Devices, methods, apparatuses, and computer readable media for machine learning functionality

Abstract

A terminal device 110 comprising receiving, from a network, a first configuration 152 comprising a set of conditions for the terminal device to store information on applicability of a machine learning

Description

[0001] Various example embodiments relate to devices, methods, apparatuses, and computer readable media for machine learning functionality. BACKGROUND

[0002] Artificial intelligence (AI) / machine learning (ML) -enabled feature refers to a feature where AI / ML may be used. AI / ML is a topic in most 3rd Generation Partnership Project (3GPP) standardization work groups. The ML models, once deployed into the functions of the cellular communication system, could potentially have a direct impact on the behavior of the system. The capability and performance of the ML model are determined by every operational phase of its lifecycle: training, testing, deployment, and inference phases. The AI / ML functionalities using these operational phases may be applied to both infrastructure network elements and terminal devices (user equipment). While the details of particular AI / ML functionalities are beyond the scope of the present disclosure, it is important that both the network and the terminal device share an understanding of capabilities of the terminal device to apply the AI / ML functionalities.

[0003] For AI / ML enhancements related to beam management (BM), two sub-use cases have been identified: beam prediction in the spatial domain, which may be referred to as BM-Casel, and beam prediction in the time domain, which may be referred to as BM-Case2. The scope of spatial beam prediction (BM-Casel) is to predict the best downlink (DL) transmit (Tx) beam and / or DL Tx / receive (Rx) beam pairs in different spatial locations. The scope of time-domain beam prediction (BM-Case2) aim to predict the best DL Tx beam and / or DL Tx / Rx beam pairs to use for next time instant. DL Tx beam prediction for both user equipment (UE)-sided model and network (NW)-sided model encompasses spatial-domain DL Tx beam prediction for Set A of beams based on measurement results of Set B of beams (“BM-Casel”) and temporal DL Tx beam prediction for Set A of beams based on the historic measurement results of Set B of beams (“BM-Case2”). The beam management is thus an example of AI / ML enhancement that may comprise one or more AI / ML functionalities executed in both the network and in the UE. The primary motivation is to support a reduced overhead and lower beam measurements and reporting latency. SUMMARY

[0004] A brief summary of exemplary embodiments is provided below to provide basic understanding of some aspects of various embodiments. It should be noted that this summary is not intended to identify key features of essential elements or define scopes of the embodiments, and its sole purpose is to introduce some concepts in a simplified form as a preamble for a more detailed description provided below.

[0005] In a first aspect, disclosed is an apparatus for a terminal device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: receive from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device; receive from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; store the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and report to the network, the applicability of the machine learning functionality in the terminal device.

[0006] In a second aspect, disclosed is an apparatus for a first network device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: transmit to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and transmit to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[0007] In a third aspect, disclosed is an apparatus for a second network device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[0008] In a fourth aspect, disclosed is an apparatus for a terminal device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: receive from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC CE cell switch message; receive from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; store the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and report to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0009] In a fifth aspect, disclosed is an apparatus for a first network device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: transmit to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and transmit to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[0010] In a sixth aspect, disclosed is an apparatus for a second network device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0011] In a seventh aspect, disclosed is an apparatus for a terminal device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: receive from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device; receive from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; store the information for at least one serving cell or at least one neighboring cell; and report to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled.

[0012] In an eighth aspect, disclosed is an apparatus for a first network device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: transmit to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and transmit to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[0013] In a ninth aspect, disclosed is an apparatus for a second network device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[0014] In a tenth aspect, disclosed is an apparatus for a terminal device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: receive from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; receive from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; store the information for at least one serving cell or at least one neighboring cell; and report to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or PUSCH or PUCCH message using MAC CE.

[0015] In an eleventh aspect, disclosed is an apparatus for a first network device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: transmit to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and transmit to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[0016] In a twelfth aspect, disclosed is an apparatus for a second network device. The apparatus may include at least one processor and at least one memory. The at least one memory may store instructions that, when executed by the at least one processor, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0017] In a thirteenth aspect, disclosed is a method performed by an apparatus for a terminal device. The method may comprise: receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device; receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and reporting to the network, the applicability of the machine learning functionality in the terminal device.

[0018] In a fourteenth aspect, disclosed is a method performed by an apparatus for a first network device. The method may comprise: transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and transmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[0019] In a fifteenth aspect, disclosed is a method performed by an apparatus for a second network device. The method may comprise: receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[0020] In a sixteenth aspect, disclosed is a method performed by an apparatus for a terminal device. The method may comprise: receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC CE cell switch message; receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and reporting to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0021] In a seventeenth aspect, disclosed is a method performed by an apparatus for a first network device. The method may comprise: transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and transmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[0022] In an eighteenth aspect, disclosed is a method performed by an apparatus for a second network device. The method may comprise: receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0023] In a nineteenth aspect, disclosed is a method performed by an apparatus for a terminal device. The method may comprise: receiving from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device; receiving from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; storing the information for at least one serving cell or at least one neighboring cell; and reporting to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled.

[0024] In a twentieth aspect, disclosed is a method performed by an apparatus for a first network device. The method may comprise: transmitting to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and transmitting to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[0025] In a twenty-first aspect, disclosed is a method performed by an apparatus for a second network device. The method may comprise: receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[0026] In a twenty-second aspect, disclosed is a method performed by an apparatus for a terminal device. The method may comprise: receiving from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; receiving from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; storing the information for at least one serving cell or at least one neighboring cell; and reporting to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or PUSCH or PUCCH message using MAC CE.

[0027] In a twenty-third aspect, disclosed is a method performed by an apparatus for a first network device. The method may comprise: transmitting to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and transmitting to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[0028] In a twenty-fourth aspect, disclosed is a method performed by an apparatus for a second network device. The method may comprise: receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0029] In a twenty-fifth aspect, disclosed is an apparatus for a terminal device. The apparatus may comprise: means for receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device; means for receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; means for storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and means for reporting to the network, the applicability of the machine learning functionality in the terminal device.

[0030] In a twenty-sixth aspect, disclosed is an apparatus for a first network device. The apparatus may comprise: means for transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and means for transmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[0031] In a twenty-seventh aspect, disclosed is an apparatus for a second network device. The apparatus may comprise: means for receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[0032] In a twenty-eighth aspect, disclosed is an apparatus for a terminal device. The apparatus may comprise: means for receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC CE cell switch message; means for receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; means for storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and means for reporting to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0033] In a twenty-ninth aspect, disclosed is an apparatus for a first network device. The apparatus may comprise: means for transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and means for transmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[0034] In a thirtieth aspect, disclosed is an apparatus for a second network device. The apparatus may comprise: means for receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0035] In a thirty-first aspect, disclosed is an apparatus for a terminal device. The apparatus may comprise: means for receiving from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device; means for receiving from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; means for storing the information for at least one serving cell or at least one neighboring cell; and means for reporting to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled.

[0036] In a thirty-second aspect, disclosed is an apparatus for a first network device. The apparatus may comprise: means for transmitting to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and means for transmitting to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[0037] In a thirty-third aspect, disclosed is an apparatus for a second network device. The apparatus may comprise: means for receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[0038] In a thirty-fourth aspect, disclosed is an apparatus for a terminal device. The apparatus may comprise: means for receiving from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; means for receiving from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; means for storing the information for at least one serving cell or at least one neighboring cell; and means for reporting to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or PUSCH or PUCCH message using MAC CE.

[0039] In a thirty-fifth aspect, disclosed is an apparatus for a first network device. The apparatus may comprise: means for transmitting to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and means for transmitting to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[0040] In a thirty-sixth aspect, disclosed is an apparatus for a second network device. The apparatus may comprise: means for receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0041] In a thirty-seventh aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a terminal device, may cause the apparatus at least to: receive from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device; receive from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; store the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and report to the network, the applicability of the machine learning functionality in the terminal device.

[0042] In a thirty-eighth aspect, a computer-readable medium is disclosed. The computer- readable medium may comprise program instructions that, when executed by an apparatus for a first network device, may cause the apparatus at least to: transmit to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and transmit to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[0043] In a thirty-ninth aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a second network device, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[0044] In a fortieth aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a terminal device, may cause the apparatus at least to: receive from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC CE cell switch message; receive from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; store the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and report to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0045] In a forty-first aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a first network device, may cause the apparatus at least to: transmit to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and transmit to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[0046] In a forty-second aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a second network device, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0047] In a forty-third aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a terminal device, may cause the apparatus at least to: receive from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device; receive from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; store the information for at least one serving cell or at least one neighboring cell; and report to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled.

[0048] In a forty-fourth aspect, a computer-readable medium is disclosed. The computer- readable medium may comprise program instructions that, when executed by an apparatus for a first network device, may cause the apparatus at least to: transmit to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and transmit to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[0049] In a forty-fifth aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a second network device, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[0050] In a forty-sixth aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a terminal device, may cause the apparatus at least to: receive from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; receive from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; store the information for at least one serving cell or at least one neighboring cell; and report to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or PUSCH or PUCCH message using MAC CE.

[0051] In a forty-seventh aspect, a computer-readable medium is disclosed. The computer- readable medium may comprise program instructions that, when executed by an apparatus for a first network device, may cause the apparatus at least to: transmit to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and transmit to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[0052] In a forty-eighth aspect, a computer-readable medium is disclosed. The computer-readable medium may comprise program instructions that, when executed by an apparatus for a second network device, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[0053] Other features and advantages of the example embodiments of the present disclosure will also be apparent from the following description of specific embodiments when read in conjunction with the accompanying drawings, which illustrate, by way of example, the principles of example embodiments of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Some example embodiments will now be described, by way of non-limiting examples, with reference to the accompanying drawings.

[0055] FIG. 1A shows an example sequence diagram according to the example embodiments of the present disclosure.

[0056] FIG. IB shows an example flow according to the example embodiments of the present disclosure.

[0057] FIG. 2 shows a flow chart illustrating an example method 200 for machine learning functionality according to the example embodiments of the present disclosure.

[0058] FIG. 3 shows a flow chart illustrating an example method 300 for machine learning functionality according to the example embodiments of the present disclosure.

[0059] FIG. 4 shows a flow chart illustrating an example method 400 for machine learning functionality according to the example embodiments of the present disclosure.

[0060] FIG. 5 shows a flow chart illustrating an example method 500 for machine learning functionality according to the example embodiments of the present disclosure.

[0061] FIG. 6 shows a flow chart illustrating an example method 600 for machine learning functionality according to the example embodiments of the present disclosure.

[0062] FIG. 7 shows a flow chart illustrating an example method 700 for machine learning functionality according to the example embodiments of the present disclosure.

[0063] FIG. 8 shows a flow chart illustrating an example method 800 for machine learning functionality according to the example embodiments of the present disclosure.

[0064] FIG. 9 shows a flow chart illustrating an example method 900 for machine learning functionality according to the example embodiments of the present disclosure.

[0065] FIG. 10 shows a flow chart illustrating an example method 1000 for machine learning functionality according to the example embodiments of the present disclosure.

[0066] FIG. 11 shows a flow chart illustrating an example method 1100 for machine learning functionality according to the example embodiments of the present disclosure.

[0067] FIG. 12 shows a flow chart illustrating an example method 1200 for machine learning functionality according to the example embodiments of the present disclosure.

[0068] FIG. 13 shows a flow chart illustrating an example method 1300 for machine learning functionality according to the example embodiments of the present disclosure.

[0069] FIG. 14 shows a block diagram illustrating an example device 1400 for machine learning functionality according to the example embodiments of the present disclosure.

[0070] FIG. 15 shows a block diagram illustrating an example device 1500 for machine learning functionality according to the example embodiments of the present disclosure.

[0071] FIG. 16 shows a block diagram illustrating an example device 1600 for machine learning functionality according to the example embodiments of the present disclosure.

[0072] FIG. 17 shows a block diagram illustrating an example apparatus 1700 for machine learning functionality according to the example embodiments of the present disclosure.

[0073] FIG. 18 shows a block diagram illustrating an example apparatus 1800 for machine learning functionality according to the example embodiments of the present disclosure.

[0074] FIG. 19 shows a block diagram illustrating an example apparatus 1900 for machine learning functionality according to the example embodiments of the present disclosure.

[0075] FIG. 20 shows a block diagram illustrating an example apparatus 2000 for machine learning functionality according to the example embodiments of the present disclosure.

[0076] FIG. 21 shows a block diagram illustrating an example apparatus 2100 for machine learning functionality according to the example embodiments of the present disclosure.

[0077] FIG. 22 shows a block diagram illustrating an example apparatus 2200 for machine learning functionality according to the example embodiments of the present disclosure.

[0078] FIG. 23 shows a block diagram illustrating an example apparatus 2300 for machine learning functionality according to the example embodiments of the present disclosure.

[0079] FIG. 24 shows a block diagram illustrating an example apparatus 2400 for machine learning functionality according to the example embodiments of the present disclosure.

[0080] FIG. 25 shows a block diagram illustrating an example apparatus 2500 for machine learning functionality according to the example embodiments of the present disclosure.

[0081] FIG. 26 shows a block diagram illustrating an example apparatus 2600 for machine learning functionality according to the example embodiments of the present disclosure.

[0082] FIG. 27 shows a block diagram illustrating an example apparatus 2700 for machine learning functionality according to the example embodiments of the present disclosure.

[0083] FIG. 28 shows a block diagram illustrating an example apparatus 2800 for machine learning functionality according to the example embodiments of the present disclosure.

[0084] Throughout the drawings, same or similar reference numbers indicate same or similar elements. A repetitive description on the same elements would be omitted. DETAILED DESCRIPTION

[0085] Herein below, some example embodiments are described in detail with reference to the accompanying drawings. The following description includes specific details for the purpose of providing a thorough understanding of various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, well known circuits, techniques and components are shown in block diagram form to avoid obscuring the described concepts and features.

[0086] Functionality may refer to an AEML-enabled Feature / Feature Group (FG) enabled by configuration(s), where configuration(s) is(are) supported based on conditions indicated by UE capability. Correspondingly, functionality-based life cycle management (LCM) operates based on, at least, one configuration of AI / ML-enabled Feature / FG or specific configurations of an AI / ML-enabled Feature / FG.

[0087] Functionality identification refers to a process / method of identifying an AI / ML functionality for the common understanding between the NW and the UE. A functionality in the present disclosure may refer to a ML enabled use-case functionality. In the present disclosure, a supported functionality may refer to a functionality that the network can configure in accordance to reported UE capabilities for the ML-enabled feature, an applicable functionality may refer to a radio configuration that can be applied at the UE to perform inference under network control, and an activated functionality may refer to a radio configuration which the UE applies to perform inference.

[0088] The AI / ML functionality may specify certain parameters defining, for example, the radio configuration under which the UE performs the phases of the AI / ML functionality: training, emulation, deployment, and inference. An activated functionality may include the inference phase while a deactivated functionality may include one, more, or all of the other phases but the inference. Accordingly, the deactivated functionality may run in the background but is not taken into use in the communication between the UE and the network. The UE may run multiple AI / ML functionalities in parallel, and they may be for the same purpose (e.g. beam management) but have different parameters, e.g. different radio configurations. In such a case, the AI / ML functionality providing the best results in the emulation may be activated. And in the present disclosure, the term “applicable functionality” can be interchangeable with a term “available functionality,” and the term “functionality” can be interchangeable with a term “model”, following the literature in the art.

[0089] In the functionality identification, if a UE performs applicable functionality reporting from the scratch after transitioning to radio resource control (RRC) connected mode from RRC idle or RRC inactive mode or after performing a cell switch, it would be inefficient. Example embodiments of the present disclosure provide solutions for indicating applicability of a machine learning functionality within the UE and to the network. According to the example embodiments of the present disclosure, the network may track the applicable / available machine learning functionality of an ML-enabled feature at the UE.

[0090] FIG. 1A shows an example sequence diagram according to the example embodiments of the present disclosure. Referring to FIG. 1 A, a UE 110 may represent any terminal device in a network, a first network device 150 may represent the network side configuring the UE 110 for storing and reporting applicability of a machine learning functionality, and a second network device 170 may represent the network side receiving the report on the applicability of the machine learning functionality. The first network device 150 and the second network device 170 may be a base station (BS), such as an Evolved Node B (eNB), a next Generation Node B (gNB), etc. From the perspective of the UE 110, the first network device 150 and the second network device 170 may be collectively referred to as the network.

[0091] In case the UE 110 reports the applicability of the machine learning functionality after transitioning to RRC connected mode from RRC idle or inactive mode or after performing an intra gNB handover (HO) or a cell measurement, if before and after the transition, the cell switch or the cell measurement the gNB serving the UE 110 does not change, the second network device 170 may be the same gNB as the first network device 150. In case the UE 110 reports the applicability of the machine learning functionality after performing an inter gNB handover or connecting to a cell served by a different gNB after the transition, in which cases before and after the transition or the cell switch the gNB serving the UE 110 changes, the second network device 170 may be different from the first network device 150. To encompass both scenarios and further similar scenarios, the first network device 150 and the second network 170 may be collectively denoted as the network to abstract the entity7 transmitting the configuration and receiving the report.

[0092] FIG. IB shows an example flow according to the example embodiments of the present disclosure. In FIG. IB, example messages which can carry the contents exchanged between the UE 110 and the network are shown, and the operations are simplified as the corresponding numerals.

[0093] Initially, at 112, the UE 110 is in RRC connected mode in a primary cell (PCell) X served by the first network device 150 and has reported UE capabilities to the first network device 150. The first network device 150 may generate a first configuration 152 for the UE 110 to store information on applicability of a machine learning functionality and a second configuration 154 for the UE 110 to report the applicability of the machine learning functionality. In some embodiments, the UE 110 may generate in the first configuration 152 a message, e.g. StoreAppConfConfig, that trigger the UE 110 to store the information on the applicability of the machine learning functionality.

[0094] In some embodiments, the first configuration 152 may comprise a set of conditions, which may also termed as criteria, and the second configuration 154 may have no conditions. In this case, for the storage, the UE 110 needs to determine whether at least one of the set of conditions is fulfilled, and in case at least one of the set of conditions is fulfilled, the UE 110 stores the information on the applicability of the machine learning functionality in the UE 110. And in this case, for the reporting, the UE 110 does not need to use the set of conditions.

[0095] In some embodiments, the second configuration 154 may comprise a set of conditions, and the first configuration 152 may have no conditions. In this case, for the storage, the UE 110 does not need to use the set of conditions. And in this case, for the reporting, the UE 110 needs to determine whether at least one of the set of conditions is fulfilled, and in case at least one of the set of conditions is fulfilled, the UE 110 reports the applicability of the machine learning functionality in the UE 110.

[0096] In some embodiments, the first configuration 152 may comprise a set of storage conditions, and the second configuration 154 may also comprise a set of reporting conditions. In this case, the UE 110 needs to determine whether at least one of the set of storage conditions is fulfilled, and in case at least one of the set of storage conditions is fulfilled, the UE 110 stores the information on the applicability of the machine learning functionality in the UE 110. And in this case, the UE 110 needs to determine whether at least one of the set of reporting conditions is fulfilled, and in case at least one of the set of reporting conditions is fulfilled, the UE 110 reports the applicability of the machine learning functionality in the UE 110. The sets of conditions for storing and for reporting may be fully or partially identical, or they may define mutually exclusive conditions.

[0097] The description of several embodiments is made in the context of a single AI / ML functionality. As described above, the UE may have multiple AI / ML functionalities with various applicability states. The embodiments described herein are directly applicable to determining, storing, and reporting the applicability for each of the multiple AI / ML functionalities. The reporting may be performed collectively, and for each of the multiple AI / ML functionalities, in a single message.

[0098] The network may use the reported applicability of machine learning functionalities of the UE 110 to configure new machine learning functionalities for the UE, to remove machine learning functionalities configured for the UE, and / or to activate the machine learning functionality for the UE 110. For example, if the UE cannot apply any of the configured machine learning functionalities for the beam management, the network may try to configure a new beam management machine learning functionality with relaxed parameters that the UE could be capable of applying. On the other hand, a machine learning functionality that has been inapplicable for the UE for a certain period of time may be dismantled.

[0099] The first network device 150 may transmit the first configuration 152 and the second configuration 154 to the UE 110. In some embodiments, the first configuration 152 and the second configuration 154 may be transmitted in a single message or in a single configuration, e g., RRC, message. In some embodiments, the first configuration 152 and the second configuration 154 may be transmitted in separate messages.

[00100] In some embodiments, the first configuration 152 may be transmitted via at least one of the following: RRC message, RRC release message, RRC reconfiguration message with or without handover command, or medium access control (MAC) control element (CE) cell switch message.

[00101] In some embodiments, the second configuration 154 may be transmitted via at least one of the following: RRC message, RRC release message, RRC reconfiguration message with or without handover command, or MAC-CE cell switch message.

[00102] Receiving the first configuration 152 and the second configuration 154, in case the first configuration 152 comprises a set of conditions, in an operation 114, the UE 110 may determine whether at least one of the set of conditions is fulfilled. In some embodiments, the set of conditions may comprise at least one of the following: the UE 110 is to transition to RRC idle mode from RRC connected mode; the UE 110 is to transition to RRC inactive mode from RRC connected mode; or the UE 110 is to perform a cell switch from PCell X to PCell A, e.g. due to UE mobility.

[00103] For example, the RRC reconfiguration message with handover command may indicate the UE 110 to perform the cell switch from PCell X to PCell A, the RRC release message may indicate the UE 110 to transition to RRC idle mode from RRC connected mode, or the RRC release message with suspend configuration may indicate the UE 110 to transition to RRC inactive mode from RRC connected mode.

[00104] In case at least one of the set of conditions in the first configuration 152 is fulfilled, or the first configuration 152 does not comprise condition for storage, in an operation 118, the UE 110 may store the information on the applicability of the machine learning functionality in the UE 110. In some embodiments, the UE 110 may store the information for at least one serving cell. In some embodiments, the UE 110 may store the information for at least one neighboring cell, e.g. a set of given neighboring cells. In some embodiments, the UE 110 may store the information for at least one serving cell and at least one neighboring cell. In some embodiments, the UE 110 may store the information per cell. In some embodiments, the UE 110 may store and identify the cell by at least one of the following: physical cell identity (PCI), frequency band, cell global identifier (CGI), or absolute radio frequency channel number (ARFCN).

[00105] In some embodiments, the UE 110 may store in the information an applicability indicator indicating that the machine learning functionality is currently applicable or not applicable in the UE 110. The indicator may be provided per machine learning functionality of the UE.

[00106] In some embodiments, the first network device 150 may request the UE 110, e.g. via the first configuration 152, the second configuration 154, or another message, to mark a machine learning functionality as inapplicable functionality if that machine learning functionality fails in performance monitoring. In some embodiments, the first network device 150 may further request the UE 110 to mark a cause value indicating a reason for the inapplicability. In some embodiments, the cause value may be generic, e.g. UE internal reason or network reason. In some embodiments, the cause value may be more specific. For example, the cause value may be, in case of the UE internal reason, e.g., high battery usage, overheating, attempt to save power, i.e. power savings, etc. or in case of network reason, e.g., resource constraints or network performance monitoring failure, etc.

[00107] In some embodiments, with or without such request from the network, in an operation 116, the UE 110 may monitor performance for the machine learning functionality via the routine cycle of the above-described phases of the AI / ML functionality. The performance monitoring may be a part of the testing phase. In the event of performance monitoring failure, the UE 110 may change the applicability indicator to indicate the machine learning functionality as inapplicable functionality, which may be performed, e.g., during the storing operation 118. In some embodiments, the UE 110 may further mark a cause value indicating the reason for the inapplicability, e.g., UE internal reason or network reason, or e.g. high battery usage, overheating, power savings, etc. in case of the UE internal reason, or network resource constraints, network performance monitoring failure, etc. in case of the network reason.

[00108] The performance monitoring failure may occur when the performance of a machine learning functionality falls below an acceptable level. The failure may correspond to a large difference between the output of a machine learning functionality and a reference value, a predefined low confidence or high uncertainty level with respect to the machine learning functionality output, or the resulting UE performance, e.g., throughput, received signal strength, etc. following an action determined by a machine learning functionality.

[00109] Then, in an operation 120, the UE 110 may transition to RRC idle or inactive mode, or may perform cell switch, e.g., handover, from PCell X to PCell A. In some embodiments, the UE 110 may transmit to the network, an acknowledgement (ACK) indicating a completion of the storage of the information. In some embodiments, the ACK may be transmitted via at least one of the following: MAC CE message, RRC reconfiguration complete message with or without a configuration for handover, RRC resume request message, RRC connection request message, or RRC connection setup complete message. For example, if the UE 110 is to transition from RRC inactive mode, RRC resume request message can be used, and if the UE 110 is to transition from RRC idle mode, RRC connection request message or RRC connection setup complete message can be used.

[00110] The UE 110 may transmit the ACK to the second network device 170. In case the transition or the cell switch does not result in the change of the gNB serving the UE 110, the second network device 170 is actually the first network device 150, although thereinafter the second network device 170 would be used. In case the transition or the cell switch results in that the gNB serving the UE 110 changes, the second network device 170, e.g., as a target gNB may be different from the first network device 150, e.g., as a source gNB.

[00111] In some embodiments, the second network device 170 may indicate using system information block (SIB), e.g., SIB1 to request UEs to set an indication if the UEs have stored applicable machine learning functionality from a previous connection to the network.

[00112] In some embodiments, the UE 110 may transmit to the second network device 170, a first indication 122 indicating existence of the stored information. The first indication 122 may be a single bit. In some embodiments, the first indication 122 may be transmitted via at least one of the following: RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00113] For example, if the UE 110 is to transition from RRC inactive mode, RRC resume request message can be used to carry the first indication 122, if the UE 110 is to transition from RRC idle mode, RRC connection request message or RRC connection setup complete message can be used to carry the first indication 122, RRC reconfiguration complete message without handover can be used to carry the first indication 122, e.g., during secondary cell addition, RRC reconfiguration complete message with handover complete indicator can be used to carry the first indication 122 e.g. in case of cell switch, or RRC measurement report message can be used to carry the first indication 122 when reporting measurement results Al, A2, etc.

[00114] In some embodiments, the second network device 170 may transmit to the UE 110, a second indication 172 indicating or requesting the UE 110 to report the applicability of the machine learning functionality in the UE 110. In some embodiments, the second indication 172 may be transmitted via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or physical downlink shared channel (PDSCH) request using MAC or downlink control information (DO). For example, RRC resume message may allow the UE 110 to transition to RRC connected mode from RRC inactive mode, and RRC setup message may allow the UE 110 to transition to RRC connected mode from RRC idle mode.

[00115] Currently the UE is in RRC connected mode. In case the second configuration 154 comprises a set of conditions, in an operation 130, the UE 110 may determine whether at least one of the set of conditions is fulfilled. In some embodiments, the set of conditions may comprise at least one of the following: (a) the UE 110 transitions to RRC connected mode from RRC idle mode; (b) the UE 110 transitions to RRC connected mode from RRC inactive mode; (c) the UE 110 performs a cell switch from PCell X to PCell A, e.g. due to UE mobility; or (d) a trigger of measurement performed by the UE 110 on a neighboring cell. For the condition (d), in some embodiments, the UE 110 may measure a given neighboring cell and determine whether that neighboring cell meets the reporting condition determined by other events, e.g., Al where serving of that neighboring cell becomes better than a threshold, or A2 where serving of that neighboring cell becomes worse than a threshold. The quality of said serving may be determined based on the measurements, e.g. signal quality measurements.

[00116] In some embodiments, the set of conditions for reporting may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively. For example, the bit mask could be 4 bit, and 1 bit for one condition, e.g., condition (a) is the leftmost bit and condition (d) is the rightmost bit. For example, value “1” may enable the corresponding condition, and value “0” may disable the corresponding condition. Thus, the network may configure one or multiple conditions for the UE 110, for example, the network may enable the condition (b) and disable the rest of the conditions.

[00117] In some embodiments, the second configuration 154 may comprise timers for the respective conditions in the set, for example, timer (a) to (d) for the reporting conditions (a) to (d), respectively. The durations of the timer (a) to (d) may be identical or different. For example, timer (a) can be set to 120 minutes, and timer (b) can be set to 180 minutes.

[00118] In some embodiments, the UE 110 may start at least one timer for an enabled condition upon receiving the second configuration 154, and the timer(s) for disabled condition(s) may not be started. Upon an expiry of the at least one timer for the enabled condition, the UE 110 may remove from the stored information, at least information associated with the enabled condition. In some embodiments, the UE 110 may remove the information associated with the enabled condition the tinier for which expires. In some embodiments, the UE 110 may remove the information associated with the enabled condition the timer for which expires and stored for one or more given serving cells or one or more neighboring cells. In some embodiments, the UE 110 may remove the information stored in the operation 118 upon the expiry of the started timer.

[00119] In case at least one of the set of conditions in the second configuration 154 is fulfilled, or the second configuration 154 does not comprise condition for reporting, in an operation 134, the UE 110 may report to the second network device 170, the applicability of the machine learning functionality in the UE 110, which is out of the stored information. In some embodiments, the UE 110 may report the applicability of the machine learning functionality in the UE 110 per cell with the respective cell identifier (ID), such as PCI, PCI and frequency, CGI, etc.

[00120] In some embodiments, the UE 110 may report the applicability of the machine learning functionality in the UE 110 via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, physical uplink shared channel (PUSCH) or physical uplink control channel (PUCCH) message using MAC CE.

[00121] For example, if the second indication 172 is received via RRC resume message, the UE 110 may report via RRC resume complete message. If the second indication 172 is received via RRC setup message, the UE 110 may report via RRC setup complete message. If the second indication 172 is received via RRC reconfiguration message, the UE 110 may report via RRC reconfiguration complete message. If the second indication 172 is received via UE capability enquiry message, the UE 110 may report via UE capability information message. If the second indication 172 is received via PDSCH request using MAC or DCI, the UE 110 may report via PUSCH or PUCCH message using MAC CE.

[00122] In some embodiments, the report on the applicability of the machine learning functionality in the UE 110 may be the stored information. In some embodiments, the report on the applicability of the machine learning functionality in the UE 110 may be a part of the stored information.

[00123] In some embodiments, the network may request the UE 110 to filter some information out, e.g., request the UE 110 to provide applicability for only those machine learning functionalities that were configured for inference to limit the size of the reporting. The network uses this report to configure and activate the machine learning functionality for the UE 110. Reporting those applicable functionalities that were also used for inference allows the network to immediately activate the applicable functionality that have been historically used for inference.

[00124] In some embodiments, the UE 110 may report the applicability indicator indicating that the machine learning functionality is currently applicable in the UE 110, and in response to the reporting of the applicability indicator, the second network device 170 may transmit to the UE 110 a message triggering activation of the machine learning functionality in the UE 110.

[00125] In some embodiments, the UE 110 may report the inapplicable functionality, in the event of the performance monitoring failure, to the second network device 170. In some embodiments, the UE 110 may report the inapplicable functionality with the cause value indicating the reason for the inapplicability.

[00126] Performance monitoring may allow the network to keep track periodically or based on an event of how an AI / ML enabled feature operates over time, and thus the network may select alternate configurations of the same AI / ML enabled feature when conditions change, e g., update of a beam codebook configuration, site configuration of antennas, etc. Receiving the report on the inapplicable functionality and the inapplicable functionality with the cause value, the network may react suitably. For example, the network can avoid selecting the inapplicable functionality for inference or can run a validation operation to ensure that the inapplicable functionality may become or remain applicable, which can provide a pre-emptive Fail-safe mechanism.

[00127] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the UE 110 comprises at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00128] The mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities may indicate a linking relationship between the supported functionalities, the applicable functionalities, and the activated functionalities. The supported functionalities comprise the applicable functionalities, and the applicable functionalities comprise the activated functionalities. In some embodiments, the mapping may be in the form of a table of ML functionalities and the statuses of the ML functionalities. In some embodiments, the status can be changed. For example, the supported ML functionalities can become the applicable functionalities, and the applicable functionalities can be activated. In some embodiments, the UE 110 is not allowed to indicate an applicable functionality, if the functionality is not (currently) supported by the UE 110 and / or the network, and the network is not allowed to activate a functionality that is reported as non-applicable by the UE 110.

[00129] The one or more radio configurations corresponding to the activated machine learning functionalities may be maintained up to date by the network as the network may reconfigure the machine learning functionalities of the UE 110 from time to time.

[00130] The supported functionalities may be tagged by the network assigned identifiers and may be determined by the network using the UE capabilities for the ML-enabled feature. For example, supported BM Case 1 with Set A / Set B dimensions and supported reference signal.

[00131] In some embodiments, the one or more identifiers may refer to a position in the list of UE capabilities feature set or feature set combinations for the ML-enabled use case. The one or more identifiers are assigned by the network and may be, e.g., a 16 bit identifier.

[00132] In some embodiments, the one or more identifiers may be indices in the one or more radio configurations corresponding to the applicable machine learning functionalities and / or the one or more radio configurations corresponding to the activated machine learning functionalities.

[00133] In some embodiments, the one or more identifiers may be checksums over the one or more radio configurations corresponding to the applicable machine learning functionalities and / or the one or more radio configurations corresponding to the activated machine learning functionalities.

[00134] In some embodiments, the network may assign the one or more identifiers to the UE 110 to refer to the elements in stored information. The one or more identifiers are generated by the first network device 150 and unambiguously refer to the elements in stored information. For example, the identifier could be an index into the elements in stored information, a 16, 24 or 32 checksum, e.g., in a manner of cyclic redundancy check (CRC), CRC16, CRC24 or CRC32, computed over the stored elements for each element in the stored information, or both the index and the checksum.

[00135] In case the second network device 170 as the target gNB is different from the first network device 150 as the source gNB, the second network device 170 may match the one or more identifiers reported from the UE 110 with those retrieved from the first network device 150. In some embodiments, the one or more identifiers are globally unique.

[00136] In some embodiments, the second network device 170 may receive from the first network device 150, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities, and verify the one or more identifiers in the report from the UE 110 based on the one or more identifiers received from the first network device 150.

[00137] As the first network device 150 may have transmitted to the second network device 170, the one or more identifiers, in the form of index and / or checksum, pointing to the stored machine learning functionalities, the second network device 170 may cross verify the one or more identifiers from the first network device 150 and those reported from the UE 110 to ensure validity of information on machine learning functionalities previously stored by the network.

[00138] In some embodiments, the first network device 150 request the UE 110 to encrypt the stored information so that the second network device 170 does not have to request the report from the UE 110 but verify the genuineness of the encryption.

[00139] An example of beam management radio configuration parameters according to the example embodiments of the present disclosure is shown below. The following Table 1 shows examples of some relevant potential configuration parameters.

[00140] Table 1 Configuration parameter Range and value Purpose SetB 8, 16, 32, 64 Input dimension to the beam management ML functionality Set A 32, 64, 128, 256 Predicted or output dimension of the beam management ML functionality Reference signal type (input) SSB, CSI-RS Type of reference signal to measure Reference signal type (output) SSB, CSI-RS Predicted RS signal type from input signal type Top K dimension 1,2,4 How many predicted beams Reporting quantity RSRP, differential RSRP Enable different type of reporting in uplink report Reporting type Option A, B A: Report always measured beam ID if predicted beam ID index is same as measured ID B: Report always predicted beam ID if predicted beam ID index is same as measured ID Time domain predicted window 200, 400, 600, 800, 1000 msec How long time window is being predicted 10

[00141] In the example embodiments of the present disclosure, the network may provide the UE with a set of radio configurations for a machine learning functionality. As an example, 6 configurations are provided to the UE: Conf 1 to Conf 6. The configurations differ from each other in the manner that the configuration parameter value would be set by the network.

[00142] Example list of supported functionalities, each Conf corresponding to a different functionality: Conf 1: Set B (16), Set A (64), Reference input signal type (SSB), Top K(2), Reporting quantity (differential RSRP), Reporting Type (B) -> spatial domain prediction Conf 2: Set B (8), Set A (64), Reference input signal type (SSB), Top K(2), Reporting quantity (RSRP), Reporting Type (B) -> spatial domain prediction Conf 3: Set B (8), Set A (64), Reference input signal type (SSB), Reference output signal type (CSI-RS), Top K(4), Reporting quantity (RSRP), Reporting Type (A) -> spatial domain prediction Conf 4: Set B (32), Set A (128), Reference input signal type (CSI-RS), Top K(4), Reporting quantity (differential RSRP), Reporting Type (B) -> spatial domain prediction Conf 5: Set B (64), Set A (256), Reference input signal type (CSI-RS), Reference output signal type (CSI-RS), Top K(2), Reporting quantity (differential RSRP), Reporting Type (B), Time domain predicted window (600 msec) -> time domain prediction Conf 6: Set B (64), Set A (256), Reference input signal type (CSI-RS), Reference output signal type (CSI-RS), Top K(2), Reporting quantity (differential RSRP), Reporting Type (B), Time domain predicted window (1000 msec) -> time domain prediction

[00143] Example list of applicable functionalities (in Cell X) Conf 1, Conf 3, Conf 4 and Conf 6 are signalled as applicable by the UE for serving cell X Conf 1, Conf 2 are signalled as applicable by the UE for neighboring cell Y Conf 2, Conf 4 are signalled as applicable by the UE for neighboring cell Z

[00144] Example list of activated functionalities (in Cell X) Conf 1 and Conf 6 are activated by the gNB in serving cell X Conf 2 is activated by the gNB in neighboring cell Y Conf 4 is activated by the gNB in neighboring cell Z

[00145] When the UE goes to RRC idle mode, RRC inactive mode, the UE stores the following: Applicable functionality for Cells X, Y and Z [{Conf 1, 3, 4 and 6}, {Conf 1, 2}, {Conf 2, 4}] Activated functionality for Cells X, Y and Z [{Conf 1 and 6}, {Conf 2}, {Conf 4}] Failed performance monitoring functionality Cells X, Y and Z [{Conf 6 (cause=battery, overheating)}, {Conf -}, {Conf -}] Configuration parameters for applicable and activated functionalities are stored with optional checksum.

[00146] “{Conf - J" means radio configuration parameters in the RRC message that were not causing an issue. Configuration 6 (Conf 6) was causing high battery consumption and also overheating at the UE, hence Conf 6 was considered as a case for performance monitoring failure.

[00147] The UE comes to Cell A from RRC inactive mode (bit mask bit was set to 1 indicating UE should report stored list of functionalities). One bit indication from the UE may indicate that the UE has stored information.

[00148] Reporting Applicable functionality for Cells X, Y and Z [{Conf 1, 3, 4 and 6}, {Conf 1, 2}, {Conf 2, 4}] Activated functionality for Cells X, Y and Z [{Conf 1 and 6}, {Conf 2}, {Conf 4}] Failed Performance Monitoring functionality Cells X, Y and Z [{Conf 6 (cause=battery, overheating)}, {Conf -}, {Conf -}] Configuration parameters for applicable and activated functionalities are stored by the gNB with optional checksum

[00149] The gNB can reuse the list and configurations Conf 1, 6, 2 or 4 if the gNB considers them applicable from gNB serving cell A point of view. The list can be updated by the UE, and the UE may require the gNB to request applicable functionalities as done previously for Cell X).

[00150] FIG. 2 shows a flow chart illustrating an example method 200 for machine learning functionality according to the example embodiments of the present disclosure. The example method 200 may be performed, for example, by an apparatus for a terminal device, such as the UE 110 above mentioned.

[00151] Referring to FIG. 2, the example method 200 may comprise: an operation 210 of receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device; an operation 220 of receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; an operation 230 of storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and an operation 240 of reporting to the network, the applicability of the machine learning functionality in the terminal device.

[00152] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00153] In some embodiments, the example method 200 may comprise: reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receiving, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00154] In some embodiments, the example method 200 may comprise: monitoring performance for the machine learning functionality; changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and reporting the inapplicable functionality to the network.

[00155] In some embodiments, the example method 200 may comprise: reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00156] In some embodiments, the example method 200 may comprise: transmitting to the network, a first indication indicating existence of the stored information.

[00157] In some embodiments, the example method 200 may comprise: receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and reporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00158] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00159] In some embodiments, the example method 200 may comprise: storing the information per cell, and reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00160] FIG. 3 shows a flow chart illustrating an example method 300 for machine learning functionality according to the example embodiments of the present disclosure. The example method 300 may be performed, for example, by an apparatus for a network device, such as the first network device 150 above mentioned.

[00161] Referring to FIG. 3, the example method 300 may comprise: an operation 310 of transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and an operation 320 of transmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[00162] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00163] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00164] FIG. 4 shows a flow chart illustrating an example method 400 for machine learning functionality according to the example embodiments of the present disclosure. The example method 400 may be performed, for example, by an apparatus for a network device, such as the second network device 170 above mentioned.

[00165] Referring to FIG. 4, the example method 400 may comprise: an operation 410 of receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[00166] In some embodiments, the example method 400 may comprise: receiving from the terminal device, a first indication indicating existence of the stored information.

[00167] In some embodiments, the example method 400 may comprise: transmitting to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and receiving from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00168] In some embodiments, the example method 400 may comprise: receiving an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and transmitting, to the terminal device in response to said applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00169] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00170] In some embodiments, the example method 400 may comprise: receiving from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and verifying the one or more identifiers in the report from the terminal device based on the one or more identifiers received from the first network device.

[00171] FIG. 5 shows a flow chart illustrating an example method 500 for machine learning functionality according to the example embodiments of the present disclosure. The example method 500 may be performed, for example, by an apparatus for a terminal device, such as the UE 110 above mentioned.

[00172] Referring to FIG. 5, the example method 500 may comprise: an operation 510 of receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC CE cell switch message; an operation 520 of receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; an operation 530 of storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and an operation 540 of reporting to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00173] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00174] In some embodiments, the example method 500 may comprise: reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receiving, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00175] In some embodiments, the example method 500 may comprise: monitoring performance for the machine learning functionality; changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and reporting the inapplicable functionality to the network.

[00176] In some embodiments, the example method 500 may comprise: reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00177] In some embodiments, the example method 500 may comprise: transmitting to the network, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00178] In some embodiments, the example method 500 may comprise: receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and reporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00179] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00180] In some embodiments, the example method 500 may comprise: storing the information per cell, and reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00181] FIG. 6 shows a flow chart illustrating an example method 600 for machine learning functionality according to the example embodiments of the present disclosure. The example method 600 may be performed, for example, by an apparatus for a network device, such as the first network device 150 above mentioned.

[00182] Referring to FIG. 6, the example method 600 may comprise: an operation 610 of transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and an operation 620 of transmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[00183] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00184] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00185] FIG. 7 shows a flow chart illustrating an example method 700 for machine learning functionality according to the example embodiments of the present disclosure. The example method 700 may be performed, for example, by an apparatus for a network device, such as the second network device 170 above mentioned.

[00186] Referring to FIG. 7, the example method 700 may comprise: an operation 710 of receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00187] In some embodiments, the example method 700 may comprise: receiving from the terminal device, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00188] In some embodiments, the example method 700 may comprise: transmitting to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and receiving from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00189] In some embodiments, the example method 700 may comprise: receiving an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and transmitting, to the terminal device in response to said applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00190] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00191] In some embodiments, the example method 700 may comprise: receiving from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and verifying the one or more identifiers in the report from the terminal device based on the one or more identifiers received from the first network device.

[00192] FIG. 8 shows a flow chart illustrating an example method 800 for machine learning functionality according to the example embodiments of the present disclosure. The example method 800 may be performed, for example, by an apparatus for a terminal device, such as the UE 110 above mentioned.

[00193] Referring to FIG. 8, the example method 800 may comprise: an operation 810 of receiving from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device; an operation 820 of receiving from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; an operation 830 of storing the information for at least one serving cell or at least one neighboring cell; and an operation 840 of reporting to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled.

[00194] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00195] In some embodiments, the example method 800 may comprise: reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receiving, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00196] In some embodiments, the example method 800 may comprise: monitoring performance for the machine learning functionality; changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and reporting the inapplicable functionality to the network.

[00197] In some embodiments, the example method 800 may comprise: reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00198] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00199] In some embodiments, the second configuration may comprise timers for the respective conditions in the set, and the example method 800 may comprise: starting at least one timer for an enabled condition upon receiving the second configuration; and removing from the information, at least information associated with the enabled condition upon an expiry of the at least one timer for the enabled condition.

[00200] In some embodiments, the example method 800 may comprise: transmitting to the network, a first indication indicating existence of the stored information.

[00201] In some embodiments, the example method 800 may comprise: receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and reporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00202] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00203] In some embodiments, the example method 800 may comprise: storing the information per cell, and reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00204] FIG. 9 shows a flow chart illustrating an example method 900 for machine learning functionality according to the example embodiments of the present disclosure. The example method 900 may be performed, for example, by an apparatus for a network device, such as the first network device 150 above mentioned.

[00205] Referring to FIG. 9, the example method 900 may comprise: an operation 910 of transmitting to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and an operation 920 of transmitting to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[00206] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00207] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00208] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00209] FIG. 10 shows a flow chart illustrating an example method 1000 for machine learning functionality according to the example embodiments of the present disclosure. The example method 1000 may be performed, for example, by an apparatus for a network device, such as the second network device 170 above mentioned.

[00210] Referring to FIG. 10, the example method 1000 may comprise: an operation 1010 of receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[00211] In some embodiments, the example method 1000 may comprise: receiving from the terminal device, a first indication indicating existence of the stored information.

[00212] In some embodiments, the example method 1000 may comprise: transmitting to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and receiving from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00213] In some embodiments, the example method 1000 may comprise: receiving an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and transmitting, to the terminal device in response to said received applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00214] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00215] In some embodiments, the example method 1000 may comprise: receiving from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and verifying the one or more identifiers reported from the terminal device based on the one or more identifiers received from the first network device.

[00216] FIG. 11 shows a flow chart illustrating an example method 1100 for machine learning functionality according to the example embodiments of the present disclosure. The example method 1100 may be performed, for example, by an apparatus for a terminal device, such as the UE 110 above mentioned.

[00217] Referring to FIG. 11, the example method 1100 may comprise: an operation 1110 of receiving from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; an operation 1120 of receiving from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; an operation 1130 of storing the information for at least one serving cell or at least one neighboring cell; and an operation 1140 of reporting to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or PUSCH or PUCCH message using MAC CE.

[00218] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00219] In some embodiments, the example method 1100 may comprise: reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receiving from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00220] In some embodiments, the example method 1100 may comprise: monitoring performance for the machine learning functionality; changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and reporting the inapplicable functionality to the network.

[00221] In some embodiments, the example method 1100 may comprise: reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00222] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00223] In some embodiments, the second configuration may comprise timers for the respective conditions in the set, and the example method 1100 may comprise: starting at least one timer for an enabled condition upon receiving the second configuration; and removing from the information, at least information associated with the enabled condition upon an expiry of the at least one timer for the enabled condition.

[00224] In some embodiments, the example method 1100 may comprise: transmitting to the network, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00225] In some embodiments, the example method 1100 may comprise: receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or physical downlink shared channel, PDSCH, request using MAC or downlink control information, DO; and reporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00226] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00227] In some embodiments, the example method 1100 may comprise: storing the information per cell, and to report the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00228] FIG. 12 shows a flow chart illustrating an example method 1200 for machine learning functionality according to the example embodiments of the present disclosure. The example method 1200 may be performed, for example, by an apparatus for a network device, such as the first network device 150 above mentioned.

[00229] Referring to FIG. 12, the example method 1200 may comprise: an operation 1210 of transmitting to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and an operation 1220 of transmitting to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[00230] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00231] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00232] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00233] FIG. 13 shows a flow chart illustrating an example method 1300 for machine learning functionality according to the example embodiments of the present disclosure. The example method 1300 may be performed, for example, by an apparatus for a network device, such as the second network device 170 above mentioned.

[00234] Referring to FIG. 13, the example method 1300 may comprise: an operation 1310 of receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00235] In some embodiments, the example method 1300 may comprise: receiving from the terminal device, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00236] In some embodiments, the example method 1300 may comprise: transmitting to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and receiving from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00237] In some embodiments, the example method 1300 may comprise: receiving an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and transmitting to the terminal device in response to said received applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00238] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00239] In some embodiments, the example method 1300 may comprise: receiving from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and verifying the one or more identifiers reported from the terminal device based on the one or more identifiers received from the first network device.

[00240] FIG. 14 shows a block diagram illustrating an example device 1400 for machine learning functionality according to the example embodiments of the present disclosure. The device, for example, may be at least part of an apparatus for a terminal device, such as the UE 110 in the above examples.

[00241] As shown in FIG. 14, the example device 1400 may include at least one processor 1410 and at least one memory 1420 that may store instructions 1430. The instructions 1430, when executed by the at least one processor 1410, may cause the device 1400 at least to perform the example method 200, 500, 800, or 1100 described above.

[00242] In various example embodiments, the at least one processor 1410 in the example device 400 may include, but is not limited to, at least one hardware processor, including at least one microprocessor such as a central processing unit (CPU), a portion of at least one hardware processor, and any other suitable dedicated processor such as those developed based on for example Field Programmable Gate Array (FPGA) and Application Specific Integrated Circuit (ASIC). Further, the at least one processor 410 may also include at least one other circuitry or element not shown in FIG. 14.

[00243] In various example embodiments, the at least one memory 1420 in the example device 1400 may include at least one storage medium in various forms, such as a transitory memory and / or a non-transitory memory. The transitory memory may include, but is not limited to, for example, a random-access memory (RAM), a cache, and so on. The non-transitory memory may include, but is not limited to, for example, a read-only memory (ROM), a hard disk, a flash memory, and so on. The term “non-transitory,” as used herein, is a limitation of the medium itself (i.e., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM). Further, the at least memory 420 may include, but is not limited to, an electric, a magnetic, an optical, an electromagnetic, an infrared, or a semiconductor system, apparatus, or device or any combination of the above.

[00244] Further, in various example embodiments, the example device 1400 may also include at least one other circuitry, element, and interface, for example at least one I / O interface, at least one antenna element, and the like.

[00245] In various example embodiments, the circuitries, parts, elements, and interfaces in the example device 1400, including the at least one processor 1410 and the at least one memory 1420, may be coupled together via any suitable connections including, but is not limited to, buses, crossbars, wiring and / or wireless lines, in any suitable ways, for example electrically, magnetically, optically, electromagnetically, and the like.

[00246] It is understood that the structure of the device on the side of the UE 110 is not limited to the above example device 1400.

[00247] FIG. 15 shows a block diagram illustrating an example device 1500 for machine learning functionality according to the example embodiments of the present disclosure. The device, for example, may be at least part of an apparatus for a network device, such as the first network device 150 in the above examples.

[00248] As shown in FIG. 15, the example device 1500 may include at least one processor 1510 and at least one memory 1520 that may store instructions 1530. The instructions 1530, when executed by the at least one processor 1510, may cause the device 1500 at least to perform the example method 300, 600, 900, or 1200 described above.

[00249] In various example embodiments, the at least one processor 1510 in the example device 1500 may include, but is not limited to, at least one hardware processor, including at least one microprocessor such as a central processing unit (CPU), a portion of at least one hardware processor, and any other suitable dedicated processor such as those developed based on for example Field Programmable Gate Array (FPGA) and Application Specific Integrated Circuit (ASIC). Further, the at least one processor 1510 may also include at least one other circuitry or element not shown in FIG. 15.

[00250] In various example embodiments, the at least one memory 1520 in the example device 1500 may include at least one storage medium in various forms, such as a transitory memory and / or a non-transitory memory. The transitory memory may include, but is not limited to, for example, a random-access memory (RAM), a cache, and so on. The non-transitory memory may include, but is not limited to, for example, a read-only memory (ROM), a hard disk, a flash memory, and so on. The term “non-transitory,” as used herein, is a limitation of the medium itself (i.e., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM). Further, the at least memory 1520 may include, but is not limited to, an electric, a magnetic, an optical, an electromagnetic, an infrared, or a semiconductor system, apparatus, or device or any combination of the above.

[00251] Further, in various example embodiments, the example device 1500 may also include at least one other circuitry, element, and interface, for example at least one VO interface, at least one antenna element, and the like.

[00252] In various example embodiments, the circuitries, parts, elements, and interfaces in the example device 1500, including the at least one processor 1510 and the at least one memory 1520, may be coupled together via any suitable connections including, but is not limited to, buses, crossbars, wiring and / or wireless lines, in any suitable ways, for example electrically, magnetically, optically, electromagnetically, and the like.

[00253] It is understood that the structure of the device on the side of the first network device 150 is not limited to the above example device 1500.

[00254] FIG. 16 shows a block diagram illustrating an example device 1600 for machine learning functionality according to the example embodiments of the present disclosure. The device, for example, may be at least part of an apparatus for a network device, such as the second network device 170 in the above examples.

[00255] As shown in FIG. 16, the example device 1600 may include at least one processor 1610 and at least one memory 1620 that may store instructions 1630. The instructions 1630, when executed by the at least one processor 1610, may cause the device 1600 at least to perform the example method 400, 700, 1000, or 1300 described above.

[00256] In various example embodiments, the at least one processor 1610 in the example device 1600 may include, but is not limited to, at least one hardware processor, including at least one microprocessor such as a central processing unit (CPU), a portion of at least one hardware processor, and any other suitable dedicated processor such as those developed based on for example Field Programmable Gate Array (FPGA) and Application Specific Integrated Circuit (ASIC). Further, the at least one processor 1610 may also include at least one other circuitry or element not shown in FIG. 16.

[00257] In various example embodiments, the at least one memory 1620 in the example device 1600 may include at least one storage medium in various forms, such as a transitory memory and / or a non-transitory memory. The transitory memory may include, but is not limited to, for example, a random-access memory (RAM), a cache, and so on. The non-transitory memory may include, but is not limited to, for example, a read-only memory (ROM), a hard disk, a flash memory, and so on. The term “non-transitory,” as used herein, is a limitation of the medium itself (i.e., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM). Further, the at least memory 1620 may include, but is not limited to, an electric, a magnetic, an optical, an electromagnetic, an infrared, or a semiconductor system, apparatus, or device or any combination of the above.

[00258] Further, in various example embodiments, the example device 1600 may also include at least one other circuitry, element, and interface, for example at least one VO interface, at least one antenna element, and the like.

[00259] In various example embodiments, the circuitries, parts, elements, and interfaces in the example device 1600, including the at least one processor 1610 and the at least one memory 1620, may be coupled together via any suitable connections including, but is not limited to, buses, crossbars, wiring and / or wireless lines, in any suitable ways, for example electrically, magnetically, optically, electromagnetically, and the like.

[00260] It is understood that the structure of the device on the side of the second network device 170 is not limited to the above example device 1600.

[00261] FIG. 17 shows a block diagram illustrating an example apparatus 1700 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a terminal device, such as the UE 110 in the above examples.

[00262] As shown in FIG. 17, the example apparatus 1700 may comprise: means 1710 for receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device; means 1720 for receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; means 1730 for storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and means 1740 for reporting to the network, the applicability of the machine learning functionality in the terminal device.

[00263] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00264] In some embodiments, the example apparatus 1700 may comprise: means for reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and means for receiving, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00265] In some embodiments, the example apparatus 1700 may comprise: means for monitoring performance for the machine learning functionality; means for changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and means for reporting the inapplicable functionality to the network.

[00266] In some embodiments, the example apparatus 1700 may comprise: means for reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00267] In some embodiments, the example apparatus 1700 may comprise: means for transmitting to the network, a first indication indicating existence of the stored information.

[00268] In some embodiments, the example apparatus 1700 may comprise: means for receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and means for reporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00269] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00270] In some embodiments, the example apparatus 1700 may comprise: means for storing the information per cell, and means for reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00271] In some example embodiments, examples of means in the example apparatus 1700 may include circuitries. For example, an example of means 1710 may include a circuitry configured to perform the operation 210 of the example method 200, an example of means 1720 may include a circuitry configured to perform the operation 220 of the example method 200, an example of means 1730 may include a circuitry configured to perform the operation 230 of the example method 200, and an example of means 1740 may include a circuitry configured to perform the operation 240 of the example method 200.

[00272] The example apparatus 1700 may further include means comprising circuitry configured to perform the example method 200. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00273] FIG. 18 shows a block diagram illustrating an example apparatus 1800 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a network device, such as the first network device 150 in the above examples.

[00274] As shown in FIG. 18, the example apparatus 1800 may comprise: means 1810 for transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and means 1820 for transmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[00275] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00276] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00277] In some example embodiments, examples of means in the example apparatus 1800 may include circuitries. For example, an example of means 1810 may include a circuitry configured to perform the operation 310 of the example method 200, and an example of means 1820 may include a circuitry configured to perform the operation 320 of the example method 300.

[00278] The example apparatus 1800 may further include means comprising circuitry configured to perform the example method 300. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00279] FIG. 19 shows a block diagram illustrating an example apparatus 1900 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a network device, such as the second network device 170 in the above examples.

[00280] As shown in FIG. 19, the example apparatus 1900 may comprise: means 1910 for receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[00281] In some embodiments, the example apparatus 1900 may comprise: means for receiving from the terminal device, a first indication indicating existence of the stored information.

[00282] In some embodiments, the example apparatus 1900 may comprise: means for transmitting to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and means for receiving from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00283] In some embodiments, the example apparatus 1900 may comprise: means for receiving an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and means for transmitting, to the terminal device in response to said applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00284] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00285] In some embodiments, the example apparatus 1900 may comprise: means for receiving from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and means for verifying the one or more identifiers in the report from the terminal device based on the one or more identifiers received from the first network device.

[00286] In some example embodiments, examples of means in the example apparatus 1900 may include circuitries. For example, an example of means 1910 may include a circuitry configured to perform the operation 410 of the example method 400.

[00287] The example apparatus 1900 may further include means comprising circuitry configured to perform the example method 400. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00288] FIG. 20 shows a block diagram illustrating an example apparatus 2000 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a terminal device, such as the UE 110 in the above examples.

[00289] As shown in FIG. 20, the example apparatus 2000 may comprise: means 2010 for receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC CE cell switch message; means 2020 for receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; means 2030 for storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and means 2040 for reporting to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00290] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00291] In some embodiments, the example apparatus 2000 may comprise: means for reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and means for receiving, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00292] In some embodiments, the example apparatus 2000 may comprise: means for monitoring performance for the machine learning functionality; means for changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and means for reporting the inapplicable functionality to the network.

[00293] In some embodiments, the example apparatus 2000 may comprise: means for reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00294] In some embodiments, the example apparatus 2000 may comprise: means for transmitting to the network, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00295] In some embodiments, the example apparatus 2000 may comprise: means for receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and means for reporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00296] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00297] In some embodiments, the example apparatus 2000 may comprise: means for storing the information per cell, and means for reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00298] In some example embodiments, examples of means in the example apparatus 2000 may include circuitries. For example, an example of means 2010 may include a circuitry configured to perform the operation 510 of the example method 500, an example of means 2020 may include a circuitry configured to perform the operation 520 of the example method 500, an example of means 2030 may include a circuitry configured to perform the operation 530 of the example method 500, and an example of means 2040 may include a circuitry configured to perform the operation 540 of the example method 500.

[00299] The example apparatus 2000 may further include means comprising circuitry configured to perform the example method 500. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00300] FIG. 21 shows a block diagram illustrating an example apparatus 2100 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a network device, such as the first network device 150 in the above examples.

[00301] As shown in FIG. 21, the example apparatus 2100 may comprise: means 2110 for transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and means 2120 for transmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[00302] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00303] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00304] In some example embodiments, examples of means in the example apparatus 2100 may include circuitries. For example, an example of means 2110 may include a circuitry configured to perform the operation 610 of the example method 600, and an example of means 2120 may include a circuitry configured to perform the operation 620 of the example method 600.

[00305] The example apparatus 2100 may further include means comprising circuitry configured to perform the example method 600. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00306] FIG. 22 shows a block diagram illustrating an example apparatus 2200 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a network device, such as the second network device 170 in the above examples.

[00307] As shown in FIG. 22, the example apparatus 2200 may comprise: means 2210 for receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00308] In some embodiments, the example apparatus 2200 may comprise: means for receiving from the terminal device, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00309] In some embodiments, the example apparatus 2200 may comprise: means for transmitting to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and means for receiving from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00310] In some embodiments, the example apparatus 2200 may comprise: means for receiving an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and means for transmitting, to the terminal device in response to said applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00311] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00312] In some embodiments, the example apparatus 2200 may comprise: means for receiving from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and means for verifying the one or more identifiers in the report from the terminal device based on the one or more identifiers received from the first network device.

[00313] In some example embodiments, examples of means in the example apparatus 2200 may include circuitries. For example, an example of means 2210 may include a circuitry configured to perform the operation 710 of the example method 700.

[00314] The example apparatus 2200 may further include means comprising circuitry configured to perform the example method 700. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00315] FIG. 23 shows a block diagram illustrating an example apparatus 2300 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a terminal device, such as the UE 110 in the above examples.

[00316] As shown in FIG. 23, the example apparatus 2300 may comprise: means 2310 for receiving from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device; means 2320 for receiving from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; means 2330 for storing the information for at least one serving cell or at least one neighboring cell; and means 2340 for reporting to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled.

[00317] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00318] In some embodiments, the example apparatus 2300 may comprise: means for reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and means for receiving, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00319] In some embodiments, the example apparatus 2300 may comprise: means for monitoring performance for the machine learning functionality; means for changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and means for reporting the inapplicable functionality to the network.

[00320] In some embodiments, the example apparatus 2300 may comprise: means for reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00321] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00322] In some embodiments, the second configuration may comprise timers for the respective conditions in the set, and the example apparatus 2300 may comprise: means for starting at least one timer for an enabled condition upon receiving the second configuration; and means for removing from the information, at least information associated with the enabled condition upon an expiry of the at least one timer for the enabled condition.

[00323] In some embodiments, the example apparatus 2300 may comprise: means for transmitting to the network, a first indication indicating existence of the stored information.

[00324] In some embodiments, the example apparatus 2300 may comprise: means for receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and means for reporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00325] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00326] In some embodiments, the example apparatus 2300 may comprise: means for storing the information per cell, and means for reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00327] In some example embodiments, examples of means in the example apparatus 2300 may include circuitries. For example, an example of means 2310 may include a circuitry configured to perform the operation 810 of the example method 800, an example of means 2320 may include a circuitry configured to perform the operation 820 of the example method 800, an example of means 2330 may include a circuitry configured to perform the operation 830 of the example method 800, and an example of means 2340 may include a circuitry configured to perform the operation 840 of the example method 800.

[00328] The example apparatus 2300 may further include means comprising circuitry configured to perform the example method 800. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00329] FIG. 24 shows a block diagram illustrating an example apparatus 2400 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a network device, such as the first network device 150 in the above examples.

[00330] As shown in FIG. 24, the example apparatus 2400 may comprise: means 2410 for transmitting to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and means 2420 for transmitting to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[00331] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00332] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00333] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00334] In some example embodiments, examples of means in the example apparatus 2400 may include circuitries. For example, an example of means 2410 may include a circuitry configured to perform the operation 910 of the example method 900, and an example of means 2420 may include a circuitry configured to perform the operation 920 of the example method 900.

[00335] The example apparatus 2400 may further include means comprising circuitry configured to perform the example method 900. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00336] FIG. 25 shows a block diagram illustrating an example apparatus 2500 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a network device, such as the second network device 170 in the above examples.

[00337] As shown in FIG. 25, the example apparatus 2500 may comprise: means 2510 for receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[00338] In some embodiments, the example apparatus 2500 may comprise: means for receiving from the terminal device, a first indication indicating existence of the stored information.

[00339] In some embodiments, the example apparatus 2500 may comprise: means for transmitting to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and means for receiving from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00340] In some embodiments, the example apparatus 2500 may comprise: means for receiving an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and means for transmitting, to the terminal device in response to said received applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00341] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00342] In some embodiments, the example apparatus 2500 may comprise: means for receiving from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and means for verifying the one or more identifiers reported from the terminal device based on the one or more identifiers received from the first network device.

[00343] In some example embodiments, examples of means in the example apparatus 2500 may include circuitries. For example, an example of means 2510 may include a circuitry configured to perform the operation 1010 of the example method 1000.

[00344] The example apparatus 2500 may further include means comprising circuitry configured to perform the example method 1000. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00345] FIG. 26 shows a block diagram illustrating an example apparatus 2600 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a terminal device, such as the UE 110 in the above examples.

[00346] As shown in FIG. 26, the example apparatus 2600 may comprise: means 2610 for receiving from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; means 2620 for receiving from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; means 2630 for storing the information for at least one serving cell or at least one neighboring cell; and means 2640 for reporting to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or PUSCH or PUCCH message using MAC CE.

[00347] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00348] In some embodiments, the example apparatus 2600 may comprise: means for reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and means for receiving from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00349] In some embodiments, the example apparatus 2600 may comprise: means for monitoring performance for the machine learning functionality; changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and means for reporting the inapplicable functionality to the network.

[00350] In some embodiments, the example apparatus 2600 may comprise: means for reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00351] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00352] In some embodiments, the second configuration may comprise timers for the respective conditions in the set, and the example apparatus 2600 may comprise: means for starting at least one timer for an enabled condition upon receiving the second configuration; and means for removing from the information, at least information associated with the enabled condition upon an expiry of the at least one timer for the enabled condition.

[00353] In some embodiments, the example apparatus 2600 may comprise: means for transmitting to the network, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00354] In some embodiments, the example apparatus 2600 may comprise: means for receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or physical downlink shared channel, PDSCH, request using MAC or downlink control information, DCI; and means for reporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00355] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00356] In some embodiments, the example apparatus 2600 may comprise: means for storing the information per cell, and means for reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00357] In some example embodiments, examples of means in the example apparatus 2600 may include circuitries. For example, an example of means 2610 may include a circuitry configured to perform the operation 1110 of the example method 1100, an example of means 2620 may include a circuitry configured to perform the operation 1120 of the example method 1100, an example of means 2630 may include a circuitry configured to perform the operation 1130 of the example method 1100, and an example of means 2640 may include a circuitry configured to perform the operation 1140 of the example method 1100.

[00358] The example apparatus 2600 may further include means comprising circuitry configured to perform the example method 1100. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00359] FIG. 27 shows a block diagram illustrating an example apparatus 2700 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a network device, such as the first network device 150 in the above examples.

[00360] As shown in FIG. 27, the example apparatus 2700 may comprise: means 2710 for transmitting to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and means 2720 for transmitting to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[00361] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00362] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00363] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00364] In some example embodiments, examples of means in the example apparatus 2700 may include circuitries. For example, an example of means 2710 may include a circuitry configured to perform the operation 1210 of the example method 1200, and an example of means 2720 may include a circuitry configured to perform the operation 1220 of the example method 1200.

[00365] The example apparatus 2700 may further include means comprising circuitry configured to perform the example method 1200. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00366] FIG. 28 shows a block diagram illustrating an example apparatus 2800 for machine learning functionality according to the example embodiments of the present disclosure. The apparatus, for example, may be at least part of a network device, such as the second network device 170 in the above examples.

[00367] As shown in FIG. 28, the example apparatus 2800 may comprise: means 2810 for receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00368] In some embodiments, the example apparatus 2800 may comprise: means for receiving from the terminal device, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00369] In some embodiments, the example apparatus 2800 may comprise: means for transmitting to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and means for receiving from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00370] In some embodiments, the example apparatus 2800 may comprise: means for receiving an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and means for transmitting to the terminal device in response to said received applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00371] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00372] In some embodiments, the example apparatus 2800 may comprise: means for receiving from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and means for verifying the one or more identifiers reported from the terminal device based on the one or more identifiers received from the first network device.

[00373] In some example embodiments, examples of means in the example apparatus 2800 may include circuitries. For example, an example of means 2810 may include a circuitry configured to perform the operation 1310 of the example method 1300.

[00374] The example apparatus 2800 may further include means comprising circuitry configured to perform the example method 1300. In some example embodiments, examples of means may also include software modules and any other suitable function entities.

[00375] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a terminal device, such as the UE 110 in the above examples, may cause the apparatus at least to: receive from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device; receive from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; store the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and report to the network, the applicability of the machine learning functionality in the terminal device.

[00376] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00377] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: report an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receive, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00378] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: monitor performance for the machine learning functionality; change the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and report the inapplicable functionality to the network.

[00379] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: report the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00380] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: transmit to the network, a first indication indicating existence of the stored information.

[00381] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and report to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00382] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00383] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: store the information per cell, and report the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00384] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a network device, such as the first network device 150 in the above examples, may cause the apparatus at least to: transmit to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and transmit to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[00385] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00386] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00387] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a network device, such as the second network device 170 in the above examples, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[00388] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the terminal device, a first indication indicating existence of the stored information.

[00389] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: transmit to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and receive from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00390] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and transmit, to the terminal device in response to said applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00391] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00392] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and verify the one or more identifiers in the report from the terminal device based on the one or more identifiers received from the first network device.

[00393] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a terminal device, such as the UE 110 in the above examples, may cause the apparatus at least to: receive from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC CE cell switch message; receive from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; store the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; and report to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00394] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00395] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: report an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receive, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00396] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: monitor performance for the machine learning functionality; change the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and report the inapplicable functionality to the network.

[00397] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: report the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00398] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: transmit to the network, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00399] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and report to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00400] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00401] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: store the information per cell, and report the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00402] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a network device, such as the first network device 150 in the above examples, may cause the apparatus at least to: transmit to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and transmit to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[00403] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device is to transition to RRC idle mode from RRC connected mode; the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

[00404] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00405] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a network device, such as the second network device 170 in the above examples, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00406] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the terminal device, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00407] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: transmit to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and receive from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00408] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and transmit, to the terminal device in response to said applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00409] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00410] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and verify the one or more identifiers in the report from the terminal device based on the one or more identifiers received from the first network device.

[00411] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a terminal device, such as the UE 110 in the above examples, may cause the apparatus at least to: receive from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device; receive from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device; store the information for at least one serving cell or at least one neighboring cell; and report to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled.

[00412] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00413] In some embodiments, the example method 800 may comprise: reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receiving, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00414] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: monitor performance for the machine learning functionality; change the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and report the inapplicable functionality to the network.

[00415] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: report the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00416] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00417] In some embodiments, the second configuration may comprise timers for the respective conditions in the set, and the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: start at least one timer for an enabled condition upon receiving the second configuration; and remove from the information, at least information associated with the enabled condition upon an expiry of the at least one timer for the enabled condition.

[00418] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: transmit to the network, a first indication indicating existence of the stored information.

[00419] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and report to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device

[00420] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00421] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: store the information per cell, and report the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00422] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a network device, such as the first network device 150 in the above examples, may cause the apparatus at least to: transmit to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell; and transmit to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device.

[00423] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00424] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00425] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00426] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a network device, such as the second network device 170 in the above examples, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network.

[00427] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the terminal device, a first indication indicating existence of the stored information.

[00428] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: transmit to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device; and receive from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00429] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and transmit, to the terminal device in response to said received applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00430] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00431] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and verify the one or more identifiers reported from the terminal device based on the one or more identifiers received from the first network device.

[00432] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a terminal device, such as the UE 110 in the above examples, may cause the apparatus at least to: receive from a network, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; receive from the network, a second configuration comprising a set of conditions for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; store the information for at least one serving cell or at least one neighboring cell; and report to the network, the applicability of the machine learning functionality in the terminal device, in case at least one of the set of conditions is fulfilled, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or PUSCH or PUCCH message using MAC CE.

[00433] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00434] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: report an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receiving from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00435] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: monitor performance for the machine learning functionality; change the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; and report the inapplicable functionality to the network.

[00436] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: report the inapplicable functionality with a cause value indicating a reason for the inapplicability.

[00437] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00438] In some embodiments, the second configuration may comprise timers for the respective conditions in the set, and the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: start at least one timer for an enabled condition upon receiving the second configuration; and remove from the information, at least information associated with the enabled condition upon an expiry of the at least one timer for the enabled condition.

[00439] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: transmit to the network, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00440] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, orPDSCH request using MAC or DCI; and report to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

[00441] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00442] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: store the information per cell, and reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

[00443] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a network device, such as the first network device 150 in the above examples, may cause the apparatus at least to: transmit to a terminal device, a first configuration for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message; and transmit to the terminal device, a second configuration comprising a set of conditions for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

[00444] In some embodiments, the set of conditions may comprise at least one of the following: the terminal device transitions to RRC connected mode from RRC idle mode; the terminal device transitions to RRC connected mode from RRC inactive mode; the terminal device performs a cell switch; or a trigger of measurement performed by the terminal device on a neighboring cell.

[00445] In some embodiments, the set of conditions may be in a form of bit mask comprising bits enabling or disabling the conditions in the set, respectively.

[00446] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00447] The example embodiments of the present disclosure also provide a computer-readable medium comprising program instructions that, when executed by an apparatus for a network device, such as the second network device 170 in the above examples, may cause the apparatus at least to: receive from a terminal device, reported applicability of a machine learning functionality in the terminal device, wherein the reported applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, UE capability information message, or PUSCH or PUCCH message using MAC CE.

[00448] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the terminal device, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

[00449] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: transmit to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or PDSCH request using MAC or DCI; and receive from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

[00450] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and transmit to the terminal device in response to said received applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

[00451] In some embodiments, at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device may comprise at least one of the following: a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities; one or more radio configurations corresponding to the applicable machine learning functionalities; one or more radio configurations corresponding to the activated machine learning functionalities; one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; a supported but inapplicable machine learning functionality; the supported but inapplicable machine learning functionality along with a cause for the inapplicability; or an identifier of at least one network device of the network.

[00452] In some embodiments, the computer-readable medium may include instructions that, when executed by the apparatus, may cause the apparatus to: receive from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; and verify the one or more identifiers reported from the terminal device based on the one or more identifiers received from the first network device.

[00453] As used herein, “at least one of the following: ” and “at least one of ” and similar wording, where the list of two or more elements are joined by “and” or “or”, mean at least any one of the elements, or at least any two or more of the elements, or at least all the elements.

[00454] The term “terminal device” refers to any end device that may be capable of wireless communication. By way of example rather than limitation, a terminal device may also be referred to as a communication device, user equipment (UE), a Subscriber Station (SS), a Portable Subscriber Station, a Mobile Station (MS), or an Access Terminal (AT). The terminal device may include, but is not limited to, a mobile phone, a cellular phone, a smart phone, voice over IP (VoIP) phones, wireless local loop phones, a tablet, a wearable terminal device, a personal digital assistant (PDA), portable computers, desktop computer, image capture terminal devices such as digital cameras, gaming terminal devices, music storage and playback appliances, vehiclemounted wireless terminal devices, wireless endpoints, mobile stations, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), USB dongles, smart devices, wireless customer-premises equipment (CPE), an Internet of Things (loT) device, a watch or other wearable, a head-mounted display (HMD), a vehicle, a drone, a medical device and applications (e.g., remote surgery), an industrial device and applications (e.g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain contexts), a consumer electronics device, a device operating on commercial and / or industrial wireless networks, and the like. The terminal device may also correspond to a Mobile Termination (MT) part of an IAB node (e.g., a relay node). In the above description, the terms “terminal device”, “communication device”, “terminal”, “user equipment” and “UE” may be used interchangeably.

[00455] The term “circuitry” throughout this disclosure may refer to one or more or all of the following: (a) hardware-only circuit implementations (such as implementations in only analog and / or digital circuitry); (b) combinations of hardware circuits and software, such as (as applicable) (i) a combination of analog and / or digital hardware circuit(s) with software / firmware and (ii) any portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a mobile phone or server, to perform various functions); and (c) hardware circuit(s) and or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation. This definition of circuitry applies to one or all uses of this term in this disclosure, including in any claims. As a further example, as used in this disclosure, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in server, a cellular network device, or other computing or network device.

[00456] Another example embodiment may relate to computer program codes or instructions which may cause an apparatus to perform at least the respective methods described above. Another example embodiment may be related to a computer-readable medium having such computer program codes or instructions stored thereon. In some embodiments, such a computer-readable medium may include at least one storage medium in various forms such as a volatile memory and / or a non-volatile memory. The volatile memory may include, but is not limited to, for example, a RAM, a cache, and so on. The non-volatile memory may include, but is not limited to, a ROM, a hard disk, a flash memory, and so on. The non-volatile memory may also include, but is not limited to, an electric, a magnetic, an optical, an electromagnetic, an infrared, or a semiconductor system, apparatus, or device or any combination of the above.

[00457] Unless the context clearly requires otherwise, throughout the description and the claims, the words “comprise,” “comprising,” and the like are to be construed in an inclusive sense, as opposed to an exclusive or exhaustive sense; that is to say, in the sense of “including, but is not limited to.” The word “coupled”, as generally used herein, refers to two or more elements that may be either directly connected, or connected by way of one or more intermediate elements. Likewise, the word “connected”, as generally used herein, refers to two or more elements that may be either directly connected, or connected by way of one or more intermediate elements. Additionally, the words “herein,” “above,” “below,” and words of similar import, when used in this application, shall refer to this application as a whole and not to any particular portions of this application. Where the context permits, words in the description using the singular or plural number may also include the plural or singular number respectively. The word “or” in reference to a list of two or more items, that word covers all of the following interpretations of the word: any of the items in the list, all of the items in the list, and any combination of the items in the list.

[00458] Moreover, conditional language used herein, such as, among others, “can,” “could,” “might,” “may,” “e.g.,” “for example,” “such as” and the like, unless specifically stated otherwise, or otherwise understood within the context as used, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements and / or states. Thus, such conditional language is not generally intended to imply that features, elements and / or states are in any way required for one or more embodiments or that one or more embodiments necessarily include logic for deciding, with or without author input or prompting, whether these features, elements and / or states are included or are to be performed in any particular embodiment.

[00459] As used herein, the term "determine / determining" (and grammatical variants thereof) can include, not least: calculating, computing, processing, deriving, measuring, investigating, looking up (for example, looking up in a table, a database or another data structure), ascertaining and the like. Also, "determining" can include receiving (for example, receiving information), accessing (for example, accessing data in a memory), obtaining and the like. Also, "determine / determining" can include resolving, selecting, choosing, establishing, and the like.

[00460] While some embodiments have been described, these embodiments have been presented by way of example, and are not intended to limit the scope of the disclosure. Indeed, the apparatus, methods, and systems described herein may be embodied in a variety of other forms; furthermore, various omissions, substitutions, and changes in the form of the methods and systems described herein may be made without departing from the spirit of the disclosure. For example, while blocks are presented in a given arrangement, alternative embodiments may perform similar functionalities with different components and / or circuit topologies, and some blocks may be deleted, moved, added, subdivided, combined, and / or modified. At least one of these blocks may be implemented in a variety of different ways. The order of these blocks may also be changed. Any suitable combination of the elements and actions of the some embodiments described above can be combined to provide further embodiments. The accompanying claims and their equivalents are intended to cover such forms or modifications as would fall within the scope and spirit of the disclosure.

[00461] Abbreviations used in the description and / or in the figures are defined as follows: 3GPP 3rd Generation Partnership Project ACK acknowledgement AI artificial intelligence ARFCN absolute radio frequency channel number BS base station BM beam management CGI cell global identifier 5 CRC cyclic redundancy check DCI downlink control information DL downlink eNB Evolved Node B FG Feature Group 10 gNB next Generation Node B HO handover ID identifier LCM life cycle management MAC medium access control 15 MAC CE MAC control element MT. machine learning NW network PCell primary cell PCI physical cell identity 20 PDSCH physical downlink shared channel PUCCH physical uplink control channel PUSCH physical uplink shared channel RRC radio resource control RS reference signal 25 CSI-RS channel state information RS RSRP reference signal receiving power Rx receive SIB system information block PBCH physical broadcast channel 30 SSB synchronization signal block Tx synchronization signal and PBCH block transmit user equipment

Claims

1. An apparatus for a terminal device, comprising:at least one processor; andat least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to:receive from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device, via at least one of the following: radio resource control, RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC, control element, CE, cell switch message;receive from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message;store the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; andreport to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or physical uplink shared channel, PUSCH, or physical uplink control channel, PUCCH, message using MAC CE.

2. The apparatus of claim 1, wherein the set of conditions comprises at least one of the following:the terminal device is to transition to RRC idle mode from RRC connected mode;the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

3. The apparatus of claim 1 or 2, wherein the apparatus is configured to report anapplicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and to receive, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

4. The apparatus of claim 3, wherein the apparatus is configured to:monitor performance for the machine learning functionality;change the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; andreport the inapplicable functionality to the network.

5. The apparatus of claim 4, wherein the apparatus is configured to: report the inapplicable functionality with a cause value indicating a reason for the inapplicability.

6. The apparatus of any of claims 1 to 5, wherein the apparatus is configured to:transmit to the network, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

7. The apparatus of claim 6, wherein the apparatus is configured to:receive from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or physical downlink shared channel, PDSCH, request using MAC or downlink control information, DCI; andreport to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

8. The apparatus of any of claims 1 to 7, wherein at least one of the stored information orthe report on the applicability of the machine learning functionality in the terminal device comprises at least one of the following:a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities;one or more radio configurations corresponding to the applicable machine learning functionalities;one or more radio configurations corresponding to the activated machine learning functionalities;one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities;a supported but inapplicable machine learning functionality;the supported but inapplicable machine learning functionality along with a cause for the inapplicability; oran identifier of at least one network device of the network.

9. The apparatus of any of claims 1 to 8, wherein the apparatus is configured to store the information per cell, and to report the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

10. An apparatus for a first network device of a network, comprising:at least one processor; andat least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to:transmit to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: radio resource control, RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC, control element, CE, cell switch message; andtransmit to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

11. The apparatus of claim 10, wherein the set of conditions comprises at least one of the following:the terminal device is to transition to RRC idle mode from RRC connected mode;the terminal device is to transition to RRC inactive mode from RRC connected mode; or the terminal device is to perform a cell switch.

12. The apparatus of claim 10 or 11, wherein at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device comprises at least one of the following:a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities;one or more radio configurations corresponding to the applicable machine learning functionalities;one or more radio configurations corresponding to the activated machine learning functionalities;one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities;a supported but inapplicable machine learning functionality;the supported but inapplicable machine learning functionality along with a cause for the inapplicability; oran identifier of at least one network device of the network.

13. An apparatus for a second network device of a network, comprising:at least one processor; andat least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to:receive from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: radio resource control, RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or physical uplink shared channel, PUSCH, or physical uplink control channel, PUCCH, message using medium access control, MAC, control element, CE.

14. The apparatus of claim 13, wherein the apparatus is configured to:receive from the terminal device, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

15. The apparatus of claim 13 or 14, wherein the apparatus is configured to:transmit to the terminal device, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or physical downlink shared channel, PDSCH, request using MAC or downlink control information, DO; andreceive from the terminal device, the reported applicability of the machine learning functionality in the terminal device, in response to the second indication.

16. The apparatus of any of claims 13 to 15, wherein the apparatus is configured to receive an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and to transmit, to the terminal device in response to saidapplicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

17. The apparatus of any of claims 13 to 16, wherein at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device comprises at least one of the following:a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities of the one or more functionalities;one or more radio configurations corresponding to the applicable machine learning functionalities;one or more radio configurations corresponding to the activated machine learning functionalities;one or more identifiers identifying at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities;a supported but inapplicable machine learning functionality;the supported but inapplicable machine learning functionality along with a cause for the inapplicability; oran identifier of at least one network device of the network.

18. The apparatus of claim 17, wherein the apparatus is configured to:receive from the first network device, one or more identifiers pointing to at least one of the following: the supported machine learning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities; andverify the one or more identifiers in the report from the terminal device based on the one or more identifiers received from the first network device.

19. A method performed by an apparatus for a terminal device, comprising:receiving from a network, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in theterminal device, via at least one of the following: radio resource control, RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC, control element, CE, cell switch message;receiving from the network, a second configuration for the terminal device to report to the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message;storing the information for at least one serving cell or at least one neighboring cell in case at least one of the set of conditions is fulfilled; andreporting to the network, the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handover complete indicator, user equipment, UE, capability information message, or physical uplink shared channel, PUSCH, or physical uplink control channel, PUCCH, message using MAC CE.

20. The method of claim 19, wherein the set of conditions comprises at least one of the following:the terminal device is to transition to RRC idle mode from RRC connected mode;the terminal device is to transition to RRC inactive mode from RRC connected mode; orthe terminal device is to perform a cell switch.

21. The method of claim 19 or 20, comprising: reporting an applicability indicator indicating that the machine learning functionality is currently applicable in the terminal device, and receiving, from the network in response to said reporting the applicability indicator, a message triggering activation of the machine learning functionality in the terminal device.

22. The method of claim 21, comprising:monitoring performance for the machine learning functionality;changing the applicability indicator to indicate the machine learning functionality as inapplicable functionality in the event of performance monitoring failure; andreporting the inapplicable functionality to the network.

23. The method of claim 22, comprising: reporting the inapplicable functionality with a cause value indicating a reason for the inapplicability.

24. The method of any of claims 19 to 23, comprising:transmitting to the network, a first indication indicating existence of the stored information, via at least one of the following: RRC, RRC resume request message, RRC connection request message, RRC connection setup complete message, RRC reconfiguration complete message with or without handover complete indicator, or RRC measurement report message.

25. The method of claim 24, comprising:receiving from the network, a second indication indicating the terminal device to report the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC resume message, RRC setup message, RRC reconfiguration message, UE capability enquiry message, or physical downlink shared channel, PDSCH, request using MAC or downlink control information, DCI; andreporting to the network in response to the second indication, the applicability of the machine learning functionality in the terminal device.

26. The method of any of claims 19 to 25, wherein at least one of the stored information or the report on the applicability of the machine learning functionality in the terminal device comprises at least one of the following:a mapping among supported machine learning functionalities, applicable machine learning functionalities, or activated machine learning functionalities;one or more radio configurations corresponding to the applicable machine learning functionalities;one or more radio configurations corresponding to the activated machine learning functionalities;one or more identifiers identifying at least one of the following: the supported machinelearning functionalities, the applicable machine learning functionalities or the activated machine learning functionalities;a supported but inapplicable machine learning functionality;the supported but inapplicable machine learning functionality along with a cause for the inapplicability; oran identifier of at least one network device of the network.

27. The method of any of claims 19 to 26, comprising: storing the information per cell, and reporting the applicability of the machine learning functionality in the terminal device per cell with the respective cell identifier.

28. A method performed by an apparatus for a first network device of a network, comprising: transmitting to a terminal device, a first configuration comprising a set of conditions for the terminal device to store information on applicability of a machine learning functionality in the terminal device for at least one serving cell or at least one neighboring cell, via at least one of the following: radio resource control, RRC, RRC release message, RRC reconfiguration message with or without handover command, or medium access control, MAC, control element, CE, cell switch message; andtransmitting to the terminal device, a second configuration for the terminal device to report to a second network device of the network the applicability of the machine learning functionality in the terminal device, via at least one of the following: RRC, RRC release message, RRC reconfiguration message with or without handover command, or MAC CE cell switch message.

29. A method performed by an apparatus for a second network device of a network, comprising:receiving from a terminal device, reported applicability of a machine learning functionality in the terminal device wherein the applicability is out of information stored by the terminal device according to a first configuration configured by a first network device of the network, via at least one of the following: radio resource control, RRC, RRC resume complete message, RRC setup complete message, RRC reconfiguration complete message with or without handovercomplete indicator, user equipment, UE, capability information message, or physical uplink shared channel, PUSCH, or physical uplink control channel, PUCCH, message using medium access control, MAC, control element, CE.5

Images