Ai / ml non-connected operation

Abstract

A user device, UE, for a wireless communication network, is configured or preconfigured with at least one Artificial Intelligence / Machine Learning model, AI / ML model, or at least one AI / ML functionality for performing one or more tasks. The UE is in a connected state, the UE is to receive an indication from the wireless communication network for activating one or more connected AI / ML models or functionalities, the one or more connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities, and if the UE is in a non-connected state, the UE is to use one or more non-connected AI / ML models or functionalities, the one or more non-connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities.

Description

[0001] AI / ML NON-CONNECTED OPERATION

[0002] Description

[0003] The present invention relates to the field of wireless communication systems or networks, more specifically a use of at least one Artificial Intelligence / Machine Learning model, AI / ML model, or at least one AI / ML functionality in a wireless communication system for performing one or more tasks. Embodiments of the present invention concern improvements and enhancements to the use of an AI / ML model or an AI / ML functionality in a user device of a wireless communication system which is in a non-connected state.

[0004] Fig. 1 is a schematic representation of an example of a terrestrial wireless network 100 including, as is shown in Fig. 1 (A), the core network, CN, 102 and one or more radio access networks RANi, RAN2, ... RANN. Fig. 1 (B) is a schematic representation of an example of a radio access network RANnthat may include one or more base stations gNBi to gNBs, each serving a specific area surrounding the base station schematically represented by respective cells IO61 to IO65. The base stations are provided to serve users within a cell. The one or more base stations may serve users in licensed and / or unlicensed bands. The term base station, BS, refers to a gNB in 5G networks, an eNB in UMTS / LTE / LTE-A / LTE- A Pro, or just a BS in other mobile communication standards. The BS may also comprise of integrated access and backhaul, IAB, nodes, e.g., an IAB Donor and / or IAB Node, consisting of a central unit, CU, as well as of a distributed unit, DU, and / or containing IAB- MTs including IAB mobile termination, MT. The term base station may refer to an access point, AP, in any of the WiFi standards, e.g., belonging to the IEEE 802.11-familiy. A user may be a stationary device or a mobile device. The wireless communication system may also be accessed by mobile or stationary loT devices which connect to a base station or to a user. The mobile or stationary devices may include physical devices, ground based vehicles, such as robots or cars, aerial vehicles, such as manned or unmanned aerial vehicles, UAVs, the latter also referred to as drones, buildings and other items or devices having embedded therein electronics, software, sensors, actuators, or the like as well as network connectivity that enables these devices to collect and exchange data across an existing network infrastructure. Fig. 1 (B) shows an exemplary view of five cells, however, the RANn may include more or less such cells, and RANnmay also include only one base station. Fig. 1 (B) shows two users UE1 and UE2, also referred to as user device or user equipment, that are in cell IO62 and that are served by base station gNB2. Another user UE3 is shown in cell IO64 which is served by base station gNB4. The arrows IO81, IO82 and IO83 schematically represent uplink / downlink connections for transmitting data from a user UE1, UE2 and UE3 to the base stations gNB2, gNB4 or for transmitting data from the base stations gNB2, gNB4 to the users UE1, UE2, UE3. This may be realized on licensed bands or on unlicensed bands. Further, Fig. 1 (B) shows two further devices 110i and HO2 in cell IO64, like loT devices, which may be stationary or mobile devices. The device 110i accesses the wireless communication system via the base station gNB4 to receive and transmit data as schematically represented by arrow 112i. The device HO2 accesses the wireless communication system via the user UE3 as is schematically represented by arrow 1122. The respective base station gNBi to gNBs may be connected to the core network 102, e.g., via the S1 interface, via respective backhaul links 114i to 114s, which are schematically represented in Fig. 1 (B) by the arrows pointing to “core”. The core network 102 may be connected to one or more external networks. The external network may be the Internet, or a private network, such as an Intranet or any other type of campus networks, e.g., a private WiFi communication system or a 4G or 5G mobile communication system. Further, some or all of the respective base station gNBi to gNBs may be connected, e.g., via the S1 or X2 interface or the XN interface in NR, with each other via respective backhaul links 116i to 1165, which are schematically represented in Fig. 1 (B) by the arrows pointing to “gNBs”. A sidelink channel allows direct communication between UEs, also referred to as device-to- device, D2D, communication. The sidelink interface in 3GPP is named PC5. Note, that the term user equipment, UE, or user device may also refer to a station, STA, as used in any of the WiFi standards, e.g., belonging to the IEEE 802.11-familiy.

[0005] For data transmission a physical resource grid may be used. The physical resource grid may comprise a set of resource elements to which various physical channels and physical signals are mapped. For example, the physical channels may include the physical downlink, uplink and sidelink shared channels, PDSCH, PLISCH, PSSCH, carrying user specific data, also referred to as downlink, uplink and sidelink payload data, the physical broadcast channel, PBCH, and the physical sidelink broadcast channel, PSBCH, carrying for example a master information block, MIB, and one or more system information blocks, SIBs, one or more sidelink information blocks, SLIBs, if supported, the physical downlink, uplink and sidelink control channels, PDCCH, GC-PDCCH, PLICCH, PSSCH, carrying for example the downlink control information, DCI, the uplink control information, UCI, and the sidelink control information, SCI, and physical sidelink feedback channels, PSFCH, carrying PC5 feedback responses. The sidelink interface may support a 2-stage SCI which refers to a first control region containing some parts of the SCI, also referred to as the 1st-stage SCI, and optionally, a second control region which contains a second part of control information, also referred to as the 2nd-stage SCI.

[0006] For the uplink, the physical channels may further include the physical random-access channel, PRACH or RACH, used by UEs for accessing the network once a UE synchronized and obtained the MIB and SIB. The physical signals may comprise reference signals or symbols, RS, synchronization signals and the like. The resource grid may comprise a frame or radio frame having a certain duration in the time domain and having a given bandwidth in the frequency domain. The frame may have a certain number of subframes of a predefined length, e.g., 1ms. Each subframe may include one or more slots of 12 or 14 OFDM symbols depending on the cyclic prefix, CP, length. A frame may also have a smaller number of OFDM symbols, e.g., when utilizing shortened transmission time intervals, sTTI, or a mini-slot / non-slot-based frame structure comprising just a few OFDM symbols.

[0007] The wireless communication system may be any single-tone or multicarrier system using frequency-division multiplexing, like the orthogonal frequency-division multiplexing, OFDM, system, the orthogonal frequency-division multiple access, OFDMA, system, or any other Inverse Fast Fourier Transform, IFFT, based signal with or without Cyclic Prefix, CP, e.g., Discrete Fourier Transform-spread-OFDM, DFT-s-OFDM. Other waveforms, like non- orthogonal waveforms for multiple access, e.g., filter-bank multicarrier, FBMC, generalized frequency division multiplexing, GFDM, or universal filtered multi carrier, LIFMC, may be used. The wireless communication system may operate, e.g., in accordance with 3GPPs LTE, LTE-Advanced, LTE-Advanced Pro, or the 5G or 5G-Advanced or 6G or 3GPPs NR, New Radio, or within LTE-ll, LTE Unlicensed or NR-U, New Radio Unlicensed, which is specified within the LTE and within NR specifications.

[0008] The wireless network or communication system depicted in Fig. 1 may be a heterogeneous network having distinct overlaid networks, e.g., a network of macro cells with each macro cell including a macro base station, like base station gNBi to gNBs, and a network of small cell base stations, not shown in Fig. 1 , like femto or pico base stations. In addition to the above-described terrestrial wireless network also non-terrestrial wireless communication networks, NTN, exist including spaceborne transceivers, like satellites, and / or airborne transceivers, like unmanned aircraft systems. The non-terrestrial wireless communication network or system may operate in a similar way as the terrestrial system described above with reference to Fig. 1 , for example in accordance with the LTE-Advanced Pro or 5G or 5G-Advanced or NR, New Radio, or a possible future 6G radio system. In mobile communication networks, for example in a network like that described above with reference to Fig. 1 , like an LTE or 5G / NR network, there may be UEs that communicate directly with each other over one or more sidelink, SL, channels, e.g., using the PC5 / PC3 interface or WiFi direct. UEs that communicate directly with each other over the sidelink may include vehicles communicating directly with other vehicles, V2V communication, vehicles communicating with other entities of the wireless communication network, V2X communication, for example roadside units, RSUs, roadside entities, like traffic lights, traffic signs, or pedestrians. An RSU may have a functionality of a BS or of a UE, depending on the specific network configuration. Other UEs may not be vehicular related UEs and may comprise any of the above-mentioned devices. Such devices may also communicate directly with each other, D2D communication, using the SL channels.

[0009] In mobile communication networks, for example in a network like that described above with reference to Fig. 1 , like an LTE or 5G / NR network, a UE may be in one of several states. These states include a connected state, like an RRC_CONNECTED state, where the UE has a connection with the wireless communication network, and at least one non-connected state where the UE has no connection with the wireless communication network, like an RRCJDLE state, or where the UE has a connection with the wireless communication network but it is not active, like an RRCJNACTIVE state.

[0010] For example, the RRCJDLE state is a state where the UE has no RRC connection with the network. This means that the UE does not have any radio resources allocated to it, and it is not aware of its cell identity or location. The UE may only receive paging messages from the network, which are used to notify the UE of incoming calls or data. The UE may also initiate a connection request by sending a random access preamble to the network.

[0011] The UE may enter the RRCJDLE state from the RRCJNACTIVE state, when there is no data or signaling activity for a long time. The UE may also enter the RRCJDLE state from the RRC_CONNECTED state, when the network releases the RRC connection due to various reasons, such as a handover failure, a radio link failure, load balancing or a network congestion. The RRCJDLE state has the lowest energy consumption for the UE, but it also has the highest latency for data transmission, as it requires a full connection setup before the data may be sent or received. For example, the RRCJNACTIVE state is a state that was introduced in 5G NR to improve the efficiency and performance of the network and the user devices or user equipments, UEs. It is a state where the UE has an RRC connection with the network, but it is not active. This means that the UE does not need to send or receive any signaling messages or data packets, and it may save power by turning off some of its radio functions. The UE may also move within a certain area, called the RAN-Based Notification Area, RNA, without informing the network about its location. This reduces the signaling overhead and the latency for the UE and the network.

[0012] The UE may enter the RRC_IN ACTIVE state from the RRC_CONNECTED state, when there is no data or signaling activity for a certain period of time. The network may configure the parameters for this transition, such as the timer value and the RNA size. The UE may also enter the RRC NACTIVE state from the RRC DLE state, when it receives a paging message from the network that contains an RRC connection setup request. The UE may exit the RRC_IN ACTIVE state and resume the RRC connection when there is a need for data or signaling transmission. This may be triggered by either the UE or the network, depending on the scenario.

[0013] The RRCJNACTIVE state is one of the features that makes 5G NR more flexible and adaptable to different use cases and scenarios, such as massive loT, enhanced mobile broadband, ultra-reliable low-latency communication, and non-terrestrial networks. It enables 5G NR to support more devices with longer battery life, lower latency, and higher mobility.

[0014] In mobile communication networks, for example in a network like that described above with reference to Fig. 1 , like an LTE or 5G / NR network, non-connected data transmissions may be implemented. For example, the 5G small data transmission is a feature that enables user equipments, UEs, to send or receive infrequent and small amounts of data without requiring a full connection setup. This may reduce the latency and the power consumption of UEs, as well as the signaling overhead and the physical resource usage of the network. 5G small data transmission is one of the features introduced in 3GPP Release 17 for 5G NR. It enables 5G to support new use cases and verticals, such as industrial automation, public safety, non-terrestrial networks and non-public networks.

[0015] There are different solutions for 5G small data transmission, depending on the state of the UE and the type of data. (1) One solution is called Early Data Transmission, EDT, which allows the UE to transmit data during the Random Access procedure. This is useful for UEs that are in the RRC DLE state or in the RRCJNACTIVE state, and need to send small data such as sensor readings or status updates. EDT may achieve a latency reduction of up to 50% compared to a conventional data transfer after the RRC connection setup.

[0016] (2) Another solution is called Small Data Transmission, SDT, which allows the UE to transmit or receive data in the RRC NACTIVE state without changing its state. This is useful for UEs that have infrequent and small data transmissions, such as smart meters or wearables. SDT may be based on Random Access Channel (RACH) or Configured Grant (CG). RACH-based SDT uses a dedicated RACH resource for the small data transmission, while CG-based SDT uses a pre-configured uplink or downlink grant for the small data transmission. CG-based SDT may achieve higher gains in terms of UE power efficiency, latency reduction, signaling overhead reduction and physical resource saving than RACH-based SDT.

[0017] In a wireless network or communication system Artificial Intelligence (Al) and Machine Learning (ML) may be employed for certain tasks. For example, according to 3GPP, AI / ML techniques and data analytics may be incorporated into the 5G system design for supporting certain tasks, e.g., for supporting network automation, data collection for various network functions, network energy savings, resource allocation and scheduling optimizations, network slicing management, load balancing, mobility optimizations, AI / ML-based services, AI / ML for the new radio (NR) air interface. For example, when considering the NR air interface, AI / ML models may be employed for one or more of the following use cases:

[0018] Channel State Information (CSI): For example, AI / ML may be used for a timedomain prediction.

[0019] Beam Management (BM): For example, AI / ML may be used for a spatial and temporal prediction.

[0020] Positioning: For example, a direct AI / ML positioning approach (e.g., fingerprinting) and an AI / ML assisted positioning approach (e.g., the output of the AI / ML model inference is an additional measurement and / or an enhancement of an existing measurement) may be implemented.

[0021] Handover: For example, the AI / ML may predict when the UE should handover to which cell to reduce unnecessary handovers.

[0022] The AI / ML model may be running at one of the two sides or at both sides of the communication link, e.g., at the gNB or the network-side, e.g., CN, and / or at the UE. Some AI / ML models may not be specified and left up to implementation, while others, e.g., enabling AI / ML for the air interface, need to be specified.

[0023] It is noted that the information in the above section is only for enhancing the understanding of the background of the invention and, therefore, it may contain information that does not form prior art that is already known to a person of ordinary skill in the art.

[0024] Starting from the above, there may be a need for improvements or enhancements to the use of AI / ML models in a wireless communication system or network.

[0025] Embodiments of the present invention are now described in further detail with reference to the accompanying drawings:

[0026] Fig. 1 (A)-(B) illustrate a wireless communication network, wherein Fig. 1 (A) is a schematic representation of an example of a terrestrial wireless network, and Fig. 1 (B) is a schematic representation of an example of a radio access network, RAN;

[0027] Fig. 2 is a schematic representation of a wireless communication system including a transmitter, like a base station, and one or more receivers, like user devices, UEs, implementing embodiments of the present invention;

[0028] Fig. 3 illustrates a user device, UE, according to an embodiment of the present invention;

[0029] Fig. 4 illustrates an embodiment of the present invention used during a random access, RACH, procedure;

[0030] Fig. 5 illustrates a use of the same non-connected AI / ML models and functionalities for the respective non-connected states according to an embodiment of the present invention;

[0031] Fig. 6 illustrates a use of different non-connected AI / ML models and functionalities for the respective non-connected states according to another embodiment of the present invention; Fig. 7 illustrates a state diagram including an additional non-connected state allowing the use of non-connected AI / ML models or functionalities according to yet another embodiment of the present invention; and

[0032] Fig. 8 illustrates an example of a computer system on which units or modules as well as the steps of the methods described in accordance with the inventive approach may execute.

[0033] Embodiments of the present invention are now described in more detail with reference to the accompanying drawings, in which the same or similar elements have the same reference signs assigned.

[0034] In a wireless communication system network, like the one described above with reference to Fig. 1 , one or more Artificial Intelligence / Machine Learning models, AI / ML models, or one or more AI / ML functionalities may be implemented in a user device or user equipment, UE, for performing one or more tasks, e.g., one or more of the following:

[0035] - AI / ML model based access to a RAN,

[0036] - AI / ML model based network energy saving,

[0037] - AI / ML model based load balancing,

[0038] - AI / ML model based mobility optimization,

[0039] - AI / ML model based use cases, like o channel state information, CSI, feedback, like a CSI compression and / or a CSI prediction, or o beam management, or o positioning, like a direct AI / ML positioning (e.g., fingerprinting) and / or an AI / ML assisted positioning,

[0040] - AI / ML model based mobility management, e.g., a handover, HO, prediction and / or conditional handover, CHO, prediction,

[0041] - AI / ML model based modulation and coding scheme, MCS, selection,

[0042] - AI / ML model based synchronization,

[0043] - AI / ML model based encoding and / or decoding and / or precoding,

[0044] - AI / ML model based modulation and / or demodulation,

[0045] - AI / ML model based positioning or ranging,

[0046] - AI / ML model based joint communication and sensing, JSAC,

[0047] - AI / ML model based feedback calculation, e.g., channel state information, CSI, channel quality indicator, CQI, preferred matrix index, PMI, rank indicator feedback, - AI / ML model based interference management,

[0048] - AI / ML model based quality of experience, QoE, and / or quality of service, QoS, predictions,

[0049] - AI / ML model based network traffic forecasting.

[0050] When implementing one or more AI / ML models or one or more AI / ML functionalities in a wireless communication network, like a 3GPP network or a WiFi network, the overall operation of the network or an efficiency of certain functions within the network may be improved. For example, the air interface in a 5G network may be enhanced using AI / ML. The respective AI / ML models when being implemented, for example within a user device, are trained on a basis of a training dataset, and the trained AI / ML model is used for performing a certain task.

[0051] To take advantage of the AI / ML models or one or more AI / ML functionalities, conventionally, AI / ML first has to be activated, e.g., by the base station or gNB, so that the UE may actually use and benefit AI / ML features / models. In other words, the UE may benefit from the AI / ML only while being in the connected state. Leaving the connected state causes the deactivation of all AI / ML features in the UE. Hence, during an initial access or while being in a non-connected state, e.g., RRCJDLE or RRCJNACTIVE, the AI / ML features / models are not be available to the UE.

[0052] For example, when a UE enters the RRC NACTIVE state or the RRCJDLE state from the RRC_CONNECTED state, conventionally, all AI / ML functionalities / models are deactivated as the UE is not expected to perform any data transmissions in these states. Also, if the UE was not in the RRC_CONNECTED state but started with the RRCJDLE state, all AI / ML functionalities / models are deactivated as the AI / ML functionalities / models are only active by the network, like a gNB, which, in turn, happens only when the UE is in the RRC_CONNECTED state. However, this leads to a situation that AI / ML is unavailable for potential transmissions by the UE being in the non-connected state, like a SDT, or a EDT or the RACH transmissions, like Msg1 (Preamble Transmission), Msg2 (Random Access Response), Msg3, Msg4 (Contention Resolution).

[0053] Therefore, there may be a need for improvements or enhancements to the use of AI / ML models or AI / ML functionalities employed in a user device of the wireless communication network which allow the UE to benefit from the AI / ML also when not being in the connected state. Embodiments of the present invention address the above problem by allowing a UE to activate certain AI / ML functionalities / models during a non-connected state, like the RRC_IN ACTIVE state, the RRCJDLE state or when being out-of-coverage of a base station. According to the present invention a user device may be configured or preconfigured with at least one Artificial Intelligence / Machine Learning model, AI / ML model, or at least one AI / ML functionality for performing one or more tasks. If the UE is in a connected state, the UE receives an indication from the wireless communication network for activating one or more connected AI / ML models or functionalities, the one or more connected AI / ML models or functionalities, which include one or more of the configured or preconfigured AI / ML models or AI / ML functionalities. If the UE is in a non-connected state, the UE uses one or more non-connected AI / ML models or functionalities, the one or more non-connected AI / ML models or functionalities, which include one or more of the configured or preconfigured AI / ML models or AI / ML functionalities.

[0054] The inventive approach is advantageous as it allows a user device, UE, to exploit or benefit from the advantages of AI / ML even when being in non-connected state. As mentioned above, in such a non-connected state, nevertheless, some communication between the UE and the network may occur, for example the mentioned small data transmissions or the transmissions performed during the RACH process during initial access or for entering the connected state. Since AI / ML is used for improving, for example, the communication between the UE and the network, allowing the UE to make use of at least some of the AI / ML it is configured or preconfigured with when being in a non-connected state, also provides for an improvement of the transmissions received at the UE and transmitted by the UE in a similar way as the AI / ML improves the transmissions between the UE and the network when the UE is in the connected state. Thereby, the inventive approach is capable to maintain the benefits of implementing AI / ML models or AI / ML functionalities in a user device also for situations in which a user device in a non-connected state.

[0055] Embodiments of the present invention may be implemented in a wireless communication system as depicted in Fig. 1 including base stations and users, like mobile terminals or loT devices. Fig. 2 is a schematic representation of a wireless communication system 310 including a transmitter 300, like a base station, and one or more receivers 302, 304, like user devices, UEs. The transmitter 300 and the receivers 302, 304 may communicate via one or more wireless communication links or channels 306a, 306b, 308, like a radio link. The transmitter 300 may include one or more antennas ANTT or an antenna array having a plurality of antenna elements, a signal processor 300a and a transceiver 300b, coupled with each other. The receivers 302, 304 include one or more antennas ANTUE or an antenna array having a plurality of antennas, a signal processor 302a, 304a, and a transceiver 302b, 304b coupled with each other. The base station 300 and the UEs 302, 304 may communicate via respective first wireless communication links 306a and 306b, like a radio link using the llu interface, while the UEs 302, 304 may communicate with each other via a second wireless communication link 308, like a radio link using the PC5 or sidelink, SL, interface. When the UEs are not served by the base station or are not connected to the base station, for example, they are not in an RRC connected state, or, more generally, when no SL resource allocation configuration or assistance is provided by a base station, the UEs may communicate with each other over the sidelink. The system or network of Fig. 2, the one or more UEs 302, 304 of Fig. 2, and the base station 300 of Fig. 2 may operate in accordance with the inventive teachings described herein.

[0056] According to aspect 1 there is provided a user device, UE, for a wireless communication network, wherein the UE is configured or preconfigured with at least one Artificial Intelligence / Machine Learning model, AI / ML model, or at least one AI / ML functionality for performing one or more tasks, wherein, if the UE is in a connected state, the UE is to receive an indication from the wireless communication network for activating one or more connected AI / ML models or functionalities, the one or more connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities, and wherein, if the UE is in a non-connected state, the UE is to use one or more non-connected AI / ML models or functionalities, the one or more non-connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities.

[0057] According to aspect 2 relating to aspect 1 , the one or more non-connected AI / ML models or functionalities comprise one or more of the following: one or more or all of the AI / ML models or AI / ML functionalities from the connected AI / ML models or functionalities, one or more or all of the AI / ML models or AI / ML functionalities from the connected AI / ML models or functionalities with different Al parameters, one or more AI / ML models or AI / ML functionalities different from the connected AI / ML models or functionalities.

[0058] According to aspect 3 relating to aspect 1 or 2, if the UE enters from the connected state into the non-connected state, the UE is to keep one or more or all of the activated AI / ML models or AI / ML functionalities as the non-connected AI / ML models or functionalities active for performing one or more of the tasks at least during part of the time the UE is in the non-connected state, or if the UE is in the non-connected state and the use of non-connected AI / ML models or functionalities is enabled, the UE is to activate the one or more non-connected AI / ML models or functionalities for performing one or more of the tasks at least during part of the time the UE is in the non-connected state.

[0059] According to aspect 4 relating to aspect 3, the UE is to activate the one or more nonconnected AI / ML models or functionalities if the UE enters the non-connected state from a state, like a LTE connected or LTE idle state, in which no AI / ML models or AI / ML functionalities had been activated in the UE by the wireless communication network, or if the UE, upon entering the non-connected state from the connected state, deactivated all connected AI / ML models or functionalities that had been activated in the UE by the wireless communication network during the connected state.

[0060] According to aspect 5 relating to any one of the preceding aspects, the connected state comprises a state, like an RRC_CONNECTED state, where the UE has a connection with the wireless communication network, and the non-connected state comprises one or more of the following states: a first non-connected state, like an RRCJDLE state, where the UE has no connection with the wireless communication network, a second non-connected state, like an RRCJNACTIVE state, where the UE has a connection with the wireless communication network but it is not active, a third non-connected state, like an RRC_NON_CONNECTED_AI state, where the UE has no connection with the wireless communication network or has a connection with the wireless communication network but it is not active, and where the UE is allowed to use the one or more non-connected AI / ML models or functionalities.

[0061] According to aspect 6 relating to aspect 5, the UE is to use the same or different nonconnected AI / ML models or functionalities in the first and second states.

[0062] According to aspect 7 relating to any one of the preceding aspects, in the non-connected state, the UE is to train the one or more non-connected AI / ML models or functionalities.

[0063] According to aspect 8 relating to any one of the preceding aspects, in the non-connected state, the UE is to monitor a performance of one or more of the non-connected AI / ML models or functionalities and / or one or more of the AI / ML models or AI / ML functionalities with which the UE is configured or preconfigured and which are inactive.

[0064] According to aspect 9 relating to aspect 8, the UE is to send a performance report

[0065] - while being in the non-connected state, e.g., using an early data transmission, EDT, or a small data transmission, SDT, or

[0066] - when returning to the connected state.

[0067] According to aspect 10 relating to aspect 8 or 9, the UE is to monitor the performance periodically, e.g., during a wake-up time of a paging cycle or a discontinuous reception, DRX, cycle, or a dedicated monitoring cycle, and / or responsive to one or more certain conditions.

[0068] According to aspect 11 relating to aspect 10, the certain condition comprises one or more of the following: a performance of the AI / ML model or AI / ML functionality is not within one or more predefined boundaries, a signaling from the wireless communication network causing the UE to monitor the performance, a change in an operational situation of the UE.

[0069] According to aspect 12 relating to aspect 11 , the signaling from the wireless communication network comprises one or more of the following: a paging message or an early paging indicator from a base station of the wireless communication network, a RAN-based Notification Area (RNA) update information, e.g., including an updated list of cells of the wireless communication network, a scheduling or non-scheduling DCI from a base station of the wireless communication network, a broadcast signaling included in a system information message, e.g., in a system information block, SIB, or in a master information block, MIB, a wakeup signal, reference signals to be used for certain procedure, e.g., a positioning procedure using sounding reference signals, SRSs.

[0070] According to aspect 13 relating to aspect 11 or 12, the change in an operational situation of the UE comprises one or more of the following: a cell reselection or selection, e.g. the UE enters an area covered by another cell than the one it is connected to, a Public Land Mobile Network, PLMN, selection, e.g. when UE is roaming and selects the most suitable PLMN, a change in measurements in neighboring cells, e.g. RSRP, RSRQ, RSSI, SINR, SNR of neighboring cells, a state transition from the connected state into the non-connected state, a state transition from the first non-connected state, like an RRCJDLE state, to second non-connected state, like an RRC_IN ACTIVE state, a change in an radio access network, RAN, based notification area update, leaving or entering a certain location, e.g., the UE may be configured with a specific location / area, responsive to receiving a positioning signal, like an SRS.

[0071] According to aspect 14 relating to any one of aspects 8 to 13, the UE is configured or preconfigured with a non-connected state monitoring configuration.

[0072] According to aspect 15 relating to aspect 14, the monitoring configuration indicates one or more of the following: how to compress the performance data, selecting some of the performance data to reduce the amount of data to be transmitted, a non-connected state monitoring periodicity, a non-connected state monitoring window configuration defining, e.g. via an RRC configuration, parameters which are to be used during the non-connected mode to generate a measurement report,

[0073] According to aspect 16 relating to any one of the preceding aspects, for using the nonconnected AI / ML models or functionalities, the UE is to activate one or more AI / ML models or AI / ML functionalities with which the UE is configured or preconfigured.

[0074] According to aspect 17 relating to aspect 16, the one or more AI / ML models or AI / ML functionalities are preconfigured in a wireless communication network specification so that certain Al types, e.g., a beam predictor Al, are activated in the non-connected state, or configured by a system information message received at the UE, like a system information block, SIB, or defined by the model, or configured by the wireless communication network to the UE while being in the connected state, e.g., using an RRC signaling, using one or more MAC CEs, in a resource pool, RP, configuration, or in a bandwidth part, BWP configuration.

[0075] According to aspect 18 relating to aspect 16 or 17, the preconfiguration or the configuration contain one or more of the following: one or more AI / ML functionalities that the UE is allowed to activate in the nonconnected state, one or more AI / ML models that the UE is allowed to activate in the non-connected state, one or more AI / ML parameters to be used by the UE while being in the nonconnected state, one or more conditions when to use AI / ML in the non-connected state, a configuration of a life cycle management, LCM, phases operation in the nonconnected state, e.g., an activation of a monitoring of the Al or a activation / deactivation of a training of the Al, one or more conditions for performing a fallback to non-AI procedures, a deactivation of some of the AI / ML models or AI / ML functionalities with which the UE is configured or preconfigured. According to aspect 19 relating to any one of the preceding aspects, the UE is to use the one or more non-connected Al models or functionalities only in case one or more conditions are met.

[0076] According to aspect 20 relating to aspect 19, the one or more conditions comprises one or more of the following: the UE is allowed to operate the one or more non-connected AI / ML models or functionalities, one or more requirements for a transmission while being in the non-connected state are met, an operation failed without AI / ML, e.g., random access channel, RACH, data is not transmitted after some attempts, an AI / ML usage duration did not surpass a predefined time limit, a time to start using Al has lapsed.

[0077] According to aspect 21 relating to aspect 20, the UE is allowed to operate the one or more non-connected Al models or functionalities when one or more of the following applies: the UE is of the type supporting the use of non-connected AI / ML models or functionalities, the UE has a sufficient battery level and / or sufficient memory storage for operating the one or more non-connected Al models or functionalities, the UE is communicating on a certain frequency band, e.g., an unlicensed or licensed band supported by AI / ML, the UE is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs supporting AI / ML, the UE operates in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, UMa, a rural microcell, RMi, a rural macrocell RMa, or indoors, one or more mobility aspects, like speed, acceleration, Doppler delay profile, are met. the UE is in a certain location or area.

[0078] According to aspect 22 relating to aspect 20 or 21 , the one or more requirements for a transmission while being in the non-connected state comprise one or more of the following: a random access channel, RACH, procedure failed for a certain number of attempts, a channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold, an interference level is below or above a certain threshold, a certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection.

[0079] According to aspect 23 relating to any one of the preceding aspects, the UE is to switch to a fallback operation or to a non-AI operation or to another AI / ML model or functionality when one or more conditions are met.

[0080] According to aspect 24 relating to aspect 23, the one or more conditions comprises one or more of the following: the UE is not capable to operate the one or more non-connected AI / ML models or functionalities, one or more requirements for a transmission while being in the non-connected state are not met, an operation using AI / ML failed, e.g., random access channel, RACH, data is not transmitted after some attempts, an Al usage duration surpassed a predefined time limit, a time to start using AI / ML has not yet lapsed.

[0081] According to aspect 25 relating to aspect 24, the UE is not capable to operate the one or more non-connected AI / ML models or functionalities when one or more of the following applies: the UE is of the type not supporting the use of non-connected AI / ML models or functionalities, the UE has an insufficient battery level and / or insufficient memory storage for operating the one or more non-connected Al models or functionalities, the UE is communicating on a certain frequency band, e.g., an unlicensed or licensed band not supported by AI / ML, the UE is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs not supporting Al, the UE does not operate in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, Uma, a rural microcell, Rmi, a rural macrocell, RMa, or indoors, one or more mobility aspects, like speed, acceleration, Doppler delay profile, are not met. the UE is not in a certain location or area. According to aspect 26 relating to aspect 24 or 25, the one or more requirements for a transmission while being in the non-connected state comprise one or more of the following: a random access channel, RACH, procedure failed for a certain number of attempts, a channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold, an interference level is below or above a certain threshold, a certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection.

[0082] According to aspect 27 relating to any one of the preceding aspects, the UE is to indicate to the wireless communication network whether or not the UE supports the use of nonconnected AI / ML models or functionalities, e.g., in a UE Capability Report.

[0083] According to aspect 28 relating to aspect 27, the UE is to indicate to the wireless communication network one or more of the following: the non-connected AI / ML models or functionalities supported by the UE, conditions for using supported non-connected AI / ML models or functionalities, scenarios in which supported non-connected AI / ML models or functionalities may be used, e.g. certain locations, SNR ranges, urban, rural, etc., a maximum duration for using the supported non-connected AI / ML models or functionalities, a maximum frequency, e.g. a smallest periodicity, with which the supported nonconnected AI / ML models or functionalities may be used, a maximum UE speed at which the supported non-connected AI / ML models or functionalities may be used, a number of simultaneously supported non-connected AI / ML models or functionalities, a number of processing units that may be simultaneously occupied by all supported non-connected AI / ML models or functionalities, a delay in activating a supported non-connected AI / ML model or functionality, e.g. 2 ms,

[0084] - whether generalization of the supported non-connected AI / ML models or functionalities is performed by the UE with or without any signaling from the wireless communication network. According to aspect 29 relating to any one of the preceding aspects, UE is configured or preconfigured with a separate set of RACH resources.

[0085] According to aspect 30 relating to any one of the preceding aspects, the UE is to switch from an E-UTRA / EPC non-connected state in which the non-connected AI / ML models or functionalities are not supported to an NR / 5GC RRC non-connected state for enabling the UE to use the non-connected Al models or functionalities.

[0086] According to aspect 31 relating to aspect 30, the UE is to switch from the E-UTRA / EPC non-connected state to the NR / 5GC RRC non-connected state only when one or more conditions are met.

[0087] According to aspect 32 relating to any one of the preceding aspects, the UE is to switch to another non-connected AI / ML model or functionality only when one or more conditions are met.

[0088] According to aspect 33 relating to aspect 32, the one or more conditions comprise one or more of the following: the UE is allowed to operate the one or more non-connected AI / ML models or functionalities, one or more requirements for a transmission while being in the non-connected state are met, an operation failed without AI / ML, e.g., random access channel, RACH, data is not transmitted after some attempts, an AI / ML usage duration did not surpass a predefined time limit, a time to start using AI / ML has lapsed.

[0089] According to aspect 34 relating to aspect 33, the UE is allowed to operate the one or more non-connected AI / ML models or functionalities when one or more of the following applies: the UE is of the type supporting the use of non-connected AI / ML models or functionalities, the UE has a sufficient battery level and / or sufficient memory storage for operating the one or more non-connected AI / ML models or functionalities, the UE is communicating on a certain frequency band, e.g., an unlicensed or licensed band supported by AI / ML, the UE is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs supporting AI / ML, the UE operates in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, UMa, a rural microcell, RMi, a rural macrocell, RMa, or indoors, one or more mobility aspects, like speed, acceleration, Doppler delay profile, are met. the UE is in a certain location or area.

[0090] According to aspect 35 relating to aspect 33 or 34, the one or more requirements for a transmission while being in the non-connected state comprise one or more of the following: a random access channel, RACH, procedure failed for a certain number of attempts, a channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold, an interference level is below or above a certain threshold, a certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection.

[0091] According to aspect 36 relating to any one of the preceding aspects, if in the non-connected state, the UE uses the one or more AI / ML models or AI / ML functionalities to perform one or more tasks associated with a data transmission by the UE, like an early data transmission, EDT, or a small data transmission, SDT.

[0092] According to aspect 37 relating to any one of the preceding aspects, the one or more of tasks comprise one or more of the following:

[0093] - AI / ML model based access to a RAN,

[0094] - AI / ML model based network energy saving,

[0095] - AI / ML model based resource allocation optimizations,

[0096] - AI / ML model based scheduling optimizations,

[0097] - AI / ML model based network slicing management,

[0098] - AI / ML model based load balancing, an AI / ML model based mobility optimization,

[0099] - AI / ML model based use cases, like o channel state information, CSI, feedback, like a CSI compression and / or a CSI prediction, or o beam management, or o positioning, like a direct AI / ML positioning (e.g., fingerprinting) and / or an AI / ML assisted positioning, o Handover like the AI / ML to predict when the UE should handover to which cell to reduce unnecessary handovers,

[0100] - AI / ML model based mobility management, e.g., a handover, HO, prediction and / or conditional handover, CHO, prediction,

[0101] - AI / ML model based modulation and coding scheme, MCS, selection,

[0102] - AI / ML model based synchronization,

[0103] - AI / ML model based encoding and / or decoding and / or precoding,

[0104] - AI / ML model based modulation and / or demodulation,

[0105] - AI / ML model based positioning or ranging,

[0106] - AI / ML model based joint communication and sensing, JSAC,

[0107] - AI / ML model based network offloading,

[0108] - AI / ML model based feedback calculation, e.g., CSI / CQI / PMI / RI feedback,

[0109] - AI / ML model based interference management,

[0110] - AI / ML model based quality of experience, QoE, and / or quality of service, QoS, predictions,

[0111] - AI / ML model based network traffic forecasting.

[0112] According to aspect 38 relating to any one of the preceding aspects, the UE comprise one or more of a power-limited UE, or a hand-held UE, like a UE used by a pedestrian, and referred to as a Vulnerable Road User, VRU, or a Pedestrian UE, P-UE, or an on-body or hand-held UE used by public safety personnel and first responders, and referred to as Public safety UE, PS-UE, or an loT UE or Ambient loT UE, e.g., a sensor, an actuator or a UE provided in a campus network to carry out repetitive tasks and requiring input from a gateway node at periodic intervals, or a mobile terminal, or a stationary terminal, or a cellular loT-UE, an industrial loT-UE, 11 oT, or a SL UE, or a vehicular UE, or a vehicular group leader UE, GL-UE, or a scheduling UE, S-UE, or an loT or narrowband loT, NB-loT, device, a NTN UE, or a WiFi device or WiFi station, STA, or a ground based vehicle, or an aerial vehicle, or a drone, or a moving base station, or road side unit, RSU, or a building, or any other item or device provided with network connectivity enabling the item / device to communicate using the wireless communication network, e.g., a sensor or actuator, or any other item or device provided with network connectivity enabling the item / device to communicate using a sidelink the wireless communication network, e.g., a sensor or actuator, or any sidelink capable network entity. According to aspect 39 there is provided a wireless communication network, like a 3rdGeneration Partnership Project, 3GPP, system, comprising a one or more user devices, UEs, of any one of the preceding aspects and one or more base stations, BSs.

[0113] According to aspect 40 relating to aspect 39, the BS comprises one or more of a macro cell base station, or a small cell base station, or a central unit of a base station, or a distributed unit of a base station, or an Integrated Access and Backhaul, IAB, node, or a road side unit, RSU, or a WiFi access point, AP, or a UE, or a SL UE, or a group leader UE, GL-LIE, or a relay or a remote radio head, or an AMF, or an SMF, or a core network entity, or mobile edge computing, MEC, entity, or a network slice as in the NR or 5G core context, or any transmission / reception point, TRP, enabling an item or a device to communicate using the wireless communication network, the item or device being provided with network connectivity to communicate using the wireless communication network.

[0114] According to aspect 41 there is provided a method for operating a user device, UE, for a wireless communication network, wherein the UE is configured or preconfigured with at least one Artificial Intelligence / Machine Learning model, AI / ML model, or at least one AI / ML functionality for performing one or more tasks, the method comprising: if the UE is in a connected state, receiving an indication from the wireless communication network for activating one or more connected AI / ML models or functionalities, the one or more connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities, and if the UE is in a non-connected state, using one or more non-connected AI / ML models or functionalities, the one or more non-connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities.

[0115] According to aspect 42 there is provided a non-transitory computer program product comprising a computer readable medium storing instructions which, when executed on a computer, perform the method of aspect 41 .

[0116] The present invention provides a computer program product comprising instructions which, when the program is executed by a computer, causes the computer to carry out one or more methods in accordance with the present invention. Embodiments of the present invention are now described in more detail with reference to the accompanying drawing. It is noted that the subsequently outlined and described aspects or embodiments may be combined such that some or all of the aspects / embodiments are implemented within one embodiment. Reference is made herein one or more AI / ML models and / or to one or more AI / ML functionalities. It is noted that when referring only to an AI / ML model, this is to be understood to refer also to an AI / ML functionality, and that when referring only to an AI / ML functionality, this it to be understood to refer also to an AI / ML model. AI / ML functionality may refer to an AI / ML-enabled Feature / Feature Group, FG, enabled by one or more configurations, where the one or more configurations may be supported based on one or more conditions indicated by a UE capability. An AI / ML-enabled Feature refers to a Feature where AI / ML may be used. It is noted that a UE may have one AI / ML model for the functionality, or the UE may have multiple AI / ML models for the functionality. Examples of use cases for AI / ML-enabled Features or Feature Groups are:

[0117] CSI feedback enhancement, e.g., overhead reduction, improved accuracy, prediction.

[0118] Beam management, e.g., beam prediction in time, and / or spatial domain for overhead and latency reduction, beam selection accuracy improvement.

[0119] Positioning accuracy enhancements for different scenarios including, e.g., those with heavy NLOS conditions.

[0120] Other examples may comprise of access to the RAN, network energy saving, NES, resource management and load balancing, mobility enhancements and optimization including handover, HO, management and / or prediction, conditional handover, CHO, management and / or prediction, modulation and coding scheme, MOS, selection, MIMO precoder calculation, general PHY-layer signal processing, e.g., synchronization, channel coding or decoding, modulation or demodulation, positioning or ranging, joint communication and sensing, JSAC, feedback calculation of CSI / CQI or PMI / RI, general MIMO processing, equalization, network offloading, interference management, quality of experience, QoE, and / or quality of service, QoS, predictions, and / or network traffic forecasting. It is noted that the AI / ML approaches for the selected sub use cases need to be diverse enough to support various requirements on the gNB-UE collaboration levels.

[0121] An AI / ML model operates based on identified models, where a model may be associated with one or more specific configurations / conditions associated with a UE capability of an AI / ML-enabled Feature / FG and additional conditions (e.g., scenarios, sites, and datasets) as determined / identified between the UE-side and the NW-side.

[0122] Fig. 3 illustrates a user device, UE, in accordance with embodiments of the present invention. The UE 400 includes a signal processing unit or signal processor 402 and one or more antennas or an antenna array 404 for communicating with other network entities over the air interface. As is depicted in Fig. 3, the UE 400 may communicate with a base station or gNB 406 using the Uu interface 408 and / or with a further UE 410 using the PC5 interface 412 for a sidelink, SL, communication. The UE 400 may use using in a connected state AI / ML models / functionalities which are referred to in the following as “connected AI / ML models or functionalities”, and in a non-connected state AI / ML models / functionalities which are referred to in the following as “non-connected AI / ML models or functionalities”. As is schematically illustrated, the UE 400 is configured or preconfigured with at least one AI / ML model or at least one AI / ML functionality 414 for performing one or more tasks. If the UE 400 is in a connected state, the UE receives an indication, like an activation signal 416, from the gNB 406 for activating one or more connected AI / ML models or functionalities 418. The one or more connected AI / ML models or functionalities 418 include one or more or all of the configured or preconfigured AI / ML models or AI / ML functionalities 414. If the UE 400 is in a non-connected state it may use one or more non-connected AI / ML models or functionalities 420. The one or more non-connected AI / ML models or functionalities 420 include one or more or all of the configured or preconfigured AI / ML models or AI / ML functionalities 414. In other words, the UE 400 when being in a non-connected state may activate one or more or all of the configured or preconfigured AI / ML models or functionalities 414 without any explicit or implicit signaling from the network side causing such an activation. Thus, in accordance with the present invention, the UE 400 may also benefit from the advantages of implementing AI / ML approaches in the UE, for example for improving the potential transmissions the UE is to transmit or receive even when being in the non-connected state.

[0123] In accordance with embodiments, the non-connected AI / ML models 420 may include AI / ML models from the connected AI / ML models 418 having the same or different parameters and / or AI / ML models different from the connected Al models 418. In accordance with embodiments, the one or more non-connected AI / ML models or functionalities 420 comprise one or more of the following:

[0124] One or more or all of the AI / ML models or AI / ML functionalities from the connected AI / ML models or functionalities 414. One or more or all of the AI / ML models or AI / ML functionalities from the connected AI / ML models or functionalities 414 with different Al parameters. In that case, the structure of the model is the same but the parameters vary, or the same model may have a different implementation (smaller / larger models).

[0125] One or more AI / ML models or AI / ML functionalities different from the connected AI / ML models or functionalities 414.

[0126] Fig. 4 illustrates an embodiment of the present invention used during a random access procedure as it may be used by the UE 400 during an initial access to the wireless network or when connecting from the non-connected state to the connected state. Fig. 4(A) illustrates the UE 400 and the base station or gNB 406. The gNB 406 broadcasts system information, like the system information block, SIB, as is indicated at 430. The SIB includes the information that the network supports the use of non-connected Al 414, i.e. , the UE 400 knows from the SIB that it is allowed to operate in accordance with the present invention, namely to use one or more non-connected AI / ML models or functionalities 420. The UE 400 is assumed to be in the non-connected state and, as is illustrated in Fig. 4(B), for entering the connected state the UE 400 initiates the RACH procedure 432. The RACH procedure is supported by AI / ML models or functionalities configured or preconfigured in the UE 400 and being part of the non-connected AI / ML models or functionalities 420. In accordance with embodiments, the RACH procedure may employ AI / ML approaches for performing or at least supporting the RACH resource selection and / or a Rx / Tx beam selection. Thus, in accordance with embodiments of the inventive approach, the UE 400 takes advantage of the benefits of the AI / ML approaches in the non-connected state thereby improving transmissions to / from the UE 400 also when being in a non-connected state.

[0127] Thus, AI / ML may be advantageous for improving the reception / transmit operations during the RACH procedure which leads to a better initial access result. However, the present invention is not limited to such use cases, rather, the AI / ML operation in the non-connected mode may be used for different use cases. In accordance with embodiments, AI / ML may be used to predict DRX cycles to adapt better to an actual traffic pattern instead of using configured or preconfigured semi-static DRX cycles. In accordance with another embodiment, the use of non-connected AI / ML models or functionality may enhance the RNA operation, in particular, the AI / ML may predict the time to trigger the RNA update, may predict a list of cells included in the RNA, may predict a list of RAN areas, for example the TAG and optionally the RAN tracking area code, or may trigger a cell selection or reselection when the UEs in the RRCJNACTIVE or idle state. In accordance with embodiments of the present invention, one may distinguish between a situation in which the UE 400 enters from a connected state, in which connected AI / ML models or functionalities had been activated, into a non-connected state, and a situation in which the UE 400 is in the non-connected state without any AI / ML models or functionalities being activated.

[0128] In the first situation, when entering from the connected state into the non-connected state, the UE 400 may maintain or keep active one or more or all of the activated AI / ML models or AI / ML functionalities as the non-connected AI / ML models or functionalities, so as to be in a position to perform the one or more tasks at least during part of the time the UE 400 is in the non-connected state.

[0129] In the second situation, when the UE is in the non-connected state, for example, when entering from a state in which no AI / ML had been activated or when initially accessing the network, the UE may determine that the network allows the use of the non-connected AI / ML models, for example from the SIB as described above with reference to Fig. 4. Thus, when the UE is in the non-connected state and determines that the use of the non-connected Al / M models or functionalities is enabled, it may activate the non-connected AI / ML models or functionalities 420 so as to be in a position to perform the one or more tasks also at least during part of the time the UE is in the non-connected state, like for performing the abovedescribed RACH procedure (see Fig. 4). The just-mentioned second scenario may apply, for example, when the UE 400 enters the non-connected state from a state in which no AI / ML models and functionalities had been activated by the wireless communication network. In other words, when entering the non-connected state from any state in which no AI / ML is used at all, the UE 400 may determine whether the use of the non-connected AI / ML models or functionalities is enabled and, if yes it may activate them. The second situation also applies for a situation in which, upon entering the non-connected state from a connected state, the UE 400 deactivated all connected AI / ML models or functionalities that had been activated in the UE. Thus, even when a transition into the non-connected state caused the deactivation of all AI / ML in the UE, and when the use of non-connected AI / ML is enabled in the network, the UE 400 may activate one or more of the non-connected AI / ML models or functionalities for performing the one or more tasks while being in the nonconnected state. In other words, when the UE is in the non-connected state and the use of the non-connected AI / ML models or functionalities is enabled, the UE may activate A l / ML models / functionalities without an activation signaling from the network side which may also be referred to as an activation of the one or more non-connected AI / ML models or functionality by the UE on its own.

[0130] In accordance with embodiments the connected state is the RRC_CONNECTED state, and there are two non-connected states, the RRCJDLE state and the RRCJNACTIVE state.

[0131] In accordance with embodiments of the present invention, there may be a one-to-one mapping between the AI / ML model states and the RRC states. Fig. 5 illustrates the RRC_CONNECTED state 436 as well as the non-connected states 438 and 440, namely the RRCJDLE state and the RRC_IN ACTIVE state. In the connected state 436, the connected AI / ML models or functionalities 418 may be employed while both non-connected states 438 may use the same non-connected AI / ML models or functionalities 420. Thus, when entering one of the non-connected states 438, 440 from the connected state 436 one set of AI / MLs models or functionalities are available to be used in these states. In Fig 5, by means of the double-headed arrows, the transitions between the respective states 436, 438 and 440 are indicated. As may be seen, the UE 400 may move from one of the three states to any other of the two remaining states.

[0132] In accordance with other embodiments, there may be different non-connected AI / ML modes for the different non-connected states. This is illustrated in Fig. 6. While Fig. 5 illustrates the use of the same non-connected AI / ML models or functionalities for the non-connected states, Fig. 6 illustrates an embodiment that distinguishes the non-connected AI / ML models or functionalities dependent on the actual non-connected state. More specifically, as is illustrated in Fig. 6, the non-connected state 438, the RRCJDLE state, may allow the use of first non-connected AI / ML models or functionalities 420a, and the second non-connected state 440, the RRCJN ACTIVE state, allows use of the non-connected AI / ML models or functionalities 420b. In accordance with embodiments, one or more of the non-connected AI / ML models or functionalities 420a, 420b may be the same. In other words, while the entire set of non-connected AI / ML models or functionalities associated with a different nonconnected states is different, some of the individual AI / ML models or functionalities in the respective sets may be the same. In accordance with yet other embodiments, an additional state into which the UE may transition may be implemented and in which the UE is in a non-connected state but is allowed to use the AI / ML models / functionality with which the UE is configured / preconfigured, i.e., is allowed to make use of the non-connected AI / ML models or functionalities 420. Fig. 7 illustrates the state diagram including the additional nonconnected state. In the embodiment of Fig. 7 a conventional RRC_CONNECTED state 436 is assumed which allows the UE 400 to make use of the connected AI / ML models or functionalities 418, while in the non-connected states RRCJDLE 438 and RRCJNACTIVE 440 the UE 400 is not enabled to make use of the configured / preconfigured AI / ML models or functionalities. Fig. 7 also illustrates the above-mentioned additional state, also referred to as an RRC_NON_CONNECTED_AI state 442. When entering the state 442 from any of the other states 436 to 440, the UE 400 is enabled to make use of the non-connected AI / ML models or functionalities 420. In accordance with embodiments, the RRC_NON_CONNECTED_AI state may be comprised of two states, namely an RRC_IDLE_AI state and an RRC_INACTIVE_AI state, thereby defining RRCJDLE and RRCJNACTIVE states in which the use of the non-connected AI / ML models or functionalities 420 is allowed.

[0133] Fig. 7 also illustrates by means of the double-headed arrows the respective transitions between the states. In accordance with embodiments, certain state transitions may not be allowed, for example, it may be required to only transition from the RRC_NON_CONNECTED_AI state to the RRCJNACTIVE state before being able to transition to the RRC_CONNECTED state. This additional state may be beneficial, as the UE would activate AI / ML features in this mode without fully connecting. This saves the battery of the UE, as in this mode it may decide to go back to the RRC-INACTIVE state, if no further communication or need for an RRC-CONNECTED mode is foreseen. Also other combinations or restrictions may be applied to the RRC state machine. For example, may be a state between the RRCJDLE and RRC_INACTIVE / RRC_CONNECTED states. In this case, the UE may be allowed to activate Al even before ever having been connected to the network.

[0134] Further, in accordance with embodiments, the RRC_NON_CONNECTED_AI state may also be a meta state for a newly designed RRC Al state machine which may be contained within the RRC_NON_CONNECTED_AI state. In this way, the Al engine within the UE 400 may be configured in a forward compatible way as this allows designing algorithms involving Al states independent from the existing RRC states not involving Al states. In accordance with further embodiments of the present invention, the UE 400, when being in the non-connected state, may train the one or more non-connected AI / ML models or functionalities. For example, the UE may continuously train its AI / ML even during the RRCJNACTIVE state or during the RRCJDLE state or, if implemented, during the RRC_NON_CONNECTED_AI state, if the UE has sufficient internal capabilities, like a sufficient battery level and / or a sufficient amount of available memory.

[0135] In accordance with further embodiments, UE 400, when being in the non-connected state 438, 440, 442may monitors the non-connected Al models and / or inactive Al models 420. In accordance with embodiments, the UE 400 may monitor a performance of one or more of the non-connected AI / ML models or functionalities 420 and / or of one or more of the AI / ML models or AI / ML functionalities 418 with which the UE 400 is configured or preconfigured and which are inactive in the non-connected state 438, 440, 442.

[0136] In accordance with embodiments, the UE 400 may send a performance report about the non-connected Al models and / or inactive Al models. The UE 400 may send the performance report while being in the non-connected state, e.g., using an early data transmission, EDT, or a small data transmission, SDT, or when returning to the connected state 436. In the non-connected state the UE 400 may monitor the non-connected Al models and / or inactive Al models periodically or when a certain condition applies. In accordance with embodiments, the UE 400 may monitor the performance periodically, e.g., during a wake-up time of a paging cycle or a discontinuous reception, DRX, cycle, or a during dedicated monitoring cycle, and / or responsive to one or more certain conditions. The conditions may include one or more of the following: a performance of the AI / ML model or AI / ML functionality is not within one or more predefined boundaries, a signaling from the wireless communication network causes the UE 400 to monitor the performance, a change in an operational situation of the UE 400.

[0137] The signaling from the wireless communication network may include one or more of the following: a paging message or an early paging indicator from a base station of the wireless communication network, a RAN-based Notification Area (RNA) update information, including an updated list of cells, a scheduling or non-scheduling DCI from a base station of the wireless communication network, a broadcast signaling included in a system information message, e.g., in a system information block, SIB, or in a master information block, MIB, a wakeup signal, reference signals to be used for certain procedure, e.g., a positioning procedure using sounding reference signals, SRSs.

[0138] The change in an operational situation of the UE 400 may include one or more of the following: a cell reselection or selection, e.g. the UE enters an area covered by another cell than the one it is connected to, a Public Land Mobile Network, PLMN, selection, e.g. when UE is roaming and selects the most suitable PLMN, a change in measurements in neighboring cells, e.g. RSRP, RSRQ, RSSI, SINR, SNR of neighboring cells, a state transition from the connected state into the non-connected state, a state transition from the first non-connected state, like an RRC DLE state, to second non-connected state, like an RRCJNACTIVE state, a change in an radio access network, RAN, based notification area update, e.g. the UE may leave the configured RNA area while being in RRCJNACTIVE, which requires to validate the AI / ML model performance for the new environment, leaving or entering a certain location, e.g., the UE may be configured with a specific location / area, responsive to receiving a positioning signal, like an SRS, e.g. after SRS configuration has been adapted by the network to ensure e.g. optimal resource allocation or interference management, UE may consider this change as a trigger to monitor AI / ML models performance.

[0139] In accordance with embodiments, the UE 400 is configured or preconfigured with a nonconnected state monitoring configuration. The monitoring configuration may indicate one or more of the following:

[0140] How to compress the performance data, e.g., by averaging over a certain averaging window to reduce the amount of data to be transmitted, by vector quantization, or by scalar quantization - A selection of some of the performance data to reduce the amount of data to be transmitted.

[0141] - A non-connected state monitoring periodicity

[0142] The non-connected state monitoring periodicity may be reduced when compared to a connected state monitoring periodicity for a monitoring in the connected state. The non-connected state monitoring periodicity may be indicated explicitly or implicitly, e.g., derived from a configured grant, CG, configuration. For example, the UE 400 may be configured with a CG configuration for data transmissions during the nonconnected mode. Then, the periodicity of this CG configuration may indicate the periodicity of the monitoring in the non-connected state, e.g. monitor always before or during or after a CG transmission.

[0143] - A non-connected state monitoring window configuration defining, e.g. via an RRC configuration, parameters which are to be used during the non-connected mode to generate a measurement report, e.g., periodicity, triggering thresholds, averaging window, KPIs to monitor, measurements to report, number or set of monitored models, maximum size of a measurement report. For example, the UE may not monitor all performance data but instead only configured performance data, possibly averaged over a configured averaging window. Also the monitoring and / or reporting may happen only after a certain trigger event happened, e.g. the performance drops below a certain threshold. This threshold may be indicated by this configuration.

[0144] In accordance with embodiments, the non-connected AI / ML models or functionalities 420 are defined by a non-connected AI / ML configuration with which the UE 400 is configured or preconfigured. For example, when using the non-connected AI / ML models or functionalities 420, the UE 400 is to activate one or more AI / ML models or AI / ML functionalities 414 with which the UE 400 is configured or preconfigured. The non-connected AI / ML configuration may be:

[0145] Preconfigured in a wireless communication network specification so that certain Al types, e.g., a beam predictor Al, are activated in the non-connected state.

[0146] Configured by a system information message received at the UE 400, like a system information block, SIB.

[0147] Defined by the model. For example, the model may include a pre-configuration indication that it is to be used in the non-connected state.

[0148] Configured by the wireless communication network to the UE while being in the connected state, e.g., using an RRC signaling, using one or more MAC CEs, in a resource pool, RP, configuration, or in a bandwidth part, BWP configuration. In accordance with embodiments, the preconfigured or configured non-connected AI / ML configuration may contain one or more of the following:

[0149] One or more AI / ML functionalities that the UE 400 is allowed to activate in the nonconnected state.

[0150] One or more AI / ML models that the UE 400 is allowed to activate in the nonconnected state.

[0151] One or more AI / ML parameters to be used by the UE 400 while being in the nonconnected state.

[0152] One or more conditions when to use AI / ML in the non-connected state.

[0153] - A configuration of a life cycle management, LCM, phases operation in the nonconnected state, e.g., an activation of a monitoring of the Al or a activation / deactivation of a training of the Al.

[0154] For example, even in a non-connected state the AI / ML model’s performance may need to be monitored. For example a „beam predictor Al“ may monitor and log the predictions it makes. This monitoring may then be aggregated and / or saved before being reported to the network later. The monitoring may also depend on a condition making sure that only relevant data gets reported.

[0155] In a further embodiment the UE may have the capability to continue .learning' even in the non-connected state. This may involve adjusting its parameters based on the data it encounters. When training of the Al is deactivated this is disabled and the model operates in a .fixed' mode, either keeping its last state or returning to a preconfigured state.

[0156] One or more conditions for performing a fallback to non-AI procedures.

[0157] - A deactivation of some of the AI / ML models or AI / ML functionalities with which the UE is configured or preconfigured.

[0158] An exemplary non-connected AI / ML configuration may be as follows:

[0159] Non-Connected-Mode-AI-Config := SEQUENCE {

[0160] Supported-AI-Functionalities SEQUENCE (SIZE(1 ..maxAIFunctionalities)) OF

[0161] Al-Functionality OPTIONAL,

[0162] Supported-AI-Models SEQUENCE (SIZE(1 ..maxAIModels)) OF model-ID

[0163] OPTIONAL,

[0164] Supporting-CellGroups SEQUENCE (SIZE(1 ..maxAICellGroups)) OF

[0165] CellGroupId OPTIONAL, Supporting-PhysCells SEQUENCE (SIZE(1 ..maxAICells)) OF PhysCellld

[0166] OPTIONAL,

[0167] Supported-RSRP-Range RSRP-Range OPTIONAL,

[0168] UnlicensedOnly ENUMERATED{enabled} OPTIONAL,

[0169] SupportedBands SEQUENCE(SIZE(1 ..maxBands)) OF

[0170] FreqBandlndicatorNR OPTIONAL,

[0171] }

[0172] For example, Supported-AI-Functionalities may be a list (of maximum length maxAl Functionalities ) of supported Al Functionalities in non-connected mode. Supported- AI-Functionalities may be a list (of maximum length maxAIModels) of supported Al models in non-connected mode. Supporting-CellGroups and Supporting-PhysCells may be list of cell group and cell IDs, respectively, that support non-connected mode Al operation. Supported-RSRP-Range may be a range of the RSRP in which non-connected mode Al is supported for a specific cell. UnlicensedOnly may indicate whether non-connected mode Al is supported only in unlicensed bands and SupportedBands may indicate the bands on which non-connected mode Al is supported.

[0173] In the embodiments, the UE 400 may use the non-connected Al models or functionalities only in case one or more conditions are met. In accordance with embodiments, the one or more conditions may include one or more of the following:

[0174] The UE 400 is allowed to operate the one or more non-connected AI / ML models or functionalities.

[0175] For example, the UE 400 may operate the one or more non-connected Al models or functionalities when one or more of the following applies: o The UE 400 is of the type supporting the use of non-connected AI / ML models or functionalities, e.g., an eMBB device, a V2X-UE, a UE with NTN support, or an loT device. o The UE 400 has a sufficient battery level and / or sufficient memory storage for operating the one or more non-connected Al models or functionalities, o The UE 400 is communicating on a certain frequency band, e.g., an unlicensed or licensed band supported by AI / ML, o The UE 400 is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs supporting AI / ML, o The UE 400 operates in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, UMa, a rural microcell, RMi, a rural macrocell RMa, or indoors, o One or more mobility aspects, like speed, acceleration, Doppler delay profile, are met. o The UE 400 is in a certain location or area.

[0176] One or more requirements for a transmission while being in the non-connected state are met.

[0177] For example, the one or more requirements for a transmission while being in the non-connected state may include one or more of the following: o A random access channel, RACH, procedure failed for a certain number of attempts. o A channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold. o An interference level is below or above a certain threshold. o A certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection.

[0178] - An operation failed without AI / ML, e.g., random access channel, RACH, data is not transmitted after some attempts.

[0179] - An AI / ML usage duration did not surpass a predefined time limit.

[0180] - A time to start using Al has lapsed. For example, to avoid unnecessary Al activations / deactivations due to a quick transition back into the connected mode after entering the non-connected mode, the UE 400 needs to stay in non-connected mode for some period of time before activating AI / ML.

[0181] In accordance with further embodiments the UE 400 may fall back to using only non-AI / ML procedures or other AI / ML models in certain situations. In accordance with embodiments, the UE 400 may switch to a fallback operation or to a non-AI operation or to another AI / ML model or functionality when one or more conditions are met. Fallback is referring to a procedure that the UE 400 applies for the same task before the AI / ML is activated. Since every AI / ML has to be activated first, all UEs are capable to perform the same tasks without AI / ML, which may involve non-AI procedures. However, in some cases it may also involve the use of AI / ML but in a way that is transparent to 3GPP, and the specification is not aware of any AI / ML usage. The one or more conditions may include one or more of the following:

[0182] The UE 400 is not capable to operate the one or more non-connected AI / ML models or functionalities.

[0183] For example, the UE 400 may not be capable to operate the one or more nonconnected Al / M L models or functionalities when one or more of the following applies: o The UE 400 is of the type not supporting the use of non-connected AI / ML models or functionalities. o The UE 400 has an insufficient battery level and / or insufficient memory storage for operating the one or more non-connected Al models or functionalities. o The UE 400 is communicating on a certain frequency band, e.g., an unlicensed or licensed band not supported by AI / ML. o The UE 400 is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs not supporting Al. o The UE 400 does not operate in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, UMa, a rural microcell, RMi, a rural macrocell RMa, or indoors. o One or more mobility aspects, like speed, acceleration, Doppler delay profile, are not met. o The UE 400 is not in a certain location or area.

[0184] One or more requirements for a transmission while being in the non-connected state are not met.

[0185] For example, the one or more requirements for a transmission while being in the non-connected state may include one or more of the following: o A random access channel, RACH, procedure failed for a certain number of attempts. o A channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold. o An interference level is below or above a certain threshold. o A certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection.

[0186] - An operation using AI / ML failed, e.g., random access channel, RACH, data is not transmitted after some attempts

[0187] - An Al usage duration surpassed a predefined time limit.

[0188] - A time to start using AI / ML has not yet lapsed.

[0189] In accordance with embodiments, the UE 400 may signal that it supports non-connected Al models. In accordance with embodiments, the UE 400 may indicate to the wireless communication network whether or not the UE 400 supports the use of non-connected AI / ML models or functionalities, e.g., in a UE Capability Report. In accordance with embodiments, the UE may indicate to the wireless communication network one or more of the following:

[0190] The non-connected AI / ML models or functionalities 420 supported by the UE 400.

[0191] The conditions for using the supported non-connected AI / ML models or functionalities 420.

[0192] The scenarios in which the supported non-connected AI / ML models or functionalities 420 may be used, e.g. certain locations, SNR ranges, urban, rural, etc.

[0193] - A maximum duration for using the supported non-connected AI / ML models or functionalities 420.

[0194] - A maximum frequency, e.g. a smallest periodicity, with which the supported nonconnected AI / ML models or functionalities 420 may be used,

[0195] - A maximum UE speed at which the supported non-connected AI / ML models or functionalities 420 may be used.

[0196] - A number of simultaneously supported non-connected AI / ML models or functionalities.

[0197] - A number of processing units that may be simultaneously occupied by all supported non-connected AI / ML models or functionalities.

[0198] - A delay in activating a supported non-connected AI / ML model or functionality, e.g. 2 ms.

[0199] Whether generalization of the supported non-connected AI / ML models or functionalities is performed by the UE 400 with or without any signaling from the wireless communication network.

[0200] In accordance with embodiments, the UE 400 may be configured or preconfigured with a separate set of RACH resources. In accordance with embodiments, the Al UE 400 may be more efficient in performing the RACH, e.g. due to a RACH resource selection Al or an enhanced beam management, etc. In that case, the efficiency is further improved if only Al UEs, possibly of the same kind, use a separate set of RACH resources.

[0201] In accordance with embodiments, the wireless communication network may be a 5G standalone, SA, network or a 5G non-standalone, NSA, network. In accordance with embodiments, for example when the wireless communication network is a 5G NSA network, the UE 400 may switch from an E-UTRA / EPC non-connected state in which the nonconnected AI / ML models or functionalities 420 are not supported to an NR / 5GC RRC nonconnected state for enabling the UE 400 to use the non-connected AI / ML models or functionalities 420. In accordance with embodiments, the UE 400 may switch from the E- UTRA / EPC non-connected state to the NR / 5GC RRC non-connected state only when one or more conditions are met.

[0202] In accordance with further embodiments, the LIE 400 may activate different ones of the nonconnected AI / ML models or functionalities 420 while being in the non-connected state. In other words, the UE 400 may switch form currently active AI / ML models or functionalities to other AI / ML models or functionalities. In accordance with embodiments, the UE 400 may switch to another non-connected AI / ML model or functionality only when one or more conditions are met.

[0203] The above-mentioned one or more conditions for switching from the E-UTRA / EPC nonconnected state to the NR / 5GC RRC non-connected state or for switching to another nonconnected AI / ML model or functionality may include one or more of the following:

[0204] The UE 400 is allowed to operate the one or more non-connected AI / ML models or functionalities.

[0205] For example, the UE 400 may not be capable to operate the one or more nonconnected AI / ML models or functionalities when one or more of the following applies:

[0206] The UE 400 is of the type supporting the use of non-connected AI / ML models or functionalities. o The UE 400 has a sufficient battery level and / or sufficient memory storage for operating the one or more non-connected AI / ML models or functionalities, o The UE 400 is communicating on a certain frequency band, e.g., an unlicensed or licensed band supported by AI / ML. o The UE 400 is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs supporting AI / ML. o The UE 400 operates in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, UMa, a rural microcell, RMi, a rural macrocell RMa, or indoors, o One or more mobility aspects, like speed, acceleration, Doppler delay profile, are met. o The UE 400 is in a certain location or area.

[0207] One or more requirements for a transmission while being in the non-connected state are met. For example, the one or more requirements for a transmission while being in the non-connected state may include one or more of the following: o A random access channel, RACH, procedure failed for a certain number of attempts. o A channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold. o An interference level is below or above a certain threshold. o A certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection,

[0208] - An operation failed without AI / ML, e.g., random access channel, RACH, data is not transmitted after some attempts.

[0209] - An AI / ML usage duration did not surpass a predefined time limit.

[0210] - A time to start using AI / ML has lapsed.

[0211] In accordance with further embodiments, in the non-connected state the UE 400 uses the one or more AI / ML models or AI / ML functionalities to perform one or more tasks associated with a data transmission by the UE, like an early data transmission, EDT, or a small data transmission, SDT.

[0212] General

[0213] Embodiments of the present invention have been described in detail above, and the respective embodiments and aspects may be implemented individually or two or more of the embodiments or aspects may be implemented in combination.

[0214] In accordance with embodiments, the wireless communication system may include a terrestrial network, or a non-terrestrial network, or networks or segments of networks using as a receiver an airborne vehicle or a space-borne vehicle, or a combination thereof. Further, the wireless communication system may by a system or network different from the above described 4G or 5G mobile communication systems, rather, embodiments of the inventive approach may also be implemented in any other wireless communication network, e.g., in a private network, such as an Intranet or any other type of campus networks, or in a WiFi communication system.

[0215] In accordance with embodiments of the present invention, a user device comprises one or more of the following: a power-limited UE, or a hand-held UE, like a UE used by a pedestrian, and referred to as a Vulnerable Road User, VRU, or a Pedestrian UE, P-UE, or an on-body or hand-held UE used by public safety personnel and first responders, and referred to as Public safety UE, PS-UE, or an loT UE, e.g., a sensor, an actuator or a UE provided in a campus network to carry out repetitive tasks and requiring input from a gateway node at periodic intervals, a mobile terminal, or a stationary terminal, or a cellular loT-UE, or a vehicular UE, or a vehicular group leader (GL) UE, or a sidelink relay, or an loT or narrowband loT, NB-loT, device, or wearable device, like a smartwatch, or a fitness tracker, or smart glasses, or a ground based vehicle, or an aerial vehicle, or a drone, or a moving base station, or road side unit (RSU), or a building, or any other item or device provided with network connectivity enabling the item / device to communicate using the wireless communication network, e.g., a sensor or actuator, or any other item or device provided with network connectivity enabling the item / device to communicate using a sidelink the wireless communication network, e.g., a sensor or actuator, or a Wi-Fi device, like a station (STA), access point (AP), node or mesh node, or mesh point, or Mesh AP, or any sidelink capable network entity.

[0216] In accordance with embodiments of the present invention, a network entity comprises one or more of the following: a macro cell base station, or a small cell base station, or a central unit of a base station, an integrated access and backhaul, IAB, node, or a distributed unit of a base station, or a road side unit (RSU), or a Wi-Fi device such as an access point (AP) or mesh node (Mesh AP), or a remote radio head, or an AMF, or a MME, or a SMF, or a core network entity, or mobile edge computing (MEC) entity, or a network slice as in the NR or 5G core context, or any transmission / reception point, TRP, enabling an item or a device to communicate using the wireless communication network, the item or device being provided with network connectivity to communicate using the wireless communication network.

[0217] Although some aspects of the described concept have been described in the context of an apparatus, it is clear, that these aspects also represent a description of the corresponding method, where a block or a device corresponds to a method step or a feature of a method step. Analogously, aspects described in the context of a method step also represent a description of a corresponding block or item or feature of a corresponding apparatus.

[0218] Various elements and features of the present invention may be implemented in hardware using analog and / or digital circuits, in software, through the execution of instructions by one or more general purpose or special-purpose processors, or as a combination of hardware and software. For example, embodiments of the present invention may be implemented in the environment of a computer system or another processing system. Fig. 8 illustrates an example of a computer system 600. The units or modules as well as the steps of the methods performed by these units may execute on one or more computer systems 600. The computer system 600 includes one or more processors 602, like a special purpose or a general-purpose digital signal processor. The processor 602 is connected to a communication infrastructure 604, like a bus or a network. The computer system 600 includes a main memory 606, e.g., a random-access memory, RAM, and a secondary memory 608, e.g., a hard disk drive and / or a removable storage drive. The secondary memory 608 may allow computer programs or other instructions to be loaded into the computer system 600. The computer system 600 may further include a communications interface 610 to allow software and data to be transferred between computer system 600 and external devices. The communication may be in the from electronic, electromagnetic, optical, or other signals capable of being handled by a communications interface. The communication may use a wire or a cable, fiber optics, a phone line, a cellular phone link, an RF link and other communications channels 612.

[0219] The terms “computer program medium” and “computer readable medium” are used to generally refer to tangible storage media such as removable storage units or a hard disk installed in a hard disk drive. These computer program products are means for providing software to the computer system 600. The computer programs, also referred to as computer control logic, are stored in main memory 606 and / or secondary memory 608. Computer programs may also be received via the communications interface 610. The computer program, when executed, enables the computer system 600 to implement the present invention. In particular, the computer program, when executed, enables processor 602 to implement the processes of the present invention, such as any of the methods described herein. Accordingly, such a computer program may represent a controller of the computer system 600. Where the disclosure is implemented using software, the software may be stored in a computer program product and loaded into computer system 600 using a removable storage drive, an interface, like communications interface 610.

[0220] The implementation in hardware or in software may be performed using a digital storage medium, for example cloud storage, a floppy disk, a DVD, a Blue-Ray, a CD, a ROM, a PROM, an EPROM, an EEPROM or a FLASH memory, having electronically readable control signals stored thereon, which cooperate or are capable of cooperating with a programmable computer system such that the respective method is performed. Therefore, the digital storage medium may be computer readable. Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperating with a programmable computer system, such that one of the methods described herein is performed.

[0221] Generally, embodiments of the present invention may be implemented as a computer program product with a program code, the program code being operative for performing one of the methods when the computer program product runs on a computer. The program code may for example be stored on a machine readable carrier.

[0222] Other embodiments comprise the computer program for performing one of the methods described herein, stored on a machine readable carrier. In other words, an embodiment of the inventive method is, therefore, a computer program having a program code for performing one of the methods described herein, when the computer program runs on a computer.

[0223] A further embodiment of the inventive methods is, therefore, a data carrier or a digital storage medium, or a computer-readable medium comprising, recorded thereon, the computer program for performing one of the methods described herein. A further embodiment of the inventive method is, therefore, a data stream or a sequence of signals representing the computer program for performing one of the methods described herein. The data stream or the sequence of signals may for example be configured to be transferred via a data communication connection, for example via the Internet. A further embodiment comprises a processing means, for example a computer, or a programmable logic device, configured to or adapted to perform one of the methods described herein. A further embodiment comprises a computer having installed thereon the computer program for performing one of the methods described herein.

[0224] In some embodiments, a programmable logic device, for example a field programmable gate array, may be used to perform some or all of the functionalities of the methods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor in order to perform one of the methods described herein. Generally, the methods are preferably performed by any hardware apparatus.

[0225] The above-described embodiments are merely illustrative for the principles of the present invention. It is understood that modifications and variations of the arrangements and the details described herein are apparent to others skilled in the art. It is the intent, therefore, to be limited only by the scope of the impending patent claims and not by the specific details presented by way of description and explanation of the embodiments herein.

Claims

CLAIMS1 . A user device, UE, for a wireless communication network, wherein the UE is configured or preconfigured with at least one Artificial Intelligence / Machine Learning model, AI / ML model, or at least one AI / ML functionality for performing one or more tasks, wherein, if the UE is in a connected state, the UE is to receive an indication from the wireless communication network for activating one or more connected AI / ML models or functionalities, the one or more connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities, and wherein, if the UE is in a non-connected state, the UE is to use one or more non-connected AI / ML models or functionalities, the one or more non-connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities.

2. The user device, UE, of claim 1 , wherein the one or more non-connected AI / ML models or functionalities comprise one or more of the following: one or more or all of the AI / ML models or AI / ML functionalities from the connectedAI / ML models or functionalities, one or more or all of the AI / ML models or AI / ML functionalities from the connected AI / ML models or functionalities with different Al parameters, one or more AI / ML models or AI / ML functionalities different from the connected AI / ML models or functionalities.

3. The user device, UE, of claim 1 or 2, wherein, if the UE enters from the connected state into the non-connected state, the UE is to keep one or more or all of the activated AI / ML models or AI / ML functionalities as the non-connected AI / ML models or functionalities active for performing one or more of the tasks at least during part of the time the UE is in the non-connected state, or if the UE is in the non-connected state and the use of non-connected AI / ML models or functionalities is enabled, the UE is to activate the one or more non-connected AI / ML models or functionalities for performing one or more of the tasks at least during part of the time the UE is in the non-connected state.

4. The user device, UE, of claim 3, wherein the UE is to activate the one or more nonconnected AI / ML models or functionalities if the UE enters the non-connected state from a state, like a LTE connected or LTE idle state, in which no AI / ML models or AI / ML functionalities had been activated in the UE by the wireless communication network, or if the UE, upon entering the non-connected state from the connected state, deactivated all connected AI / ML models or functionalities that had been activated in the UE by the wireless communication network during the connected state.

5. The user device, UE, of any one of the preceding claims, wherein the connected state comprises a state, like an RRC_CONNECTED state, where the UE has a connection with the wireless communication network, and the non-connected state comprises one or more of the following states: a first non-connected state, like an RRCJDLE state, where the UE has no connection with the wireless communication network, a second non-connected state, like an RRCJNACTIVE state, where the UE has a connection with the wireless communication network but it is not active, a third non-connected state, like an RRC_NON_CONNECTED_AI state, where the UE has no connection with the wireless communication network or has a connection with the wireless communication network but it is not active, and where the UE is allowed to use the one or more non-connected AI / ML models or functionalities.

6. The user device, UE, of claim 5, wherein the UE is to use the same or different nonconnected AI / ML models or functionalities in the first and second states.

7. The user device, UE, of any one of the preceding claims, wherein, in the nonconnected state, the UE is to train the one or more non-connected AI / ML models or functionalities.

8. The user device, UE, of any one of the preceding claims, wherein, in the nonconnected state, the UE is to monitor a performance of one or more of the non-connected AI / ML models or functionalities and / or one or more of the AI / ML models or AI / ML functionalities with which the UE is configured or preconfigured and which are inactive.

9. The UE of claim 8, wherein the UE is to send a performance report- while being in the non-connected state, e.g., using an early data transmission, EDT, or a small data transmission, SDT, or- when returning to the connected state.

10. The user device, UE, of claim 8 or 9, wherein the UE is to monitor the performance periodically, e.g., during a wake-up time of a paging cycle or a discontinuous reception, DRX, cycle, or a dedicated monitoring cycle, and / or responsive to one or more certain conditions.

11. The user device, UE, of claim 10, wherein the certain condition comprises one or more of the following: a performance of the AI / ML model or AI / ML functionality is not within one or more predefined boundaries, a signaling from the wireless communication network causing the UE to monitor the performance, a change in an operational situation of the UE.

12. The user device, UE, of claim 11 , wherein the signaling from the wireless communication network comprises one or more of the following: a paging message or an early paging indicator from a base station of the wireless communication network, a RAN-based Notification Area (RNA) update information, e.g., including an updated list of cells of the wireless communication network, a scheduling or non-scheduling DCI from a base station of the wireless communication network, a broadcast signaling included in a system information message, e.g., in a system information block, SIB, or in a master information block, MIB, a wakeup signal, reference signals to be used for certain procedure, e.g., a positioning procedure using sounding reference signals, SRSs.

13. The user device, UE, of claim 11 or 12, wherein the change in an operational situation of the UE comprises one or more of the following:a cell reselection or selection, e.g. the UE enters an area covered by another cell than the one it is connected to, a Public Land Mobile Network, PLMN, selection, e.g. when UE is roaming and selects the most suitable PLMN, a change in measurements in neighboring cells, e.g. RSRP, RSRQ, RSSI, SINR, SNR of neighboring cells, a state transition from the connected state into the non-connected state, a state transition from the first non-connected state, like an RRCJDLE state, to second non-connected state, like an RRC_IN ACTIVE state, a change in an radio access network, RAN, based notification area update, leaving or entering a certain location, e.g., the UE may be configured with a specific location / area, responsive to receiving a positioning signal, like an SRS.

14. The user device, UE, of any one of claims 8 to 13, wherein the UE is configured or preconfigured with a non-connected state monitoring configuration.

15. The user device, UE, of claim 14, wherein the monitoring configuration indicates one or more of the following: how to compress the performance data, selecting some of the performance data to reduce the amount of data to be transmitted, a non-connected state monitoring periodicity, a non-connected state monitoring window configuration defining, e.g. via an RRC configuration, parameters which are to be used during the non-connected mode to generate a measurement report.

16. The user device, UE, of any one of the preceding claims, wherein, for using the nonconnected AI / ML models or functionalities, the UE is to activate one or more AI / ML models or AI / ML functionalities with which the UE is configured or preconfigured.

17. The user device, UE, of claim 16, wherein the one or more AI / ML models or AI / ML functionalities are preconfigured in a wireless communication network specification so that certain Al types, e.g., a beam predictor Al, are activated in the non-connected state, orconfigured by a system information message received at the UE, like a system information block, SIB, or defined by the model, or configured by the wireless communication network to the UE while being in the connected state, e.g., using an RRC signaling, using one or more MAC CEs, in a resource pool, RP, configuration, or in a bandwidth part, BWP configuration.

18. The user device, UE, of claim 16 or 17, wherein the preconfiguration or the configuration contain one or more of the following: one or more AI / ML functionalities that the UE is allowed to activate in the nonconnected state, one or more AI / ML models that the UE is allowed to activate in the non-connected state, one or more AI / ML parameters to be used by the UE while being in the nonconnected state, one or more conditions when to use AI / ML in the non-connected state, a configuration of a life cycle management, LCM, phases operation in the nonconnected state, e.g., an activation of a monitoring of the Al or a activation / deactivation of a training of the Al, one or more conditions for performing a fallback to non-AI procedures, a deactivation of some of the AI / ML models or AI / ML functionalities with which the UE is configured or preconfigured.

19. The user device, UE, of any one of the preceding claims, wherein the UE is to use the one or more non-connected Al models or functionalities only in case one or more conditions are met.

20. The user device, UE, of claim 19, wherein the one or more conditions comprises one or more of the following: the UE is allowed to operate the one or more non-connected AI / ML models or functionalities, one or more requirements for a transmission while being in the non-connected state are met, an operation failed without AI / ML, e.g., random access channel, RACH, data is not transmitted after some attempts, an AI / ML usage duration did not surpass a predefined time limit,a time to start using Al has lapsed.

21. The user device, UE, of claim 20, wherein the UE is allowed to operate the one or more non-connected Al models or functionalities when one or more of the following applies: the UE is of the type supporting the use of non-connected AI / ML models or functionalities, the UE has a sufficient battery level and / or sufficient memory storage for operating the one or more non-connected Al models or functionalities, the UE is communicating on a certain frequency band, e.g., an unlicensed or licensed band supported by AI / ML, the UE is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs supporting AI / ML, the UE operates in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, UMa, a rural microcell, RMi, a rural macrocell RMa, or indoors, one or more mobility aspects, like speed, acceleration, Doppler delay profile, are met. the UE is in a certain location or area.

22. The user device, UE, of claim 20 or 21 , wherein the one or more requirements for a transmission while being in the non-connected state comprise one or more of the following: a random access channel, RACH, procedure failed for a certain number of attempts, a channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold, an interference level is below or above a certain threshold, a certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection.

23. The user device, UE, of any one of the preceding claims, wherein the UE is to switch to a fallback operation or to a non-AI operation or to another AI / ML model or functionality when one or more conditions are met.

24. The user device, UE, of claim 23, wherein the one or more conditions comprises one or more of the following: the UE is not capable to operate the one or more non-connected AI / ML models or functionalities,one or more requirements for a transmission while being in the non-connected state are not met, an operation using AI / ML failed, e.g., random access channel, RACH, data is not transmitted after some attempts, an Al usage duration surpassed a predefined time limit, a time to start using AI / ML has not yet lapsed.

25. The user device, UE, of claim 24, wherein the UE is not capable to operate the one or more non-connected AI / ML models or functionalities when one or more of the following applies: the UE is of the type not supporting the use of non-connected AI / ML models or functionalities, the UE has an insufficient battery level and / or insufficient memory storage for operating the one or more non-connected Al models or functionalities, the UE is communicating on a certain frequency band, e.g., an unlicensed or licensed band not supported by AI / ML, the UE is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs not supporting Al, the UE does not operate in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, Uma, a rural microcell, Rmi, a rural macrocell, RMa, or indoors, one or more mobility aspects, like speed, acceleration, Doppler delay profile, are not met. the UE is not in a certain location or area.

26. The user device, UE, of claim 24 or 25, wherein the one or more requirements for a transmission while being in the non-connected state comprise one or more of the following: a random access channel, RACH, procedure failed for a certain number of attempts, a channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold, an interference level is below or above a certain threshold, a certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection.

27. The user device, UE, of any one of the preceding claims, wherein the UE is to indicate to the wireless communication network whether or not the UE supports the use of non-connected AI / ML models or functionalities, e.g., in a UE Capability Report.

28. The user device, UE, of claim 27, wherein the UE is to indicate to the wireless communication network one or more of the following: the non-connected AI / ML models or functionalities supported by the UE, conditions for using supported non-connected AI / ML models or functionalities, scenarios in which supported non-connected AI / ML models or functionalities may be used, e.g. certain locations, SNR ranges, urban, rural, etc., a maximum duration for using the supported non-connected AI / ML models or functionalities, a maximum frequency, e.g. a smallest periodicity, with which the supported nonconnected AI / ML models or functionalities may be used, a maximum UE speed at which the supported non-connected AI / ML models or functionalities may be used, a number of simultaneously supported non-connected AI / ML models or functionalities, a number of processing units that may be simultaneously occupied by all supported non-connected AI / ML models or functionalities, a delay in activating a supported non-connected AI / ML model or functionality, e.g. 2 ms,- whether generalization of the supported non-connected AI / ML models or functionalities is performed by the UE with or without any signaling from the wireless communication network.

29. The user device, UE, of any one of the preceding claims, wherein the UE is configured or preconfigured with a separate set of RACH resources.

30. The user device, UE, of any one of the preceding claims, wherein the UE is to switch from an E-UTRA / EPC non-connected state in which the non-connected AI / ML models or functionalities are not supported to an NR / 5GC RRC non-connected state for enabling the UE to use the non-connected Al models or functionalities.31 . The user device, UE, of claim 30, wherein the UE is to switch from the E-UTRA / EPC non-connected state to the NR / 5GC RRC non-connected state only when one or more conditions are met.

32. The user device, UE, of any one of the preceding claims, wherein the UE is to switch to another non-connected AI / ML model or functionality only when one or more conditions are met.

33. The user device, UE, of claim 32, wherein the one or more conditions comprise one or more of the following: the UE is allowed to operate the one or more non-connected AI / ML models or functionalities, one or more requirements for a transmission while being in the non-connected state are met, an operation failed without AI / ML, e.g., random access channel, RACH, data is not transmitted after some attempts, an AI / ML usage duration did not surpass a predefined time limit, a time to start using AI / ML has lapsed.

34. The user device, UE, of claim 33, wherein the UE is allowed to operate the one or more non-connected AI / ML models or functionalities when one or more of the following applies: the UE is of the type supporting the use of non-connected AI / ML models or functionalities, the UE has a sufficient battery level and / or sufficient memory storage for operating the one or more non-connected AI / ML models or functionalities, the UE is communicating on a certain frequency band, e.g., an unlicensed or licensed band supported by AI / ML, the UE is connected to or transmits to a certain base station, e.g., to a gNB out of a preconfigured or configured list indicating gNBs supporting AI / ML, the UE operates in a certain scenario, e.g., in an urban microcell, UMi, in an urban macrocell, UMa, a rural microcell, RMi, a rural macrocell, RMa, or indoors, one or more mobility aspects, like speed, acceleration, Doppler delay profile, are met. the UE is in a certain location or area.

35. The user device, UE, of claim 33 or 34, wherein the one or more requirements for a transmission while being in the non-connected state comprise one or more of the following: a random access channel, RACH, procedure failed for a certain number of attempts,a channel quality is sufficient for the transmission, e.g., a reference signal received power, RSRP, is above a certain threshold, an interference level is below or above a certain threshold, a certain channel condition applies, i.e., there is a line-of-sight, LOS, connection or a non-LOS, NLOS, connection.

36. The user device, UE, of any one of the preceding claims, wherein, if in the nonconnected state, the UE uses the one or more AI / ML models or AI / ML functionalities to perform one or more tasks associated with a data transmission by the UE, like an early data transmission, EDT, or a small data transmission, SDT.

37. The user device, UE, of any one of the preceding claims, wherein the one or more of tasks comprise one or more of the following:- AI / ML model based access to a RAN,- AI / ML model based network energy saving,- AI / ML model based resource allocation optimizations,- AI / ML model based scheduling optimizations,- AI / ML model based network slicing management,- AI / ML model based load balancing, an AI / ML model based mobility optimization,- AI / ML model based use cases, like o channel state information, CSI, feedback, like a CSI compression and / or a CSI prediction, or o beam management, or o positioning, like a direct AI / ML positioning (e.g., fingerprinting) and / or an AI / ML assisted positioning, o Handover like the AI / ML to predict when the UE should handover to which cell to reduce unnecessary handovers,- AI / ML model based mobility management, e.g., a handover, HO, prediction and / or conditional handover, CHO, prediction,- AI / ML model based modulation and coding scheme, MCS, selection,- AI / ML model based synchronization,- AI / ML model based encoding and / or decoding and / or precoding,- AI / ML model based modulation and / or demodulation,- AI / ML model based positioning or ranging,- AI / ML model based joint communication and sensing, JSAC,- AI / ML model based network offloading,- AI / ML model based feedback calculation, e.g., CSI / CQI / PMI / RI feedback,- AI / ML model based interference management,- AI / ML model based quality of experience, QoE, and / or quality of service, QoS, predictions,- AI / ML model based network traffic forecasting.

38. The user device, UE, of any of the preceding claims, wherein the UE comprise one or more of a power-limited UE, or a hand-held UE, like a UE used by a pedestrian, and referred to as a Vulnerable Road User, VRU, or a Pedestrian UE, P-UE, or an on-body or hand-held UE used by public safety personnel and first responders, and referred to as Public safety UE, PS-UE, or an loT UE or Ambient loT UE, e.g., a sensor, an actuator or a UE provided in a campus network to carry out repetitive tasks and requiring input from a gateway node at periodic intervals, or a mobile terminal, or a stationary terminal, or a cellular loT-UE, an industrial loT-UE, 11 oT, or a SL UE, or a vehicular UE, or a vehicular group leader UE, GL-UE, or a scheduling UE, S-UE, or an loT or narrowband loT, NB-loT, device, a NTN UE, or a WiFi device or WiFi station, STA, or a ground based vehicle, or an aerial vehicle, or a drone, or a moving base station, or road side unit, RSU, or a building, or any other item or device provided with network connectivity enabling the item / device to communicate using the wireless communication network, e.g., a sensor or actuator, or any other item or device provided with network connectivity enabling the item / device to communicate using a sidelink the wireless communication network, e.g., a sensor or actuator, or any sidelink capable network entity.

39. A wireless communication network, like a 3rdGeneration Partnership Project, 3GPP, system, comprising a one or more user devices, UEs, of any one of the preceding claims and one or more base stations, BSs.

40. The wireless communication network of claim 39, wherein the BS comprises one or more of a macro cell base station, or a small cell base station, or a central unit of a base station, or a distributed unit of a base station, or an Integrated Access and Backhaul, IAB, node, or a road side unit, RSU, or a WiFi access point, AP, or a UE, or a SL UE, or a group leader UE, GL-UE, or a relay or a remote radio head, or an AMF, or an SMF, or a core network entity, or mobile edge computing, MEC, entity, or a network slice as in the NR or 5G core context, or any transmission / reception point, TRP, enabling an item or a device tocommunicate using the wireless communication network, the item or device being provided with network connectivity to communicate using the wireless communication network.

41. A method for operating a user device, UE, for a wireless communication network, wherein the UE is configured or preconfigured with at least one Artificial Intelligence / Machine Learning model, AI / ML model, or at least one AI / ML functionality for performing one or more tasks, the method comprising: if the UE is in a connected state, receiving an indication from the wireless communication network for activating one or more connected AI / ML models or functionalities, the one or more connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities, and if the UE is in a non-connected state, using one or more non-connected AI / ML models or functionalities, the one or more non-connected AI / ML models or functionalities comprising one or more of the configured or preconfigured AI / ML models or AI / ML functionalities.

42. A non-transitory computer program product comprising a computer readable medium storing instructions which, when executed on a computer, perform the method of claim 41.

Images