipboard equipment, pedestrian user equipment (PUE), smart home devices (home appliances with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), game consoles, personal computers (PCs), ATMs, or self-service machines, etc. Wearable devices include: smartwatches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart chains, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among these, in-vehicle devices can also be referred to as in-vehicle terminals, in-vehicle controllers, in-vehicle modules, in-vehicle components, in-vehicle chips, or in-vehicle units, etc. It should be noted that the specific type of terminal 11 is not limited in this application embodiment. Network-side equipment 12 can include access network equipment or core network equipment, wherein access network equipment can also be referred to as Radio Access Network (RAN) equipment, radio access network function, or radio access network unit. Access network equipment can include base stations, Wireless Local Area Network (WLAN) access points (APs), or Wireless Fidelity (WiFi) nodes, etc.In this context, a base station may be referred to as a Node B (NB), an Evolved Node B (eNB), a Next Generation Node B (gNB), a New Radio Node B (NR Node B), an Access Point, a Relay Base Station (RBS), a Serving Base Station (SBS), a Base Transceiver Station (BTS), a Radio Base Station, a Radio Transceiver, a Basic Service Set (BSS), an Extended Service Set (ESS), a Home Node B (HNB), a Home Evolved Node B, a Transmit / Receive Point (TRP), or any other suitable term in the relevant field, as long as the same technical effect is achieved. The base station is not limited to any specific technical terminology. It should be noted that in this application embodiment, only a base station in an NR system is used as an example for introduction, and the specific type of base station is not limited.

[0036] Core network equipment, also known as core network nodes, core network functions, or core network elements, includes, but is not limited to, at least one of the following: Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (or L-NEF), and Binding Support Function. Support Functions (BSF), Application Functions (AF), Location Management Functions (LMF), Gateway Mobile Location Centres (GMLC), and Network Data Analytics Functions (NWDAF), etc. It should be noted that this application embodiment only uses core network equipment in the NR system as an example for description, and does not limit the specific type of core network equipment. If the name of the core network equipment mentioned in this application embodiment changes in subsequent protocol versions (e.g., 6G), it will also be within the scope of protection of this application.

[0037] 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).

[0038] Model monitoring refers to the process of monitoring the model inference of an AI unit. For example, in related technologies, when the model inference performance of an AI unit deteriorates due to a shift in the distribution of inference data or other reasons, model monitoring can promptly detect this performance degradation, thus preventing excessive negative impacts on the performance of the communication system. However, the model monitoring schemes for AI units provided in related technologies still suffer from problems such as the accuracy of AI unit state determination results and high monitoring latency, affecting the performance of the communication system. To address these issues, this application provides an AI unit state determination scheme to improve the accuracy of AI unit state determination results, reduce monitoring latency, and improve the performance of the communication system.

[0039] It should be noted that the AI ​​unit described in this application may also be referred to as an AI model, machine learning (ML) model, ML unit, AI structure, AI function, AI characteristic, neural network, neural network function, neural network functionality, etc. Alternatively, the AI ​​unit may refer to a processing unit capable of implementing specific algorithms, formulas, processing flows, capabilities, etc., related to AI. It may also be a processing method, algorithm, function, module, or unit for a specific dataset. Furthermore, the AI ​​unit may be a processing method, algorithm, function, module, or unit running on AI / ML related hardware such as GPUs, NPUs, TPUs, and ASICs. This application does not impose specific limitations in these areas. Optionally, the specific dataset includes the input and / or output of the AI ​​unit.

[0040] Optionally, the identifier of the AI ​​unit may be an AI model identifier, AI structure identifier, AI algorithm identifier, AI function identifier, ML model identifier, ML function identifier, ML unit identifier, etc., or it may be 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 / ML, or the identifier of a function, characteristic, capability, or module related to the AI / ML. This application does not make any specific limitations in this regard.

[0041] It should be noted that the monitoring metric mentioned in this application can also be understood as monitoring key performance indicators (KPIs) or KPIs, etc.

[0042] The beam mentioned in this application may also be referred to as spatial filtering, synchronization signal and PBCH block (SSB), channel state information reference signal (CSI-RS), etc.

[0043] The beam identifier mentioned in this application may include, but is not limited to, at least one of the following: beam angle, beam identifier, beam resource identifier, beam resource set identifier, beam index, beam resource index, and beam resource set index.

[0044] The beam quality mentioned in this application can be understood as the signal quality corresponding to spatial filtering, which may include, but is not limited to, at least one of the following: Reference Signal Receiving Power (RSRP), Reference Signal Receiving Quality (RSRQ), and Signal to Interference plus Noise Ratio (SINR).

[0045] The ground-truth mentioned in this application can also be called a label, and there is no limitation on that.

[0046] The technical solutions provided in this application will be described in detail below with reference to the accompanying drawings and through some embodiments and application scenarios.

[0047] like Figure 2 The diagram shown illustrates a flowchart of a state determination method 200 for an AI unit provided in an exemplary embodiment of this application. This method 200 can be executed by, but is not limited to, a terminal, specifically by hardware and / or software installed in the terminal. In this embodiment, the method 200 may include at least the following steps.

[0048] S210, the terminal obtains the first information from the network-side device.

[0049] The first information is used to indicate monitoring-related information of the AI ​​unit in the first RRC state, and the first RRC state includes at least one of connected state, disconnected state, idle state, and inactive state.

[0050] S220, the terminal determines the state of the AI ​​unit in the first RRC state based on the first information.

[0051] The state of the AI ​​unit includes at least one of the following: available state, unavailable state, active state, and deactivated state.

[0052] In this embodiment, the terminal determines the state of the AI ​​unit in the first RRC state, such as at least one of available, unavailable, active, or deactivated states, by utilizing monitoring-related information of the AI ​​unit indicated by the network-side device in the first RRC state. This improves the accuracy of the AI ​​unit state determination during model monitoring, reduces monitoring latency, and ensures the performance of the communication system.

[0053] In some embodiments, the first information may include, but is not limited to, at least one of the following 11)-13).

[0054] 11) First indication information, used to indicate that the terminal is allowed to determine the state of the AI ​​unit.

[0055] In this embodiment, the first indication information allows the terminal to determine the AI ​​unit in the first RRC state. This can be understood as the network-side device enabling the terminal to determine the state of the AI ​​unit in the first RRC state through autonomous actual measurement, such as determining whether to activate the AI ​​unit. This avoids the problem of excessive monitoring latency caused by the terminal reporting acquired monitoring indicators to the network-side device, which then determines the AI ​​unit's state based on the monitoring indicators and instructs the terminal accordingly. This reduces monitoring latency. Especially when the AI ​​unit's performance deteriorates, it further avoids the problem of communication performance degradation due to excessive monitoring latency not being addressed in time, ensuring the performance of the communication system.

[0056] Optionally, the first indication information can be an explicit indication or an implicit indication. For example, assuming an explicit indication using 1 bit, then "1" can indicate that the terminal is allowed to determine the state of the AI ​​unit, and "0" can indicate that the terminal is not allowed to determine the state of the AI ​​unit; there is no limitation here.

[0057] 12) Second indication information, used to indicate the monitoring-related thresholds of the AI ​​unit in the first RRC state.

[0058] Specifically, by using the second indication information to indicate the monitoring-related thresholds of the AI ​​unit in the first RRC state of the terminal, on the one hand, the terminal can monitor the measured quantities itself, while the network-side device can indirectly control the state of the AI ​​unit by controlling the monitoring-related thresholds. This avoids the problem of unrestricted terminal-side information caused by relying solely on information from the terminal (such as a terminal in a non-connected state) for model monitoring in related technologies, thus improving the accuracy of the AI ​​unit state determination results during the model monitoring process. On the other hand, it also improves the timeliness of the AI ​​unit state determination results during the model monitoring process, avoiding the problem of excessive monitoring latency when the terminal (such as a terminal in a connected state) needs to report the obtained monitoring indicators to the network-side device, and then the network-side device determines the state of the AI ​​unit based on the monitoring indicators and instructs it to the terminal. Especially when the performance of the AI ​​unit deteriorates, it can further avoid the problem of negatively impacting the communication system due to the AI ​​unit's performance deteriorating but not being activated in time. This achieves effective management of the AI ​​unit in the first RRC state of the communication system, ensuring the performance of the communication system.

[0059] Optionally, the monitoring-related thresholds of the AI ​​unit indicated by the second indication information may include, but are not limited to, at least one of the thresholds of the first measurement quantity, the second measurement quantity, and the third measurement quantity.

[0060] Wherein, the first measurement quantity is a measured value (i.e., an actual measured value) and is related to the input of the AI ​​unit. In this embodiment, when the monitoring-related threshold of the AI ​​unit indicated by the second indication information includes the threshold of the first measurement quantity, it can be understood that: when the terminal cannot obtain the truth value, when the terminal judges whether the input data distribution of the AI ​​unit matches the available input characteristics of the AI ​​unit through the first measurement quantity to indirectly judge the inference performance of the AI ​​unit, the terminal can indirectly judge the inference performance of the AI ​​unit according to the threshold of the first measurement quantity configured by the network-side device. This is beneficial for determining the state of the AI ​​unit in a wireless channel environment lacking the truth value, thereby realizing the effective management of the AI ​​unit and ensuring the performance of the communication system.

[0061] Optionally, in this embodiment, the first measurement may include, but is not limited to, at least one of the following: beam quality measurement, timing advance (TA) measurement, channel quality measurement, and specific measurement. The specific measurement is determined based on the channel quality measurement, such as a processed value of the channel measurement, such as eigenvalue decomposition or time delay Doppler transformation.

[0062] Optionally, the first measurement may be an absolute value, a relative value, a distribution-related parameter value, etc., and there are no restrictions on this.

[0063] The second measurement is related to the prediction result (or inference result) of the AI ​​unit. In this embodiment, when the monitoring-related threshold of the AI ​​unit indicated by the second indication information includes the threshold of the second measurement, it can be understood that: when the terminal cannot obtain the true value, and the terminal indirectly judges the inference performance of the AI ​​unit by judging whether the output data distribution of the AI ​​unit matches the available output characteristics of the AI ​​unit through the second measurement, the terminal can indirectly judge the inference performance according to the threshold of the second measurement configured by the network-side device. This is beneficial for more accurately determining the state of the AI ​​unit in a wireless channel environment lacking the true value, thereby achieving effective management of the AI ​​unit and ensuring the performance of the communication system.

[0064] Optionally, in this embodiment, the second measurement may include, but is not limited to, at least one of the following: first beam quality, first beam identifier, predicted synchronization value, predicted channel quality, predicted channel processing value, predicted time of arrival (TOA), and predicted distance from the access network device to the user.

[0065] Wherein, the first beam identifier is one or more predicted beam identifiers. For example, the first beam identifier may be the beam identifier of the predicted beam with the strongest beam quality, or it may be the beam identifier of the predicted top S (S greater than 1) beams with the strongest beam quality, etc.

[0066] In this embodiment, the first beam identifier can be used for at least one of the following a)-c).

[0067] a) Determine the random access timing (RACH Occasion, RO).

[0068] In cases where there are multiple first beam identifiers, the RO can be determined by selecting one from multiple identifiers through protocol agreement (e.g., the beam with the strongest beam quality), default, or random selection, without any restrictions.

[0069] b) Determine the transmission beam corresponding to the random access message.

[0070] The random access message may be one or more of messages such as, but not limited to, message 1 (Msg 1), Msg A, and Msg 3.

[0071] Optionally, if there are multiple first beam identifiers, the transmitting beam corresponding to the random access message can be determined from among them by means of protocol agreement, default or random selection, without any restrictions.

[0072] c) Determine the receive beam corresponding to the random access message.

[0073] The random access message may be one or more of Msg 2, Msg B, Msg 4, etc., but is not limited to.

[0074] Optionally, if there are multiple first beam identifiers, the receiving beam corresponding to the random access message can be determined from among them by means of protocol agreement, default or random selection, without any limitation.

[0075] The first beam quality is the strongest beam quality among the predicted multiple beam qualities.

[0076] Optionally, the second measurement can be an absolute value, a relative value, a distribution-related parameter value, etc., and there are no restrictions on this.

[0077] The third measurement is related to the monitoring index of the AI ​​unit. The monitoring index is determined based on the true value, such as the monitoring index being determined based on the true value and the prediction result corresponding to the true value (such as the second measurement). In this embodiment, when the monitoring-related threshold of the AI ​​unit indicated by the second indication information includes the threshold of the third measurement, it can be understood that: when the terminal can obtain the true value, when the terminal judges the inference performance of the AI ​​unit through the third measurement, the terminal can judge the inference performance of the AI ​​unit according to the threshold of the third measurement configured by the network-side device. This is beneficial for more accurately determining the state of the AI ​​unit in a wireless channel environment, thereby achieving effective management of the AI ​​unit and ensuring the performance of the communication system.

[0078] Optionally, in this embodiment, the third measurement may include, but is not limited to, at least one of the following 121)-128).

[0079] 121) First accuracy rate, the first accuracy rate is the proportion of the first target number in the total number of inferences used for monitoring by the AI ​​unit, the first target number is the number of times that the beam identifier corresponding to the strongest beam quality predicted in N1 inferences is the same as the beam identifier corresponding to the strongest beam quality actually measured, N1 is an integer greater than or equal to 1.

[0080] In this embodiment, the first accuracy rate can also be understood as or replaced by the top 1 accuracy rate.

[0081] In some embodiments, the number of bits occupied by the first accuracy threshold in the second indication information is .

[0082] 122) Second accuracy, the second accuracy is the proportion of the second target number of inferences in the total number of inferences used for monitoring the AI ​​unit, the second target number of inferences is the number of times in N2 inferences that the beam identifiers corresponding to the measured top K strongest beam qualities belong to the beam identifiers corresponding to the predicted top K strongest beam qualities, or the second target number of inferences is the number of times the beam identifiers corresponding to the predicted top K strongest beam qualities belong to the beam identifiers corresponding to the measured top K strongest beam qualities, where K is an integer greater than 1 and N2 is an integer greater than or equal to 1.

[0083] In some embodiments, the second accuracy rate may also be understood as or replaced by topK accuracy rate.

[0084] It is worth noting that the number of bits occupied by the second accuracy threshold in the second indication information is... .

[0085] 123) The difference between the first beam quality and the second beam quality, wherein the first beam quality is the strongest beam quality among the predicted multiple beam qualities, and the second beam quality is the measured beam quality corresponding to the first beam identifier. For the relevant description of the first beam identifier, please refer to the aforementioned relevant description, which will not be repeated here.

[0086] 124) The difference between the second beam quality and the third beam quality, wherein the third beam quality is the measured strongest beam quality.

[0087] 125) The difference between the predicted downlink synchronization and the measured downlink synchronization.

[0088] 126) The difference between downlink synchronization in non-anchor cells and downlink synchronization in anchor cells.

[0089] 127) The number of times the preamble is sent during random access.

[0090] 128) Access latency during random access.

[0091] 13) Third indication information, used to indicate whether the AI ​​unit is enabled in the first RRC state.

[0092] Specifically, the instruction via the third indication information to indicate whether the terminal enables the AI ​​unit in the first RRC state can be understood as: the network-side device decides whether to enable the AI ​​unit in the first RRC state and instructs the terminal accordingly, thereby enabling the network-side device to effectively control the AI ​​unit in the first RRC state. For example, the network-side device can control the state of the AI ​​unit in connected, disconnected, idle, or inactive states, thereby reducing the latency in determining the state of the AI ​​unit while expanding the control range of the network-side device.

[0093] In some embodiments, when the terminal in S220 determines the state of the AI ​​unit in the first RRC state based on the first information, the RRC state of the terminal (such as connected state, disconnected state, idle state, or inactive state) may be the same as or different from the first RRC state, thereby improving the flexibility of the state determination of the AI ​​unit.

[0094] For example, when the terminal is in a connected state, it determines the state of the AI ​​unit in the connected state based on the first information; or, when the terminal is in a disconnected state, it determines the state of the AI ​​unit in the disconnected state based on the first information; or, when the terminal is in a connected state, it determines the state of the AI ​​unit in the disconnected state based on the first information; ... and so on, without limitation.

[0095] In some embodiments, depending on the first information described above, there may be multiple ways to implement the determination of the state of the AI ​​unit in the first RRC state in S220 based on the first information. For example, in this embodiment, the implementation may include, but is not limited to, any one of the following methods 1 to 4.

[0096] Method 1: When the first information includes the first indication information and the second indication information, the state of the AI ​​unit in the first RRC state is determined according to the monitoring-related threshold of the AI ​​unit indicated by the second indication information, such as at least one of the following: available state, unavailable state, active state, and deactivated state.

[0097] Method 2: When the first information includes the second indication information, determine the state of the AI ​​unit in the first RRC state according to the monitoring-related threshold of the AI ​​unit indicated by the second indication information, such as at least one of the following: available state, unavailable state, active state, and deactivated state.

[0098] Method 3: When the first information includes the third indication information and the third indication information is used to indicate that the AI ​​unit is enabled in the first RRC state, the state of the AI ​​unit in the first RRC state is determined to be at least one of the available state and the active state according to the third indication information.

[0099] Method 4: When the first information includes the third indication information and the third indication information is used to indicate that the AI ​​unit is not enabled in the first RRC state, the state of the AI ​​unit in the first RRC state is determined to be at least one of the unavailable state and the deactivated state according to the third indication information.

[0100] It is worth noting that when the terminal uses the first information to determine the state of the AI ​​unit, the terminal may be camped in the cell where the network-side device that provides the first information is located, or it may no longer be camped in the cell where the network-side device is located; there is no restriction on this.

[0101] In some embodiments, the process by which the terminal in the aforementioned methods 1 and / or 2 determines the state of the AI ​​unit in the first RRC state based on the monitoring-related threshold of the AI ​​unit indicated by the second indication information may include, but is not limited to: the terminal determining, based on the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit, that the state of the AI ​​unit in the first RRC state is at least one of the available state, unavailable state, active state, and deactivated state. The target measurement quantity includes at least one of the first measurement quantity, the second measurement quantity, and the third measurement quantity.

[0102] Wherein, assuming that the target measurement quantity includes the first measurement quantity, that is, the monitoring-related threshold of the AI ​​unit in the first RRC state indicated by the second indication information includes the threshold of the first measurement quantity, then the terminal can obtain the first measurement quantity by means of reference signal measurement, etc.

[0103] Assuming that the target measurement includes the second measurement, that is, the monitoring-related threshold of the AI ​​unit in the first RRC state indicated by the second indication information includes the threshold of the second measurement, then the terminal can obtain the second measurement by obtaining the inference result of the AI ​​unit, etc.

[0104] Assuming the target measurement includes the third measurement, that is, the monitoring-related threshold of the AI ​​unit in the first RRC state indicated by the second indication information includes the threshold of the third measurement, then the way the terminal obtains the third measurement may include, but is not limited to: when the terminal is in a connected state, it may obtain the prediction result through AI unit inference or other means, and obtain the true value by measuring the reference signal configured by the network-side device; or, when the terminal is in a disconnected state, it may obtain the prediction result through AI unit inference or other means, and obtain the true value by on-demand SSB scanning, without any restrictions.

[0105] Based on this, as an optional implementation, the process by which the terminal determines the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state according to the magnitude relationship between the target measurement and the monitoring-related threshold of the AI ​​unit can include: when the magnitude relationship between the target measurement and the monitoring-related threshold of the AI ​​unit satisfies a first condition, i.e., it is determined that the performance model inference performance of the AI ​​unit has not deteriorated, the terminal determines that the state of the AI ​​unit in the first RRC state is at least one of the available state and the active state. The first condition includes at least one of the following conditions 11-118.

[0106] Condition 101: If the measured value of the beam quality is greater than the first threshold, it can be determined that the data distribution of the measured value of the beam quality matches the available input characteristics of the AI ​​unit, thereby indirectly determining that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, the terminal can determine that the state of the AI ​​unit in the first RRC state is an available state and / or an active state.

[0107] Condition 102: If the measured value of TA is less than the second threshold, it can be indirectly determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0108] Condition 103: If the first beam quality is greater than the third threshold, it can be determined that the data distribution of the first beam quality matches the available output characteristics of the AI ​​unit, thereby indirectly determining that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0109] Condition 104: If the predicted synchronization value is less than the fourth threshold, it can be indirectly determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0110] Condition 105: If the first accuracy is greater than the fifth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0111] Condition 106: If the second accuracy is greater than the sixth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0112] Condition 107: If the difference between the first beam quality and the second beam quality is less than the seventh threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0113] Condition 108: If the difference between the second beam quality and the third beam quality is less than the eighth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0114] Condition 109: If the deviation between the predicted downlink synchronization and the measured downlink synchronization is less than the ninth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an available state and / or an active state.

[0115] Condition 110: If the difference between the downlink synchronization of the non-anchor cell and the downlink synchronization of the anchor cell is less than the tenth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is available and / or active.

[0116] Condition 111: If the number of times the preamble is sent during random access is less than the eleventh threshold, it can be determined that the inference performance of the AI ​​unit is good or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an available state and / or an active state.

[0117] Condition 112: If the access delay during random access is less than the twelfth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an available state and / or an active state.

[0118] Condition 113: If the measured value of the channel quality is greater than the thirteenth threshold, it can be determined that the data distribution of the measured value of the channel quality matches the available input characteristics of the AI ​​unit, thereby indirectly determining that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an available state and / or an active state.

[0119] Condition 114: If the prediction error of the specific measurement value is less than the fourteenth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0120] Condition 115: If the prediction error of the channel quality is less than the fifteenth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an available state and / or an active state.

[0121] Condition 116: If the prediction error of the channel processing value is less than the sixteenth threshold, it can be determined that the inference performance of the AI ​​unit is good or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an available state and / or an active state.

[0122] Condition 117: If the prediction error of the TOA is less than the seventeenth threshold, it can be determined that the inference performance of the AI ​​unit is better or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is a usable state and / or an active state.

[0123] Condition 118: If the prediction error of the distance from the access network device to the user is less than the eighteenth threshold, it can be determined that the inference performance of the AI ​​unit is good or has not deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is available and / or active.

[0124] It is understood that for the aforementioned conditions 101-118, if there is an equality condition, the state of the AI ​​unit in the first RRC state can be determined to be at least one of the following: available state, unavailable state, active state, or deactivated state. For example, taking condition 101 as an example, if the measured value of the beam quality is equal to the first threshold, then the state of the AI ​​unit can be determined to be available state, unavailable state, active state, or deactivated state.

[0125] Furthermore, the aforementioned first to eighteenth thresholds and subsequent nineteenth to thirty-sixth thresholds are at least a portion of the monitoring-related thresholds of the AI ​​unit indicated by the aforementioned second instruction information.

[0126] As another optional implementation, the process by which the aforementioned terminal determines the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit can include: determining the state of the AI ​​unit in the first RRC state as at least one of the unavailable state and the deactivated state when the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit satisfies a second condition. The second condition includes at least one of the following conditions 201-218.

[0127] Condition 201: If the measured value of the beam quality is less than the nineteenth threshold, it can be determined that the data distribution of the measured value of the beam quality does not match the input characteristics available to the AI ​​unit, thereby indirectly determining that the inference performance of the AI ​​unit has deteriorated. Therefore, the terminal can determine that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0128] The nineteenth threshold can be less than or equal to the first threshold in the first condition.

[0129] Condition 202: If the measured value of TA is greater than the twentieth threshold, it can be indirectly determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unusable state and / or a deactivated state.

[0130] The twentieth threshold can be greater than or equal to the second threshold in the first condition.

[0131] Condition 203: If the first beam quality is less than the twenty-first threshold, it can be determined that the data distribution of the first beam quality does not match the output characteristics available to the AI ​​unit, thereby indirectly determining that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unusable state and / or a deactivated state.

[0132] The 21st threshold can be less than or equal to the 3rd threshold in the first condition.

[0133] Condition 204: If the predicted synchronization value is greater than the twenty-second threshold, it can be indirectly determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0134] The twelfth threshold can be greater than or equal to the fourth threshold in the first condition.

[0135] Condition 205: If the first accuracy is less than the twenty-third threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unusable state and / or a deactivated state.

[0136] The 23rd threshold can be less than or equal to the 5th threshold in the first condition.

[0137] Condition 206: If the second accuracy is less than the twenty-fourth threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0138] The 24th threshold can be less than or equal to the 6th threshold in the first condition.

[0139] Condition 207: If the difference between the first beam quality and the second beam quality is greater than the twenty-fifth threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unusable state and / or a deactivated state.

[0140] The 25th threshold can be greater than or equal to the 7th threshold in the first condition.

[0141] Condition 208: If the difference between the second beam quality and the third beam quality is greater than the twenty-sixth threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0142] The 26th threshold can be greater than or equal to the 8th threshold in the first condition.

[0143] Condition 209: If the deviation between the predicted downlink synchronization and the measured downlink synchronization is greater than the twenty-seventh threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0144] The 27th threshold can be greater than or equal to the 9th threshold in the first condition.

[0145] Condition 210: If the difference between the downlink synchronization of the non-anchor cell and the downlink synchronization of the anchor cell is greater than the twenty-eighth threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is unavailable and / or deactivated.

[0146] The 28th threshold can be greater than or equal to the 10th threshold in the first condition.

[0147] Condition 211: If the number of preamble transmissions during random access is greater than the 29th threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is unavailable and / or deactivated.

[0148] The 29th threshold can be greater than or equal to the 11th threshold in the first condition.

[0149] Condition 212: If the access delay during random access is greater than the 30th threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is unavailable and / or deactivated.

[0150] The thirtieth threshold can be greater than or equal to the twelfth threshold in the first condition.

[0151] Condition 213: If the measured value of the channel quality is less than the thirty-first threshold, it can be determined that the data distribution of the measured value of the channel quality does not match the input characteristics available to the AI ​​unit, and thus indirectly determine that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0152] The thirty-first threshold can be less than or equal to the thirteenth threshold in the first condition.

[0153] Condition 214: If the prediction error of the specific measurement value is greater than the thirty-second threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0154] The thirty-second threshold can be greater than or equal to the fourteenth threshold in the first condition.

[0155] Condition 215: If the prediction error of the channel quality is greater than the 33th threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0156] The thirty-third threshold can be greater than or equal to the fifteenth threshold in the first condition.

[0157] Condition 216: If the prediction error of the channel processing value is greater than the thirty-fourth threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unavailable state and / or a deactivated state.

[0158] The thirty-fourth threshold can be greater than or equal to the sixteenth threshold in the first condition.

[0159] Condition 217: If the prediction error of the TOA is greater than the thirty-fifth threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is an unusable state and / or a deactivated state.

[0160] The thirty-fifth threshold can be greater than or equal to the seventeenth threshold in the first condition.

[0161] Condition 218: If the prediction error of the distance from the access network device to the user is greater than the thirty-sixth threshold, it can be determined that the inference performance of the AI ​​unit has deteriorated. Therefore, it can be determined that the state of the AI ​​unit in the first RRC state is unavailable and / or deactivated.

[0162] The thirty-sixth threshold can be greater than or equal to the eighteenth threshold in the first condition.

[0163] Similar to the first condition mentioned above, in this embodiment, the judgment of the second condition can determine that the inference performance of the AI ​​unit has deteriorated, so as to manage the AI ​​unit.

[0164] Furthermore, for conditions 201-218 mentioned above, if there is an equality condition, the state of the AI ​​unit in the first RRC state can be determined to be at least one of the following: available state, unavailable state, active state, or deactivated state. For example, taking condition 201 as an example, if the measured value of the beam quality is equal to the first threshold, the state of the AI ​​unit can be determined to be available state, unavailable state, active state, or deactivated state.

[0165] In some embodiments, after the terminal determines the state of the AI ​​unit in the first RRC state, the method embodiment 200 may further include Figure 3 The content of S230 shown is as follows.

[0166] S230, perform at least one of the first operation and the second operation according to the state of the AI ​​unit.

[0167] The implementation process of S230 can be varied. For example, when the terminal determines that the state of the AI ​​unit in the first RRC state is at least one of the available state and the active state, it performs a first operation, which includes activating the AI ​​unit or maintaining the active state of the AI ​​unit. That is, when the terminal determines that the inference performance of the AI ​​unit is good or has not deteriorated, it can enable the AI ​​unit to improve the performance of the communication system.

[0168] For example, if the state of the AI ​​unit in the first RRC state is determined to be at least one of the unavailable state and the deactivated state, the terminal performs a second operation. The second operation includes at least one of the following: deactivating the AI ​​unit, maintaining the deactivated state of the AI ​​unit, reverting to a non-AI algorithm, or switching to another AI unit other than the AI ​​unit in question. In other words, when the terminal determines that the inference performance of the AI ​​unit has deteriorated, it shuts down the AI ​​unit to prevent the degraded AI unit from negatively impacting the communication system.

[0169] In some embodiments, the process by which the terminal obtains the first information from the network-side device in S210 may include, but is not limited to: the terminal obtaining the first information from the network-side device in a third RRC state. The third RRC state may include, but is not limited to, at least one of a connected state, a disconnected state, an idle state, and an inactive state.

[0170] In other words, in this embodiment, the RRC state when the terminal obtains the first information can be the same as or different from the RRC state when determining the state of the AI ​​unit, thereby adapting to the state determination of the AI ​​unit under different circumstances.

[0171] For example, assuming that the third RRC state is the same as the first RRC state, such as both being disconnected states, then the terminal can obtain first information from the network-side device in the disconnected state, and determine the state of the AI ​​unit in the disconnected state based on the first information.

[0172] For example, assuming that the third RRC state is different from the first RRC state, such as the third RRC state being connected and the first RRC state being disconnected, then the terminal can obtain first information from the network-side device in the connected state, and determine the state of the AI ​​unit in the disconnected state based on the first information.

[0173] For example, assuming that the third RRC state is the same as the first RRC state, such as both being connected states, then the terminal can obtain first information from the network-side device in the connected state, and determine the state of the AI ​​unit in the connected state based on the first information.

[0174] In some embodiments, when the third RRC state is a disconnected state, an idle state, or an inactive state, the method by which the terminal obtains the first information from the network-side device may include: the terminal receiving a broadcast message from the network-side device, the broadcast message including the first information. In this embodiment, the network-side device transmits the first information via broadcast, which can instruct all terminals receiving the broadcast message to determine the state of the AI ​​unit based on the first information, thereby improving the management efficiency of the AI ​​unit.

[0175] When the third RRC state is connected, the terminal may obtain the first information from the network-side device in the following ways: the terminal receives an RRC message from the network-side device, the RRC message including the first information, and the RRC message including at least one of an RRC Release message, an RRC Resume message, and an RRC Reconfiguration message. The RRC Release message may also be referred to as or replaced by an RRC Release with Suspended Indication message.

[0176] In some embodiments, when the third RRC state is a connected state, the terminal may also request the first information from the network-side device, that is, the first information is sent by the network-side device according to the terminal's request.

[0177] In some embodiments, where the first information includes the third indication information and the third indication information is used to indicate whether the AI ​​unit is enabled in the second RRC state, the RRC message includes an RRC release message; wherein the second RRC state includes at least one of the disconnected state, idle state, and inactive state in the first RRC state.

[0178] In some embodiments, the method embodiment 200 may further include: the terminal sending second information to a network-side device, wherein the second information is used to request monitoring of the AI ​​unit in the first RRC state during the connected state. Then, after receiving the second information, the network-side device may send the first information to the terminal to indicate monitoring-related information of the AI ​​unit in the first RRC state.

[0179] In other words, in this embodiment, the terminal can request the network-side device to determine and indicate the monitoring-related indicators of the AI ​​unit in the first RRC state based on the monitoring indicators of the connected state. This effectively utilizes the information from the network-side device to determine the state of the AI ​​unit, improves the accuracy of the state determination results of the AI ​​unit during the model monitoring process, and reduces monitoring latency. At the same time, by performing model monitoring of the AI ​​unit in advance in the connected state, it is beneficial for the network-side device to obtain the inference performance of the AI ​​unit before the model inference of the AI ​​unit, and better manage the state of the AI ​​unit in the first RRC state.

[0180] Furthermore, when the first RRC state is in a disconnected, idle, or inactive state, the network-side device can determine whether to enable the AI ​​unit in the disconnected, idle, or inactive state based on the monitoring indicators of the connected state, and instruct the terminal through third indication information. This allows the network-side device to determine the state of the AI ​​unit, ensuring that the network-side device can manage the AI ​​unit in real time.

[0181] Optionally, the second information may include, but is not limited to, at least one of a fourth indication information and a fifth indication information. The fourth indication information is used to indicate the AI ​​unit monitored by the terminal in the connected state. The fifth indication information is used to indicate the identifier of the AI ​​unit monitored by the terminal in the connected state.

[0182] In this embodiment, the network-side device can clearly identify which AI unit the terminal is requesting to monitor through the fourth and / or fifth indication information. This ensures that the first information indicated by the network-side device to the terminal matches the terminal's needs. For example, it can ensure that the network-side device configures monitoring reference measurement resources that match the AI ​​unit requested by the terminal, so that the terminal can obtain monitoring indicators and report them to the network-side device. This allows the network-side device to obtain the monitoring indicators of the corresponding AI unit and then send the first information that matches the terminal's AI capabilities to the terminal.

[0183] In some embodiments, the first information may be determined by the network-side device based on a third measurement reported by the terminal. Based on this, the method embodiment 200 may further include: the network-side device sending reporting configuration information to the terminal. The reporting configuration information is used to configure the monitoring indicators to be reported.

[0184] Optionally, the reported configuration information may include, but is not limited to, at least one of the following 21)-26).

[0185] 21) First measurement resource indication, wherein the first measurement resource is the measurement resource corresponding to the input data of the AI ​​unit.

[0186] 22) Second measurement resource indication, wherein the second measurement resource is the measurement resource corresponding to the truth value or label of the AI ​​unit.

[0187] 23) Association indication of the first and second measurement resources.

[0188] 24) Activation indication of the first measurement resource.

[0189] 25) Activation indication of the second measurement resource.

[0190] 26) Indication of the third measurement quantity.

[0191] Based on this, in some embodiments, after the terminal obtains the third measurement based on the reported configuration measurement, it sends the third information to the network-side device, the third information being used to indicate the third measurement.

[0192] In other words, when determining the state of the AI ​​unit in the second RRC state, the terminal can first perform reference signal measurement and AI unit inference in the connected state, and then report the third measurement obtained from the inference to the network-side device. This is used by the network-side device to determine the monitoring-related indicators of the AI ​​unit in the second RRC state (such as the disconnected state, idle state, or inactive state), i.e., the first information. Therefore, by performing model monitoring in advance in the connected state, it is beneficial for the network-side device to obtain the inference performance of the AI ​​unit before the model inference of the AI ​​unit, which is conducive to better management of the state of the AI ​​unit in the second RRC state.

[0193] In some embodiments, in order to ensure that the first information configured by the network-side device for the terminal or the reported configuration matches the terminal's capabilities, in this embodiment, the terminal may also report its capabilities to the network-side device, wherein the terminal capabilities are used to indicate the capabilities of the AI ​​units or AI capabilities supported by the terminal.

[0194] Optionally, the terminal capabilities may include, but are not limited to, at least one of the sixth indication information, the seventh indication information, and the eighth indication information.

[0195] The sixth indication information is used to indicate whether AI-based prediction is supported in the first RRC state (such as disconnected state, idle state, inactive state).

[0196] The seventh indication information is used to indicate the applicable AI unit functions of the terminal in the first RRC state (such as disconnected state, idle state, inactive state).

[0197] The eighth indication information is used to indicate the applicable AI unit identifier of the terminal in the first RRC state (such as disconnected state, idle state, inactive state).

[0198] Based on the description of the aforementioned method embodiment 200, for ease of understanding, the state determination scheme of the AI ​​unit provided in this application embodiment will be further explained below with reference to Examples 1-3, as follows.

[0199] Example 1 Assuming the first RRC state is the same as the third RRC state, and the terminal obtains the first information in the third RRC state (e.g., disconnected, idle, or inactive in Example 1), and determines the state of the AI ​​unit in the first RRC state (e.g., disconnected, idle, or inactive in Example 1), then, as Figure 4a As shown, its implementation process is as follows.

[0200] S411, the network-side device broadcasts the first message.

[0201] Wherein, the first information includes first indication information and second indication information, or the first information includes second indication information, the first indication information is used to indicate that the terminal is allowed to determine the state of the AI ​​unit, and the second indication information is used to indicate the monitoring-related thresholds of the AI ​​unit in the first RRC state.

[0202] S412, the terminal receives the first information from the network-side device in the third RRC state.

[0203] S413, the terminal determines the state of the AI ​​unit in the first RRC state based on the first information, such as available state, unavailable state, active state, deactivated state, etc.

[0204] S414, the terminal performs a first operation or a second operation based on the state of the AI ​​unit.

[0205] For example, when the AI ​​unit is in an available state and / or an active state, a first operation is performed, such as activating the AI ​​unit or continuing to maintain the active state of the AI ​​unit.

[0206] For example, if the AI ​​unit is in an unavailable state and / or a deactivated state, a second operation is performed, such as deactivating the AI ​​unit, maintaining the deactivated state of the AI ​​unit, reverting to a non-AI algorithm, or switching to another AI unit other than the AI ​​unit.

[0207] In Example 1, monitoring-related thresholds for AI unit activation or deactivation are broadcast by the network-side device, thereby enabling the network side to indirectly control the AI ​​unit on the terminal side and ensure the accuracy of the AI ​​unit's state determination.

[0208] Furthermore, compared to network-side devices directly instructing the terminal on the status of the AI ​​unit, in Example 1, since the terminal can obtain monitoring-related measurements in real time, it can quickly determine the status of the AI ​​unit based on these real-time measurements and the monitoring-related thresholds broadcast by the network-side devices, resulting in higher timeliness of AI unit management.

[0209] It is understood that the implementation process of each step in this Example 1 can refer to the relevant description in the aforementioned method embodiment 200, and will not be repeated here. In addition, this Example 1 may include more or fewer steps than the aforementioned S411-S414, and there is no limitation here.

[0210] Example 2 Assuming the first RRC state is different from the third RRC state, and the terminal obtains the first information in the third RRC state (connected state in Example 2) and determines the state of the AI ​​unit in the first RRC state (disconnected state, idle state, or inactive state in Example 2), then, as Figure 4b As shown, its implementation process is as follows.

[0211] S421, the terminal reports its capabilities to the network-side device.

[0212] The terminal capabilities may include, but are not limited to, at least one of the sixth instruction information, the seventh instruction information, and the eighth instruction information.

[0213] The sixth indication information is used to indicate whether AI-based prediction is supported in the first RRC state.

[0214] The seventh indication information is used to indicate the AI ​​unit functions that are applicable to the terminal in the first RRC state.

[0215] The eighth indication information is used to indicate the applicable AI unit identifier of the terminal in the first RRC state.

[0216] S422, the terminal sends the second information to the network-side device.

[0217] The second information is used to request monitoring of the AI ​​unit in the first RRC state during the connected state.

[0218] S422 is an optional step.

[0219] S423, the network-side device sends configuration information to the terminal.

[0220] The reporting configuration information is used to configure the monitoring metrics to be reported. The reporting configuration information includes at least one of the following.

[0221] The first measurement resource is the measurement resource corresponding to the input data of the AI ​​unit.

[0222] The second measurement resource is the measurement resource corresponding to the truth value or label of the AI ​​unit.

[0223] Association indication of the first and second measurement resources.

[0224] Activation indication of the first measurement resource.

[0225] Activation indication for the second measurement resource.

[0226] The indication of the third measurement quantity.

[0227] S424, the terminal sends third information to the network-side device.

[0228] The third information is used to indicate a third measurement quantity to the network-side device.

[0229] S425, the network-side device sends an RRC release message to the terminal.

[0230] The RRC release message includes first information, which includes third indication information. The third indication information is used to indicate whether the AI ​​unit is enabled in the second RRC state (in this example 2, the disconnected state, idle state, or inactive state).

[0231] S426, the terminal receives an RRC release message from the network-side device while in connected state.

[0232] S427, the terminal determines the state of the AI ​​unit in the second RRC state, such as available state, unavailable state, active state, deactivated state, etc., based on the first information in the RRC release message.

[0233] For example, when the third indication information is used to indicate that the AI ​​unit is enabled in the second RRC state, the state of the AI ​​unit in the first RRC state is determined to be at least one of the available state and the active state; For example, when the third indication information is used to indicate that the AI ​​unit is not enabled in the second RRC state, the state of the AI ​​unit in the first RRC state is determined to be at least one of the unavailable state and the deactivated state.

[0234] S428, the terminal performs a first operation or a second operation based on the state of the AI ​​unit.

[0235] For example, when the AI ​​unit is in an available state and / or an active state, a first operation is performed, such as activating the AI ​​unit or continuing to maintain the active state of the AI ​​unit.

[0236] For example, if the AI ​​unit is in an unavailable state and / or a deactivated state, a second operation is performed, such as deactivating the AI ​​unit, maintaining the deactivated state of the AI ​​unit, reverting to a non-AI algorithm, or switching to another AI unit other than the AI ​​unit.

[0237] In Example 2, before activating the AI ​​unit in the first RRC state (connected, idle, or inactive state in this example), the terminal first informs the network-side device of its AI capabilities through capability reporting in the connected state. Then, the network-side device configures the monitoring reference signal according to the terminal's AI capabilities. Next, the terminal performs reference signal measurement and AI inference to obtain a third measurement quantity and reports it. Finally, the network-side device determines the state of the AI ​​unit based on the third measurement quantity and instructs the terminal through third indication information, such as whether to enable the AI ​​unit in the second RRC state. That is, in Example 2, the network-side device decides whether the AI ​​unit is activated or available, thus enabling real-time model management by the network-side device.

[0238] It is understood that the implementation process of each step in this Example 2 can refer to the relevant description in the aforementioned method embodiment 200, and will not be repeated here. In addition, this Example 2 may include more or fewer steps than the aforementioned S421-S428, and there is no limitation here.

[0239] Example 3 Assuming the first RRC state is the same as the third RRC state, and the terminal obtains the first information in the third RRC state (connected state in example 3), and determines the state of the AI ​​unit in the first RRC state (connected state in example 3) based on the first information, then, as Figure 4c As shown, its implementation process is as follows.

[0240] S431, the network-side device sends the first message.

[0241] Wherein, the first information includes first indication information and second indication information, or the first information includes second indication information, wherein the first indication information is used to indicate that the terminal is allowed to determine the state of the AI ​​unit, and the second indication information is used to indicate the monitoring-related thresholds of the AI ​​unit in the connected state.

[0242] Optionally, the second indication information may also include an indication of a third measurement.

[0243] S432, the terminal receives the first information from the network-side device in the third RRC state (connected state in this embodiment).

[0244] S433, the network-side device sends monitoring-related resource configuration information to the terminal.

[0245] The monitoring-related resource configuration is used to configure the measurement resources monitored by the AI ​​unit. The monitoring-related resource configuration includes at least one of the following.

[0246] The first measurement resource is the measurement resource corresponding to the input data of the AI ​​unit.

[0247] The second measurement resource is the measurement resource corresponding to the truth value or label of the AI ​​unit.

[0248] Association indication of the first and second measurement resources.

[0249] Activation indication of the first measurement resource.

[0250] Activation indication for the second measurement resource.

[0251] The resource configuration information in S433 and the first information in S432 can be carried in the same signaling message or they can be different signaling messages; this is not a restriction.

[0252] S434, the terminal determines the state of the AI ​​unit (such as available, unavailable, activated, deactivated, etc.) in the first RRC state (connected state in this embodiment) based on the first information.

[0253] S435, the terminal executes the first operation or the second operation based on the state of the AI ​​unit.

[0254] For example, when the AI ​​unit is in an available state and / or an active state, a first operation is performed, such as activating the AI ​​unit or continuing to maintain the active state of the AI ​​unit.

[0255] For example, if the AI ​​unit is in an unavailable state and / or a deactivated state, a second operation is performed, such as deactivating the AI ​​unit, maintaining the deactivated state of the AI ​​unit, reverting to a non-AI algorithm, or switching to another AI unit other than the AI ​​unit.

[0256] In Example 3, monitoring-related thresholds for activating or deactivating the AI ​​unit are configured through network-side devices, thereby enabling the network side to indirectly control the AI ​​unit on the terminal side and ensuring the accuracy of the AI ​​unit's state determination.

[0257] Furthermore, compared to network-side devices directly instructing the terminal on the status of the AI ​​unit, in Example 3, since the terminal can obtain monitoring-related measurements in real time, it can quickly determine the status of the AI ​​unit based on these real-time measurements and the monitoring-related thresholds configured on the network-side devices, resulting in higher timeliness of AI unit management.

[0258] It is understood that the implementation process of each step in this Example 3 can refer to the relevant description in the aforementioned method embodiment 200, and will not be repeated here. In addition, this Example 3 may include more or fewer steps than the aforementioned S431-S435, and there is no limitation here.

[0259] like Figure 5 The diagram shown illustrates a flowchart of a state determination method 500 for an AI unit provided in an exemplary embodiment of this application. This method 500 can be executed by, but is not limited to, a network-side device, specifically by hardware and / or software installed in the network-side device. In this embodiment, the method 500 may include at least the following steps.

[0260] S510, the network-side device sends the first information to the terminal.

[0261] The first information is used to indicate monitoring-related information of the AI ​​unit in the first RRC state, wherein the first RRC state includes at least one of connected state, disconnected state, idle state, and inactive state.

[0262] In some embodiments, the first information includes at least one of the following: first indication information for indicating that the terminal is allowed to determine the state of the AI ​​unit; second indication information for indicating the monitoring-related thresholds of the AI ​​unit in the first RRC state; and third indication information for indicating whether the AI ​​unit is enabled in the first RRC state. The monitoring-related thresholds of the AI ​​unit include at least one of the following: a threshold for a first measurement quantity, wherein the first measurement quantity is a measured value and is related to the input of the AI ​​unit; a threshold for a second measurement quantity, wherein the second measurement quantity is related to the prediction result of the AI ​​unit; and a threshold for a third measurement quantity, wherein the third measurement quantity is related to the monitoring index of the AI ​​unit, and the monitoring index is determined based on a true value.

[0263] In some embodiments, the first measurement includes at least one of a beam quality measurement, a timing advance (TA) measurement, a channel quality measurement, and a specific measurement determined based on the channel quality measurement.

[0264] In some embodiments, the second measurement includes at least one of a first beam quality, a first beam identifier, a predicted synchronization value, a predicted channel quality, a predicted channel processing value, a predicted time of arrival (TOA), and a predicted distance from the access network device to the user; wherein the first beam identifier is a predicted beam identifier, and the first beam quality is the strongest beam quality among a plurality of predicted beam qualities.

[0265] In some embodiments, the third measurement includes at least one of the following: a first accuracy rate, wherein the first accuracy rate is the proportion of a first target number of inferences used for monitoring the AI ​​unit, and the first target number of inferences is the number of times, in N1 inferences, the beam identifier corresponding to the predicted strongest beam quality is the same as the beam identifier corresponding to the measured strongest beam quality, where N1 is an integer greater than or equal to 1; a second accuracy rate, wherein the second accuracy rate is the proportion of a second target number of inferences used for monitoring the AI ​​unit, and the second target number of inferences is the number of times, in N2 inferences, the beam identifier corresponding to the measured top K strongest beam qualities belongs to the beam identifier corresponding to the predicted top K strongest beam qualities, or, the second target number of inferences is the number of times, in N2 inferences, the beam identifier corresponding to the predicted top K strongest beam qualities is the same as the beam identifier corresponding to the measured top K strongest beam qualities. The beam identifier is the number of times the beam identifier corresponds to the K strongest measured beam qualities, where K is an integer greater than 1 and N2 is an integer greater than or equal to 1; the difference between the first beam quality and the second beam quality, where the first beam quality is the strongest among multiple predicted beam qualities and the second beam quality is the measured beam quality corresponding to the first beam identifier, where the first beam identifier is a predicted beam identifier; the difference between the second beam quality and the third beam quality, where the third beam quality is the strongest measured beam quality; the difference between the predicted downlink synchronization and the measured downlink synchronization; the difference between the downlink synchronization of a non-anchor cell and the downlink synchronization of an anchor cell; the number of times the preamble is sent during random access; and the access delay during random access.

[0266] In some embodiments, the first beam identifier is used for at least one of the following: determining the random access timing; determining the transmit beam corresponding to the random access message; and determining the receive beam corresponding to the random access message.

[0267] In some embodiments, the network-side device sends first information to the terminal, including any one of the following: the network-side device sends a broadcast message, the broadcast message including the first information; the network-side device sends an RRC message to the terminal, the RRC message including the first information, the RRC message including at least one of an RRC release message, an RRC recovery message, and an RRC reconfiguration message; the network-side device sends the first information to the terminal according to the terminal's request.

[0268] In some embodiments, where the first information includes the third indication information and the third indication information is used to indicate whether the AI ​​unit is enabled in the second RRC state, the RRC message includes an RRC release message; wherein the second RRC state includes at least one of the disconnected state, idle state, and inactive state in the first RRC state.

[0269] In some embodiments, the method further includes: the network-side device receiving second information from the terminal; wherein the second information is used to request monitoring of the AI ​​unit in the first RRC state in a connected state.

[0270] In some embodiments, the second information includes at least one of the following: fourth indication information for indicating the AI ​​unit that the terminal monitors in the connected state; and fifth indication information for indicating the identifier of the AI ​​unit that the terminal monitors in the connected state.

[0271] It is understood that each implementation in this method embodiment 500 has the same or corresponding technical features as the aforementioned method embodiment 200 or 300. Therefore, the implementation of each implementation in this method embodiment 500 can be referred to the relevant descriptions in the aforementioned method embodiment 200 or 300 to achieve the same or corresponding technical effects. To avoid repetition, it will not be described again here.

[0272] The AI ​​unit state determination method provided in this application embodiment can be executed by an AI unit state determination entity. Device In this embodiment, the state of the AI ​​unit is used for determination. Device Taking the execution of the AI ​​unit state determination method as an example, the state determination device of the AI ​​unit provided in the embodiments of this application is described.

[0273] This application provides a state determination device for an AI unit. As an example, the state determination device for the AI ​​unit can be a communication device or a component within a communication device, such as a chip. The communication device can be a terminal, a network-side device, or a server, etc. Exemplarily, the terminal can include, but is not limited to, the type of terminal 11 listed above, and the network-side device can include, but is not limited to, the type of network-side device 12 listed above. This application does not impose specific limitations.

[0274] The state determination device for the AI ​​unit includes a transmission module (such as a receiving module and a transmitting module) and a processing module. The receiving module, transmitting module, and processing module can be implemented in software or hardware. When implemented in hardware, the processing module can be implemented by a processor. For example, the processor can include general-purpose processors, special-purpose processors, such as a Central Processing Unit (CPU), microprocessor, Digital Signal Processor (DSP), Artificial Intelligence (AI) processor, Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC), Network Processor (NP), Field Programmable Gate Array (FPGA), or other programmable logic devices, gate circuits, transistors, discrete hardware components, etc. The receiving module and transmitting module can be implemented by a communication interface, which can include one or more of the following: transceiver, pins, circuits, bus, radio frequency unit, etc.

[0275] For details, see Figure 6 When the AI ​​unit's state determination device is a terminal or a component within a terminal, the AI ​​unit's state determination device 600 includes a transmission module 610 for acquiring first information from a network-side device, wherein the first information is used to indicate monitoring-related information of the AI ​​unit in a first RRC state; and a processing module 620 for determining the state of the AI ​​unit in the first RRC state based on the first information; wherein the first RRC state includes at least one of a connected state, a disconnected state, an idle state, and an inactive state, and the state of the AI ​​unit includes at least one of an available state, an unavailable state, an active state, and a deactivated state.

[0276] In some embodiments, the first information includes at least one of the following: first indication information for indicating that the terminal is allowed to determine the state of the AI ​​unit; second indication information for indicating the monitoring-related thresholds of the AI ​​unit in the first RRC state; and third indication information for indicating whether the AI ​​unit is enabled in the first RRC state. The monitoring-related thresholds of the AI ​​unit include at least one of the following: a threshold for a first measurement quantity, wherein the first measurement quantity is a measured value and is related to the input of the AI ​​unit; a threshold for a second measurement quantity, wherein the second measurement quantity is related to the prediction result of the AI ​​unit; and a threshold for a third measurement quantity, wherein the third measurement quantity is related to the monitoring index of the AI ​​unit, and the monitoring index is determined based on a true value.

[0277] In some embodiments, determining the state of the AI ​​unit in the first RRC state based on the first information includes any one of the following: when the first information includes the first indication information and the second indication information, determining the state of the AI ​​unit in the first RRC state based on the monitoring-related threshold of the AI ​​unit indicated by the second indication information; when the first information includes the second indication information, determining the state of the AI ​​unit in the first RRC state based on the monitoring-related threshold of the AI ​​unit indicated by the second indication information; when the first information includes the third indication information and the third indication information is used to indicate that the AI ​​unit is enabled in the first RRC state, determining the state of the AI ​​unit in the first RRC state as at least one of the available state and the active state based on the third indication information; when the first information includes the third indication information and the third indication information is used to indicate that the AI ​​unit is not enabled in the first RRC state, determining the state of the AI ​​unit in the first RRC state as at least one of the unavailable state and the deactivated state based on the third indication information.

[0278] In some embodiments, determining the state of the AI ​​unit in the first RRC state according to the monitoring-related threshold of the AI ​​unit indicated by the second indication information includes: determining the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit; wherein the target measurement quantity includes at least one of the first measurement quantity, the second measurement quantity, and the third measurement quantity.

[0279] In some embodiments, the first measurement includes at least one of a beam quality measurement, a timing advance (TA) measurement, a channel quality measurement, and a specific measurement determined based on the channel quality measurement.

[0280] In some embodiments, the second measurement includes at least one of a first beam quality, a first beam identifier, a predicted synchronization value, a predicted channel quality, a predicted channel processing value, a predicted time of arrival (TOA), and a predicted distance from the access network device to the user; wherein the first beam identifier is a predicted beam identifier, and the first beam quality is the strongest beam quality among a plurality of predicted beam qualities.

[0281] In some embodiments, the third measurement includes at least one of the following: a first accuracy rate, which is the proportion of a first target number of inferences used for monitoring the AI ​​unit, wherein the first target number of inferences is the number of times, in N1 inferences, the beam identifier corresponding to the predicted strongest beam quality is the same as the beam identifier corresponding to the measured strongest beam quality, and N1 is an integer greater than or equal to 1; and a second accuracy rate, which is the proportion of a second target number of inferences used for monitoring the AI ​​unit, wherein the second target number of inferences is the number of times, in N2 inferences, the beam identifiers corresponding to the measured top K strongest beam qualities belong to the beam identifiers corresponding to the predicted top K strongest beam qualities. The number of times, or the second target number of times, is the number of times the beam identifier corresponding to the predicted top K strongest beam qualities belongs to the measured top K strongest beam qualities, where K is an integer greater than 1 and N2 is an integer greater than or equal to 1; the difference between the first beam quality and the second beam quality, where the first beam quality is the strongest beam quality among the predicted multiple beam qualities, and the second beam quality is the measured beam quality corresponding to the first beam identifier, where the first beam identifier is a predicted beam identifier; the difference between the second beam quality and the third beam quality, where the third beam quality is the measured strongest beam quality; and the difference between the predicted downlink synchronization and the measured downlink synchronization. The difference between downlink synchronization in non-anchor cells and downlink synchronization in anchor cells; the number of times the preamble is sent during random access; and the access delay during random access.

[0282] In some embodiments, the first beam identifier is used for at least one of the following: determining the random access timing; determining the transmit beam corresponding to the random access message; and determining the receive beam corresponding to the random access message.

[0283] In some embodiments, determining the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit includes: determining the state of the AI ​​unit in the first RRC state as at least one of the available state and the active state when the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit satisfies a first condition; wherein the first condition includes at least one of the following: the measured value of the beam quality is greater than a first threshold; the measured value of the TA is less than a second threshold; the first beam quality is greater than a third threshold; the predicted synchronization value is less than a fourth threshold; the first accuracy is greater than a fifth threshold; the second accuracy is greater than a sixth threshold; the first... The difference between the beam quality and the second beam quality is less than the seventh threshold; the difference between the second beam quality and the third beam quality is less than the eighth threshold; the deviation between the predicted downlink synchronization and the measured downlink synchronization is less than the ninth threshold; the difference between the downlink synchronization of the non-anchor cell and the downlink synchronization of the anchor cell is less than the tenth threshold; the number of preamble transmissions during random access is less than the eleventh threshold; the access delay during random access is less than the twelfth threshold; the measured value of the channel quality is greater than the thirteenth threshold; the prediction error of the specific measurement value is less than the fourteenth threshold; the prediction error of the channel quality is less than the fifteenth threshold; the prediction error of the channel processing value is less than the sixteenth threshold; the prediction error of the TOA is less than the seventeenth threshold; and the prediction error of the distance from the access network equipment to the user is less than the eighteenth threshold.

[0284] In some embodiments, determining the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit includes: determining the state of the AI ​​unit in the first RRC state as at least one of the unavailable state and the deactivated state when the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit satisfies a second condition; wherein the second condition includes at least one of the following: the measured value of the beam quality is less than the nineteenth threshold; the measured value of the TA is greater than the twentieth threshold; the first beam quality is less than the twenty-first threshold; the predicted synchronization value is greater than the twenty-second threshold; the first accuracy is less than the twenty-third threshold; and the second accuracy is less than the twenty-fourth threshold; The following conditions are met: the difference between the quality of the first beam and the quality of the second beam is greater than the twenty-fifth threshold; the difference between the quality of the second beam and the quality of the third beam is greater than the twenty-sixth threshold; the deviation between the predicted downlink synchronization and the measured downlink synchronization is greater than the twenty-seventh threshold; the difference between the downlink synchronization of the non-anchor cell and the downlink synchronization of the anchor cell is greater than the twenty-eighth threshold; the number of preamble transmissions during random access is greater than the twenty-ninth threshold; the access delay during random access is greater than the thirtieth threshold; the measured value of the channel quality is less than the thirty-first threshold; the prediction error of the specific measurement value is greater than the thirty-second threshold; the prediction error of the channel quality is greater than the thirty-third threshold; the prediction error of the channel processing value is greater than the thirty-fourth threshold; the prediction error of the TOA is greater than the thirty-fifth threshold; and the prediction error of the distance from the access network equipment to the user is greater than the thirty-sixth threshold.

[0285] In some embodiments, the processing module 620 is further configured to perform at least one of the following: when it is determined that the state of the AI ​​unit in the first RRC state is at least one of the available state and the active state, the first operation includes activating the AI ​​unit or continuing to maintain the active state of the AI ​​unit; If the state of the AI ​​unit in the first RRC state is determined to be at least one of the unavailable state and the deactivated state, the second operation is performed, the second operation including at least one of deactivating the AI ​​unit, continuing to maintain the deactivated state of the AI ​​unit, falling back to a non-AI algorithm, and switching to another AI unit other than the AI ​​unit.

[0286] In some embodiments, obtaining the first information from the network-side device includes any one of the following: receiving a broadcast message from the network-side device, the broadcast message including the first information; receiving an RRC message from the network-side device, the RRC message including the first information, the RRC message including at least one of an RRC release message, an RRC recovery message, and an RRC reconfiguration message.

[0287] In some embodiments, where the first information includes the third indication information and the third indication information is used to indicate whether the AI ​​unit is enabled in the second RRC state, the RRC message includes an RRC release message; wherein the second RRC state includes at least one of the disconnected state, idle state, and inactive state in the first RRC state.

[0288] In some embodiments, the transmission module 610 is further configured to: send second information to a network-side device; wherein the second information is configured to request monitoring of the AI ​​unit in the first RRC state in the connected state.

[0289] In some embodiments, the second information includes at least one of the following: fourth indication information for indicating the AI ​​unit that the terminal monitors in the connected state; and fifth indication information for indicating the identifier of the AI ​​unit that the terminal monitors in the connected state.

[0290] The state determination device 600 for the AI ​​unit provided in this application embodiment can achieve... Figure 2 or Figure 3 The various processes implemented in the method embodiments achieve the same technical effect, and will not be described again here to avoid repetition.

[0291] See Figure 7 When the state determination device of the AI ​​unit is a network-side device or a component in a network-side device, the state determination device 700 of the AI ​​unit includes a processing module 710 for determining the first information; and a transmission module 720 for sending the first information to the terminal; wherein the first information is used to indicate monitoring-related information of the AI ​​unit in the first Radio Resource Control (RRC) state, wherein the first RRC state includes at least one of connected state, disconnected state, idle state, and inactive state.

[0292] In some embodiments, the first information includes at least one of the following: first indication information for indicating that the terminal is allowed to determine the state of the AI ​​unit; second indication information for indicating the monitoring-related thresholds of the AI ​​unit in the first RRC state; and third indication information for indicating whether the AI ​​unit is enabled in the first RRC state. The monitoring-related thresholds of the AI ​​unit include at least one of the following: a threshold for a first measurement quantity, wherein the first measurement quantity is a measured value and is related to the input of the AI ​​unit; a threshold for a second measurement quantity, wherein the second measurement quantity is related to the prediction result of the AI ​​unit; and a threshold for a third measurement quantity, wherein the third measurement quantity is related to the monitoring index of the AI ​​unit, and the monitoring index is determined based on a true value.

[0293] In some embodiments, the first measurement includes at least one of a beam quality measurement, a timing advance (TA) measurement, a channel quality measurement, and a specific measurement determined based on the channel quality measurement.

[0294] In some embodiments, the second measurement includes at least one of a first beam quality, a first beam identifier, a predicted synchronization value, a predicted channel quality, a predicted channel processing value, a predicted time of arrival (TOA), and a predicted distance from the access network device to the user; wherein the first beam identifier is a predicted beam identifier, and the first beam quality is the strongest beam quality among a plurality of predicted beam qualities.

[0295] In some embodiments, the third measurement includes at least one of the following: a first accuracy rate, wherein the first accuracy rate is the proportion of a first target number of inferences used for monitoring the AI ​​unit, and the first target number of inferences is the number of times, in N1 inferences, the beam identifier corresponding to the predicted strongest beam quality is the same as the beam identifier corresponding to the measured strongest beam quality, where N1 is an integer greater than or equal to 1; a second accuracy rate, wherein the second accuracy rate is the proportion of a second target number of inferences used for monitoring the AI ​​unit, and the second target number of inferences is the number of times, in N2 inferences, the beam identifier corresponding to the measured top K strongest beam qualities belongs to the beam identifier corresponding to the predicted top K strongest beam qualities, or, the second target number of inferences is the number of times, in N2 inferences, the beam identifier corresponding to the predicted top K strongest beam qualities is the same as the beam identifier corresponding to the measured top K strongest beam qualities. The beam identifier is the number of times the beam identifier corresponds to the K strongest measured beam qualities, where K is an integer greater than 1 and N2 is an integer greater than or equal to 1; the difference between the first beam quality and the second beam quality, where the first beam quality is the strongest among multiple predicted beam qualities and the second beam quality is the measured beam quality corresponding to the first beam identifier, where the first beam identifier is a predicted beam identifier; the difference between the second beam quality and the third beam quality, where the third beam quality is the strongest measured beam quality; the difference between the predicted downlink synchronization and the measured downlink synchronization; the difference between the downlink synchronization of a non-anchor cell and the downlink synchronization of an anchor cell; the number of times the preamble is sent during random access; and the access delay during random access.

[0296] In some embodiments, the first beam identifier is used for at least one of the following: determining the random access timing; determining the transmit beam corresponding to the random access message; and determining the receive beam corresponding to the random access message.

[0297] In some embodiments, sending the first information to the terminal includes any one of the following: sending a broadcast message, the broadcast message including the first information; sending an RRC message to the terminal, the RRC message including the first information, the RRC message including at least one of an RRC release message, an RRC recovery message, and an RRC reconfiguration message; the network-side device sending the first information to the terminal according to the terminal's request.

[0298] In some embodiments, where the first information includes the third indication information and the third indication information is used to indicate whether the AI ​​unit is enabled in the second RRC state, the RRC message includes an RRC release message; wherein the second RRC state includes at least one of the disconnected state, idle state, and inactive state in the first RRC state.

[0299] In some embodiments, the transmission module 720 is further configured to: receive second information from the terminal; wherein the second information is configured to request monitoring of the AI ​​unit in the first RRC state in the connected state.

[0300] In some embodiments, the second information includes at least one of the following: fourth indication information for indicating the AI ​​unit that the terminal monitors in the connected state; and fifth indication information for indicating the identifier of the AI ​​unit that the terminal monitors in the connected state.

[0301] The state determination device 700 for the AI ​​unit provided in this application embodiment can achieve... Figure 5 The various processes implemented in the method embodiments achieve the same technical effect, and will not be described again here to avoid repetition.

[0302] like Figure 8 As shown, this application embodiment also provides a communication device 800, including a processor 801 and a memory 802. The memory 802 stores programs or instructions that can run on the processor 801. For example, when the communication device 800 is a terminal, when the program or instructions are executed by the processor 801, they implement the various steps of the state determination method embodiment 200 or 300 of the AI ​​unit described above, and achieve the same technical effect. When the communication device 800 is a network-side device, when the program or instructions are executed by the processor 801, they implement the various steps of the state determination method embodiment 500 of the AI ​​unit described above, and achieve the same technical effect. To avoid repetition, further details are omitted here.

[0303] This application embodiment also provides a terminal, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement, for example... Figure 2 or Figure 3 The steps in the method embodiment shown are illustrated. This terminal embodiment corresponds to the above-described terminal-side method embodiment. All implementation processes and methods of the above-described method embodiments can be applied to this terminal embodiment and achieve the same technical effect. The terminal can be... Figure 6 The AI ​​unit's state determination device 600 is shown. Specifically, Figure 9 A schematic diagram of the hardware structure of a terminal to implement an embodiment of this application. The terminal 900 includes, but is not limited to, at least some of the following components: radio frequency unit 901, network module 902, audio output unit 903, input unit 904, sensor 905, display unit 906, user input unit 907, interface unit 908, memory 909, and processor 910.

[0304] Those skilled in the art will understand that the terminal 900 may also include a power supply (such as a battery) for supplying power to various components. The power supply may be logically connected to the processor 910 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. Figure 9 The terminal structure shown does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0305] It should be understood that, in this embodiment, the input unit 904 may include a graphics processor 9041 and a microphone 9042. The graphics processor 9041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 906 may include a display panel 9061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit 907 includes at least one of a touch panel 9071 and other input devices 9072. The touch panel 9071 is also called a touch screen. The touch panel 9071 may include two parts: a touch detection device and a touch controller. Other input devices 9072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.

[0306] In this embodiment, after receiving downlink data from the network-side device, the radio frequency unit 901 can transmit it to the processor 910 for processing; in addition, the radio frequency unit 901 can send uplink data to the network-side device. Typically, the radio frequency unit 901 includes, but is not limited to, antennas, amplifiers, transceivers, couplers, low-noise amplifiers, duplexers, etc.

[0307] The memory 909 can be used to store software programs or instructions, as well as various data. The memory 909 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 909 may include volatile memory or non-volatile memory. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 909 in the embodiments of this application includes, but is not limited to, these and any other suitable types of memory.

[0308] Processor 910 may include one or more processing units; optionally, processor 910 integrates an application processor and a modem processor, wherein the application processor mainly handles operations involving the operating system, user interface, and applications, and the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into processor 910.

[0309] The radio frequency unit 901 is used to acquire first information from the network-side device, wherein the first information is used to indicate monitoring-related information of the AI ​​unit in the first RRC state; the processor 910 is used to determine the state of the AI ​​unit in the first RRC state according to the first information; wherein the first RRC state includes at least one of connected state, disconnected state, idle state, and inactive state, and the state of the AI ​​unit includes at least one of available state, unavailable state, active state, and deactivated state.

[0310] In some embodiments, the first information includes at least one of the following: first indication information for indicating that the terminal is allowed to determine the state of the AI ​​unit; second indication information for indicating the monitoring-related thresholds of the AI ​​unit in the first RRC state; and third indication information for indicating whether the AI ​​unit is enabled in the first RRC state. The monitoring-related thresholds of the AI ​​unit include at least one of the following: a threshold for a first measurement quantity, wherein the first measurement quantity is a measured value and is related to the input of the AI ​​unit; a threshold for a second measurement quantity, wherein the second measurement quantity is related to the prediction result of the AI ​​unit; and a threshold for a third measurement quantity, wherein the third measurement quantity is related to the monitoring index of the AI ​​unit, and the monitoring index is determined based on a true value.

[0311] In some embodiments, determining the state of the AI ​​unit in the first RRC state based on the first information includes any one of the following: when the first information includes the first indication information and the second indication information, determining the state of the AI ​​unit in the first RRC state based on the monitoring-related threshold of the AI ​​unit indicated by the second indication information; when the first information includes the second indication information, determining the state of the AI ​​unit in the first RRC state based on the monitoring-related threshold of the AI ​​unit indicated by the second indication information; when the first information includes the third indication information and the third indication information is used to indicate that the AI ​​unit is enabled in the first RRC state, determining the state of the AI ​​unit in the first RRC state as at least one of the available state and the active state based on the third indication information; when the first information includes the third indication information and the third indication information is used to indicate that the AI ​​unit is not enabled in the first RRC state, determining the state of the AI ​​unit in the first RRC state as at least one of the unavailable state and the deactivated state based on the third indication information.

[0312] In some embodiments, determining the state of the AI ​​unit in the first RRC state according to the monitoring-related threshold of the AI ​​unit indicated by the second indication information includes: determining the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit; wherein the target measurement quantity includes at least one of the first measurement quantity, the second measurement quantity, and the third measurement quantity.

[0313] In some embodiments, the first measurement includes at least one of a beam quality measurement, a timing advance (TA) measurement, a channel quality measurement, and a specific measurement determined based on the channel quality measurement.

[0314] In some embodiments, the second measurement includes at least one of a first beam quality, a first beam identifier, a predicted synchronization value, a predicted channel quality, a predicted channel processing value, a predicted time of arrival (TOA), and a predicted distance from the access network device to the user; wherein the first beam identifier is a predicted beam identifier, and the first beam quality is the strongest beam quality among a plurality of predicted beam qualities.

[0315] In some embodiments, the third measurement includes at least one of the following: a first accuracy rate, wherein the first accuracy rate is the proportion of a first target number of inferences used for monitoring the AI ​​unit, and the first target number of inferences is the number of times, in N1 inferences, the beam identifier corresponding to the predicted strongest beam quality is the same as the beam identifier corresponding to the measured strongest beam quality, where N1 is an integer greater than or equal to 1; a second accuracy rate, wherein the second accuracy rate is the proportion of a second target number of inferences used for monitoring the AI ​​unit, and the second target number of inferences is the number of times, in N2 inferences, the beam identifier corresponding to the measured top K strongest beam qualities belongs to the beam identifier corresponding to the predicted top K strongest beam qualities, or, the second target number of inferences is the number of times, in N2 inferences, the beam identifier corresponding to the predicted top K strongest beam qualities is the same as the beam identifier corresponding to the measured top K strongest beam qualities. The beam identifier is the number of times the beam identifier corresponds to the K strongest measured beam qualities, where K is an integer greater than 1 and N2 is an integer greater than or equal to 1; the difference between the first beam quality and the second beam quality, where the first beam quality is the strongest among multiple predicted beam qualities and the second beam quality is the measured beam quality corresponding to the first beam identifier, where the first beam identifier is a predicted beam identifier; the difference between the second beam quality and the third beam quality, where the third beam quality is the strongest measured beam quality; the difference between the predicted downlink synchronization and the measured downlink synchronization; the difference between the downlink synchronization of a non-anchor cell and the downlink synchronization of an anchor cell; the number of times the preamble is sent during random access; and the access delay during random access.

[0316] In some embodiments, the first beam identifier is used for at least one of the following: determining the random access timing; determining the transmit beam corresponding to the random access message; and determining the receive beam corresponding to the random access message.

[0317] In some embodiments, determining the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit includes: determining the state of the AI ​​unit in the first RRC state as at least one of the available state and the active state when the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit satisfies a first condition; wherein the first condition includes at least one of the following: the measured value of the beam quality is greater than a first threshold; the measured value of the TA is less than a second threshold; the first beam quality is greater than a third threshold; the predicted synchronization value is less than a fourth threshold; the first accuracy is greater than a fifth threshold; the second accuracy is greater than a sixth threshold; the first... The difference between the beam quality and the second beam quality is less than the seventh threshold; the difference between the second beam quality and the third beam quality is less than the eighth threshold; the deviation between the predicted downlink synchronization and the measured downlink synchronization is less than the ninth threshold; the difference between the downlink synchronization of the non-anchor cell and the downlink synchronization of the anchor cell is less than the tenth threshold; the number of preamble transmissions during random access is less than the eleventh threshold; the access delay during random access is less than the twelfth threshold; the measured value of the channel quality is greater than the thirteenth threshold; the prediction error of the specific measurement value is less than the fourteenth threshold; the prediction error of the channel quality is less than the fifteenth threshold; the prediction error of the channel processing value is less than the sixteenth threshold; the prediction error of the TOA is less than the seventeenth threshold; and the prediction error of the distance from the access network equipment to the user is less than the eighteenth threshold.

[0318] In some embodiments, determining the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit includes: determining the state of the AI ​​unit in the first RRC state as at least one of the unavailable state and the deactivated state when the magnitude relationship between the target measurement quantity and the monitoring-related threshold of the AI ​​unit satisfies a second condition; wherein the second condition includes at least one of the following: the measured value of the beam quality is less than the nineteenth threshold; the measured value of the TA is greater than the twentieth threshold; the first beam quality is less than the twenty-first threshold; the predicted synchronization value is greater than the twenty-second threshold; the first accuracy is less than the twenty-third threshold; and the second accuracy is less than the twenty-fourth threshold; The following conditions are met: the difference between the quality of the first beam and the quality of the second beam is greater than the twenty-fifth threshold; the difference between the quality of the second beam and the quality of the third beam is greater than the twenty-sixth threshold; the deviation between the predicted downlink synchronization and the measured downlink synchronization is greater than the twenty-seventh threshold; the difference between the downlink synchronization of the non-anchor cell and the downlink synchronization of the anchor cell is greater than the twenty-eighth threshold; the number of preamble transmissions during random access is greater than the twenty-ninth threshold; the access delay during random access is greater than the thirtieth threshold; the measured value of the channel quality is less than the thirty-first threshold; the prediction error of the specific measurement value is greater than the thirty-second threshold; the prediction error of the channel quality is greater than the thirty-third threshold; the prediction error of the channel processing value is greater than the thirty-fourth threshold; the prediction error of the TOA is greater than the thirty-fifth threshold; and the prediction error of the distance from the access network equipment to the user is greater than the thirty-sixth threshold.

[0319] In some embodiments, the processor 910 is further configured to: perform a first operation when it is determined that the state of the AI ​​unit in the first RRC state is at least one of the available state and the active state, the first operation including activating the AI ​​unit or continuing to maintain the active state of the AI ​​unit; If the state of the AI ​​unit in the first RRC state is determined to be at least one of the unavailable state and the deactivated state, the second operation is performed, the second operation including at least one of deactivating the AI ​​unit, continuing to maintain the deactivated state of the AI ​​unit, falling back to a non-AI algorithm, and switching to another AI unit other than the AI ​​unit.

[0320] In some embodiments, obtaining the first information from the network-side device includes any one of the following: receiving a broadcast message from the network-side device, the broadcast message including the first information; receiving an RRC message from the network-side device, the RRC message including the first information, the RRC message including at least one of an RRC release message, an RRC recovery message, and an RRC reconfiguration message.

[0321] In some embodiments, where the first information includes the third indication information and the third indication information is used to indicate whether the AI ​​unit is enabled in the second RRC state, the RRC message includes an RRC release message; wherein the second RRC state includes at least one of the disconnected state, idle state, and inactive state in the first RRC state.

[0322] In some embodiments, the radio frequency unit 901 is further configured to: send second information to a network-side device; wherein the second information is configured to request monitoring of the AI ​​unit in the first RRC state during the connected state.

[0323] In some embodiments, the second information includes at least one of the following: fourth indication information for indicating the AI ​​unit that the terminal monitors in the connected state; and fifth indication information for indicating the identifier of the AI ​​unit that the terminal monitors in the connected state.

[0324] It is understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description of method embodiment 200 or 300 and achieve the same or corresponding technical effects. To avoid repetition, it will not be described again here.

[0325] This application embodiment also provides a network-side device, 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, for example... Figure 5 The steps of the method embodiment shown are illustrated. 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.

[0326] Specifically, embodiments of this application also provide a network-side device, which can be... Figure 7 The AI ​​unit's state determination device 700 is shown. For example... Figure 10As shown, the network-side device 1000 includes: an antenna 1001, a radio frequency (RF) device 1002, a baseband device 1003, a processor 1004, and a memory 1005. The antenna 1001 is connected to the RF device 1002. In the uplink direction, the RF device 1002 receives information through the antenna 1001 and transmits the received information to the baseband device 1003 for processing. In the downlink direction, the baseband device 1003 processes the information to be transmitted and sends it to the RF device 1002. The RF device 1002 processes the received information and transmits it through the antenna 1001.

[0327] The method executed by the network-side device 1000 in the above embodiments can be implemented in the baseband device 1003, which includes a baseband processor.

[0328] The baseband device 1003 may include, for example, at least one baseband board on which multiple chips are disposed, such as... Figure 10 As shown, one of the chips is, for example, a baseband processor, which is connected to the memory 1005 via a bus interface to call the program in the memory 1005 and execute the network-side device operations shown in the above method embodiment.

[0329] The network-side device 1000 may also include a network interface 1006, such as a Common Public Radio Interface (CPRI).

[0330] Specifically, the network-side device 1000 in this application embodiment further includes: instructions or programs stored in memory 1005 and executable on processor 1004, wherein processor 1004 calls the instructions or programs in memory 1005 to execute. Figure 7 The methods executed by each module shown achieve the same technical effect, and to avoid repetition, they will not be described in detail here.

[0331] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described AI unit state determination method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0332] The processor mentioned above is the processor in the terminal described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk. In some examples, the readable storage medium may be a non-transient readable storage medium.

[0333] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described AI unit state determination method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0334] 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.

[0335] This application also provides a computer program / program product, which is stored in a storage medium and executed by at least one processor to implement the various processes of the above-described AI unit state determination method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0336] This application also provides a wireless communication system, including a terminal and a network-side device. The terminal can be used to implement various processes of the above-mentioned AI unit state determination method embodiment 200 or 300, and the network-side device can be used to implement various processes of the above-mentioned AI unit state determination method embodiment 500, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0337] 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.

[0338] From the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of computer software products plus necessary general-purpose hardware platforms, and of course, they can also be implemented by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes several instructions to cause the terminal or network-side device to execute the methods described in the various embodiments of this application.

[0339] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other implementations under the guidance of this application without departing from the spirit and scope of the claims. All of these implementations are within the protection scope of this application.

Claims

1. A method for determining the state of an artificial intelligence (AI) unit, characterized in that, include: The terminal obtains first information from the network-side device, wherein the first information is used to indicate monitoring-related information of the AI ​​unit in the first Radio Resource Control (RRC) state; The terminal determines the state of the AI ​​unit in the first RRC state based on the first information; The first RRC state includes at least one of connected state, disconnected state, idle state, and inactive state, and the state of the AI ​​unit includes at least one of available state, unavailable state, active state, and deactivated state.

2. The method as described in claim 1, characterized in that, The first information includes at least one of the following: The first instruction information is used to indicate that the terminal is allowed to determine the state of the AI ​​unit; The second indication information is used to indicate the monitoring-related thresholds of the AI ​​unit in the first RRC state; The third indication information is used to indicate whether the AI ​​unit is enabled in the first RRC state; The monitoring-related thresholds of the AI ​​unit include at least one of the following: A threshold for a first measurement, wherein the first measurement is a measured value and is related to the input of the AI ​​unit; A threshold for a second measurement, which is related to the prediction result of the AI ​​unit; The threshold of the third measurement quantity is related to the monitoring index of the AI ​​unit, and the monitoring index is determined based on the true value.

3. The method as described in claim 2, characterized in that, Determining the state of the AI ​​unit in the first RRC state based on the first information includes any one of the following: When the first information includes the first indication information and the second indication information, the state of the AI ​​unit in the first RRC state is determined according to the monitoring-related threshold of the AI ​​unit indicated by the second indication information; If the first information includes the second indication information, the state of the AI ​​unit in the first RRC state is determined according to the monitoring-related threshold of the AI ​​unit indicated by the second indication information; When the first information includes the third indication information, and the third indication information is used to indicate that the AI ​​unit is enabled in the first RRC state, the state of the AI ​​unit in the first RRC state is determined to be at least one of the available state and the active state according to the third indication information; When the first information includes the third indication information, and the third indication information is used to indicate that the AI ​​unit is not enabled in the first RRC state, the state of the AI ​​unit in the first RRC state is determined to be at least one of the unavailable state and the deactivated state according to the third indication information.

4. The method as described in claim 3, characterized in that, Determining the state of the AI ​​unit in the first RRC state based on the monitoring-related thresholds of the AI ​​unit indicated by the second indication information includes: Based on the relationship between the target measurement and the monitoring-related threshold of the AI ​​unit, the state of the AI ​​unit in the first RRC state is determined to be at least one of the available state, unavailable state, active state, and deactivated state; The target measurement quantity includes at least one of the first measurement quantity, the second measurement quantity, and the third measurement quantity.

5. The method as described in claim 4, characterized in that, The first measurement includes at least one of the following: beam quality measurement, timing advance (TA) measurement, channel quality measurement, and a specific measurement, wherein the specific measurement is determined based on the channel quality measurement.

6. The method as described in claim 4, characterized in that, The second measurement includes at least one of the following: first beam quality, first beam identifier, predicted synchronization value, predicted channel quality, predicted channel processing value, predicted time of arrival (TOA), and predicted distance from the access network device to the user. Wherein, the first beam identifier is a predicted beam identifier, and the first beam quality is the strongest beam quality among multiple predicted beam qualities.

7. The method as described in claim 4, characterized in that, The third measurement quantity includes at least one of the following: The first accuracy rate is the proportion of the first target number of inferences in the total number of inferences used for monitoring the AI ​​unit. The first target number of inferences is the number of times that the beam identifier corresponding to the strongest beam quality predicted in N1 inferences is the same as the beam identifier corresponding to the strongest beam quality measured in N1 inferences. N1 is an integer greater than or equal to 1. The second accuracy rate is the proportion of the second target number of inferences used for monitoring the AI ​​unit. The second target number of inferences is the number of times in N2 inferences that the beam identifiers corresponding to the measured top K strongest beam qualities belong to the beam identifiers corresponding to the predicted top K strongest beam qualities. Alternatively, the second target number of inferences is the number of times the beam identifiers corresponding to the predicted top K strongest beam qualities belong to the beam identifiers corresponding to the measured top K strongest beam qualities. Here, K is an integer greater than 1, and N2 is an integer greater than or equal to 1. The difference between a first beam quality and a second beam quality, wherein the first beam quality is the strongest beam quality among a plurality of predicted beam qualities, and the second beam quality is the measured beam quality corresponding to the first beam identifier, wherein the first beam identifier is a predicted beam identifier; The difference between the second beam quality and the third beam quality, wherein the third beam quality is the measured strongest beam quality; The difference between the predicted downlink synchronization and the measured downlink synchronization; The difference between downlink synchronization in non-anchor cells and downlink synchronization in anchor cells; The number of times the preamble is sent during random access; Access latency during random access.

8. The method as described in claim 6 or 7, characterized in that, The first beam identifier is used for at least one of the following: Determine the timing of random access; Determine the transmission beam corresponding to the random access message; Determine the receiving beam corresponding to the random access message.

9. The method according to any one of claims 4-8, characterized in that, The step of determining the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the relationship between the target measurement and the monitoring-related threshold of the AI ​​unit includes: If the magnitude relationship between the target measurement and the monitoring-related threshold of the AI ​​unit satisfies a first condition, the state of the AI ​​unit in the first RRC state is determined to be at least one of the available state and the active state; The first condition includes at least one of the following: The measured value of the beam quality is greater than the first threshold; The measured value of TA is less than the second threshold; The quality of the first beam is greater than the third threshold; The predicted synchronization value is less than the fourth threshold; The first accuracy rate is greater than the fifth threshold; The second accuracy rate is greater than the sixth threshold; The difference between the quality of the first beam and the quality of the second beam is less than the seventh threshold; The difference between the second beam quality and the third beam quality is less than the eighth threshold; The deviation between the predicted downlink synchronization and the measured downlink synchronization is less than the ninth threshold; The difference between the downlink synchronization of the non-anchor cell and the downlink synchronization of the anchor cell is less than the tenth threshold. The number of times the preamble is sent during random access is less than the eleventh threshold; The access delay during random access is less than the twelfth threshold; The measured value of the channel quality is greater than the thirteenth threshold; The prediction error for the specific measurement value is less than the fourteenth threshold; The prediction error of the channel quality is less than the fifteenth threshold; The prediction error of the channel processing value is less than the sixteenth threshold; The prediction error of the TOA is less than the seventeenth threshold; The prediction error of the distance from the access network device to the user is less than the eighteenth threshold.

10. The method according to any one of claims 2-9, characterized in that, The step of determining the state of the AI ​​unit in the first RRC state as at least one of the available state, unavailable state, active state, and deactivated state based on the relationship between the target measurement and the monitoring-related threshold of the AI ​​unit includes: If the relationship between the target measurement and the monitoring-related threshold of the AI ​​unit satisfies the second condition, the state of the AI ​​unit in the first RRC state is determined to be at least one of the unavailable state and the deactivated state; The second condition includes at least one of the following: The measured value of the beam quality is less than the nineteenth threshold; The measured value of TA is greater than the twentieth threshold; The quality of the first beam is less than the twenty-first threshold; The predicted synchronization value is greater than the twenty-second threshold; The first accuracy rate is less than the twenty-third threshold; The second accuracy rate is less than the twenty-fourth threshold; The difference between the quality of the first beam and the quality of the second beam is greater than the twenty-fifth threshold; The difference between the second beam quality and the third beam quality is greater than the twenty-sixth threshold; The deviation between the predicted downlink synchronization and the measured downlink synchronization is greater than the 27th threshold; The difference between the downlink synchronization of the non-anchor cell and the downlink synchronization of the anchor cell is greater than the 28th threshold; During random access, the number of times the preamble is sent exceeds the 29th threshold; The access latency during random access exceeds the 30th threshold. The measured value of the channel quality is less than the thirty-first threshold; The prediction error for the specific measurement value is greater than the thirty-second threshold; The prediction error for the channel quality is greater than the thirty-three threshold; The prediction error of the channel processing value is greater than the thirty-fourth threshold; The prediction error of the TOA is greater than the thirty-fifth threshold; The prediction error of the distance from the access network device to the user is greater than the thirty-sixth threshold.

11. The method according to any one of claims 1-10, characterized in that, The method further includes at least one of the following: If the state of the AI ​​unit in the first RRC state is determined to be at least one of the available state and the active state, the terminal performs a first operation, the first operation including activating the AI ​​unit or continuing to maintain the active state of the AI ​​unit; If the terminal determines that the state of the AI ​​unit in the first RRC state is at least one of the unavailable state and the deactivated state, the terminal performs a second operation, the second operation including at least one of deactivating the AI ​​unit, continuing to maintain the deactivated state of the AI ​​unit, falling back to a non-AI algorithm, and switching to another AI unit other than the AI ​​unit.

12. The method according to any one of claims 1-11, characterized in that, The terminal obtains first information from the network-side device, including any one of the following: The terminal receives a broadcast message from the network-side device, the broadcast message including the first information; The terminal receives an RRC message from the network-side device. The RRC message includes the first information and includes at least one of an RRC release message, an RRC recovery message, and an RRC reconfiguration message.

13. The method as described in claim 12, characterized in that, When the first information includes the third indication information, and the third indication information is used to indicate whether the AI ​​unit is enabled in the second RRC state, the RRC message includes an RRC release message; The second RRC state includes at least one of the following in the first RRC state: disconnected state, idle state, and inactive state.

14. The method according to any one of claims 1-13, characterized in that, The method further includes: The terminal sends the second information to the network-side device; The second information is used to request monitoring of the AI ​​unit in the first RRC state during the connected state.

15. The method as described in claim 14, characterized in that, The second information includes at least one of the following: The fourth instruction information is used to instruct the AI ​​unit that monitors the terminal in the connected state; The fifth indication information is used to indicate the identifier of the AI ​​unit that the terminal monitors in the connected state.

16. A method for determining the state of an artificial intelligence (AI) unit, characterized in that, include: The network-side device sends the first information to the terminal; The first information is used to indicate monitoring-related information of the AI ​​unit in the first Radio Resource Control (RRC) state, wherein the first RRC state includes at least one of connected state, disconnected state, idle state, and inactive state.

17. The method as described in claim 16, characterized in that, The first information includes at least one of the following: The first instruction information is used to indicate that the terminal is allowed to determine the state of the AI ​​unit; The second indication information is used to indicate the monitoring-related thresholds of the AI ​​unit in the first RRC state; The third indication information is used to indicate whether the AI ​​unit is enabled in the first RRC state; The monitoring-related thresholds of the AI ​​unit include at least one of the following: A threshold for a first measurement, wherein the first measurement is a measured value and is related to the input of the AI ​​unit; A threshold for a second measurement, which is related to the prediction result of the AI ​​unit; The threshold of the third measurement quantity is related to the monitoring index of the AI ​​unit, and the monitoring index is determined based on the true value.

18. The method as described in claim 17, characterized in that, The first measurement includes at least one of the following: beam quality measurement, timing advance (TA) measurement, channel quality measurement, and a specific measurement, wherein the specific measurement is determined based on the channel quality measurement.

19. The method as described in claim 17, characterized in that, The second measurement includes at least one of the following: first beam quality, first beam identifier, predicted synchronization value, predicted channel quality, predicted channel processing value, predicted time of arrival (TOA), and predicted distance from the access network device to the user. Wherein, the first beam identifier is a predicted beam identifier, and the first beam quality is the strongest beam quality among multiple predicted beam qualities.

20. The method as described in claim 17, characterized in that, The third measurement quantity includes at least one of the following: The first accuracy rate is the proportion of the first target number of inferences in the total number of inferences used for monitoring the AI ​​unit. The first target number of inferences is the number of times that the beam identifier corresponding to the strongest beam quality predicted in N1 inferences is the same as the beam identifier corresponding to the strongest beam quality measured in N1 inferences. N1 is an integer greater than or equal to 1. The second accuracy rate is the proportion of the second target number of inferences used for monitoring the AI ​​unit. The second target number of inferences is the number of times in N2 inferences that the beam identifiers corresponding to the measured top K strongest beam qualities belong to the beam identifiers corresponding to the predicted top K strongest beam qualities. Alternatively, the second target number of inferences is the number of times the beam identifiers corresponding to the predicted top K strongest beam qualities belong to the beam identifiers corresponding to the measured top K strongest beam qualities. Here, K is an integer greater than 1, and N2 is an integer greater than or equal to 1. The difference between a first beam quality and a second beam quality, wherein the first beam quality is the strongest beam quality among a plurality of predicted beam qualities, and the second beam quality is the measured beam quality corresponding to the first beam identifier, wherein the first beam identifier is a predicted beam identifier; The difference between the second beam quality and the third beam quality, wherein the third beam quality is the measured strongest beam quality; The difference between the predicted downlink synchronization and the measured downlink synchronization; The difference between downlink synchronization in non-anchor cells and downlink synchronization in anchor cells; The number of times the preamble is sent during random access; Access latency during random access.

21. The method as described in claim 19 or 20, characterized in that, The first beam identifier is used for at least one of the following: Determine the timing of random access; Determine the transmission beam corresponding to the random access message; Determine the receiving beam corresponding to the random access message.

22. The method according to any one of claims 16-21, characterized in that, The network-side device sends first information to the terminal, including any one of the following: The network-side device sends a broadcast message, the broadcast message including the first information; The network-side device sends an RRC message to the terminal. The RRC message includes the first information and includes at least one of an RRC release message, an RRC recovery message, and an RRC reconfiguration message. The network-side device sends the first information to the terminal according to the terminal's request.

23. The method as described in claim 22, characterized in that, When the first information includes the third indication information, and the third indication information is used to indicate whether the AI ​​unit is enabled in the second RRC state, the RRC message includes an RRC release message; The second RRC state includes at least one of the following in the first RRC state: disconnected state, idle state, and inactive state.

24. The method according to any one of claims 16-23, characterized in that, The method further includes: The network-side device receives second information from the terminal; The second information is used to request monitoring of the AI ​​unit in the first RRC state during the connected state.

25. The method as described in claim 24, characterized in that, The second information includes at least one of the following: The fourth instruction information is used to instruct the AI ​​unit that monitors the terminal in the connected state; The fifth indication information is used to indicate the identifier of the AI ​​unit that the terminal monitors in the connected state.

26. A state determination device for an artificial intelligence (AI) unit, characterized in that, include: The transmission module is used to acquire first information from the network-side device, wherein the first information is used to indicate monitoring-related information of the AI ​​unit in the first Radio Resource Control (RRC) state; The processing module is configured to determine the state of the AI ​​unit in the first RRC state based on the first information; The first RRC state includes at least one of connected state, disconnected state, idle state, and inactive state, and the state of the AI ​​unit includes at least one of available state, unavailable state, active state, and deactivated state.

27. The apparatus as claimed in claim 26, characterized in that, The first information includes at least one of the following: The first instruction information is used to indicate that the terminal is allowed to determine the state of the AI ​​unit; The second indication information is used to indicate the monitoring-related thresholds of the AI ​​unit in the first RRC state; The third indication information is used to indicate whether the AI ​​unit is enabled in the first RRC state; The monitoring-related thresholds of the AI ​​unit include at least one of the following: A threshold for a first measurement, wherein the first measurement is a measured value and is related to the input of the AI ​​unit; A threshold for a second measurement, which is related to the prediction result of the AI ​​unit; The threshold of the third measurement quantity is related to the monitoring index of the AI ​​unit, and the monitoring index is determined based on the true value.

28. The apparatus as claimed in claim 27, characterized in that, Determining the state of the AI ​​unit in the first RRC state based on the first information includes any one of the following: When the first information includes the first indication information and the second indication information, the state of the AI ​​unit in the first RRC state is determined according to the monitoring-related threshold of the AI ​​unit indicated by the second indication information; If the first information includes the second indication information, the state of the AI ​​unit in the first RRC state is determined according to the monitoring-related threshold of the AI ​​unit indicated by the second indication information; When the first information includes the third indication information, and the third indication information is used to indicate that the AI ​​unit is enabled in the first RRC state, the state of the AI ​​unit in the first RRC state is determined to be at least one of the available state and the active state according to the third indication information; When the first information includes the third indication information, and the third indication information is used to indicate that the AI ​​unit is not enabled in the first RRC state, the state of the AI ​​unit in the first RRC state is determined to be at least one of the unavailable state and the deactivated state according to the third indication information.

29. A state determination device for an artificial intelligence (AI) unit, characterized in that, include: The transmission module is used to send the first information to the terminal; The first information is used to indicate monitoring-related information of the AI ​​unit in the first Radio Resource Control (RRC) state, wherein the first RRC state includes at least one of connected state, disconnected state, idle state, and inactive state.

30. The apparatus as claimed in claim 29, characterized in that, The first information includes at least one of the following: The first instruction information is used to indicate that the terminal is allowed to determine the state of the AI ​​unit; The second indication information is used to indicate the monitoring-related thresholds of the AI ​​unit in the first RRC state; The third indication information is used to indicate whether the AI ​​unit is enabled in the first RRC state; The monitoring-related thresholds of the AI ​​unit include at least one of the following: A threshold for a first measurement, wherein the first measurement is a measured value and is related to the input of the AI ​​unit; A threshold for a second measurement, which is related to the prediction result of the AI ​​unit; The threshold of the third measurement quantity is related to the monitoring index of the AI ​​unit, and the monitoring index is determined based on the true value.

31. A terminal, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the method as described in any one of claims 1 to 15.

32. A network-side device, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the method as described in any one of claims 16 to 25.

33. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the method as claimed in any one of claims 1 to 15, or the steps of the method as claimed in any one of claims 16 to 25.

Images