User equipment ai computing in RRC idle or RRC inactive states

Abstract

In one aspect, a method performed by user equipment (UE) in communication with a network includes in response to one or more first conditions being satisfied, transitioning to a first sub-state of a radio resource control (RRC) inactive state or an RRC idle state, and in response to one or more second conditions being satisfied, transitioning to a second sub-state of the RRC inactive or the RRC idle state. In the first sub-state, the UE decodes a message associated with an artificial intelligence (AI) model when the message is received from the network. In the second sub-state, the UE ignores the message associated with the AI model when the message is received from the network.

Description

USER EQUIPMENT Al COMPUTING IN RRC IDLE OR RRC INACTIVE STATESCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims benefit and priority to U.S. Provisional Application No. 63 / 516,807, filed July 31, 2023, entitled “User Equipment Al Computing In RRC Idle Or RRC Inactive States.” The content of which is incorporated herein by reference in its entirety.TECHNICAL FIELD

[0002] This invention relates generally to wireless technology and more particularly to communications involving artificial intelligence (Al) model in an efficient manner.BACKGROUND

[0003] Fifth generation mobile network (5G) is a wireless standard that aims to improve upon data transmission speed, reliability, availability, and more. The wireless standard includes numerous procedures that may be implemented by a transmitting device or a receiving device that improves the latency, the speed, and the reliability of uplink and downlink transmissions.SUMMARY

[0004] Aspects of the present disclosure relate to 5G new radio (NR) operating in the licensed spectrum or in the shared and unlicensed spectrum (NR-U).

[0005] In an aspect, a method performed by user equipment (UE) in communication with a network comprises in response to one or more first conditions being satisfied, transitioning to a first sub-state of a radio resource control (RRC) inactive state or an RRC idle state, and in response to one or more second conditions being satisfied, transitioning to a second sub-state of the RRC inactive or the RRC idle state. In the first sub-state, the UE decodes a message associated with an artificial intelligence (Al) model when the message is received from the network. In the second sub-state, the UE ignores the message associated with the Al model when the message is received from the network.

[0006] In an aspect, a method performed by user equipment (UE) in communication with a network, comprises performing an operation associated with an artificial intelligence (Al) model, and receiving an indication from the network that is associatedwith completion of the operation. When the UE is in an RRC idle or RRC inactive state at the completion of the operation, the UE may transition to an RRC connected state based on the indication, and reporting a result of the completed operation that is associated with the Al model in the RRC connected state.

[0007] In an aspect, a method performed by user equipment (UE) in communication with a network, comprises performing an operation associated with an Al model. When the UE is in an RRC idle or RRC inactive state at completion of the operation, the UE may transition to an RRC connected state based on whether a condition is satisfied. The UE may report a result of the completed operation that is associated with the Al model in the RRC connected state.

[0008] In an aspect, a base station comprises a transceiver configured to communicate with a user equipment (UE), and a processor communicatively coupled to the transceiver and configured to perform the methods described herein from the perspective of the network.

[0009] In an aspect, a processor (e.g., a baseband processor) of a UE is configured to perform the methods described herein from the perspective of the UE.

[0010] Other technical features may be readily apparent to one skilled in the art from the following figures, descriptions, and claims.BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In the figures of the accompanying drawings, aspects described are illustrated by way of example and not by limitation.

[0012] FIG. 1 illustrates an example wireless communication system according to some aspects.

[0013] FIG. 2 illustrates uplink and downlink communications according to some aspects.

[0014] FIG. 3 illustrates an example block diagram of a user equipment (UE) according to some aspects.

[0015] FIG. 4 illustrates an example block diagram of a base station (BS) according to some aspects.

[0016] FIG. 5 illustrates an example block diagram of cellular communication circuitry, according to some aspects.

[0017] FIG. 6 shows an example diagram of a UE in communication with a network to communicate with respect to an Al operation, in accordance with some aspects.

[0018] FIG. 7 shows a diagram for UE states and sub-states that accommodate Al compute status of a UE, in accordance with some aspects.

[0019] FIG. 8 shows an example workflow of a UE that is in communication with a network with respect to Al-related paging, in accordance with some aspects.

[0020] FIG. 9 shows a method performed by a UE having sub-states in RRC idle or RRC inactive states, in accordance with some aspects.

[0021] FIG. 10 shows an example workflow performed by a UE or network for an Al operation, in accordance with some aspects.

[0022] FIG. 11 shows a method performed by a UE to report a result of an Al operation based on an indication from a network, in accordance with some aspects.

[0023] FIG. 12 shows an example workflow of a UE that is in communication with a network with respect to Al-related paging, in accordance with some aspects.

[0024] FIG. 13 shows a method performed by a UE to report a result of an Al operation based on internal conditions, in accordance with some aspects.DETAILED DESCRIPTION

[0025] A method and apparatus of a device that communicates wirelessly with a network (e.g., a 5G network) may perform artificial intelligence (Al) operations in an intelligent and efficient manner. It will be apparent, however, to one skilled in the art, that aspects of the present disclosure may be practiced without these specific details. In other instances, well-known components, structures, and techniques have not been shown in detail in order not to obscure the understanding of this description.

[0026] Reference in the specification to “some aspects” or “an aspect” means that a particular feature, structure, or characteristic described in connection with the aspect can be included in at least one aspect of the disclosure. The appearances of the phrase “in some aspects” in various places in the specification do not necessarily all refer to the same aspect.

[0027] In the following description and claims, the terms “coupled” and “connected,” along with their derivatives, may be used. “Coupled” is used to indicate that two or more elements, which may or may not be in direct physical or electrical contact with each other,co-operate or interact with each other. “Connected” is used to indicate the establishment of communication between two or more elements that are coupled with each other.

[0028] The processes depicted in the figures that follow, are performed by processing logic that comprises hardware (e.g., circuitry, dedicated logic, etc.), software (such as is run on a general-purpose computer system or a dedicated machine), or a combination of both. Although the processes are described below in terms of some sequential operations, it should be appreciated that some of the operations described may be performed in different order. Moreover, some operations may be performed in parallel rather than sequentially.

[0029] The terms “server,” “client,” and “device” are intended to refer generally to data processing systems rather than specifically to a particular form factor for the server, client, and / or device.

[0030] FIG. 1 illustrates a simplified example wireless communication system, according to some aspects. It is noted that the system of FIG. 1 is merely one example of a possible system, and that features of this disclosure may be implemented in any of various systems, as desired.

[0031] As shown, the example wireless communication system includes a base station 102A which communicates over a transmission medium with one or more user devices 106A, 106B, etc., through 106N. Each of the user devices may be referred to as a “user equipment” (UE).

[0032] The base station (BS) 102A may be a base transceiver station (BTS) or cell site (a “cellular base station”) and may include hardware that enables wireless communication with the UEs 106 A through 106N.

[0033] The communication area (or coverage area) of the base station may be referred to as a “cell.” The base station 102 A and the UEs 106 may be configured to communicate over the transmission medium using any of various radio access technologies (RATs), also referred to as wireless communication technologies, or telecommunication standards, such as GSM, UMTS (associated with, for example, WCDMA or TD-SCDMA air interfaces), LTE, LTE-Advanced (LTE-A), 5G new radio (5G NR), HSPA, 3GPP2 CDMA2000 (e.g, IxRTT, IxEV-DO, HRPD, eHRPD), etc. Note that if the base station 102A is implemented in the context of LTE, it may alternately be referred to as an ‘eNodeB’ or ‘eNB’. Note that if the base station 102 A is implemented in the context of 5G NR, it may alternately be referred to as ‘gNodeB’ or ‘gNB’.

[0034] As shown, the base station 102A may also be equipped to communicate with a network 100 (e.g., a core network of a cellular service provider, a telecommunication network such as a public switched telephone network (PSTN), and / or the Internet, among various possibilities). Thus, the base station 102A may facilitate communication between the user devices and / or between the user devices and the network 100. In particular, the cellular base station 102A may provide UEs 106 with various telecommunication capabilities, such as voice, SMS and / or data services.

[0035] Base station 102A and other similar base stations (such as base stations 102B . . . 102N) operating according to the same or a different cellular communication standard may thus be provided as a network of cells, which may provide continuous or nearly continuous overlapping service to UEs 106A-N and similar devices over a geographic area via one or more cellular communication standards.

[0036] Thus, while base station 102A may act as a “serving cell” for UEs 106A-N as illustrated in FIG. 1, each UE 106 may also be capable of receiving signals from (and possibly within communication range of) one or more other cells (which might be provided by base stations 102B-N and / or any other base stations), which may be referred to as “neighboring cells”. Such cells may also be capable of facilitating communication between user devices and / or between user devices and the network 100. Such cells may include “macro” cells, “micro” cells, “pico” cells, and / or cells which provide any of various other granularities of service area size. For example, base stations 102A-B illustrated in FIG. 1 might be macro cells, while base station 102N might be a micro cell. Other configurations are also possible.

[0037] In some aspects, base station 102 A may be a next generation base station, e.g., a 5G New Radio (5G NR) base station, or “gNB”. In some aspects, a gNB may be connected to a legacy evolved packet core (EPC) network and / or to a NR core (NRC) network. In addition, a gNB cell may include one or more transition and reception points (TRPs). In addition, a UE capable of operating according to 5G NR may be connected to one or more TRPs within one or more gNBs.

[0038] Note that a UE 106 may be capable of communicating using multiple wireless communication standards. For example, the UE 106 may be configured to communicate using a wireless networking (e.g., Wi-Fi) and / or peer-to-peer wireless communication protocol (e.g., Bluetooth, Wi-Fi peer-to-peer, etc.) in addition to at least one cellular communication protocol (e.g., GSM, UMTS (associated with, for example, WCDMA orTD-SCDMA air interfaces), LTE, LTE-A, 5G NR, HSPA, 3GPP2 CDMA2000 (e.g., IxRTT, IxEV-DO, HRPD, eHRPD), etc.). The UE 106 may also or alternatively be configured to communicate using one or more global navigational satellite systems (GNSS, e.g., GPS or GLONASS), one or more mobile television broadcasting standards (e.g., ATSC-M / H or DVB-H), and / or any other wireless communication protocol, if desired. Other combinations of wireless communication standards (including more than two wireless communication standards) are also possible.

[0039] FIG. 2 illustrates UE 106A that can be in communication with a base station 102 through uplink and downlink communications, according to some aspects. The UEs may each be a device with cellular communication capability such as a mobile phone, a hand-held device, a computer or a tablet, or virtually any type of wireless device.

[0040] The UE may include a processor that is configured to execute program instructions stored in memory. The UE may perform any of the method aspects described herein by executing such stored instructions. Alternatively, or in addition, the UE may include a programmable hardware element such as an FPGA (field-programmable gate array) that is configured to perform any of the method aspects described herein, or any portion of any of the method aspects described herein.

[0041] The UE may include one or more antennas for communicating using one or more wireless communication protocols or technologies. In some aspects, the UE may be configured to communicate using, for example, CDMA2000 (IxRTT / lxEV- DO / HRPD / eHRPD) or LTE using a single shared radio and / or GSM or LTE using the single shared radio. The shared radio may couple to a single antenna, or may couple to multiple antennas (e.g., for MIMO) for performing wireless communications. In general, a radio may include any combination of a baseband processor, analog RF signal processing circuitry (e.g., including filters, mixers, oscillators, amplifiers, etc.), or digital processing circuitry (e.g., for digital modulation as well as other digital processing). Similarly, the radio may implement one or more receive and transmit chains using the aforementioned hardware. For example, the UE 106 may share one or more parts of a receive and / or transmit chain between multiple wireless communication technologies, such as those discussed above.

[0042] In some aspects, the UE may include separate transmit and / or receive chains (e.g., including separate antennas and other radio components) for each wireless communication protocol with which it is configured to communicate. As a furtherpossibility, the UE may include one or more radios which are shared between multiple wireless communication protocols, and one or more radios which are used exclusively by a single wireless communication protocol. For example, the UE might include a shared radio for communicating using either of LTE or 5G NR (or LTE or IxRTTor LTE or GSM), and separate radios for communicating using each of Wi-Fi and Bluetooth. Other configurations are also possible.

[0043] FIG. 3 illustrates an example simplified block diagram of a communication device 106, according to some aspects. It is noted that the block diagram of the communication device of FIG. 3 is only one example of a possible communication device. According to aspects, communication device 106 may be a UE device, a mobile device or mobile station, a wireless device or wireless station, a desktop computer or computing device, a mobile computing device (e.g., a laptop, notebook, or portable computing device), a tablet and / or a combination of devices, among other devices. As shown, the communication device 106 may include a set of components 300 configured to perform core functions. For example, this set of components may be implemented as a system on chip (SOC), which may include portions for various purposes. Alternatively, this set of components 300 may be implemented as separate components or groups of components for the various purposes. The set of components 300 may be coupled (e.g., communicatively; directly or indirectly) to various other circuits of the communication device 106.

[0044] For example, the communication device 106 may include various types of memory (e.g., including NAND flash 310), an input / output interface such as connector I / F 320 (e.g., for connecting to a computer system; dock; charging station; input devices, such as a microphone, camera, keyboard; output devices, such as speakers; etc.), the display 360, which may be integrated with or external to the communication device 106, and cellular communication circuitry 330 such as for 5G NR, LTE, GSM, etc., and short to medium range wireless communication circuitry 329 (e.g., Bluetooth™ and WLAN circuitry). In some aspects, communication device 106 may include wired communication circuitry (not shown), such as a network interface card, e.g., for Ethernet.

[0045] The cellular communication circuitry 330 may couple (e.g., communicatively; directly or indirectly) to one or more antennas, such as antennas 335 and 336 as shown. The short to medium range wireless communication circuitry 329 may also couple (e.g., communicatively; directly or indirectly) to one or more antennas, such as antennas 337and 338 as shown. Alternatively, the short to medium range wireless communication circuitry 329 may couple (e.g., communicatively; directly or indirectly) to the antennas 335 and 336 in addition to, or instead of, coupling (e.g., communicatively; directly or indirectly) to the antennas 337 and 338. The short to medium range wireless communication circuitry 329 and / or cellular communication circuitry 330 may include multiple receive chains and / or multiple transmit chains for receiving and / or transmitting multiple spatial streams, such as in a multiple-input multiple output (MIMO) configuration.

[0046] In some aspects, as further described below, cellular communication circuitry 330 may include dedicated receive chains (including and / or coupled to, e.g., communicatively; directly or indirectly, dedicated processors and / or radios) for multiple radio access technologies (RATs) (e.g., a first receive chain for LTE and a second receive chain for 5G NR). In addition, in some aspects, cellular communication circuitry 330 may include a single transmit chain that may be switched between radios dedicated to specific RATs. For example, a first radio may be dedicated to a first RAT, e.g., LTE, and may be in communication with a dedicated receive chain and a transmit chain shared with an additional radio, e.g., a second radio that may be dedicated to a second RAT, e.g., 5G NR, and may be in communication with a dedicated receive chain and the shared transmit chain.

[0047] The communication device 106 may also include and / or be configured for use with one or more user interface elements. The user interface elements may include any of various elements, such as display 360 (which may be a touchscreen display), a keyboard (which may be a discrete keyboard or may be implemented as part of a touchscreen display), a mouse, a microphone and / or speakers, one or more cameras, one or more buttons, and / or any of various other elements capable of providing information to a user and / or receiving or interpreting user input.

[0048] The communication device 106 may further include one or more smart cards 345 that include SIM (Subscriber Identity Module) functionality, such as one or more UICC(s) (Universal Integrated Circuit Card(s)) cards 345.

[0049] As shown, the SOC 300 may include processor(s) 302, which may execute program instructions for the communication device 106 and display circuitry 304, which may perform graphics processing and provide display signals to the display 360. The processor(s) 302 may also be coupled to memory management unit (MMU) 340, which may be configured to receive addresses from the processor(s) 302 and translate thoseaddresses to locations in memory (e.g., memory 306, read only memory (ROM) 350, NAND flash memory 310) and / or to other circuits or devices, such as the display circuitry 304, short range wireless communication circuitry 229, cellular communication circuitry 330, connector I / F 320, and / or display 360. The MMU 340 may be configured to perform memory protection and page table translation or set up. In some aspects, the MMU 340 may be included as a portion of the processor(s) 302.

[0050] As noted above, the communication device 106 may be configured to communicate using wireless and / or wired communication circuitry. The communication device 106 may also be configured to determine a physical downlink shared channel scheduling resource for a user equipment device and a base station. Further, the communication device 106 may be configured to group and select CCs from the wireless link and determine a virtual CC from the group of selected CCs. The wireless device may also be configured to perform a physical downlink resource mapping based on an aggregate resource matching patterns of groups of CCs.

[0051] As described herein, the communication device 106 may include hardware and software components for implementing the above features for determining a physical downlink shared channel scheduling resource for a communications device 106 and a base station. The processor 302 of the communication device 106 may be configured to implement part or all of the features described herein, e.g., by executing program instructions stored on a memory medium (e.g., a non-transitory computer-readable memory medium). Alternatively (or in addition), processor 302 may be configured as a programmable hardware element, such as an FPGA (Field Programmable Gate Array), or as an ASIC (Application Specific Integrated Circuit). Alternatively (or in addition) the processor 302 of the communication device 106, in conjunction with one or more of the other components 300, 304, 306, 310, 320, 329, 330, 340, 345, 350, 360 may be configured to implement part or all of the features described herein.

[0052] In addition, as described herein, processor 302 may include one or more processing elements. Thus, processor 302 may include one or more integrated circuits (ICs) that are configured to perform the functions of processor 302. In addition, each integrated circuit may include circuitry (e.g., first circuitry, second circuitry, etc.) configured to perform the functions of processor(s) 302.

[0053] Further, as described herein, cellular communication circuitry 330 and short- range wireless communication circuitry 329 may each include one or more processingelements. In other words, one or more processing elements may be included in cellular communication circuitry 330 and, similarly, one or more processing elements may be included in short range wireless communication circuitry 329. Thus, cellular communication circuitry 330 may include one or more integrated circuits (ICs) that are configured to perform the functions of cellular communication circuitry 330. In addition, each integrated circuit may include circuitry (e.g., first circuitry, second circuitry, etc.) configured to perform the functions of cellular communication circuitry 230. Similarly, the short-range wireless communication circuitry 329 may include one or more ICs that are configured to perform the functions of short-range wireless communication circuitry 32. In addition, each integrated circuit may include circuitry (e.g., first circuitry, second circuitry, etc.) configured to perform the functions of short-range wireless communication circuitry 329.

[0054] FIG. 4 illustrates an example block diagram of a base station 102, according to some aspects. It is noted that the base station of FIG. 4 is merely one example of a possible base station. As shown, the base station 102 may include processor(s) 404 which may execute program instructions for the base station 102. The processor(s) 404 may also be coupled to memory management unit (MMU) 440, which may be configured to receive addresses from the processor(s) 404 and translate those addresses to locations in memory (e.g., memory 460 and read only memory (ROM) 450) or to other circuits or devices.

[0055] The base station 102 may include at least one network port 470. The network port 470 may be configured to couple to a telephone network and provide a plurality of devices, such as UE devices 106, access to the telephone network as described above in FIGS. 1 and 2.

[0056] The network port 470 (or an additional network port) may also or alternatively be configured to couple to a cellular network, e.g., a core network of a cellular service provider. The core network may provide mobility related services and / or other services to a plurality of devices, such as UE devices 106. In some cases, the network port 470 may couple to a telephone network via the core network, and / or the core network may provide a telephone network (e.g., among other UE devices serviced by the cellular service provider).

[0057] In some aspects, base station 102 may be a next generation base station, e.g., a 5G New Radio (5G NR) base station, or “gNB”. In such aspects, base station 102 may be connected to a legacy evolved packet core (EPC) network and / or to a NR core (NRC)network. In addition, base station 102 may be considered a 5G NR cell and may include one or more transition and reception points (TRPs). In addition, a UE capable of operating according to 5G NR may be connected to one or more TRPs within one or more gNBs. In some aspects, the base station can operate in 5G NR-U mode.

[0058] The base station 102 may include at least one antenna 434, and possibly multiple antennas. The at least one antenna 434 may be configured to operate as a wireless transceiver and may be further configured to communicate with UE devices 106 via radio 430. The antenna 434 communicates with the radio 430 via communication chain 432. Communication chain 432 may be a receive chain, a transmit chain or both. The radio 430 may be configured to communicate via various wireless communication standards, including, but not limited to, 5GNR, 5GNR-U, LTE, LTE-A, GSM, UMTS, CDMA2000, Wi-Fi, etc.

[0059] The base station 102 may be configured to communicate wirelessly using multiple wireless communication standards. In some instances, the base station 102 may include multiple radios, which may enable the base station 102 to communicate according to multiple wireless communication technologies. For example, as one possibility, the base station 102 may include an LTE radio for performing communication according to LTE as well as a 5G NR radio for performing communication according to 5G NR and 5G NR-U. In such a case, the base station 102 may be capable of operating as both an LTE base station and a 5G NR base station. As another possibility, the base station 102 may include a multi-mode radio which is capable of performing communications according to any of multiple wireless communication technologies (e.g., 5G NR and Wi-Fi, LTE and Wi-Fi, LTE and UMTS, LTE and CDMA2000, UMTS and GSM, etc ).

[0060] As described further subsequently herein, the BS 102 may include hardware and software components for implementing or supporting implementation of features described herein. The processor 404 of the base station 102 may be configured to implement or support implementation of part or all of the methods described herein, e.g., by executing program instructions stored on a memory medium (e.g., a non-transitory computer-readable memory medium). Alternatively, the processor 404 may be configured as a programmable hardware element, such as an FPGA (Field Programmable Gate Array), or as an ASIC (Application Specific Integrated Circuit), or a combination thereof. Alternatively (or in addition) the processor 404 of the BS 102, in conjunction with one ormore of the other components 430, 432, 434, 440, 450, 460, 470 may be configured to implement or support implementation of part or all of the features described herein.

[0061] In addition, as described herein, processor(s) 404 may be comprised of one or more processing elements. In other words, one or more processing elements may be included in processor(s) 404. Thus, processor(s) 404 may include one or more integrated circuits (ICs) that are configured to perform the functions of processor(s) 404. In addition, each integrated circuit may include circuitry (e.g., first circuitry, second circuitry, etc.) configured to perform the functions of processor(s) 404.

[0062] Further, as described herein, radio 430 may be comprised of one or more processing elements. In other words, one or more processing elements may be included in radio 430. Thus, radio 430 may include one or more integrated circuits (ICs) that are configured to perform the functions of radio 430. In addition, each integrated circuit may include circuitry (e.g., first circuitry, second circuitry, etc.) configured to perform the functions of radio 430.

[0063] FIG. 5 illustrates an example simplified block diagram of cellular communication circuitry, according to some aspects. It is noted that the block diagram of the cellular communication circuitry of FIG. 5 is only one example of a possible cellular communication circuit. According to aspects, cellular communication circuitry 330 may be included in a communication device, such as communication device 106 described above. As noted above, communication device 106 may be a user equipment (UE) device, a mobile device or mobile station, a wireless device or wireless station, a desktop computer or computing device, a mobile computing device (e.g., a laptop, notebook, or portable computing device), a tablet and / or a combination of devices, among other devices.

[0064] The cellular communication circuitry 330 may couple (e.g., communicatively; directly or indirectly) to one or more antennas, such as antennas 335 a-b and 336 as shown (in FIG. 3). In some aspects, cellular communication circuitry 330 may include dedicated receive chains (including and / or coupled to, e.g., communicatively; directly or indirectly, dedicated processors and / or radios) for multiple RATs (e.g., a first receive chain for LTE and a second receive chain for 5G NR). For example, as shown in FIG. 5, cellular communication circuitry 330 may include a modem 510 and a modem 520. Modem 510 may be configured for communications according to a first RAT, e.g., such as LTE or LTE-A, and modem 520 may be configured for communications according to a second RAT, e.g., such as 5G NR.

[0065] As shown, modem 510 may include one or more processors 512 and a memory 516 in communication with processors 512. Modem 510 may be in communication with a radio frequency (RF) front end 530. RF front end 530 may include circuitry for transmitting and receiving radio signals. For example, RF front end 530 may include receive circuitry (RX) 532 and transmit circuitry (TX) 534. In some aspects, receive circuitry 532 may be in communication with downlink (DL) front end 550, which may include circuitry for receiving radio signals via antenna 335a.

[0066] Similarly, modem 520 may include one or more processors 522 and a memory 526 in communication with processors 522. Modem 520 may be in communication with an RF front end 540. RF front end 540 may include circuitry for transmitting and receiving radio signals. For example, RF front end 540 may include receive circuitry 542 and transmit circuitry 544. In some aspects, receive circuitry 542 may be in communication with DL front end 560, which may include circuitry for receiving radio signals via antenna 335b.

[0067] In some aspects, a switch 570 may couple transmit circuitry 534 to uplink (UL) front end 572. In addition, switch 570 may couple transmit circuitry 544 to UL front end 572. UL front end 572 may include circuitry for transmitting radio signals via antenna 336. Thus, when cellular communication circuitry 330 receives instructions to transmit according to the first RAT (e.g., as supported via modem 510), switch 570 may be switched to a first state that allows modem 510 to transmit signals according to the first RAT (e.g., via a transmit chain that includes transmit circuitry 534 and UL front end 572). Similarly, when cellular communication circuitry 330 receives instructions to transmit according to the second RAT (e.g., as supported via modem 520), switch 570 may be switched to a second state that allows modem 520 to transmit signals according to the second RAT (e.g., via a transmit chain that includes transmit circuitry 544 and UL front end 572).

[0068] As described herein, the modem 510 may include hardware and software components for implementing the above features or for determining a physical downlink shared channel scheduling resource for a user equipment device and a base station, as well as the various other techniques described herein. The processors 512 may be configured to implement part or all of the features described herein, e.g., by executing program instructions stored on a memory medium (e.g., a non-transitory computer-readable memory medium). Alternatively (or in addition), processor 512 may be configured as a programmable hardware element, such as an FPGA (Field Programmable Gate Array), oras an ASIC (Application Specific Integrated Circuit). Alternatively (or in addition) the processor 512, in conjunction with one or more of the other components 530, 532, 534, 550, 570, 572, 335 and 336 may be configured to implement part or all of the features described herein.

[0069] In addition, as described herein, processors 512 may include one or more processing elements. Thus, processors 512 may include one or more integrated circuits (ICs) that are configured to perform the functions of processors 512. In addition, each integrated circuit may include circuitry (e.g., first circuitry, second circuitry, etc.) configured to perform the functions of processors 512.

[0070] As described herein, the modem 520 may include hardware and software components for implementing the above features for determining a physical downlink shared channel scheduling resource for a user equipment device and a base station, as well as the various other techniques described herein. The processors 522 may be configured to implement part or all of the features described herein, e.g., by executing program instructions stored on a memory medium (e.g., a non-transitory computer-readable memory medium). Alternatively (or in addition), processor 522 may be configured as a programmable hardware element, such as an FPGA (Field Programmable Gate Array), or as an ASIC (Application Specific Integrated Circuit). Alternatively (or in addition) the processor 522, in conjunction with one or more of the other components 540, 542, 544, 550, 570, 572, 335 and 336 may be configured to implement part or all of the features described herein.

[0071] In addition, as described herein, processors 522 may include one or more processing elements. Thus, processors 522 may include one or more integrated circuits (ICs) that are configured to perform the functions of processors 522. In addition, each integrated circuit may include circuitry (e.g., first circuitry, second circuitry, etc.) configured to perform the functions of processors 522.

[0072] 5G supports multi-antenna transmission, beam-forming, and simultaneous transmission from multiple geographically separates sites. Channels of different antenna ports that are relevant for a UE may differ, for example, in terms of radio channel properties. QCL antenna port may be geographically separated.

[0073] 5G physical channels provide flexible communication between the 5G base stations and the UEs. 5G NR has specified the physical channels for 5G networks that can be used either for Downlink or Uplink communication. 5G NR physical channels used foruplink communication includes the physical uplink shared channel (PUSCH), the physical uplink control channel (PUCCH), and the physical random-access channel (PRACH). Uplink signals such as DM-RS, PT-RS, and SRS are also supported. 5G NR supports the simultaneous transmission on PUSCH and PUCCH. PUSCH is typically used to carry the user data and optionally, can carry uplink control information (UCI).

[0074] PDSCH stands for Physical Downlink Shared Channel and is a channel used to deliver data from the base station (e.g., gNb) to the user equipment (UE) in the downlink direction. PDSCH supports high data rates and low latency for a wide range of applications and services. It uses advanced modulation and coding schemes, as well as multiple antenna techniques such as MIMO (Multiple Input Multiple Output), to maximize spectral efficiency and improve the overall performance of the network. PDSCH is also used in conjunction with other channels, such as the Physical Downlink Control Channel (PDCCH) and Physical Hybrid ARQ Indicator Channel (PHICH), to support features such as channel state information reporting, scheduling and retransmission of data packets, and HARQ (Hybrid Automatic Repeat Request) feedback. PDSCH enables the delivery of high-speed data and low-latency services to users in the downlink direction, and supports a range of advanced features and capabilities that promote efficient and reliable operation of the network.

[0075] Radio Resource Control (RRC) is a protocol layer that exists between the radio access network (RAN) and the core network in 5G. RRC is used to manage radio resources, including the establishment, maintenance, and release of radio connections between the user equipment (UE) and the base station. The RRC protocol performs several essential functions including connection establishment, connection maintenance, and connection release. RRC handles the procedures for establishing a connection between the UE and the base station. This includes procedures such as random access, initial cell search, and authentication. Once a connection is established, RRC is responsible for maintaining the connection and managing the radio resources efficiently. It ensures the appropriate quality of service (QoS) for the UE and handles mobility-related events, such as handovers between cells or handovers between different types of 5G networks (e.g., from 5G New Radio to 5G Core). When the UE no longer needs the connection or if there are other reasons for releasing it, RRC handles the procedures for releasing the connection and freeing up the associated radio resources. The RRC protocol operates on top of the physical layer and the medium access control (MAC) layer in the protocol stack. It communicateswith other layers in the 5G system, such as the packet data convergence protocol (PDCP), radio link control (RLC), and the user plane, to support efficient and reliable communication between the UE and the network.

[0076] RRC protocol defines several states that a UE can be in. These states determine the level of connectivity and resource allocation between the UE and the network. The RRC states in 5G are RRC IDLE, RRC INACTIVE, and RRC CONNECTED.

[0077] In RRC Idle (RRC IDLE) state, the UE is not actively connected to the network. It is not assigned any dedicated resources, and its radio interface is deactivated. The UE periodically monitors system information broadcasts from the network to stay informed about available cells and other relevant network parameters. When the UE needs to establish a connection or perform a service request, it transitions to a RRC connected state.

[0078] When the UE is in RRC inactive (RRC INACTIVE) state, it has limited 5G resources allocated and uses primarily the LTE network for connectivity. In this state, the UE can quickly transition to the RRC CONNECTED state when one or more triggering events occur, as defined by one or more standards.

[0079] RRC Connected (RRC CONNECTED) state represents an active connection between the UE and the network. It can be further divided into sub-states: a. RRC Connected lnit: The initial sub-state where the UE has just established an RRC connection with the network. b. RRC Connected: The UE is in a stable connected state, and it can exchange data and signaling with the network (e.g., over physical uplink and physical downlink channels). It can receive and transmit data, perform handovers, and execute various procedures. c. RRC Connected Reconfiguring: The UE is undergoing a reconfiguration of its RRC connection parameters, such as changing the radio bearer configuration or updating network configuration information. d. RRC Connected Suspend: The UE temporarily suspends its RRC connection while maintaining radio resources for a specified period. It is typically used in powersaving scenarios to conserve energy. e. RRC Connected Release: The UE is releasing its RRC connection, terminating the active connection with the network.

[0080] Each RRC state has its own specific behaviors performed by the UE and network which results in specific power consumption characteristics and signaling procedures for the state (or sub-state). The network and the UE manage RRC state transitions based on numerous factors such as network conditions, service requirements, mobility, and power optimization. As discussed, there may be additional state transition management and additional sub-states to account for artificial intelligence concerns and operations thereof.

[0081] In the present disclosure, unless otherwise indicated, the terms artificial intelligence (Al) and machine learning (ML) may be interchangeable, and an ML model may refer to an Al model, and vice versa.

[0082] Artificial intelligence (Al) refers to the development of computer systems that can perform tasks that typically involve human intelligence. These systems are designed to simulate human cognitive abilities, such as learning, reasoning, problem-solving, and perception. Al encompasses various subfields, including machine learning, natural language processing, computer vision, and robotics.

[0083] Machine learning is a subset of artificial intelligence (Al) that focuses on developing algorithms and models that allow computer systems to learn and improve from data without being explicitly programmed. In machine learning, instead of relying on explicit instructions, the system learns patterns and relationships from the available data and uses that knowledge to make predictions or take actions. An Al model may refer to an algorithm and its trained parameters that take an input and generate an output based on the input. Different models may include different algorithms. In some aspects described, the Al or ML model may include an artificial neural network.

[0084] The process of machine learning may involve training a model using a dataset, which consists of input data and corresponding output labels or desired outcomes. The model learns from this training data by identifying patterns, correlations, and statistical relationships. Once trained, the model can then be used to make predictions or decisions on new, unseen data.

[0085] There are various types of machine learning algorithms, including supervised learning, unsupervised learning, and reinforcement learning. Supervised learning involves providing the model with labeled training data, where the input and output pairs are explicitly provided. Unsupervised learning involves discovering patterns or structures inunlabeled data. Reinforcement learning involves training an agent to interact with an environment and learn optimal actions based on receiving rewards or penalties.

[0086] An Al model may include an artificial neural network (ANN). An ANN machine learning algorithm may consist of interconnected artificial neurons, also called nodes or units, organized into layers. Each node receives one or more inputs, performs a mathematical operation on these inputs, and produces an output. These inputs may be weighted, and the weights may dictate the strength or importance of each input. The outputs from the nodes in one layer become inputs to the nodes in the next layer, forming a network of interconnected nodes. An artificial neural network may include an input layer, one or more hidden layers, and an output layer. The input layer receives the initial data or features, the hidden layers perform computations on the inputs, and the output layer produces the final output or prediction. During training, the weights of the connections between nodes are adjusted to optimize the neural network's performance on a specific task. This adjustment is done using a process called backpropagation, which calculates the network's error with respect to the weights and updates those weights to reduce the error. The objective of training is to adjust those weights to minimize the difference between the network's output and the desired output for a given input. Once the difference satisfies a threshold, the neural network may be said to be trained. More generally, an Al model may be trained in such a manner to minimize the difference between the model's output and a desired output given training data. Once trained, the Al model may be used for inferencing - by providing inputs and using the Al model to infer a desired output.

[0087] Al and ML techniques are deemed to be an integral part of future wireless networks such as 5G-Advanced and 6G. In particular, it is envisioned that Al may play a role for Optimization of Network Functionalities and ALas-a-Service (AlaaS).

[0088] With regard to Optimization of Network Functionalities, Al models that are trained by the Al agent(s) in a network environment (e.g., a 5G or 6G platform) may be used to facilitate certain decision making in a network (e.g., a radio access network, UE and Core network). A UE may use an Al model to infer outputs that are used in various network tasks such as beam management, CSI enhancement, Positioning, Resource allocation, Network management (0AM), Route Selection, Energy Saving, Load Balancing, or other operations performed by UE. For example, the UE may provide input to a trained Al model to obtain an output, and based on the output, select an appropriatebeam to use for uplink or downlink, or to manage processes in view of current energy storage levels.

[0089] With regard to Al-as-a-Service (AlaaS), Al services (e.g., inferencing) can be consumed by applications initiated or being executed at either user or network side, and the service provider can be any Al agent reachable in the network, including the UE. Some of the inferencing use-cases of the Al models may include: one-shot object recognition at the UE based on Al model split, with model split executed at another network entity, Al model download from network / edge to the UE for video recognition in an AR display / gaming session, or federated learning based services with multiple UEs participating in a global Al training in a given session, i.e., multi-view object recognition.

[0090] In the disclosure, features are described that support UE performance of Al computing tasks including Al model training, Al inferencing, or both. To perform Al computing tasks, the UE may need to obtain data or information from another entity in the network (e.g., the gNB, or another UE).

[0091] For example, for Al model training, the UE may need to get new or updated training datasets from the network or another UE. For Al inferencing based on trained models, the UE may need to obtain an updated or refined Al model (or a respective model ID) from the network or other UE(s).

[0092] When the UE is operating in RRC-CONNECTED state, obtaining the Al-related data (e.g., the Al model, the training dataset, the Al model ID, the training dataset ID, or a combination thereof) may be performed in a direct manner. The UE may receive the Al model or training dataset or a respective ID for an Al model that is used for a respective Al task (e.g., inferencing or training) through one or more operations. For example, the dataset, the Al model, or a respective ID may be communicated by the network to the UE through the application layer within a standard application layer session. The session may be dedicated for the Al-related data.

[0093] When the UE is in RRC-IDLE / INACTIVE state, the mechanism in which a UE may obtain such Al-related data is unresolved. The UE should first return to RRC- CONNECTED state in order to retrieve the connectivity and receive the information required for Al computing, however, currently the procedure for RRC connection establishment is not based on the need of Al computing - there is no RRC establishment based on UE transitioning into AI / ML model training or inference stages. Instead, under current protocols, the transition to RRC connected state can be triggered by paging fromthe network (e.g., some downlink (DL) data intended to the UE becomes available), or by arrival of uplink (UL) data at the UE buffer. The UE initializes the random access procedures to establish the necessary UL and DL timing and identification protocols. Thus, there is an absence of solutions for the UE to obtain Al-related data when not in RRC- connected state.

[0094] In the present disclosure, problems are identified for the scenarios where a UE needs to perform Al computing tasks, which may include Al model training and inferencing, when it is in RRC inactive or RRC idle state.

[0095] A first problem considers how to characterize the UE with respect to its Al related compute status. This characterization facilitates the network (e.g., a gNB) to perform configuration procedures related to the UE Al compute operations, which in turn, may guide the UE and network to coordinate Al operations in a manner that responsibly utilizes energy storage and network resources.

[0096] A second problem considers how the network (e.g., gNB) and UE can negotiate with each other in order to optimize the trade-off between resource efficiency and Al computing requirement. For example, the latency of an Al model transfer requirement may be used to configure UE RRC behavior when the UE is in RRC-IDLE / INACTIVE state. Solutions such problems, as well as others, are disclosed.

[0097] FIG. 6 shows an example diagram of a UE in communication with a network to communicate with respect to an Al operation, in accordance with some aspects. Although each figure shows one or more aspects, it should be understood that an operations or detail in one aspect may overlap with an operation or detail in another aspect. For example, operations or details described with respect to FIG. 6 may overlap with or extend to other figures or aspects.

[0098] In FIG. 6, EU 602 may be in communication with a network 610 which may be a 5G NR network. UE 602 may obtain an Al model 604 from network 610, or Al training data 606 from network 610, or both. UE 602 may perform a training or inferencing operation 612 with the Al model 604 or the Al training data 606, or both.

[0099] In some examples, UE 602 may obtain an Al model 604 which may be sufficient to perform a desired inferencing task. The UE 602 may use Al model 604 to perform the inferencing operation at operation 612 by providing the Al model 604 with a given input, and obtaining a desired output from the Al model 604. As described, using the Al model(e.g., inferencing) may be performed with as a network functionality operation, or an Al as a service operation.

[0100] Additionally, or alternatively, UE 602 may determine a need to train an Al model (e.g., Al model 604 or a different Al model). UE 602 may obtain Al training data 606 from network 610 and use that Al training data to train the Al model at operation 612. Training of the Al model may include one or more rounds of providing a training data to the Al model 604 as input, comparing the output of the Al model to a corresponding desired result of the training data, and adjusting parameters of the Al model (e.g., weights) to reduce a difference between the output and the desired result. An Al model may be used for training or inferencing at operation 612 for the application layer, or the protocol layer, or both.

[0101] UE 602 may perform a Report Al result 608 operation to the network 610. This result may include a training result (e.g., ‘Al model “x” has been successfully trained', or ‘Al model “y” failed training’). The result 608 may include an inferencing result such as an output of the Al model 604, or a response that is determined based on the Al model inference. For example, the result 608 may include an inferred parameter used to optimize network functionality (e.g., beam ‘x’ should be used for uplink). In some examples, the result 608 may include a trained version of Al model 604. For example, report Al result 608 may include sending the network an entire trained Al model, which may in some instances include a neural network with connected layers of nodes and connection strengths (e.g., weight connections that are tuned through training) between the nodes. Alternatively, the UE 602 may send just trained model parameters (e.g., weights) to the network 610 which may then be used to reconstruct the trained model.

[0102] The UE 602 may provide Al related status 614 to the network 610. This may include an Al compute status of the UE 602 (e.g., whether or not it is currently performing an inferencing or training operation, an amount or percentage of the operation left to be performed, etc.).

[0103] At Al feedback Al-feedback operation 616, the network 610 may use the Report Al result 608 or the Al related status 614, or both, for various operations, such as scheduling of UEs, network resource allocation, localization of UEs, and / or other operations. Aspects described may vary from one Al application to another.

[0104] As discussed, the UE may perform training or inferencing operation 612 in any RRC state. The UE, however, may have one or more additional sub-states to more efficiently perform these Al related operations when in RRC idle or RRC inactive states.

[0105] FIG. 7 shows a diagram for UE states and sub-states that accommodate Al compute status of a UE, in accordance with some aspects. Aspects described may address the first problem of how a UE may characterize itself with respect to its Al compute status.

[0106] In FIG. 7, UE 712 may perform the training of one or more Al models 716. This training may be performed in any of the RRC states (e.g., RRC-Connected State 702, RRC idle RRC-Idle State 710, and / or RRC inactive RRC-Inactive State 704). Similarly, instructions or configuration of Al model training may be sent from network 714 to UE 712 any of the RRC states. Once the model 716 is trained by UE 712, it may be used by UE 712 or the network 714 for various functionalities. Reporting to the network may not be needed.

[0107] While the UE 712 is in RRC-Idle State 710 or RRC-Inactive State 704, the network 714 can send certain information or data to the UE 712 for the purposes of training the Al model 716. This Al-related data 718 may be delivered via mechanisms such as, for example, Multicast / Broadcast Services (MBS), Paging, Mobile-Terminated Small Data Transmission (MT-SDT), and / or Synchronization signals. If the UE completes the Al operation while in RRC-connected state, the UE may report the result via an uplink channel. If, however, the Al operation is completed by the UE 712 while the UE is still in RRC-Inactive State 704 or RRC-Idle State 710, then there may be some conditions when the UE should report this result to the network immediately, and other conditions in which the UE should wait.

[0108] The UE 712 may i) continue to monitor or decode the messages (Al-related data 718) facilitating or conveying data for Al training, and / or ii) report the trained Al model 716 to the network, if reporting is needed. Both operations may result in excessive and unnecessary energy expenditure by the UE 712. Without these operations when the UE is in RRC-Inactive State 704 or RRC-Idle State 710, the network 714 may not know whether the UE is still performing Al model training. As a result, the network 714 may page the UE to perform Al computing unnecessarily.

[0109] Aspects of the present disclosure include Al computing sub-states for UE 712 in RRC-Inactive State 704 and RRC-Idle State 710. The UE 712 may implement these Al computing sub-states (Al model computing active state 706 and Al model computinginactive state 708) to characterize the UE's need for Al-related data 718 for performing Al computing (e.g., operation 612). As such, the UE 712 and network 714 may coordinate Al operations around these sub-states.

[0110] UE 712 may include Al model computing active state 706. In this state, the UE 712 may have at least one of the following behaviors:1. the UE decodes the DL messages / signals (e.g., MBS, paging) dedicated to AI / ML model computing,2. the UE proactively transfers data / information (by e.g., resume RRC connection) relating to AI / ML model computing,3. the UE stores all configuration, information and / or data set to be used for AI / ML model training, and / or4. the UE is available for AI / ML training and inference (e.g., operation 612) if a need is triggered by an application or network protocol.

[0111] Additionally, UE 712 may include Al model computing inactive state 708. In this state, the UE may have at least one of the following behaviors:1. the UE does not decode the DL messages / signals (e.g., MBS, paging) dedicated to AI / ML model computing, for example, the UE ignores paging messages or paging early indication that is identified by network 714 to be used for Al model computing purposes, even if the UE is within the paging record,2. the UE does not proactively transfer data / information (by e.g., resume RRC connection) relating to AI / ML model computing (e.g., even if some data / information relating to AI / ML model computing are available at the UE, the UE would not transmit them until it returns to RRC-CONNECTED for other reasons), and / or3. the UE discards all configuration, information and / or data set to be used for AI / ML model training.

[0112] The UE 712 may transition into and out of each sub-state in response to different conditions. The UE may transition into Al model computing inactive state 708 (e.g., from Al model computing active state 706 or as an initial state) based on one or more of the following events:1. the UE receives an instruction from the network to perform computing state transition,2. the UE receives an instruction from the network to halt / terminate all the computing tasks,3. UE device telemetries (e.g., computing resource availability or the battery level of the UE) are less than a respective threshold, and / or4. the UE has completed all the Al model computing operation.

[0113] Similarly, the UE may transition into Al model computing active state 706 (e.g., from Al model computing inactive state 708 or as an initial state) based on one or more of the following events:1. the UE receives an instruction from the network to perform computing state transition,2. the UE receives an instruction from the network to resume / initiate some computing tasks,3. the UE device telemetries of the UE are greater than a respective threshold, and / or4. the UE is currently performing an Al compute operation.

[0114] The UE may send a notification (e.g., Al status 720) to the NW in response to each time the UE enters into one of the compute states. For example, if the UE transitions to Al model computing active state 706, it may indicate in Al status 720 that the UE is now in Al model computing active state 706. Similarly, if the UE transitions to Al model computing inactive state 708, the UE may send Al status 720 to the network 714 indicating that it is now in Al model computing inactive state 708.

[0115] The notification (e.g., Al status 720) may be sent from the UE 712 to network 714 through RRC signaling or MAC CE signaling. In an example, the notification can be transmitted with small data transmission (SDT) mechanism. Alternatively, the UE may resume RRC connection (e.g., temporarily transition to RRC-Connected State 702) to send such notification, and then return to RRC-Inactive State 704 or RRC-Idle State 710. The UE may further include one or more “reasons” of the state change in the notification. For example, the reason may indicate an application or network protocol that is to use the Al model 716 for inferencing, or an indication that the Al model is being trained and a reason for the training. An Al model may be trained due to one or more reasons such as, for example, the UE's location, a timestamp that is associated with the Al model may indicate that the Al model needs to be re-trained, new training data may be available for an Al model, or various other reasons.

[0116] FIG. 8 shows an example workflow of a UE that is in communication with a network with respect to Al-related paging, in accordance with some aspects.

[0117] In some aspects, the network may page a UE while the UE is not in RRC connected state (e.g., when it is in RRC-Inactive State 704 or RRC-Idle State 710). Paging may refer to signaling from the network to the UE in which Network notifies the UE that data (e.g., information or an instruction) is available for the UE, typically when the UE is idle or inactive. In response, the UE may decode the content of the page signaling. The content of the page signal may include the cause of the page (e.g., available information or instruction to the UE), and the UE may determine how it will respond to the page based on the content and other logic.

[0118] At block 802, a UE may be in an RRC inactive state or RRC idle state. At block 804, the UE may receive, from the network, a paging or paging early indication (PEI) signal from the network. This signal may include a request from the network that the UE perform an Al-related operation. The request may include an indication that training data or an Al model is available on the network for the UE to download.

[0119] In an aspect, paging message or the paging early indication (PEI) from the network may include the information relating to IDs of Al model or IDs of training data for a particular Al model. The paging targets (e.g., a UE) may provide additional information or data for training of certain Al models. The network may perform this paging for various reasons such as, for example, the location of the UE, a time stamp associated with the UE, or model number or type of the UE, or other reason.

[0120] At block 806, the UE may determine if it is interested in performing the Al operation (e.g., Al training or Al model download) that is indicated in the paging or PEI signal. If the UE has completed training of the Al model mentioned in the paging or PEI signal, then the UE may determine that it has no interest in training the Al model. In some examples, the UE may check a timestamp of its stored Al model and if it has been trained within a threshold time period, the UE may determine that the Al model has been trained and there is no need to train the Al model, however, if the timestamp indicates that the threshold time period is exceeded, then the UE may determine there is a need to perform the Al task. Additionally, or alternatively, the UE may determine that the UE is not responsible for training the Al model (e.g., based on a lookup table, or based on the UE's location, or other condition). In another example, the UE may determine that the Al model that is available for the UE to download (as presented by the paging or PEI signaling) is not needed by the UE at least based on the UE's location, or current cell, or UE type, or other condition.

[0121] In response to determining that the UE has interest in the Al model or task, the UE may proceed to block 810 and respond to the paging or PEI signal. For example, the UE may temporarily transition to an RRC connected state (e.g., through an RRC resume procedure) and then respond to the network to obtain additional Al model data (e.g., an Al model, Al training data, Al performance requirements, etc.).

[0122] In response to determining that the UE does not have an interest in the Al model or task, the UE may proceed to block 808 and ignore the paging or PEI signal. In this case, the UE does not transition out of RRC idle or RRC inactive, and does not attempt to or obtain the relevant Al-related data (e.g., the Al model, training data, Al performance requirements, etc.).

[0123] Therefore, even for a UE that actively performs an Al based operation (e.g., in Al model computing active state 706), the UE may still ignore the paging by the network 610 if, for example, the UE has already completed the training of this model or the UE is not responsible or interested in training this model. This may save the UE compute resources and energy.

[0124] FIG. 9 shows a method performed by a UE having sub-states in RRC idle or RRC inactive states, according to some aspects. The method 900 may be performed by processing logic of a UE, which may be coupled to a transceiver, where the processor may execute instructions stored on computer-readable memory to perform the method described.

[0125] Method 900 may be performed by processing logic that may comprise hardware (e.g., circuitry, dedicated logic, programmable logic, a processor, a processing device, a central processing unit (CPU), a system-on-chip (SoC), etc.), software (e.g., instructions running / executing on a processing device), firmware (e.g., microcode), or a combination thereof. In some aspects, at least a portion of method 900 may be performed by a UE that is in communication with a network (e.g., a 5G network) as described in other sections. Some of the operations described with respect to method 900 may correspond to operations described with respect to FIG. 6, FIG. 7, and FIG. 8.

[0126] The method illustrates example functions used by various aspects. Although specific function blocks ("blocks") are disclosed in the method, such blocks are examples. That is, aspects are well suited to performing various other blocks or variations of the blocks recited in the method. It is appreciated that the blocks in the method may be performed in an order different than presented.

[0127] At block 902, the method includes: in response to one or more first conditions being satisfied, transitioning to a first sub-state of a radio resource control (RRC) inactive state or an RRC idle state.

[0128] At block 904, the method includes: in response to one or more second conditions being satisfied, transitioning to a second sub-state of the RRC inactive or the RRC idle state, wherein in the first sub-state, the UE decodes a message associated with an artificial intelligence (Al) model when the message is received from the network, and wherein in the second sub-state, the UE ignores the message associated with the Al model when the message is received from the network.

[0129] The first sub-state may correspond to Al model computing active state 706 and the second sub-state may correspond to Al model computing inactive state 708.

[0130] In the first sub-state, the UE may perform a training or inferencing operation with the Al model, and in the second sub-state, the UE does not perform the training or the inferencing operation with the Al model (e.g., the UE ignores Al-related messages from the network).

[0131] In the first sub-state, the UE may receive an identifier (ID) that is associated with a second Al model from the network, and determine that the UE has already trained the second Al model or has no use for the second Al model. The UE may ignore a second message that is associated with the second Al model in response to determining that the UE has already trained the second Al model or does not have the use for the second Al model. This is described in detail with respect to FIG. 8.

[0132] In the first sub-state (e.g., Al model computing active state 706), the UE may transition to an RRC connected state to transmit data associated with the Al model to the network, and in the second sub-state (e.g., Al model computing inactive state 708), the UE does not transition to the RRC connected state to transmit the data associated with the Al model to the network. For example, when the UE receives the message from the network, the UE may initiate RRC resume to transition to the connected state when in the first substate, but in the second sub-state, the UE ignores the message and remains in RRC idle or RRC inactive state.

[0133] In the first sub-state, the UE may store data associated with the Al model and in the second sub-state, the UE may discard the data associated with the Al model. The sub-states help the UE manage memory resources efficiently with respect to Al-related data.

[0134] The UE may transition to the second sub-state in response to at least one of: receiving an instruction from the network to perform the transition to the second sub-state, receiving an instruction from the network to halt or terminate all compute tasks, determining that a computing resource of the UE is less than a threshold, determining that a battery level of the UE is less than a second threshold, or completion of pending operations associated with the Al model. In one aspect, each of these conditions will cause the UE to transition to the second sub-state (e.g., Al model computing inactive state 708).

[0135] The UE may transition to the first sub-state in response to at least one of: receiving an instruction from the network to transition to the first sub-state, receiving an instruction from the network to resume or initiate an operation associated with the Al model, determining that a computing resource of the UE is greater than a threshold, or determining that a battery level of the UE is greater than second threshold. In one aspect, each of these conditions will cause the UE to transition to the first sub-state (e.g., Al model computing active state 706).

[0136] Method 900 may further comprise sending a state change notification to the network in response to the UE transitioning from between the first sub-state and the second sub-state (e.g., from Al model computing active state 706 to Al model computing inactive state 708 or from Al model computing inactive state 708 to Al model computing active state 706).

[0137] In an example, sending the state change notification to the network may be performed using a small data transfer (SDT) protocol. Small Data Transmission (SDT) is a procedure in LTE or 5G that allows data transmission or signaling between the network and UE, while remaining the UE remains in RRC inactive state (i.e., without transitioning to RRC CONNECTED state). SDT is typically enabled on a radio bearer basis and the SDT procedure may be initiated by the UE under certain conditions such as, for example, less than a specified amount of uplink (UL) data awaits transmission across all radio bearers for which SDT is enabled, the downlink (DL) reference signal received power (RSRP) is above a configured threshold, and a valid SDT resource is available. Details regarding SDT configuration and resource management may vary depending on application.

[0138] Alternatively, or additionally, sending the state change notification to the network may be performed using a resumed RRC connection. For example, the UE maytransition in RRC connected (through an RRC resume procedure), and send the state change notification to the network over an UL channel in the RRC connected state.

[0139] The UE may include a reason for the transition between the first sub-state and the second sub-state in the state change notification. As discussed, this may include an application or network protocol that is to use the Al model for inferencing or training.

[0140] In an aspect, method 900 may be performed from the perspective of the network. For example, the method may be performed by a network (e.g., a gNB) in communication with a UE. The method may include sending a message associated with an Al model to the UE in a first a first sub-state of a radio resource control (RRC) inactive state or an RRC idle state that is decoded by the UE. The method may further include sending the message associated with the Al model to UE in the second sub-state of the RRC inactive or the RRC idle state that is ignored by the UE. Other aspects described above with respect to method 900 may performed from the perspective of the network.

[0141] FIG. 10 shows an example workflow performed by a UE or network for an Al operation, in accordance with some aspects.

[0142] Workflow 1000 shows operations or methods performed by a UE 1004 or network 1006 to coordinate an Al operation. The network 1006 may, at operation 1018, provide an Al instruction to the UE 1004. The Al instruction may instruct UE 1004 to train or update one or more Al models. Additionally, or alternatively, the Al instruction 1018 may request that the UE perform inferencing for specific model (e.g., to avoid sharing the UE data with the network). The Al model could already be available to the UE, or the model IDs can be transmitted to the UE by the NW for the UE to reference the appropriate Al model. The UE 1004 may perform the instructed Al operation in any of the RRC states, including RRC Idle or RRC inactive state. Similarly, the Al instruction 1018 may be sent in any of the RRC states.

[0143] In response to receiving the Al instruction 1018, the UE may start Al model operation 1012 as instructed by the network 1006. As discussed, however, upon completion of Al model operation at 1014, the UE 1004 may be in a non-connected RRC state such as RRC idle or RRC inactive. For example, the UE may have entered RRC idle or RRC inactive state at operation 1010, during or prior to performing the Al model operation. Once one of the Al model is completely trained or the inference result is obtained by the UE (e.g., at operation 1014), the UE may report the result of the operation, which may include a completely trained Al model, a training result or performance, aninference result with the inference performance, or other Al operation result. This may be reported to the network 1006 which may include a base station. In some examples, network 1006 may include another UE.

[0144] If, however, the Al operation is completed when the UE is in RRC- IDLE / INACTIVE state, the UE may implement logic (at operation 1002) to determine if the UE should resume RRC connection (i.e., return to RRC connected state) to report the results of the Al operation. For example, if the UE does not resume RRC connection proactively for Al model reporting, the trained Al model or inference outcome may eventually become outdated. If the UE resumes RRC connection proactively solely for Al model reporting, this may result in higher UE power consumption.

[0145] Therefore, aspects described may resolve this trade-off between timely Al model training or inference reporting and UE power saving. When the network instructs the UE to perform an Al instruction at operation 1018, the network 1006 may send Al model configuration information to the UE at operation 1008 that provides guidance to the UE as to whether or not to resume RRC connection after completion of the Al model operation, to report the result of the Al operation. Generally, completion of an Al operation may include when an Al model has generated an output (e.g., an inference) based on input data, or when an Al operation has completed training.

[0146] The Al model configuration (received by the UE at operation 1008) may include conditions in which the result of the Al operation is to be reported immediately back to network 1006. For each Al model in Al instruction 1018, the Al model configuration may include whether or not the operation associated with that Al model should be reported back to the network 1006 if completed in RRC idle or RRC inactive state. Additionally, the Al model configuration may include, for each Al model, a quality of service (QoS) for each Al model, a life expectancy of the Al model, or additional information.

[0147] In an aspect, the Al model configuration may include, for each Al model in an Al operation instructed through Al instruction 1018, an Al model ID (e.g., a name or series of symbols that uniquely identifies an Al model), a binary flag, a life expectancy of the Al model, and / or an event trigger.

[0148] For example, the Al model configuration may include a list of Al model IDs which the UE should or should not proactively resume RRC connections for reporting, for example, “Al model X, and Y are to be reported by resuming RRC connection and model Z is not”.

[0149] Additionally, or alternatively, Al model configuration may include a binary flag that is associated with each Al model may indicate whether the UE should or should not proactively resume the RRC connection if the operation is completed in RRC idle or RRC inactive state.

[0150] Additionally, or alternatively, Al model configuration may include the life expectancy of each of the Al models to be trained, or a period or schedule of how often each Al model is to be re-trained.

[0151] Additionally, or alternatively, the Al model configuration may include an event trigger may include one or more performance requirements (e.g., QoS) that the Al operation is to satisfy.

[0152] At operation 1002, the UE determines whether or not to trigger the RRC resume procedure based on the configuration data received in the Al instruction 1018. For example, after completion of the Al model operation 1014 with a given Al model, if the binary flag of the Al model that was trained or used for inferencing is set, or if the Al model used in the operation is named on the Al model list, or if the life expectancy of the Al model that is trained or used for inferencing is less than a threshold, or if the Al model that is used for inferencing or training is less than a threshold, or if the operation (e.g., inferencing or training) satisfies the performance requirements of the configuration data, then the UE may proceed to perform the RRC resume procedure at operation 1016. The Al model configuration 1008 may include any combination of such conditions.

[0153] Once the UE transitions to RRC connected state, the UE may report Al operation result at operation 1020. This may include transmitting the entire trained Al model, or an indication that training of the Al model completed successfully, or providing Al model inferencing results (e.g., an output of an Al model) which may be used by network 1006 for various network-related functionality.

[0154] At operation 1002, the UE may determine that the condition or conditions of the Al model configuration 1008 are not satisfied. In response, the UE may proceed to operation 1022 and does not initiate an RRC resume procedure. The UE may remain in RRC idle or RRC inactive state. At operation 1022, the UE may store the Al operation result (e.g., a trained Al model or models, or the inferencing result). The UE may report the result to the network 1006 at a later time, for example, when the UE resumes RRC connection due to another triggering event.

[0155] FIG. 11 shows a method performed by a UE to report a result of an Al operation based on an indication from a network, in accordance with some aspects. The method 1100 may be performed by processing logic of a UE, which may be coupled to a transceiver, where the processor may execute instructions stored on computer-readable memory to perform the method described. The performed by user equipment (UE) in communication with a network.

[0156] Method 1100 may be performed by processing logic that may comprise hardware (e.g., circuitry, dedicated logic, programmable logic, a processor, a processing device, a central processing unit (CPU), a system-on-chip (SoC), etc.), software (e.g., instructions running / executing on a processing device), firmware (e.g., microcode), or a combination thereof. In some aspects, at least a portion of method 900 may be performed by a UE that is in communication with a network (e.g., a 5G network) as described in other sections. Some of the operations described with respect to method 1100 may correspond to operations described in other sections such as, for example, with respect to FIG. 10.

[0157] The method illustrates example functions used by various aspects. Although specific function blocks ("blocks") are disclosed in the method, such blocks are examples. That is, aspects are well suited to performing various other blocks or variations of the blocks recited in the method. It is appreciated that the blocks in the method may be performed in an order different than presented.

[0158] At block 1102, the method may include performing an operation (e.g., an Al operation 612) associated with an artificial intelligence (Al) model. This operation may be initiated by the network (e.g., through Al instruction 1018).

[0159] At block 1104, the method may include receiving an indication from the network that is associated with completion of the operation. For example, the network may provide the indication to the network by configuring the UE with signaling (e.g., through Al model configuration operation 1008) which may part of the Al instruction 1018 or separate. Thus, the indication may include any of the described conditions included in the Al model configuration 1008.

[0160] At block 1106, the method may include, when the UE is in an RRC idle or RRC inactive state at the completion of the operation, transitioning to an RRC connected state based on the indication. For example, if the condition or conditions in the indication are satisfied, the UE may transition to RRC connected. Otherwise, the UE may remain in its current state (e.g., RRC idle or RRC inactive state).

[0161] At block 1108, the method may include reporting a result of the completed operation that is associated with the Al model in the RRC connected state. The result may include an inferencing result, or a complete trained Al model, or other result.

[0162] The indication from the network may include one or more IDs that are each associated with the one or more respective Al models. For example, the network may send to the UE one or more Al model ID that uniquely identifies a given Al model. The UE and network may use this list of Al model IDs as a ‘to report’ list. When the UE completes the Al operation, it may refer to this list and if the completed operation with Al model ‘x’ is associated with an Al model on that list, the UE may go ahead and transition to the RRC connected state and report this Al model as described with respect to operation 1002, 1016, and 1020. If, however, the Al model used in connection with the completed Al operation is not on the Al model ID list, the UE does not transition to the RRC connected state (and does not report the operation result to the network at that moment), as described with respect to operations 1002, 1022.

[0163] In an aspect, the indication from the network comprises a flag that is associated with the Al model. The UE may transition to the RRC connected state is performed in response to the flag having a first state (e.g., set). Conversely, the UE may remain in RRC idle or RRC inactive (and not transition to RRC connected state) in response to the flag having a second state (e.g., not set).

[0164] In an aspect, the indication from the network comprises an expiration time that is associated with the Al model. In such a case, the UE may transition to the RRC connected state in response to the expiration time of the Al model not being exceeded. For example, if an expiration time associated with Al model ‘x’ is ‘y’ number of hours, and the time since the Al model was last updated is less than ‘y’ number of hours upon completion of the Al operation, the UE may transition to RRC connected state and report the result. Similarly, if the expiration time is exceeded (e.g., the time since the Al model was last updated / trained is the same as or greater than ‘y’ number of hours), the UE may remain in RRC idle or inactive state and not report the result of the operation. If the Al model expires due to not being updated or trained within a threshold time, then the Al model may conserve energy by remaining in RRC idle or inactive state.

[0165] In an aspect, in response to the UE not transitioning to the RRC connected state based on the indication, the UE stores the result of the operation that is associated with the Al model when in the RRC idle or the RRC inactive state. For example, as described aboveand with respect to FIG. 10, if it is determined that the one or more conditions received in the configuration information (e.g., through operation 1008) are not satisfied, then the network may store the result of the Al operation. In some aspects, the UE may send the stored result (e.g., the stored Al model or inferencing result) to the network the next time the UE returns to RRC connected state. In this manner, the UE may still convey the result of the Al operation to the network, but does not actively transition to RRC connected when the configured conditions are not satisfied, thereby conserving energy.

[0166] In an aspect, method 1100 may be performed from the perspective of the network. For example, a method may include sending, by the network, an indication (e.g., one or more conditions) associated with an Al operation to the UE. The indication may include the one or more conditions used by the network to configure the UE for determining whether to transition to RRC connected state upon the completion of the Al operation. The method may include receiving, by the network, a result of the completed operation (e.g., a complete trained Al model, and / or an inferencing result) from the UE upon the UE's transition to RRC connected state. Other aspects described with respect to method 1100 also apply from the perspective of the network.

[0167] FIG. 12 shows an example workflow of a UE that is in communication with a network with respect to Al-related paging, in accordance with some aspects. The workflow 1200 may be performed by a UE in communication with a network.

[0168] A UE may, at block 1202, transition into RRC idle or RRC inactive state. While in this RRC idle or RRC inactive state, the UE may complete an Al model operation at block 1204.

[0169] At decision block 1206, the UE may check whether one or more Al model operation conditions are satisfied. The Al model operation conditions may be internal, unlike the network-dependent configuration signaling mechanism described with respect to FIG. 10. If one or more of the Al model operation conditions are satisfied, the UE may proceed to block 1212 and resume the RRC connection (e.g., initiate an RRC resume procedure to transition to RRC connected state). If the one or more conditions are not satisfied, the UE may proceed to block 1208 and remain in RRC idle or RRC inactive state.

[0170] The Al model operation conditions may be referred to as RRC resume conditions. In an aspect, determining whether or not Al model operation conditions are satisfied includes determining whether determining whether or not Al model operation conditions relate to the Al model operation.

[0171] In an aspect, the Al model condition or conditions may include one or more of:1. when the number of completed but not yet reported Al models exceeds a threshold,2. when the total data volume of completed (but not yet reported) Al models exceeds a threshold,3. when the time elapsed since the completion of training of one or more Al models which have not yet been reported has exceeded one or more thresholds, and / or4. when the UE application layer triggers the need for an Al model transmission.

[0172] At block 1212, the UE may resume RRC connection through initiating an RRC resume procedure with the network. Upon successful completion of the RRC resume procedure, the UE may transition to the RRC connected state and proceed to block 1214. At block 1214, the UE may report the trained Al model or inference result. For example, the UE may send, in an uplink data channel, the trained Al model (or parameters thereof), or an output of the Al model (e.g., an inferencing result).

[0173] At block 1208, the UE may remain in RRC idle or in RRC inactive state. In this case, the UE does not initiate the RRC resume procedure. The UE may proceed to block 1210 and store the trained Al model or the inferencing result such as the output of the Al model as a result of the Al operation.

[0174] More generally, in the various aspects described, rather than resuming RRC connection, the UE may determine to trigger small data transmission (SDT) in order to report the result of the Al operation (e.g., return the trained model or result of the trained model or result of inferencing. For example, the UE may trigger SDT operations and report the result of the Al operation through SDT in response to instructions received by the network (e.g., through Al model configuration 1008) or in response to the one or more conditions described above, or both.

[0175] In an aspect, when the UE determines to trigger RRC resume for any of the reasons described in the present disclosure, the UE may provide an indication to the network (e.g., gNB) that it has stored one or more trained and unreported Al models. The network may configure appropriate resource (e.g., radio bearer and / or resource allocation) for Al model reporting.

[0176] In another aspect, a periodic timer and feedback for the Al model status reporting between the UE and gNB can be introduced. Depending on the performance of the Al model, the network (e.g., gNB) can instruct the UE to stop or resume training the Al model.

[0177] FIG. 13 shows a method performed by a UE to report a result of an Al operation based on internal conditions, in accordance with some aspects. The method 1100 may be performed by processing logic of a UE, which may be coupled to a transceiver, where the processor may execute instructions stored on computer-readable memory to perform the method described. The performed by user equipment (UE) in communication with a network.

[0178] Method 1300 may be performed by processing logic that may comprise hardware (e.g., circuitry, dedicated logic, programmable logic, a processor, a processing device, a central processing unit (CPU), a system-on-chip (SoC), etc.), software (e.g., instructions running / executing on a processing device), firmware (e.g., microcode), or a combination thereof. In some aspects, at least a portion of method 1300 may be performed by a UE that is in communication with a network (e.g., a 5G network) as described in other sections. Some of the operations described with respect to method 1300 may correspond to operations described in other sections such as, for example, with respect to FIG. 12.

[0179] The method illustrates example functions used by various aspects. Although specific function blocks ("blocks") are disclosed in the method, such blocks are examples. That is, aspects are well suited to performing various other blocks or variations of the blocks recited in the method. It is appreciated that the blocks in the method may be performed in an order different than presented.

[0180] At block 1302, the method includes performing an operation associated with an Al model. This Al operation may include inferencing with the Al model or training the Al model.

[0181] At block 1304, the method includes when the UE is in an RRC idle or RRC inactive state at completion of the operation, transitioning to an RRC connected state based on whether a condition is satisfied. This condition may be an internal UE condition which is separate from the network or independent from configuration by the network. For example, the UE may determine, without input from the network, whether or not to transition to RRC connected state.

[0182] At block 1306, the method includes reporting, to the network, the completion of the operation that is associated with the Al model in the RRC connected state. For example, the UE may transition to RRC connected state and then send the result of the Al operation (e.g., the inferencing result, or the entire trained Al model) to the network.

[0183] If at block 1304, the condition is not satisfied, then the UE may remain in RRC idle or RRC inactive state. Similar to other examples, the UE may store the result of the completed operation and wait opportunistically until the UE transitions to RRC connected state (due to another Al model operation or another non- Al trigger event) to then report the result.

[0184] The UE may implement internal logic such as checking the status of stored fields or performing API calls to determine if one or more conditions are satisfied. In an example, the condition is satisfied when a number of completed operations, or a total data size of the completed operations is greater than a threshold. These completed Al operations may include those where the result has not been reported to the network.

[0185] In an example, the condition is satisfied when a time elapsed since the completion of the operation is greater than a threshold. For example, it could be that, upon completion of the operation, the results of the Al model are no longer deemed to be reliable. Thus, the UE may conserve energy and network resources by not reporting the completion of the operation (e.g., a result) to the network.

[0186] In an example, the condition comprises a UE application layer trigger. For example, an application may indicate to the UE to report an Al operation result where that Al operation either trains an Al model for the application or provides an inferencing result for the application in the application layer.

[0187] As discussed, in some aspects, rather than transitioning to RRC connected, the UE may report the result of the Al operation through small data transmission (SDT) in response to the condition being satisfied (e.g., in method 1300) or in response to a preconfiguration of the UE by the network (e.g., in method 1100).

[0188] In an example, the method further comprises, at block 1306 orblock 1108, when the UE determines that it will transition to RRC connected state and report the result of the Al operation to the network, the UE may send an indication to the network that the Al model is trained and stored at the UE. The UE may receive a configuration instruction from the network to perform the reporting of the completion of the operation. For example, the network may configure appropriate network resources (e.g., radio resource allocation) for the Al model reporting.

[0189] In an example, the method comprises, at block 1306 or block 1108, sending status of the Al operation that is associated with the Al model to the network periodically, and receiving an instruction from the network to stop or resume the Al operation that isassociated with the Al model. For example, the UE may implement a periodic timer and feedback for the Al model status reporting. The network (e.g., gNB) may instruct the UE to stop or resume the Al operation based on if the performance of the Al model satisfies one or more thresholds.

[0190] In an aspect, the method 1300 may be performed from the perspective of the network. For example, the method may comprise receiving a result of a completed operation that is associated with an Al model from a UE in RRC connected state (or through SDT in a non-RRC connected state), wherein the UE performs the operation that is associated with the Al model and then transitions to the RRC connected state based on whether a condition is satisfied. Other aspects described with respect to method 1300 also apply from the perspective of the network.

[0191] There are a number of example embodiments described herein.

[0192] Example 1 is a method performed by user equipment (UE) in communication with a network comprising: in response to one or more first conditions being satisfied, transitioning to a first sub-state of a radio resource control (RRC) inactive state or an RRC idle state; and in response to one or more second conditions being satisfied, transitioning to a second sub-state of the RRC inactive or the RRC idle state, wherein in the first substate, the UE decodes a message associated with an artificial intelligence (Al) model when the message is received from the network, and wherein in the second sub-state, the UE ignores the message associated with the Al model when the message is received from the network.

[0193] Example 2 is the method of Example 1 that may optionally include that in the first sub-state, the UE performs a training or inferencing operation with the Al model, and in the second sub-state, the UE does not perform the training or the inferencing operation with the Al model.

[0194] Example 3 is the method of Example 1 that may optionally include in the first sub-state, receiving an identifier (ID) that is associated with a second Al model from the network, determining that the UE has already trained the second Al model or has no use for the second Al model, and ignoring a second message that is associated with the second Al model in response to determining that the UE has already trained the second Al model or does not have the use for the second Al model.

[0195] Example 4 is the method of Example 1 that may optionally include that in the first sub-state, the UE transitions to the RRC connected state to transmit data associatedwith the Al model to the network, and in the second sub-state, the UE does not transition to the RRC connected state to transmit the data associated with the Al model to the network.

[0196] Example 5 is the method of Example 1 that may optionally include that in the first sub-state, the UE stores data associated with the Al model and in the second sub-state, the UE discards the data associated with the Al model.

[0197] Example 6 is the method of Example 1 that may optionally include that the UE transitions to the second sub-state in response to at least one of: receiving an instruction from the network to perform the transition to the second sub-state, receiving an instruction from the network to halt or terminate all compute operations, determining that a computing resource of the UE is less than a threshold, determining that a battery level of the UE is less than a second threshold, or completion of pending operations associated with the Al model.

[0198] Example 7 is the method of Example 1 that may optionally include that the UE transitions to the first sub-state in response to at least one of: receiving an instruction from the network to transition to the first sub-state, receiving an instruction from the network to resume or initiate an operation associated with the Al model, determining that a computing resource of the UE is greater than a threshold, or determining that a battery level of the UE is greater than second threshold.

[0199] Example 8 is the method of Example 1 that may optionally include sending a state change notification to the network in response to the UE transitioning from between the first sub-state and the second sub-state.

[0200] Example 9 is the method of Example 8 that may optionally include that sending the state change notification to the network is performed using a small data transfer (SDT) protocol.

[0201] Example 10 is the method of Example 8 that may optionally include that sending the state change notification to the network is performed using a resumed RRC connection.

[0202] Example 11 is the method of Example 8 that may optionally include that the UE includes a reason for the transition between the first sub-state and the second sub-state in the state change notification.

[0203] Example 12 is a method performed by user equipment (UE) in communication with a network, comprising: performing an operation associated with an artificial intelligence (Al) model; receiving an indication from the network that is associated with completion of the operation; when the UE is in an RRC idle or RRC inactive state at the completion of the operation, transitioning to an RRC connected state based on the indication; and reporting a result of the completed operation that is associated with the Al model in the RRC connected state.

[0204] Example 13 is the method of Example 12 that may optionally include that the indication from the network comprises one or more IDs that are each associated with one or more respective Al models, and wherein the UE does not transition to the RRC connected state in response to none of the one or more IDs being associated with the Al model that is associated with the completed operation.

[0205] Example 14 is the method of Example 12 that may optionally include that the indication from the network comprises a flag that is associated with the Al model, and wherein transitioning to the RRC connected state is performed in response to the flag having a first state, and wherein transitioning to the RRC connected state is not performed in response to the flag having a second state.

[0206] Example 15 is the method of Example 12 that may optionally include that the indication from the network comprises an expiration time that is associated with the Al model, and wherein transitioning to the RRC connected state is performed in response to an expiration time of the Al model not being exceeded, and wherein transitioning to the RRC connected state is not performed in response to the expiration time of the Al model being exceeded.

[0207] Example 16 is the method of Example 12 that may optionally include that in response to the UE not transitioning to the RRC connected state, the UE stores the result of the operation that is associated with the Al model when in the RRC idle or the RRC inactive state.

[0208] Example 17 is method performed by user equipment (UE) in communication with a network, comprising: performing an operation associated with an Al model; when the UE is in an RRC idle or RRC inactive state at completion of the operation, transitioning to an RRC connected state based on whether a condition is satisfied; and reporting a result of the completed operation that is associated with the Al model in the RRC connected state.

[0209] Example 18 is the method of Example 17 that may optionally include that the condition is satisfied when a number of total of completed operations, or a total data size of the total completed operations is greater than a threshold.

[0210] Example 19 is the method of Example 17 that may optionally include that

[0211] the condition is satisfied when a time elapsed since the completion of the operation is greater than a threshold.

[0212] Example 20 is the method of Example 17 that may optionally include that the condition comprises a UE application layer trigger.

[0213] Example 21 is a user equipment (UE) comprising one or more processors configured to perform the method of any of examples 1 to 11.

[0214] Example 22 is a non-transitory machine-readable medium having executable instructions to cause one or more processing units to perform the method of any of examples 1 to 11.

[0215] Example 23 is a user equipment (UE) comprising one or more processors configured to perform the method of any of examples 12 to 16.

[0216] Example 24 is a non-transitory machine-readable medium having executable instructions to cause one or more processing units to perform the method of any of examples 12 to 16.

[0217] Example 25 is a user equipment (UE) comprising one or more processors configured to perform the method of any of examples 17 to 20.

[0218] Example 26 is a non-transitory machine-readable medium having executable instructions to cause one or more processing units to perform the method of any of examples 17 to 20.

[0219] Portions of what was described above may be implemented with logic circuitry such as a dedicated logic circuit or with a microcontroller or other form of processing core that executes program code instructions. Thus, processes taught by the discussion above may be performed with program code such as machine-executable instructions that cause a machine that executes these instructions to perform certain functions. In this context, a “machine” may be a machine that converts intermediate form (or “abstract”) instructions into processor specific instructions (e.g., an abstract execution environment such as a “virtual machine” (e.g., a Java Virtual Machine), an interpreter, a Common Language Runtime, a high-level language virtual machine, etc.), and / or, electronic circuitry disposedon a semiconductor chip (e.g., “logic circuitry” implemented with transistors) designed to execute instructions such as a general-purpose processor and / or a special-purpose processor. Processes taught by the discussion above may also be performed by (in the alternative to a machine or in combination with a machine) electronic circuitry designed to perform the processes (or a portion thereof) without the execution of program code.

[0220] The present invention also relates to an apparatus for performing the operations described herein. This apparatus may be specially constructed for the required purpose, or it may comprise a general -purpose computer selectively activated or reconfigured by a computer program stored in the computer. Such a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, and magnetic-optical disks, read-only memories (ROMs), RAMs, EPROMs, EEPROMs, magnetic or optical cards, or any type of media suitable for storing electronic instructions, and each coupled to a computer system bus.

[0221] A machine-readable medium includes any mechanism for storing or transmitting information in a form readable by a machine (e.g., a computer). For example, a machine-readable medium includes read only memory (“ROM”); random access memory (“RAM”); magnetic disk storage media; optical storage media; flash memory devices; etc.

[0222] In any of the examples described, a processor may include a baseband processor (also known as baseband radio processor, BP, or BBP) is a device (a chip or part of a chip) in a network interface that manages radio functions, such as communicating (e.g., TX and RX) over an antenna.

[0223] An article of manufacture may be used to store program code. An article of manufacture that stores program code may be embodied as, but is not limited to, one or more memories (e.g., one or more flash memories, random access memories (static, dynamic, or other)), optical disks, CD-ROMs, DVD ROMs, EPROMs, EEPROMs, magnetic or optical cards or other type of machine-readable media suitable for storing electronic instructions. Program code may also be downloaded from a remote computer (e.g., a server) to a requesting computer (e.g., a client) by way of data signals embodied in a propagation medium (e.g., via a communication link (e.g., a network connection)).

[0224] The preceding detailed descriptions are presented in terms of algorithms and symbolic representations of operations on data bits within a computer memory. These algorithmic descriptions and representations are the tools used by those skilled in the dataprocessing arts to most effectively convey the substance of their work to others skilled in the art. An algorithm is here, and generally, conceived to be a self-consistent sequence of operations leading to a desired result. The operations are those requiring physical manipulations of physical quantities. Usually, though not necessarily, these quantities take the form of electrical or magnetic signals capable of being stored, transferred, combined, compared, and otherwise manipulated. It has proven convenient at times, principally for reasons of common usage, to refer to these signals as bits, values, elements, symbols, characters, terms, numbers, or the like.

[0225] It should be kept in mind, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise as apparent from the above discussion, it is appreciated that throughout the description, discussions utilizing terms such as “transmitting”, “sending”, “selecting,” “determining,” “receiving,” “forming,” “grouping,” “aggregating,” “generating,” “removing,” or the like, refer to the action and processes of a computer system, or similar electronic computing device, that manipulates and transforms data represented as physical (electronic) quantities within the computer system's registers and memories into other data similarly represented as physical quantities within the computer system memories or registers or other such information storage, transmission or display devices. Reference to some aspects or examples may be referred to as one or more aspects or one or more examples.

[0226] The processes and displays presented herein are not inherently related to any particular computer or other apparatus. Various general-purpose systems may be used with programs in accordance with the teachings herein, or it may prove convenient to construct a more specialized apparatus to perform the operations described. The required structure for a variety of these systems will be evident from the description below. In addition, the present invention is not described with reference to any particular programming language. It will be appreciated that a variety of programming languages may be used to implement the teachings of the invention as described herein.

[0227] It is well understood that the use of personally identifiable information should follow privacy policies and practices that are generally recognized as meeting or exceeding industry or governmental requirements for maintaining the privacy of users. In particular, personally identifiable information data should be managed and handled so as to minimizerisks of unintentional or unauthorized access or use, and the nature of authorized use should be clearly indicated to users.

[0228] The foregoing discussion merely describes some exemplary aspects of the present invention. One skilled in the art will readily recognize from such discussion, the accompanying drawings and the claims that various modifications can be made without departing from the spirit and scope of the invention.

Claims

CLAIMSWhat is claimed is:

1. A method performed by user equipment (UE) in communication with a network comprising: in response to one or more first conditions being satisfied, transitioning to a first sub-state of a radio resource control (RRC) inactive state or an RRC idle state; and in response to one or more second conditions being satisfied, transitioning to a second sub-state of the RRC inactive or the RRC idle state, wherein in the first sub-state, the UE decodes a message associated with an artificial intelligence (Al) model when the message is received from the network, and wherein in the second sub-state, the UE ignores the message associated with the Al model when the message is received from the network.

2. The method of claim 1, wherein in the first sub-state, the UE performs a training or inferencing operation with the Al model, and in the second sub-state, the UE does not perform the training or the inferencing operation with the Al model.

3. The method of claim 1, further comprising: in the first sub-state, receiving an identifier (ID) that is associated with a second Al model from the network, determining that the UE has already trained the second Al model or has no use for the second Al model, and ignoring a second message that is associated with the second Al model in response to determining that the UE has already trained the second Al model or does not have the use for the second Al model.

4. The method of claim 1, wherein in the first sub-state, the UE transitions to the RRC connected state to transmit data associated with the Al model to the network, and in the second sub-state, the UE does not transition to the RRC connected state to transmit the data associated with the Al model to the network.

5. The method of claim 1, wherein in the first sub-state, the UE stores data associated with the Al model and in the second sub-state, the UE discards the data associated with the Al model.

6. The method of claim 1, wherein the UE transitions to the second sub-state in response to at least one of: receiving an instruction from the network to perform the transition to the second sub-state, receiving an instruction from the network to halt or terminate all compute operations, determining that a computing resource of the UE is less than a threshold, determining that a battery level of the UE is less than a second threshold, or completion of pending operations associated with the Al model.

7. The method of claim 1, wherein the UE transitions to the first sub-state in response to at least one of: receiving an instruction from the network to transition to the first sub-state, receiving an instruction from the network to resume or initiate an operation associated with the Al model, determining that a computing resource of the UE is greater than a threshold, or determining that a battery level of the UE is greater than second threshold.

8. The method of claim 1, further comprising sending a state change notification to the network in response to the UE transitioning from between the first sub-state and the second sub -state.

9. The method of claim 8, wherein sending the state change notification to the network is performed using a small data transfer (SDT) protocol.

10. The method of claim 8, wherein sending the state change notification to the network is performed using a resumed RRC connection.

11. The method of claim 8, wherein the UE includes a reason for the transition between the first sub-state and the second sub-state in the state change notification.

12. A method performed by user equipment (UE) in communication with a network, comprising: performing an operation associated with an artificial intelligence (Al) model; receiving an indication from the network that is associated with completion of the operation; when the UE is in an RRC idle or RRC inactive state at the completion of the operation, transitioning to an RRC connected state based on the indication; and reporting a result of the completed operation that is associated with the Al model in the RRC connected state.

13. The method of claim 12, wherein the indication from the network comprises one or more IDs that are each associated with one or more respective Al models, and wherein the UE does not transition to the RRC connected state in response to none of the one or more IDs being associated with the Al model that is associated with the completed operation.

14. The method of claim 12, wherein the indication from the network comprises a flag that is associated with the Al model, and wherein transitioning to the RRC connected state is performed in response to the flag having a first state, and wherein transitioning to the RRC connected state is not performed in response to the flag having a second state.

15. The method of claim 12, wherein the indication from the network comprises an expiration time that is associated with the Al model, and wherein transitioning to the RRC connected state is performed in response to an expiration time of the Al model not being exceeded, and wherein transitioning to the RRC connected state is not performed in response to the expiration time of the Al model being exceeded.

16. The method of claim 12, wherein in response to the UE not transitioning to the RRC connected state, the UE stores the result of the operation that is associated with the Al model when in the RRC idle or the RRC inactive state.

17. A method performed by user equipment (UE) in communication with a network, comprising: performing an operation associated with an Al model; when the UE is in an RRC idle or RRC inactive state at completion of the operation, transitioning to an RRC connected state based on whether a condition is satisfied; and reporting a result of the completed operation that is associated with the Al model in the RRC connected state.

18. The method of claim 17, wherein the condition is satisfied when a number of total of completed operations, or a total data size of the total completed operations is greater than a threshold.

19. The method of claim 17, wherein the condition is satisfied when a time elapsed since the completion of the operation is greater than a threshold.

20. The method of claim 17, wherein the condition comprises a UE application layer trigger.

21. A user equipment (UE) comprising one or more processors configured to perform the method of any of claim 1 to claim 11.

22. A non-transitory machine-readable medium having executable instructions to cause one or more processing units to perform the method of any of claim 1 to claim 11.

23. A user equipment (UE) comprising one or more processors configured to perform the method of any of claim 12 to claim 16.

24. A non-transitory machine-readable medium having executable instructions to cause one or more processing units to perform the method of any of claim 12 to claim 16.

25. A user equipment (UE) comprising one or more processors configured to perform the method of any of claim 17 to claim 20.

26. A non-transitory machine-readable medium having executable instructions to cause one or more processing units to perform the method of any of claim 17 to claim 20.