Measurement accuracy requirements for testability AI / ML beam prediction

Abstract

An apparatus for communication, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: determine a beam management mode based on a configuration indicative of the beam management mode; switching to the beam management mode; performing at least one measurement related to artificial intelligence or machine learning beam management while operating in the beam management mode; and transmitting the at least one measurement value to a test device or a network. Related methods and the like are also provided.

Description

Technical Field

[0001] Examples and non-limiting example embodiments relate generally to communications, and more particularly to measurement accuracy requirements for AI / ML (artificial intelligence / machine learning) beam prediction with testability. Background Art

[0002] It is well known that devices use transmit and receive beams to communicate in a communication network. Summary of the invention

[0003] According to the present invention, there is provided a device for communication, comprising:

[0004] at least one processor; and

[0005] at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least:

[0006] determining the beam management mode based on a configuration indicating the beam management mode;

[0007] Switching to the beam management mode;

[0008] performing at least one measurement related to artificial intelligence or machine learning beam management while operating in the beam management mode; and

[0009] The at least one measurement value is transmitted to a test device or a network.

[0010] Preferably, the mode includes one of the following:

[0011] Artificial intelligence or machine learning models, or

[0012] Traditional mode.

[0013] Preferably, the mode comprises artificial intelligence or machine learning operations for at least one of:

[0014] The spatial domain beam identifier prediction for the top beam, or

[0015] Spatial domain beam identifier prediction for multiple beams, or

[0016] spatial domain layer 1 reference signal received power prediction, or

[0017] The time domain beam identifier prediction for the top beam, or

[0018] Time domain beam identifier prediction for multiple beams, or

[0019] Time domain layer 1 reference signal received power prediction, or

[0020] Downlink transmit receive beam pair prediction.

[0021] Preferably, when the instructions are executed by the at least one processor, the device is caused to at least:

[0022] Receiving a command from the test equipment or the network for the apparatus to switch to an artificial intelligence or machine learning beam management mode, wherein the beam management mode comprises the artificial intelligence or machine learning beam management mode.

[0023] Preferably, when the instructions are executed by the at least one processor, the device is caused to at least:

[0024] Detect changes in radio conditions in the environment; and

[0025] The beam management mode is determined based on changes in the radio condition environment.

[0026] Preferably, when the instructions are executed by the at least one processor, the device is caused to at least:

[0027] When the change in the radio condition environment includes performance improvement, determining the beam management mode as an artificial intelligence or machine learning mode.

[0028] Preferably, when the instructions are executed by the at least one processor, the device is caused to at least:

[0029] When the change in the radio condition environment includes performance degradation, the beam management mode is determined to be a conventional mode.

[0030] Preferably, the at least one measurement is used as input for beam prediction using artificial intelligence or machine learning.

[0031] Preferably, the at least one measurement value comprises at least one of the following:

[0032] The layer 1 reference signal received power measurement, or

[0033] one or more locations of a user's device, or

[0034] One or more transmit beam angles, or

[0035] Non-line-of-sight (nLoS) information, or

[0036] Non-line-of-sight (nLoS) indication, or

[0037] Line of sight (LoS) information, or

[0038] Line of sight (LoS) indication.

[0039] Preferably, when the error associated with the at least one measurement is within the error range of the artificial intelligence or machine learning beam management mode, or within the layer 1 reference signal received power (L1-RSRP) measurement error range, the measurement accuracy requirement of the artificial intelligence or machine learning mode for beam management is met.

[0040] Preferably, the measurement accuracy requirement of the conventional mode for beam management is met when the error associated with the at least one measurement is within the conventional mode error range, or within the layer 1 reference signal received power (L1-RSRP) error range.

[0041] Preferably, when the instructions are executed by the at least one processor, the device is caused to at least:

[0042] An indication is transmitted to a test device or a network that the apparatus is operating in the beam management mode.

[0043] Preferably, the at least one measured value meets or does not meet a measurement accuracy requirement.

[0044] Preferably, the measurement accuracy requirement comprises a value.

[0045] Preferably:

[0046] satisfying the measurement accuracy requirement of an artificial intelligence or machine learning mode for beam management when the at least one measured value is within a first tolerance value of the value; and

[0047] The measurement accuracy requirement for a conventional mode of beam management is met when the at least one measured value is within a second tolerance value of the value.

[0048] Preferably, the first tolerance value is smaller than the second tolerance value.

[0049] Preferably, the first tolerance value is greater than the second tolerance value.

[0050] Preferably, when the instructions are executed by the at least one processor, the device is caused to at least:

[0051] An artificial intelligence or machine learning model for beam management is trained based on the at least one measurement.

[0052] Preferably, when the instructions are executed by the at least one processor, the device is caused to at least:

[0053] An indication is received from the test device or the network that the at least one measurement value satisfies a measurement accuracy requirement.

[0054] Preferably, when the instructions are executed by the at least one processor, the device is caused to at least:

[0055] Based on whether the at least one measurement value meets the measurement accuracy requirement, an instruction to switch to the beam management mode is received from the test device or the network.

[0056] According to the present invention, there is provided a device for communication, comprising:

[0057] at least one processor; and

[0058] at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least:

[0059] determining a configuration indicating a beam management mode;

[0060] receiving, from a user equipment, at least one measurement value related to artificial intelligence or machine learning beam management based on the beam management mode;

[0061] selecting an accuracy target for the beam management mode; and

[0062] A determination is made based on the at least one measurement whether the accuracy target is met.

[0063] According to the present invention, there is provided a method for communication, comprising:

[0064] determining the beam management mode based on a configuration indicating the beam management mode;

[0065] Switching to the beam management mode;

[0066] performing at least one measurement related to artificial intelligence or machine learning beam management while operating in the beam management mode; and

[0067] The at least one measurement value is transmitted to a test device or a network.

[0068] According to the present invention, there is provided a method for communication, comprising:

[0069] determining a configuration indicating a beam management mode;

[0070] receiving, from a user equipment, at least one measurement value related to artificial intelligence or machine learning beam management based on the beam management mode;

[0071] selecting an accuracy target for the beam management mode; and

[0072] A determination is made based on the at least one measurement whether the accuracy target is met.

[0073] According to the present invention, a computer program product is provided, comprising instructions for causing a device to execute any one of the above methods.

[0074] According to the present invention, a computer-readable medium is provided, comprising instructions for causing a device to execute any one of the above methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In the following description, the above-mentioned aspects and other features are explained in conjunction with the accompanying drawings.

[0076] Figure 1 is a flow chart of the solution described in this article.

[0077] Figure 2 The signaling for measurement requirements between TE and UE in command mode is shown.

[0078] Figure 3 Shows the signaling used for the measurement requirement test between the TE and the UE when they are in simulation mode.

[0079] Figure 4 is a block diagram of one possible and non-limiting system in which example embodiments may be practiced.

[0080] Figure 5 is an example device configured to implement the examples described herein.

[0081] Figure 6 A representation of an example of a non-volatile storage medium for storing instructions to implement the examples described herein is shown.

[0082] Figure 7 is an example method based on the examples described herein.

[0083] Figure 8 is an example method based on the examples described herein. DETAILED DESCRIPTION

[0084] To support new AI / ML-enabled radio interfaces for next-generation cellular systems, 3GPP is currently studying a new study item (RP-213599) for Release 18. AI / ML-based beam management targets spatial and / or temporal beam prediction to reduce overhead and latency.

[0085] The study aims to identify areas where AI / ML can improve the performance of air interface functions. The regulatory impact will be evaluated to improve the overall understanding of what is required to implement AI / ML technologies for air interface. Beam management use cases for spatial and / or temporal beam prediction are further studied. The scope of spatial beam prediction (BM-Case1) is to predict the best Tx / Rx beams for different spatial locations. In contrast, time domain beam prediction (BM-Case2) aims to predict the most likely beam to be used for the next batch of moments, such as used for beam prediction in the spatial domain (BM-Case 1).

[0086] Therefore, the examples described herein relate to a 3GPP study item for AI ML technologies in 3GPP (Study on Artificial Intelligence (AI) / Machine Learning (ML) for NR Air Interface RP-213599), which aims to study a 3GPP framework for AI / ML over the air interface corresponding to each target use case, with respect to aspects such as performance, complexity, and potential regulatory impact. One use case to focus on is beam management, e.g., beam prediction in the time and / or space domain to reduce overhead and latency, and improve beam selection accuracy [RAN1].

[0087] There is no testing mechanism for measurement data for AI / ML based beam management use cases that have impact on specification requirements (such as 3GPP) and have an impact on UE and TE interfaces.

[0088] 3GPP is currently studying NW-side and / or UE-side model training and model inference for AI / ML beam management. One aspect is data collection for model training / inference. For AI / ML beam management, the measurement accuracy of measured quantities such as L1-RSRP needs to be verified.

[0089] Thus, the examples described herein relate to the possibility of enhancing performance requirements, tests, use cases, procedures, functions, and measurements using AI / ML-based methods. In addition, the examples described herein relate to improving the data quality of data collection for network-side AI / ML models, as well as improving the accuracy specifications for training data sets and training data generation.

[0090] Currently, L1-RSRP accuracy requirements are specified in 38.133 [3GPP TS 38.133 L1-RSRP measurement requirements] for beam management without AI / ML. For AI / ML-based beam prediction, the accuracy requirements may vary based on the performance required by the model, for example, the current measurement accuracy requirements for legacy mode may be relaxed, with an L1-RSRP error range of [-6dB, +6dB].

[0091] Although the NW has configured the input parameters to the UE, there may be further limitations associated with the provided ML model and input parameters. Erroneous results or high error inputs or invalid input parameters may lead to inaccurate or invalid predictions.

[0092] What is designed and described in this paper is a test framework including some impacts on the interface between UE and test equipment (TE) / NW to verify and validate L1-RSRP measurement requirements in situations where the UE needs to comply with different accuracy requirements in different functional modes.

[0093] The proposed framework consists of the following steps:

[0094] Step 1: The goal of this step is to ensure that the functional modes on the UE side are well known on the TE side. This can be achieved by issuing a command from the TE to the UE to operate in a specific mode. For example, the TE sends a command to the UE to enable an AI / ML based beam management use case. The UE can then confirm the activation of the requested mode. The TE can also enforce a certain functional mode to the UE by simulating certain radio conditions. The changed radio conditions in a controlled environment force the performance degradation / improvement of the activated functional mode. For example, in case of performance degradation, the UE is forced to enter the legacy mode (without AI / ML) through the Lifecycle Management (LCM) process of the AI / ML function.

[0095] Step 2: At this step, after the requested mode is activated, the UE starts measurements to ensure the required accuracy for the used mode.

[0096] Step 3: At this step, the UE reports the measurement values ​​to the TE / NW.

[0097] Step 4: The TE / NW already knows the mode used at the UE side, either through a command sent to the UE or by simulating radio conditions suitable for a specific mode, and selects the corresponding accuracy target for the configured mode.

[0098] Step 5: Finally, TE verifies whether the accuracy target for the configured mode has been achieved. If the accuracy target for both legacy and AI / ML modes is verified for the corresponding configuration, the measurement accuracy requirement for L1-RSRP is verified.

[0099] Figure 1A flow chart of the method 100 described herein for implementing the solution is shown. At 110, the method starts. At 120, the UE 10 is configured (e.g., by the NW or TE (where the NW or TE corresponds to the gNB 70 of item 202 or the core network 90 having one or more network functions 99) to have an AI / ML mode or a conventional mode for beam management. At 130, the UE starts measurements to ensure the required accuracy for the mode used. At 140, the TE selects a target accuracy for verifying the selected mode. At 150, the TE verifies whether the accuracy target for the selected mode is achieved. At 160, it is determined whether the error is less than the target error. If it is determined at 160 that the error is less than the target error (e.g., "yes"), the method goes to 170. If it is determined at 160 that the error is greater than or equal to the target error (e.g., "no"), the method goes to 180. At 170, the L1 RSRP measurement accuracy is verified. At 180, the L1 RSRP measurement accuracy is not verified.

[0100] Therefore, this document describes a testing framework for measurement accuracy requirements of AI / ML-based functions for beam management use cases, such as AI / ML beam management for Top-K (K is a number, such as an integer) beam prediction in spatial / time domains, AI / ML beam management for L1-RSRP prediction in spatial domains, and AI / ML beam management for Tx-Rx beam pair prediction.

[0101] We assume that the UE performs training and inference to obtain the predictions. For the AI / ML beam management use case and the AI / ML positioning use case, the UE can use the L1-RSRP measurements as input for the neural network to perform the predictions. The measurement data needs to be precise enough for high prediction accuracy.

[0102] The measurement requirements for AI / ML beam prediction, such as the measurement accuracy requirements for AI / ML beam management, should be more stringent than the measurement accuracy of traditional modes.

[0103] For AI / ML beam prediction based on different functions, the measurement accuracy requirement of AI / ML beam prediction is the same. For example, even though the UE performs Top-K beam prediction in the spatial domain or Top-1 beam prediction or Top-KDL Tx Rx beam pair prediction in the time domain, as long as the UE operates in AI / ML beam prediction mode, the measurement accuracy requirement will be the same.

[0104] In one embodiment, the measurement accuracy requirements for different use cases may require different accuracy requirements, for example, the L1-RSRP measurement accuracy for AI / ML beam management may have different requirements than the L1-RSRP measurement accuracy for AI / ML positioning use cases.

[0105] like Figure 2 and Figure 3 As shown in , measurement requirements and test mechanisms are proposed in two main areas. Figure 2 In the embodiment, the TE-UE is in a command mode, where the test equipment (TE) 202 (or NW 70, 99) can send a command to the UE 10. For example, if the L1-RSRP measurement error is higher than the AI / ML BM measurement requirement, the TE 202 or the NW (70, 99) can send a command to the UE 10 to operate in the AI / ML BM mode or switch the UE to the traditional mode.

[0106] In one embodiment, if the L1-RSRP measurement error is higher than the AI / ML positioning measurement requirement, the TE 202 or NW (70, 99) may send a command to the UE 10 to operate in the AI / ML positioning mode or switch the UE to the legacy mode.

[0107] exist Figure 3 In the embodiment, when the UE 10 is in AI / ML BM mode, the TE-UE is in simulation mode, and the measurement data set is collected under a specific radio condition environment (e.g., urban macro (UMa) radio conditions with low / static mobility UEs, or UMa radio conditions with high mobility UEs or urban micro (UMi)). The TE 202 (or NW 70, 99) simulates the radio conditions from the UE, and if the measurement requirements are not met, the UE can be switched to operate in the legacy mode.

[0108] Please note: According to 3GPP TS 38.133, the ground truth of the L1-RSRP value is known at the NW. In clauses 10.1.19 and 10.1.20 of [3GPP TS 38.133 L1-RSRP measurement requirements], the L1-RSRP accuracy requirements for FR1 and FR2 are described respectively. For FR1, the SSB-based L1-RSRP requirements and the CSI-RS-based L1-RSRP requirements are given in 10.1.19.1 and 10.1.19.2, respectively. For FR2, the SSB-based L1-RSRP requirements and the CSI-RS-based L1-RSRP requirements are given in 10.1.20.1 and 10.1.20.2, respectively. Therefore, the minimum and maximum requirements for L1-RSRP are defined (for legacy).

[0109] Figure 2 The details of the measurement requirements and test mechanisms when the TE-UE is in command mode are shown in Figure 1. Figure 3 As shown in .

[0110] Figure 2 The details of option 1 (TE / NW send command) shown in the following are described as follows:

[0111] 1: TE or NW configures the function and sends a command to switch to the configured AI / ML function-based operation to the UE 10. Function-based AI / ML beam management can be, for example, spatial domain Top-K or Top-1 beam ID prediction, spatial domain L1-RSRP prediction and / or time domain beam ID prediction and / or DL ​​Tx-Rx beam pair prediction.

[0112] 2: (Optional) The UE may be configured to send a confirmation that it is operating in a capability-based mode / operation (e.g. AI / ML beam management mode).

[0113] 3: The UE sends detailed information about the functionality of running this mode (e.g., an indication of whether the UE performs Top-K or Top-1 beam ID prediction in the spatial domain and / or time domain, or whether the UE performs L1-RSRP prediction in the spatial domain or time domain, or whether the UE performs DL Tx-Rx beam pair prediction).

[0114] 4: The UE performs measurements in AI / ML BM mode with respect to its corresponding functions, such as L1-RSRP measurements.

[0115] 5: The UE reports the AI / ML measurement values ​​based on the AI / ML BM function to the TE / NW.

[0116] 6: TE or NW AI / ML BM measurement accuracy requirements, for example, L1-RSRP+Error (dB): The error (dB) needs to be in the range of [-x, +x] dB to verify the measurement data based on the AI / ML BM mode function.

[0117] Please note: TE 202 or NW (70, 99) knows the L1-RSRP measurement accuracy (ground truth), 3GPP TS38.133. As mentioned in the previous paragraph, in clauses 10.1.19 and 10.1.20. Therefore, TE or NW can use the ground truth of L1-RSRP to verify L1-RSRP + Error (dB).

[0118] 7: If the error (dB) is not within the range of [-x, +x]dB, it means that the measurement data set fails the test.

[0119] 8: TE sends a command to UE to switch to legacy mode.

[0120] 9: Then, the UE switches to the legacy mode and reports the measurement data for the legacy mode to the TE / NW.

[0121] 10: TE / NW verifies the measurement data with traditional requirements, such as L1-RSRP+Error (dB), where Error (dB) needs to be in the range of [-y, +y] dB.

[0122] Please note: The error range [-x, +x]dB for the corresponding function for AI / ML BM use cases is tighter than [-y, +y]dB for the traditional mode.

[0123] Figure 3 The details of Option 2 (TE simulation performance degradation / improvement) shown in the figure are as follows:

[0124] 1: TE / NW (202, 70, 99) will change the radio condition environment according to UE 10, for example, TE / NW changes the channel profile, for example (UMI or UMA condition with UE mobility).

[0125] 2: The UE detects that the performance criterion switches to AI / ML mode (AI / ML BM mode), and the UE switches to AI / ML mode accordingly.

[0126] 3: The UE sends detailed information about the functionality of running this mode (e.g., whether the UE performs Top-K or Top-1 beam ID prediction in the spatial domain and / or time domain, or whether the UE performs L1-RSRP prediction in the spatial domain or time domain, or whether the UE performs DL Tx-Rx beam pair prediction).

[0127] 4: The UE performs measurements in AI / ML BM mode with respect to its corresponding function, such as L1-RSRP measurement.

[0128] 5: The UE reports the AI / ML measurement value of this function to the TE / NW.

[0129] 6: TW or NW, with AI / ML BM measurement accuracy requirements, for example L1-RSRP+Error (dB): The Error (dB) needs to be in the range of [-x, +x] dB to verify the measurement data based on the AI / ML BM mode function.

[0130] 7: If the error (dB) is not within the range of [-x, +x] dB, it means that the measurement data set fails the test.

[0131] 8: The UE then detects (or is informed of) the performance degradation and the UE switches back to the legacy mode.

[0132] 9: Then, the UE reports the measurement data of the traditional mode to the TE or NW.

[0133] 10: TE / NW verifies the measurement data with traditional requirements, such as L1-RSRP+error (dB), where the error (dB, dB) needs to be in the range of [-y, +y] dB.

[0134] Figure 4 A block diagram of one possible and non-limiting example of a cellular network 1 connected to a user equipment (UE) 10 is shown. Figure 4 A number of network elements are shown in the cellular network of FIG: a base station 70; and a core network 90.

[0135] exist Figure 4 In the present invention, a user equipment (UE) 10 wirelessly communicates with a base station 70 of a cellular network 1 via a radio link 11. The UE 10 is a wireless communication device such as a mobile device configured to access a cellular network. The UE 10 is shown as having one or more antennas 28. The ellipsis 2 indicates that there may be multiple UEs 10 wirelessly communicating with the base station 70 via a radio link. The UE 10 includes one or more processors 13, one or more memories 15, and other circuits 16. The other circuits 16 include one or more receivers (Rx) 17 and one or more transmitters (Tx) 18. Programs 12 are used to cause the UE 10 to perform the operations described herein. For the UE 10, the other circuits 16 may include circuits such as for user interface elements (not shown) such as a display.

[0136] The base station 70, as a network element of the cellular network 1, provides access to the cellular network 1 and the data network 91 to the UE 10 via the core network 90 (e.g., via the user plane function (UPF) of the core network 90). The base station 70 is shown as having one or more antennas 58. In general, the base station 70 is referred to as a RAN node 70 in this document. An example of a RAN node 70 is a gNB. However, there are many other examples of RAN nodes, including eNB (LTE base station) or a transmit receive point (TRP). The base station 70 includes one or more processors 73, one or more memories 75 and other circuits 76. The other circuits 76 include one or more receivers (Rx) 77 and one or more transmitters (Tx) 78. The program 72 is used to enable the base station 70 to perform the operations described herein.

[0137] It should be noted that base station 70 may instead be implemented via other wireless technologies, such as Wi-Fi (a wireless network protocol used by devices to communicate without a direct cable connection). In the case of Wi-Fi, link 11 may be characterized as a wireless link.

[0138] Two or more base stations 70 communicate using, for example, a link 79. Link 79 may be wired or wireless, or both, and may implement, for example, an Xn interface for fifth generation (5G), an X2 interface for LTE, or other suitable interfaces for other standards.

[0139] The cellular network 1 may include a core network 90 as a third illustrated element or elements, which may include core network functions and provide a connection to a data network 91 such as a telephone network and / or a data communications network (e.g., the Internet) via one or more links 81. The core network 90 includes one or more processors 93, one or more memories 95, and other circuits 96. The other circuits 96 include one or more receivers (Rx) 97 and one or more transmitters (Tx) 98. Programs 92 are used to enable the core network 90 to perform the operations described herein.

[0140] The core network 90 may be a 5GC (5G core network). The core network 90 may implement or include a plurality of network functions (NFs) 99, and the program 92 may include one or more NFs 99. The 5G core network may use hardware such as memory, processors, and virtualization layers. It may be a single independent computing system, a distributed computing system, or a cloud computing system. The NF 99 as a network element of the core network may be a container or a virtual machine running on the hardware of the computing system constituting the core network 90.

[0141] The core network functions for 5G may include access and mobility management functions, such as access and mobility control functions (AMF), provided by network functions 99; and session management functions, such as session management functions (SMF), provided by network functions. The core network functions for access and mobility management in LTE networks may be provided by MME (mobility management entity) and / or SGW (serving gateway) functions that route data to the data network. Many other functions are also possible, such as Figure 4 31 . The examples shown in FIG. 31 include: AMF; SMF; MME; SGW; Gateway Mobile Location Center (GMLC); Location Management Function (LMF); Unified Data Management (UDM); Unified Data Repository (UDR); Network Repository Function (NRF); and / or Evolved Serving Mobile Location Center (E-SMLC). These are merely exemplary core network functions that may be provided by the core network 90, and note that both 5G and LTE core network functions may be provided by the core network 90. ​​A radio access network (RAN) node 70 is coupled to the core network 90 via a backhaul link 31. The RAN node 70 and the core network 90 may include an NG interface for 5G, an S1 interface for LTE, or other suitable interfaces for other radio access technologies for communicating via the backhaul link 31.

[0142] In the data network 91, there is a computer-readable medium 94. The computer-readable medium 94 contains instructions, which, when downloaded and installed into the memory 15, 75 or 95 of the corresponding UE 10, base station 70 and / or core network element 90 and executed by the processor 13, 73 or 93, enable the corresponding device to perform the corresponding actions described herein. The computer-readable medium 94 can be implemented in other forms, such as via an optical disk or a memory stick.

[0143] Programs 12, 72 and 92 include instructions stored by corresponding one or more memories 15, 75 or 95. These instructions, when executed by corresponding one or more processors 13, 73 or 93, cause corresponding devices 10, 70 or 90 to perform operations described herein. Computer-readable memory 15, 75 or 95 may be of any type suitable for the local technical environment and may be implemented using any suitable data storage technology, such as semiconductor-based storage devices, flash memory, firmware, magnetic storage devices and systems, optical storage devices and systems, fixed memory and removable memory. Computer-readable memory 15, 75 and 95 may be a device for performing storage functions. As non-limiting examples, processors 13, 73 and 93 may be of any type suitable for the local technical environment and may include one or more of a general-purpose computer, a special-purpose computer, a microprocessor, a digital signal processor (DSP) and a processor based on a multi-core processor architecture. Processors 13, 73 and 93 may be devices for causing their respective devices to perform functions such as those described herein.

[0144] Receivers 17, 77 and 97 and transmitters 18, 78 and 98 may implement a wired or wireless interface. Receivers and transmitters may be combined together as a transceiver.

[0145] Figure 5An example apparatus 500 is provided, which may be implemented in hardware and configured to implement the examples described herein. The apparatus 500 includes at least one processor 502 (e.g., an FPGA and / or a CPU), one or more memories 504 including computer program code 505, the computer program code 504 having instructions for executing the methods described herein, wherein at least one memory 504 and the computer program code 505 are configured to utilize at least one processor 502 to cause the apparatus 500 to implement circuits, processes, components, modules, or functions (implemented using a control module 506) to implement the examples described herein, including measurement accuracy requirements for AI / ML beam prediction with testability. The memory 504 may be a non-transitory memory, a temporary memory, a volatile memory (e.g., a RAM), or a non-volatile storage (e.g., a ROM). The configuration or determination mode 530 of the control module and the execution or reception of control measurements 540 implement aspects described herein related to measurement accuracy requirements for AI / ML beam prediction with testability. Optionally, the included satisfied requirements 550 are configured to determine whether measurement accuracy requirements are satisfied (eg, satisfied requirements 550 are performed by TE / NW 202 , 70 , 99 ).

[0146] The device 500 includes a display and / or I / O interface 508, which includes user interface (UI) circuits and components that can be used to display various aspects or states of the methods described herein (e.g., when one of the methods is being executed or at a subsequent time), or to receive input from a user, for example using a keyboard, camera, touch screen, touch area, microphone, biometrics, one or more sensors, etc. The device 500 includes one or more communication interfaces, such as a network (N / W) interface (I / F) 510. The communication I / F 510 can be wired and / or wireless, and communicates over the Internet / other network via any communication technology (including via one or more links 524). The link 524 can be Figure 5 Links 11 and / or 79 and / or 31 and / or 81 in. Figure 5 The links 11 and / or 79 and / or 31 and / or 81 in FIG. 5 may also be implemented using the transceiver 516 and the corresponding wireless link 526. The communication I / F 510 may include one or more transmitters or one or more receivers.

[0147] The transceiver 516 includes one or more transmitters 518 and one or more receivers 520. The transceiver 516 and / or the communication I / F 510 may include standard well-known components, such as amplifiers, filters, frequency converters, (de)modulators and encoder / decoder circuits, and one or more antennas, such as the antenna 514 used for communication via the wireless link 526.

[0148] The control module 506 of the device 500 includes one or both of the components 506-1 and / or 506-2 that can be implemented in a variety of ways. The control module 506 can be implemented in hardware as the control module 506-1, such as being implemented as part of one or more processors 502. The control module 506-1 can also be implemented as an integrated circuit or implemented by other hardware such as a programmable gate array. In another example, the control module 506 can be implemented as a control module 506-2, which is implemented as a computer program code (with corresponding instructions) 505 and executed by one or more processors 502. For example, one or more memories 504 store instructions that, when executed by one or more processors 502, cause the device 500 to perform one or more operations as described herein. In addition, one or more processors 502, one or more memories 504, and example algorithms (e.g., as flow charts and / or signaling diagrams) encoded as instructions, programs, or codes are means for performing the operations described herein.

[0149] The device 500 implementing the function of the control 506 may be a UE 10, a base station 70 (e.g., a gNB 70), or a core network 90 including one or more network functions 99. Therefore, the processor 502 may correspond to the processor 13, the processor 73, and / or the processor 93, the memory 504 may correspond to one or more memories 15, one or more memories 75, and / or one or more memories 95, the computer program code 505 may correspond to the program 12, the program 72, or the program 92, the communication I / F 510 and / or the transceiver 516 may correspond to other circuits 16, other circuits 76, or other circuits 96, and the antenna 514 may correspond to the antenna 28 or the antenna 58. The device 500 may also correspond to the test device 202.

[0150] Alternatively, apparatus 500 and its elements may not correspond to UE 10, base station 70 or core network and their respective elements including one or more network functions 99, because apparatus 500 may be part of a self-organizing / optimizing network (SON) node or other node such as a node in a cloud.

[0151] The apparatus 500 may also be distributed throughout the network (eg, 91 ), including within and between the apparatus 500 and any network elements (such as the core network 90 and / or base stations 70 and / or UEs 10 ).

[0152] like Figure 5As shown in , interface 512 enables data communication and signaling between the various items of device 500. For example, interface 512 can be one or more buses, such as address, data or control buses, and can include any interconnection mechanism, such as a series of lines on a motherboard or integrated circuit, optical fiber or other optical communication equipment, etc. The computer program code (e.g., instructions) 505 including control 506 may include object-oriented software that is configured to pass data or messages between objects within computer program code 505. Device 500 does not necessarily include each of the features mentioned, or may also include other features. The various components of device 500 may reside at least partially in a common housing 528, or subsets of the various components of device 500 may be at least partially located in different housings that may include housing 528.

[0153] Figure 6 Schematic diagrams of non-volatile storage media 600a (e.g., a computer / compact disk (CD) or digital versatile disk (DVD)) and 600b (e.g., a universal serial bus (USB) memory stick) and 600c (e.g., cloud storage for downloading instructions and / or parameters 602 or receiving instructions and / or parameters 602 via email) storing instructions and / or parameters 602 are shown, which, when executed by a processor, allow the processor to perform one or more steps of the methods described herein.

[0154] Figure 7 An example method 700 based on example embodiments described herein is provided. At 710, the method includes determining a beam management mode based on a configuration indicating a beam management mode. At 720, the method includes switching to the beam management mode. At 730, the method includes performing at least one measurement related to artificial intelligence or machine learning beam management while operating in the beam management mode. At 740, the method includes transmitting at least one measurement value to a test device or a network. The method 700 may be performed using the UE 10 or the apparatus 500.

[0155] Figure 8 An example method 800 based on various example embodiments described herein. At 810, the method includes determining a configuration indicating a beam management mode. At 820, the method includes receiving at least one measurement value related to artificial intelligence or machine learning beam management from a user equipment based on the beam management mode. At 830, the method includes selecting an accuracy target for the beam management mode. At 840, the method includes determining whether the accuracy target is met based on the at least one measurement value. The method 800 may be performed using the test device 202, the base station 70, the network function 99 performed using a network entity, or the apparatus 500.

[0156] The following examples are provided and described in this article.

[0157] Example 1

[0158] A device comprises: at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, causes the device to at least: determine a beam management mode based on a configuration indicating a beam management mode; switch to the beam management mode; when operating in the beam management mode, perform at least one measurement related to artificial intelligence or machine learning beam management; and transmit the at least one measurement value to a test device or a network.

[0159] Example 2

[0160] An apparatus as described in Example 1, wherein the mode comprises one of: an artificial intelligence or machine learning mode, or a traditional mode.

[0161] Example 3

[0162] An apparatus as described in any of Examples 1 to 2, wherein the pattern includes an artificial intelligence or machine learning operation for at least one of: a spatial domain beam identifier prediction for a top beam, or a spatial domain beam identifier prediction for multiple top beams, or a spatial domain layer 1 reference signal received power prediction, or a time domain beam identifier prediction for a top beam, or a time domain beam identifier prediction for multiple top beams, or a time domain layer 1 reference signal received power prediction, or a downlink transmit receive beam pair prediction.

[0163] Example 4

[0164] An apparatus as described in any of Examples 1 to 3, wherein the instructions, when executed by the at least one processor, cause the apparatus to at least: receive a command from the test device or the network for the apparatus to switch to an artificial intelligence or machine learning beam management mode, wherein the beam management mode includes the artificial intelligence or machine learning beam management method.

[0165] Example 5

[0166] An apparatus as described in any one of Examples 1 to 4, wherein, when the instruction is executed by the at least one processor, causes the apparatus to at least: detect changes in the radio condition environment; and determine the beam management mode based on the changes in the radio condition environment.

[0167] Example 6

[0168] An apparatus as described in Example 5, wherein, when the instruction is executed by the at least one processor, causes the apparatus to at least: determine the beam management mode as an artificial intelligence or machine learning mode when the change in the radio condition environment includes performance improvement.

[0169] Example 7

[0170] An apparatus as described in any one of Examples 5 to 6, wherein, when the instruction is executed by the at least one processor, causes the apparatus to at least: determine the beam management mode as a traditional mode when the change in the radio condition environment includes performance degradation.

[0171] Example 8

[0172] An apparatus as described in any one of Examples 1 to 7, wherein the at least one measurement value is used as input for beam prediction using artificial intelligence or machine learning.

[0173] Example 9

[0174] An apparatus as described in any of Examples 1 to 8, wherein the at least one measurement value includes at least one of: a layer 1 reference signal received power measurement value, or one or more locations of a user equipment, or one or more transmit beam angles, or non-line-of-sight (nLoS) information, or a non-line-of-sight (nRoS) indication, or line-of-sight (LoS) information, or a line-of-sight (LoS) indication.

[0175] Example 10

[0176] An apparatus as described in any of Examples 1 to 9, wherein the measurement accuracy requirements of the artificial intelligence or machine learning mode for beam management are met when the error associated with the at least one measurement value is within an artificial intelligence or machine learning beam management mode error range, or within a layer 1 reference signal received power (L1-RSRP) measurement error range.

[0177] Example 11

[0178] An apparatus as described in any of Examples 1 to 10, wherein the measurement accuracy requirements of the traditional mode for beam management are met when the error associated with the at least one measurement value is within the traditional mode error range or within the layer 1 reference signal received power (L1-RSRP) error range.

[0179] Example 12

[0180] An apparatus as described in any of Examples 1 to 11, wherein, when the instructions are executed by the at least one processor, the apparatus at least: transmits an indication to the test equipment or the network that the apparatus is operating in a beam management mode.

[0181] Example 13

[0182] A device as described in any one of Examples 1 to 12, wherein the at least one measurement value meets or does not meet the measurement accuracy requirement.

[0183] Example 14

[0184] An apparatus as described in Example 13, wherein the measurement accuracy requirement includes a value.

[0185] Example 15

[0186] An apparatus as described in Example 14, wherein: when the at least one measurement value is within a first tolerance value of the value, the measurement accuracy requirement of the artificial intelligence or machine learning mode for beam management is met; and when the at least one measurement value is within a second tolerance value of the value, the measurement accuracy requirement of the traditional mode for beam management is met.

[0187] Example 16

[0188] An apparatus according to Example 15, wherein the first tolerance value is less than the second tolerance value.

[0189] Example 17

[0190] The apparatus of any one of Examples 15 to 16, wherein the first tolerance value is greater than the second tolerance value.

[0191] Example 18

[0192] An apparatus as described in any of Examples 1 to 17, wherein, when the instruction is executed by the at least one processor, the apparatus at least: trains an artificial intelligence or machine learning model for beam management based on the at least one measurement value.

[0193] Example 19

[0194] An apparatus as described in any one of Examples 1 to 18, wherein, when the instructions are executed by the at least one processor, the apparatus is caused to at least: receive an indication from the test device or the network that the at least one measurement value meets the measurement accuracy requirement.

[0195] Example 20

[0196] An apparatus as described in any of Examples 1 to 19, wherein, when the instruction is executed by the at least one processor, causes the apparatus to at least: receive an instruction from the test device or the network to switch to the beam management mode based on whether the at least one measurement value meets the measurement accuracy requirement.

[0197] Example 21

[0198] A device comprises: at least one processor; and at least one memory storing instructions, which, when executed by the at least one processor, causes the device to at least: determine a configuration indicating a beam management mode; based on the beam management mode, receive at least one measurement value related to artificial intelligence or machine learning beam management from a user device; select an accuracy target for the beam management mode; and determine whether the accuracy target is met based on the at least one measurement value.

[0199] Example 22

[0200] An apparatus as described in Example 21, wherein the mode comprises one of: an artificial intelligence or machine learning mode, or a traditional mode.

[0201] Example 23

[0202] An apparatus as described in any of Examples 21 to 22, wherein the pattern includes an artificial intelligence or machine learning operation for at least one of: a spatial domain beam identifier prediction for a top beam, or a spatial domain beam identifier prediction for multiple top beams, or a spatial domain layer 1 reference signal received power prediction, or a time domain beam identifier prediction for a top beam, or a time domain beam identifier prediction for multiple top beams; or a time domain layer 1 reference signal received power prediction, or a downlink transmit receive beam pair prediction.

[0203] Example 24

[0204] An apparatus as described in any of Examples 21 to 23, wherein, when the instruction is executed by the at least one processor, the apparatus at least: transmits a command to the user equipment for switching the user equipment to an artificial intelligence or machine learning beam management mode, wherein the beam management mode includes the artificial intelligence or machine learning beam management mode.

[0205] Example 25

[0206] An apparatus as described in any one of Examples 21 to 24, wherein, when the instruction is executed by the at least one processor, causes the apparatus to at least: change the radio condition environment based on the user equipment; wherein the configuration indicating the beam management mode is based on changes in the radio condition environment according to the user equipment.

[0207] Example 26

[0208] As in the apparatus of Example 25, wherein when the change in the radio condition environment includes performance improvement, the configuration indicates that the beam management mode is an artificial intelligence or machine learning mode.

[0209] Example 27

[0210] An apparatus as in any one of Examples 25 to 26, wherein when the change in the radio condition environment includes performance degradation, the configuration indicates that the beam management mode is a legacy mode.

[0211] Example 28

[0212] An apparatus as described in any of Examples 21 to 27, wherein the at least one measurement value is used as input for beam prediction using artificial intelligence or machine learning.

[0213] Example 29

[0214] An apparatus as described in any of Examples 21 to 28, wherein the at least one measurement value includes at least one of: a layer 1 reference signal received power measurement value, or one or more locations of a user equipment, or one or more transmit beam angles, or non-line-of-sight (nLoS) information, or a non-line-of-sight indication, or line-of-sight information, or a line-of-sight indication.

[0215] Example 30

[0216] An apparatus as described in any of Examples 21 to 29, wherein, when the instructions are executed by the at least one processor, the apparatus at least: determines a measurement accuracy requirement for an artificial intelligence or machine learning mode for beam management; and determines that the measurement accuracy requirement for the artificial intelligence or machine learning mode is met when an error associated with the at least one measurement value is within an artificial intelligence or machine learning beam management mode error range or within a layer 1 reference signal received power (L1-RSRP) measurement error range.

[0217] Example 31

[0218] An apparatus as described in any one of Examples 21 to 30, wherein, when the instruction is executed by the at least one processor, at least the apparatus is caused to: determine the measurement accuracy requirement of the traditional mode for beam management; and when the error associated with the at least one measurement value is within the traditional mode error range or within the layer 1 reference signal received power (L1-RSRP) error range, determine that the measurement accuracy requirement of the traditional mode for beam management is met.

[0219] Example 32

[0220] An apparatus as described in any one of Examples 21 to 31, wherein, when the instructions are executed by the at least one processor, the apparatus at least: receives an indication from the user equipment that the user equipment is operating in a beam management mode.

[0221] Example 33

[0222] An apparatus as described in any of Examples 21 to 32, wherein, when the instruction is executed by the at least one processor, the apparatus is caused to at least: determine whether the at least one measurement value meets the measurement accuracy requirement.

[0223] Example 34

[0224] An apparatus as in Example 33, wherein the measurement accuracy requirement includes a value.

[0225] Example 35

[0226] An apparatus as described in Example 34, wherein, when the instruction is executed by the at least one processor, causes the apparatus to at least: determine that the measurement accuracy requirement of an artificial intelligence or machine learning mode for beam management is met when the at least one measurement value is within a first tolerance value of the value; and determine that the measurement accuracy requirement of a traditional mode for beam management is met when the at least one measurement value is within a second tolerance value of the value.

[0227] Example 36

[0228] An apparatus as described in Example 35, wherein the first tolerance value is less than the second tolerance value.

[0229] Example 37

[0230] An apparatus as described in any of Examples 35 to 36, wherein the first tolerance value is greater than the second tolerance value.

[0231] Example 38

[0232] An apparatus as described in any of Examples 21 to 37, wherein, when the instruction is executed by the at least one processor, causes the apparatus to at least: train an artificial intelligence or machine learning model for beam management based on the at least one measurement value.

[0233] Example 39

[0234] An apparatus as described in any of Examples 21 to 38, wherein, when the instruction is executed by the at least one processor, the apparatus at least: determines whether the at least one measurement value meets the measurement accuracy requirement; and transmits an indication to the user equipment that the at least one measurement value meets the measurement accuracy requirement.

[0235] Example 40

[0236] An apparatus as described in any of Examples 21 to 39, wherein, when the instruction is executed by the at least one processor, the apparatus at least: determines whether the at least one measurement value meets the measurement accuracy requirement; and based on whether the at least one measurement value meets the measurement accuracy requirement, transmits an indication to the user equipment to switch to the beam management mode.

[0237] Example 41

[0238] A method comprising: determining a beam management mode based on a configuration indicating the beam management mode; switching to the beam management mode; performing at least one measurement related to artificial intelligence or machine learning beam management while operating in the beam management mode; and transmitting the at least one measurement value to a test device or a network.

[0239] Example 42

[0240] A method comprising: determining a configuration indicating a beam management mode; receiving at least one measurement related to artificial intelligence or machine learning beam management from a user device based on the beam management mode; selecting an accuracy target for the beam management mode; and determining whether the accuracy target is met based on the at least one measurement value.

[0241] Example 43

[0242] A device comprising: a device for determining a beam management mode based on a configuration indicating the beam management mode; a device for switching to the beam management mode; a device for performing at least one measurement related to artificial intelligence or machine learning beam management when operating in the beam management mode; and a device for transmitting the at least one measurement value to a test device or a network.

[0243] Example 44

[0244] A device comprising: a device for determining a configuration indicating a beam management mode; a device for receiving at least one measurement value related to artificial intelligence or machine learning beam management from a user device based on the beam management mode; a device for selecting an accuracy target for the beam management mode; and a device for determining whether the accuracy target is met based on the at least one measurement value.

[0245] Example 45

[0246] A machine-readable, non-transitory program storage device tangibly embodies a program of instructions executable by the machine for performing operations, the operations comprising: determining a beam management mode based on a configuration indicating the beam management mode; switching to the beam management mode; performing at least one measurement related to artificial intelligence or machine learning beam management while operating in the beam management mode; and transmitting the at least one measurement value to a test device or a network.

[0247] Example 46

[0248] A machine-readable, non-transitory program storage device tangibly embodies a program of instructions executable by the machine for performing operations, the operations comprising: determining a configuration indicating a beam management mode; receiving at least one measurement value related to artificial intelligence or machine learning beam management from a user device based on the beam management mode; selecting an accuracy target for the beam management mode; and determining whether the accuracy target is met based on the at least one measurement value.

[0249] References to "computers", "processors", etc. should be understood to cover not only computers with different architectures such as single / multi-processor architectures and sequential or parallel architectures, but also special-purpose circuits such as field programmable gate arrays (FPGAs), application-specific circuits (ASICs), signal processing devices and other processing circuits. References to computer programs, instructions, code, etc. should be understood to cover software or firmware for programmable processors, such as, for example, the programmable content of hardware devices, whether instructions for a processor, or configuration settings for a fixed-function device, gate array, or programmable logic device, etc.

[0250] The memory described herein may be implemented using any suitable data storage technology, such as semiconductor-based memory devices, flash memory, magnetic memory devices and systems, optical memory devices and systems, non-transitory memory, temporary memory, fixed memory, and removable memory. The memory may include a database for storing data.

[0251] As used in this application, "circuitry" may refer to one or more or all of the following: (a) hardware circuit implementations, such as implementations in analog and / or digital circuits; and (b) combinations of circuitry and software (and / or firmware), such as (as applicable) (i) a processor and / or (ii) a portion of a processor / software (including a digital signal processor), a combination of software and memory, which work together to enable the device to perform various functions; and (c) circuitry, such as a microprocessor or a portion of a microprocessor, even though it need not be physically present. This definition of circuitry applies to this disclosure, including one or all uses of the term in any claim. As a further example, as used in this disclosure, the term circuitry also covers implementations of only a processor (or multiple processors) or a processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular element, a baseband integrated circuit or an application processor integrated circuit for a mobile phone or a similar integrated circuit in a server, cellular network device, or other network device.

[0252] It should be understood that the above description is illustrative only. Various alternatives and modifications may be devised by those skilled in the art. For example, the features described in the various dependent claims may be combined with each other in any suitable combination. In addition, the features of the different example embodiments discussed above may be selectively combined into new example embodiments. Therefore, this specification is intended to cover all such alternatives, modifications and variations that fall within the scope of the appended claims.

[0253] The following acronyms and abbreviations that may be found in the specification and / or drawings are given below (abbreviations and acronyms may be appended to each other, or may be appended to each other using dashes, hyphens, slashes or other characters such as numbers, or in some examples, the text "s" to indicate a plurality, and may not be case sensitive):

[0254] 3GPP Third Generation Partnership Project

[0255] 4G Fourth Generation

[0256] 5G Fifth Generation

[0257] 5GC 5G Core Network

[0258] AI Artificial Intelligence

[0259] AMF access and mobility management function

[0260] ASIC

[0261] BM Beam Management

[0262] CD / Computer CD

[0263] CPU Central Processing Unit

[0264] CSI-RS channel state information reference signal

[0265] DL Downlink

[0266] DSP Digital Signal Processor

[0267] DVD Digital Versatile Disc

[0268] eNB Evolved Node B (e.g. LTE base station)

[0269] EPC Evolved Packet Core

[0270] E-SMLC Evolution Service Mobile Location Center

[0271] FPGA Field Programmable Gate Array

[0272] FR frequency range

[0273] GMLC Gateway Mobile Location Center

[0274] gNB is a base station for 5G / NR, which provides NR user plane and control plane protocol termination to UE and is connected to 5GC via NG interface.

[0275] I / F Interface

[0276] I / O Input / Output

[0277] L1 Layer 1

[0278] LCM Lifecycle Management

[0279] LMF location management function

[0280] LoS Line of Sight

[0281] LTE Long Term Evolution (4G)

[0282] ML Machine Learning

[0283] MME mobility management entity

[0284] NF Network Function

[0285] NG New Generation

[0286] NG-RAN Next Generation Radio Access Network

[0287] nLoS Non Line of Sight

[0288] NR New Radio

[0289] NRF Network Repository Features

[0290] N / W Network

[0291] NW Network

[0292] RAM Random Access Memory

[0293] RAN Radio Access Network

[0294] RAN1 Radio Layer 1

[0295] ROM Read Only Memory

[0296] RP RAN Plenary Meeting

[0297] RSRP Reference Signal Received Power

[0298] Rx Receiver or receiving

[0299] S1 Interface connecting eNB and EPC

[0300] SGW Service Gateway

[0301] SMF session management functions

[0302] SON self-organizing / optimizing network

[0303] SSB Synchronization Signal Block or Synchronization Signal and Physical Broadcast Channel Block TE Test Equipment

[0304] TRP Transmit Receiving Point

[0305] TS Technical Specifications

[0306] Tx transmitter or transmitter

[0307] UDM Unified Data Management

[0308] UDR Unified Data Repository

[0309] UE User Equipment (e.g., wireless, usually mobile device) UI User Interface

[0310] UMa City Macro

[0311] UMA City Macro

[0312] UMi City Micro

[0313] UMI City Micro

[0314] UPF User Plane Function

[0315] USB Universal Serial Bus

[0316] Wi-Fi A wireless networking protocol used by devices to communicate without a direct cable connection

[0317] X2 Network interface between RAN nodes and between RAN and core network

[0318] Xn Network interface between NG-RAN nodes

Claims

1. A device for communication, comprising: at least one processor; as well as at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: determining the beam management mode based on a configuration indicating the beam management mode; Switching to the beam management mode; performing at least one measurement related to artificial intelligence or machine learning beam management while operating in the beam management mode; as well as The at least one measurement value is transmitted to a test device or a network.

2. The device according to claim 1, wherein: The mode includes one of the following: Artificial intelligence or machine learning models, or Traditional mode.

3. The device according to any one of claims 1 to 2, wherein: The mode includes artificial intelligence or machine learning operations for at least one of: The spatial domain beam identifier prediction for the top beam, or Spatial domain beam identifier prediction for multiple beams, or spatial domain layer 1 reference signal received power prediction, or The time domain beam identifier prediction for the top beam, or Time domain beam identifier prediction for multiple beams, or Time domain layer 1 reference signal received power prediction, or Downlink transmit receive beam pair prediction.

4. The device according to any one of claims 1 to 3, wherein: When the instructions are executed by the at least one processor, the device at least: Receiving a command from the test equipment or the network for the apparatus to switch to an artificial intelligence or machine learning beam management mode, wherein the beam management mode comprises the artificial intelligence or machine learning beam management mode.

5. The device according to any one of claims 1 to 4, wherein: When the instructions are executed by the at least one processor, the device at least: Detect changes in radio conditions in the environment; and The beam management mode is determined based on changes in the radio condition environment.

6. The device according to claim 5, wherein: When the instructions are executed by the at least one processor, the device at least: When the change in the radio condition environment includes performance improvement, determining the beam management mode as an artificial intelligence or machine learning mode.

7. The device according to any one of claims 5 to 6, wherein: When the instructions are executed by the at least one processor, the device at least: When the change in the radio condition environment includes performance degradation, the beam management mode is determined to be a conventional mode.

8. The device according to any one of claims 1 to 7, wherein: The at least one measurement is used as input for beam prediction using artificial intelligence or machine learning.

9. The device according to any one of claims 1 to 8, wherein: The at least one measurement value comprises at least one of the following: The layer 1 reference signal received power measurement, or one or more locations of a user's device, or One or more transmit beam angles, or Non-line-of-sight (nLoS) information, or Non-line-of-sight (nLoS) indication, or Line of sight (LoS) information, or Line of sight (LoS) indication.

10. The device according to any one of claims 1 to 9, wherein: When the error associated with the at least one measurement is within the artificial intelligence or machine learning beam management mode error range, or within the layer 1 reference signal received power (L1-RSRP) measurement error range, the measurement accuracy requirement of the artificial intelligence or machine learning mode for beam management is met.

11. The device according to any one of claims 1 to 10, wherein: When the error associated with the at least one measurement is within the legacy mode error range, or within the layer 1 reference signal received power (L1-RSRP) error range, the measurement accuracy requirement of the legacy mode for beam management is met.

12. The device according to any one of claims 1 to 11, wherein: When the instructions are executed by the at least one processor, the device at least: An indication is transmitted to a test device or a network that the apparatus is operating in the beam management mode.

13. The device according to any one of claims 1 to 12, wherein: The at least one measurement value meets or does not meet the measurement accuracy requirement.

14. The device of claim 13, wherein: The measurement accuracy requirement includes a value.

15. The apparatus of claim 14, wherein: satisfying the measurement accuracy requirement of an artificial intelligence or machine learning mode for beam management when the at least one measured value is within a first tolerance value of the value; and The measurement accuracy requirement for a conventional mode of beam management is met when the at least one measured value is within a second tolerance value of the value.

16. The device of claim 15, wherein: The first tolerance value is smaller than the second tolerance value.

17. The device according to any one of claims 15 to 16, wherein: The first tolerance value is greater than the second tolerance value.

18. The device according to any one of claims 1 to 17, wherein: When the instructions are executed by the at least one processor, the device at least: An artificial intelligence or machine learning model for beam management is trained based on the at least one measurement.

19. The device according to any one of claims 1 to 18, wherein: When the instructions are executed by the at least one processor, the device at least: An indication is received from the test device or the network that the at least one measurement value satisfies a measurement accuracy requirement.

20. The device according to any one of claims 1 to 19, wherein: When the instructions are executed by the at least one processor, the device at least: Based on whether the at least one measurement value meets the measurement accuracy requirement, an instruction to switch to the beam management mode is received from the test device or the network.

21. An apparatus for communication, comprising: at least one processor; as well as at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus to at least: determining a configuration indicating a beam management mode; receiving, from a user equipment, at least one measurement value related to artificial intelligence or machine learning beam management based on the beam management mode; selecting an accuracy target for the beam management mode; as well as A determination is made based on the at least one measurement whether the accuracy target is met.

22. A method for communication, comprising: determining the beam management mode based on a configuration indicating the beam management mode; Switching to the beam management mode; performing at least one measurement related to artificial intelligence or machine learning beam management while operating in the beam management mode; and The at least one measurement value is transmitted to a test device or a network.

23. A method for communication, comprising: determining a configuration indicating a beam management mode; receiving, from a user equipment, at least one measurement value related to artificial intelligence or machine learning beam management based on the beam management mode; selecting an accuracy target for the beam management mode; as well as A determination is made based on the at least one measurement whether the accuracy target is met.

24. A computer program product comprising instructions for causing an apparatus to perform the method according to any one of claims 22-23.

25. A computer readable medium comprising instructions for causing an apparatus to perform the method according to any one of claims 22-23.

Images