Model discovery and selection for cooperative machine learning in cellular networks

Abstract

The OAM core network can receive a request for the ML / NN model and the features associated with the ML / NN procedure. The OAM core network can determine, based on the request, the latest updates to the ML / NN model and the features, and generate a response to the request indicating the latest updates to the ML / NN model and the features. In aspects, a base station can initiate the request for the ML / NN model and the features by sending the request to the OAM core network. The base station can receive the generated response of the OAM core network based on the sent request. In further aspects, a UE can initiate the request for the ML / NN model and the features by sending the request to the base station, where the UE can receive the ML / NN model and the features from the base station based on the sent request.

Description

[0001] Cross-reference to related applications

[0002] This application claims the benefit of U.S. Patent Application No. 17 / 111,346, filed on December 3, 2020, entitled “MODEL DISCOVERY AND SELECTION FOR COOPERATIVE MACHINE LEARNING IN CELLULAR NETWORKS,” which is expressly incorporated herein by reference in its entirety. Technical Field

[0003] In summary, this disclosure relates to wireless communication systems, and more specifically, to model discovery and selection for collaborative machine learning (ML) in cellular networks. Background Technology

[0004] Wireless communication systems are widely deployed to provide a variety of telecommunications services such as telephone, video, data, messaging, and broadcasting. Typical wireless communication systems employ multiple access technologies capable of supporting communication with multiple users by sharing available system resources. Examples of such multiple access technologies include Code Division Multiple Access (CDMA) systems, Time Division Multiple Access (TDMA) systems, Frequency Division Multiple Access (FDMA) systems, Orthogonal Frequency Division Multiple Access (OFDMA) systems, Single Carrier Frequency Division Multiple Access (SC-FDMA) systems, and Time Division Synchronous Code Division Multiple Access (TD-SCDMA) systems.

[0005] These multiple access technologies have been adopted in various telecommunications standards to provide a common protocol enabling different wireless devices to communicate at the city, country, region, and even global levels. An example telecommunications standard is 5G New Radio (NR). 5G NR is part of the continuous evolution of mobile broadband released by the 3rd Generation Partnership Project (3GPP) to meet new requirements associated with latency, reliability, security, scalability (e.g., in conjunction with the Internet of Things (IoT),) and others. 5G NR includes services associated with enhanced mobile broadband (eMBB), massive machine-type communications (mMTC), and ultra-reliable low-latency communications (URLLC). Some aspects of 5G NR can be based on the 4G Long Term Evolution (LTE) standard. There is a need for further improvements to 5G NR technology. These improvements can also be applied to other multiple access technologies and telecommunications standards that adopt them. Summary of the Invention

[0006] The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects, and is intended to neither identify key or critical elements of all aspects nor delineate the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that is presented later.

[0007] In one aspect of the disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus can transmit, to an interface of a core network, a request for at least one of a model or a feature associated with at least one of a machine learning (ML) procedure or a neural network (NN) procedure; determine, via the interface of the core network, the at least one of the model or the feature based on the request, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature; and receive, from the interface of the core network, a response to the request for the at least one of the model or the feature, the response to the request indicating the latest update to the at least one of the model or the feature.

[0008] In another aspect of the disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus can determine to initiate a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure; transmit, to a core network, the request; and receive, from the core network based on the request, the at least one of the model or the feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature.

[0009] In yet another aspect of the disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus can transmit, to a base station, a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure; and receive, from the base station based on the request, the at least one of the model or the feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature.

[0010] To the accomplishment of the foregoing and related aspects, one or more aspects comprise the features recited in the following claims, the full scope of which should be accorded to support the validity of the principles of the one or more aspects. The foregoing and the following detailed description are indicative of certain illustrative features as a basis of the one or more aspects, one or more implementations of the one or more aspects, and the BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 FIG. 1 is a diagram illustrating an example of a wireless communications system and an access network.

[0012] Figure 2AFIG. 1 is a diagram illustrating an example of a first subframe, in accordance with various aspects of the present disclosure.

[0013] Figure 2B FIG. 2 is a diagram illustrating an example of DL channels within a subframe, in accordance with various aspects of the present disclosure.

[0014] Figure 2C FIG. 3 is a diagram illustrating an example of a second frame, in accordance with various aspects of the present disclosure.

[0015] Figure 2D FIG. 4 is a diagram illustrating an example of UL channels within a subframe, in accordance with various aspects of the present disclosure.

[0016] Figure 3 FIG. 5 is a diagram illustrating an example of a base station and user equipment (UE) in an access network.

[0017] Figure 4 A diagram showing a network architecture for machine learning (ML) and neural network (NN) model discovery and management techniques is shown.

[0018] Figures 5A-5B A call flow diagram showing ML / NN procedures based on signaling initiated by an operations and management (OAM) core network is shown.

[0019] Figure 6 A call flow diagram showing ML / NN procedures based on base station initiated signaling is shown.

[0020] Figure 7 A call flow diagram for UE initiated model and feature signaling is shown.

[0021] Figure 8 A flow diagram of a method of wireless communication of an OAM core network is shown.

[0022] Figure 9 A flow diagram of a method of wireless communication of a base station is shown.

[0023] Figure 10 A flow diagram of a method of wireless communication of a UE is shown.

[0024] Figure 11 FIG. 1 is a diagram illustrating an example of a first subframe, in accordance with various aspects of the present disclosure.

[0025] Figure 12 FIG. 1 is a diagram illustrating an example of a first subframe, in accordance with various aspects of the present disclosure.

[0026] Figure 13 FIG. 1 is a diagram illustrating an example of a first subframe, in accordance with various aspects of the present disclosure. DETAILED DESCRIPTION

[0027] The detailed description set forth below, in connection with the appended drawings, is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein can be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of various concepts. However, it will be apparent to those skilled in the art that these concepts can be practiced without these specific details. In some instances, well-known structures and components are shown in block diagram form, rather than in detail, in order to avoid obscuring the concepts.

[0028] Several aspects of telecommunication systems will now be presented with reference to various apparatus and methods. These apparatus and methods will be described in the following detailed description and illustrated in the accompanying drawings by various blocks, components, circuits, processes, algorithms, etc. (collectively referred to as "elements"). These elements can be implemented using electronic hardware, computer software, or any combination thereof. Whether such elements are implemented as hardware or software depends on the particular application and design constraints imposed on the overall system.

[0029] By way of example, an element, or any portion of an element, or any combination of elements can be implemented as a "processing system" that includes one or more processors. Examples of processors include microprocessors, microcontrollers, graphics processing units (GPUs), central processing units (CPUs), application processors, digital signal processors (DSPs), reduced instruction set computing (RISC) processors, systems on a chip (SoC), baseband processors, field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gated logic, discrete hardware circuits, and other suitable hardware configured to perform the various functionality described throughout this disclosure. One or more processors in the processing system can execute software. Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software components, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0030] Accordingly, in one or more example embodiments, the functions described can be implemented in hardware, software, or any combination thereof. If implemented in software, the functions can be stored on or encoded as one or more instructions or code on a computer-readable medium. Computer-readable media includes computer storage media. Storage media can be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise a random-access memory (RAM), a read-only memory (ROM), an electrically erasable programmable ROM (EEPROM), compact disk ROM (CD-ROM) or other optical disk storage, magnetic disk storage or other magnetic storage devices, combinations of the aforementioned types of computer-readable media, or any other medium that can be used to store computer executable code in the form of instructions or data structures that can be accessed by a computer.

[0031] Figure 1 FIG. 1 is a diagram illustrating an example of a wireless communications system and an access network 100. The wireless communications system (also referred to as a wireless wide area network (WWAN)) includes base stations 102, UEs 104, an Evolved Packet Core (EPC) 160, and another core network 190 (e.g., a 5G Core (5GC)). The base stations 102 can include macro cells (high power cellular base stations) and / or small cells (low power cellular base stations). The macro cells can include base stations. The small cells can include femtocells, picocells, and microcells.

[0032] The base stations 102 configured for 4G LTE (collectively referred to as the Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network (E-UTRAN)) can interface with the EPC 160 through first backhaul links 132 (e.g., S I interface). The base stations 102 configured for 5G NR (collectively referred to as the Next Generation RAN (NG-RAN)) can interface with the core network 190 through second backhaul links 184. In addition to other functions, the base stations 102 can perform one or more of the following functions: transmission of user data, radio channel encryption and decryption, integrity protection, header compression, mobility control functions (e.g., handover, dual connectivity), inter-cell interference coordination, connection setup and release, load balancing, distribution of paging information, NAS node selection, synchronization, radio access network (RAN) sharing, multimedia broadcast multicast service (MBMS), user and device tracking, RAN information management (RIM), paging, positioning, and delivery of warning messages. The base stations 102 can communicate directly or indirectly (e.g., through the EPC 160 or core network 190) with each other over third backhaul links 134 (e.g., X2 interface). The first, second, and third backhaul links 132, 184, and 134 can be wired or wireless.

[0033] The base stations 102 can wirelessly communicate with the UEs 104. Each of the base stations 102 can provide communication coverage for a respective geographic coverage area 110. There can be overlapping geographic coverage areas 110. For example, a small cell 102' can have a coverage area 110' that overlaps with one or more macrocells 102. A network that includes both small cell and macrocells can be known as a heterogeneous network. A heterogeneous network can also include Home Evolved Node Bs (eNBs) (HeNBs), which can provide service to a restricted group known as a closed subscriber group (CSG). The communication links 120 between the base stations 102 and the UEs 104 can include uplink (UL) (also referred to as reverse link) transmissions from a UE 104 to a base station 102 and / or downlink (DL) (also referred to as forward link) transmissions from a base station 102 to a UE 104. The communication links 120 can use multiple-input and multiple-output (MIMO) antenna technology, including spatial multiplexing, beamforming, and / or transmit diversity. The communication links can be through one or more carriers, and each carrier can be a band of frequency waves having a predetermined width and can be used to transmit data between base stations 102 and UEs 104. The base stations 102 / UEs 104 can use spectrum up to Y megahertz (MHz) (e.g., 5, 10, 15, 20, 100, 400, etc. MHz) bandwidth per carrier allocated in a carrier aggregation of up to a total of Yx MHz (x component carriers) used for transmission in each direction. The carriers can or can not be adjacent to each other. The allocation of carriers can be asymmetric with respect to DL and UL (e.g., more or less carriers can be allocated for DL than for UL). The component carriers can include a primary component carrier and one or more secondary component carriers. A primary component carrier can be referred to as a primary cell (PCell) and a secondary component carrier can be referred to as a secondary cell (SCell).

[0034] Certain UEs 104 can communicate with each other using device-to-device (D2D) communication link 158. The D2D communication link 158 can use DL / UL WWAN spectrum. The D2D communication link 158 can use one or more sidelink channels, such as a physical sidelink broadcast channel (PSBCH), a physical sidelink discovery channel (PSDCH), a physical sidelink shared channel (PSSCH), and a physical sidelink control channel (PSCCH). D2D communication can be through a variety of wireless D2D communications systems, such as for example, WiMedia, Bluetooth, ZigBee, Wi-Fi based on Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards, LTE, or NR.

[0035] The wireless communications system can also include a Wi-Fi access point (AP) 150 in communication with Wi-Fi stations (STAs) 152 via communication links 154, e.g., in a 5 GHz unlicensed frequency spectrum. When communicating in an unlicensed frequency spectrum, the STAs 152 / AP 150 can perform a clear channel assessment (CCA) prior to communicating in order to determine whether the channel is available.

[0036] The small cells 102' can operate in a licensed and / or an unlicensed frequency spectrum. When operating in an unlicensed frequency spectrum, the small cells 102' can employ NR and use the same 5 GHz unlicensed frequency spectrum as used by the Wi-Fi AP 150, for example. The small cells 102' employing NR in an unlicensed frequency spectrum can boost coverage of the access network and / or increase capacity of the access network.

[0037] The electromagnetic spectrum is often subdivided based on frequency / wavelength into various classes, bands, channels, etc. In 5G NR, two initial operating bands have been identified as frequency range designations FR1 (410 MHz - 7. 125 GHz) and FR2 (24.25 GHz - 52.6 GHz). The frequencies between FR1 and FR2 are often referred to as mid-band frequencies. Although a portion of FR1 is greater than 6 GHz, FR1 is often referred to as a “sub-6 GHz” band in various documents and articles. A similar nomenclature issue sometimes occurs with respect to FR2, which is often referred to as a “millimeter wave” band in documents and articles, despite being different from the extremely high frequency (EHF) band (30 GHz - 300 GHz) which is identified by the International Telecommunications Union (ITU) as a “millimeter wave” band.

[0038] With the above in mind, unless specifically stated otherwise, it should be appreciated that the term “sub-6 GHz” or the like is used herein to generically refer to frequencies that can be less than 6 GHz, can be within FR1, or can include mid-band frequencies. Furthermore, unless specifically stated otherwise, it should be appreciated that the term “millimeter wave” or the like is used herein to generically refer to frequencies that can include mid-band frequencies, can be within FR2, or can be within the EHF band.

[0039] The base stations 102, whether small cell 102' or large cell (e.g., macro base station), can include and / or be referred to as an eNB, gNodeB (gNB), gNodeB (gNB), or another type of base station. Some base stations, such as gNB 180 can operate in a traditional sub 6 GHz spectrum, in millimeter wave frequencies, and / or near millimeter wave frequencies in communication with the UEs 104. When the gNB 180 operates in millimeter wave frequencies, the gNB 180 can be referred to as a millimeter wave base station. Millimeter wave base stations 180 can utilize beamforming 182 in

[0040] The base station 180 can transmit a beamformed signal to the UE 104 in one or more transmit directions 182'. The UE 104 can receive the beamformed signal from the base station 180 in one or more receive directions 182". The UE 104 can also transmit a beamformed signal to the base station 180 in one or more transmit directions. The base station 180 can receive the beamformed signal from the UE 104 in one or more receive directions. The base station 180 / UE 104 can perform beam training to determine the best receive and transmit directions for each of the base station 180 / UE 104. The transmit and receive directions for the base station 180 can or can not be the same. The transmit and receive directions for the UE 104 can or can not be the same.

[0041] The EPC 160 can include a mobility management entity (MME) 162, other MMEs 164, a serving gateway 166, a multimedia broadcast multicast service (MBMS) gateway 168, a broadcast multicast service center (BM-SC) 170, and a packet data network (PDN) gateway 172. The MME 162 can be in communication with a home subscriber server (HSS) 174. The MME 162 is the control node that processes the signaling between the UEs 104 and the EPC 160. Generally, the MME 162 provides bearer and connection management. All user Internet protocol (IP) packets are transferred through the serving gateway 166, which itself is connected to the PDN gateway 172. The PDN gateway 172 provides UE IP address allocation as well as other functions. The PDN gateway 172 and the BM-SC 170 are connected to the IP services 176. The IP services 176 can include the Internet, an intranet, an IP multimedia subsystem (IMS), a PS streaming service, and / or other IP services. The BM-SC 170 can provide functions for MBMS user service provisioning and

[0042] The core network 190 can include an access and mobility management function (AMF) 192, other AMFs 193, a session management function (SMF) 194, and a user plane function (UPF) 195. The AMF 192 can be in communication with a unified data management (UDM) 196. The AMF 192 is the control node that processes the signaling between the UEs 104 and the core network 190. Generally, the AMF 192 provides QoS flow and session management. All user Internet protocol (IP) packets are transferred

[0043] A base station can include and / or be referred to as a gNB, NodeB, eNB, access point, base transceiver station, radio base station, radio transceiver, transceiver function, basic service set (BSS), extended service set (ESS), transmit reception point (TRP), or some other suitable terminology. The base station 102 provides an access point to the EPC 160 or core network 190 for a UE 104. Examples of UEs 104 include a cellular phone, a smart phone, a session initiation protocol (SIP) phone, a laptop, a personal digital assistant (PDA), a satellite radio, a global positioning system, a multimedia device, a video device, a digital audio player (e.g., MP3 player), a camera, a game console, a tablet, a smart device, a wearable device, a vehicle, an electric meter, a gas pump, a large or small kitchen appliance, a healthcare device, an implant, a sensor / actuator, a display, or any other similar functional device. Some of the UEs 104 can be referred to as IoT devices (e.g., a parking meter, gas pump, toaster, vehicle, heart monitor, etc.). The UE 104 can also be referred to as a station, a mobile station, a subscriber station, a mobile unit, a subscriber unit, a wireless unit, a remote unit, a mobile device, a wireless device, a wireless communications device, a remote device, a mobile subscriber station, an access terminal, a mobile terminal, a wireless terminal, a remote terminal, a handset, a user agent, a mobile client, a client, or some other suitable terminology.

[0044] Referring again to Figure 1In certain aspects, the core network 190 can include an operations and management (OAM) initiator component 191 configured to transmit, to an interface of the core network, a request for at least one of a model or a feature associated with at least one of a machine learning (ML) procedure or a neural network (NN) procedure, determine, via the interface of the core network, the at least one of the model or the feature based on the request, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature, and receive, from the interface of the core network, a response to the request for the at least one of the model or the feature, the response to the request indicating the latest update to the at least one of the model or the feature. In certain aspects, the base station 180 can include a base station initiator component 199 configured to determine to initiate a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure, transmit, to the core network, the request, and receive, from the core network based on the request, the at least one of the model or the feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature. In certain aspects, the UE 104 can include a UE initiator component 198 configured to transmit, to the base station, a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure, and receive, from the base station based on the request, the at least one of the model or the feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature. Although the following description can focus on 5G NR, the concepts described herein can be applicable to other similar areas, such as LTE, LTE-A, CDMA, GSM, and other wireless technologies.

[0045] Figure 2A FIG. 200 is a diagram 200 illustrating an example of a first subframe within a 5G NR frame structure. Figure 2B FIG. 230 is a diagram 230 illustrating an example of DL channels within a 5G NR subframe. Figure 2C FIG. 250 is a diagram 250 illustrating an example of a second subframe within a 5G NR frame structure. Figure 2D FIG. 280 is a diagram 280 illustrating an example of UL channels within a 5G NR subframe. The 5G NR frame structure can be frequency-division duplexed (FDD) in which Figure 2A 、 2CIn the examples provided, a 5G NR frame structure is assumed to be TDD with subframe 4 configured with slot format 28 (with mostly DL), where D is DL, U is UL, and F is flexible to use between DL / UL, and subframe 3 configured with slot format 1 (with all UL). While subframes 3, 4 are shown with slot formats 1, 28, respectively, any particular subframe can be configured with any of the various available slot formats 0-61. Slot formats 0, 1 are all DL, all UL, respectively. Other slot formats 2-61 include a mix of DL, UL, and flexible symbols. UEs are configured with a slot format (dynamically through DL control information (DCI) or semi-statically / statically through radio resource control (RRC) signaling) through a received slot format indicator (SFI). Note that the following description also applies for a 5G NR frame structure that is TDD.

[0046] Other wireless communication technologies can have different frame structures and / or different channels. A frame (10 ms) can be divided into 10 equally sized subframes (1 ms). Each subframe can include one or more slots. A subframe can also include mini-slots, which can contain 7, 4, or 2 symbols. Each slot can contain 7 or 14 symbols, depending on the slot configuration. For slot configuration 0, each slot can contain 14 symbols, while for slot configuration 1, each slot can contain 7 symbols. Symbols on the DL can be μ cyclic prefix (CP) orthogonal frequency division multiplexing (OFDM) (CP-OFDM) symbols. Symbols on the UL can be CP-OFDM symbols (for high throughput scenarios) or discrete Fourier transform (DFT) spread OFDM (DFT-s-OFDM) symbols (also known as single carrier frequency division multiple access (SC-FDMA) symbols) (for power limited scenarios; limited to single stream transmission). The number of slots within a subframe can be dependent on the slot configuration and the numerology. For slot configuration 0, different numerologies m0to 4 allow for 1, 2, 4, 8, and 16 slots per subframe, respectively. For slot configuration 1, different numerologies 0 to 2 allow for 2, 4, and 8 slots per subframe, respectively. Accordingly, for slot configuration 0 and numerology m, there are 14 symbols / slot and 2 μ *15 kHz, where m is the numerology 0 to 4. Thus, numerology m = 0 has a subcarrier spacing of 15 kHz, and numerology m = 4 has a subcarrier spacing of 240 kHz. The symbol length / duration is inversely related to the subcarrier spacing. Figures 2A-2DExamples are provided for slot configuration 0 with 14 symbols per slot and digital scheme μ=2 with 4 slots per subframe. The slot duration is 0.25 ms, the subcarrier spacing is 60 kHz, and the symbol duration is approximately 16.67 μs. Within the frame set, one or more different bandwidth portions (BWPs) that can be frequency-division multiplexed can exist (see Figure 2). Each BWP can have a specific digital scheme.

[0047] A resource grid can be used to represent frame structure. Each time slot includes a resource block (RB) (also known as a physical RB (PRB)), which extends 12 consecutive subcarriers. The resource grid is divided into multiple resource elements (REs). The number of bits carried by each RE depends on the modulation scheme.

[0048] like Figure 2A As shown, some of the REs carry reference (pilot) signals (RS) for the UE. RS may include demodulation RS (DM-RS) for channel estimation at the UE (indicated as R for a specific configuration, but other DM-RS configurations are possible) and channel state information reference signals (CSI-RS). RS may also include beam measurement RS (BRS), beam refinement RS (BRRS), and phase tracking RS (PT-RS).

[0049] Figure 2BAn example of various DL channels are shown. The physical downlink control channel (PDCCH) carries DCI within one or more control channel elements (CCEs) (e.g., 1, 2, 4, 8, or 16 CCEs), each CCE including six RE groups (REGs), each REG including 12 consecutive REs in one OFDM symbol of an RB. A PDCCH within one BWP can be referred to as a control resource set (CORESET). A UE is configured to monitor PDCCH candidates in a PDCCH search space (e.g., common search space, UE-specific search space) during PDCCH monitoring occasions on the CORESET, where the PDCCH candidates have different DCI formats and different aggregation levels. An additional BWP can be located at a greater and / or lower frequency across the channel bandwidth. A primary synchronization signal (PSS) can be within symbol 2 of particular subframes of a frame. The PSS is used by a UE 104 to determine subframe / symbol timing and a physical layer identity. A secondary synchronization signal (SSS) can be within symbol 4 of particular subframes of a frame. The SSS is used by a UE to determine a physical layer cell identity group number and radio frame timing. Based on the physical layer identity and the physical layer cell identity group number, the UE can determine a physical cell identifier (PCI). Based on the PCI, the UE can determine locations of the aforementioned DM-RS. The physical broadcast channel (PBCH), which carries a master information block (MIB) that provides primary synchronization signal (PSS) and secondary synchronization signal (SSS) configuration information, can be logically grouped with the PSS and SSS to form a synchronization signal (SS) / PBCH block (also referred to as SS block (SSB)). The MIB provides a number of RBs in the system bandwidth and a system frame number (SFN). The physical downlink shared channel (PDSCH) carries user data, broadcast system information (e.g., system information blocks (SIBs)), and paging messages.

[0050] As illustrated in Figure 2C Some of the REs carry DM-RS (indicated as R for one particular configuration, but other DM-RS configurations are possible) for channel estimation at the base station. The UE can transmit DM-RS for the physical uplink control channel (PUCCH) and DM-RS for the physical uplink shared channel (PUSCH). The PUSCH DM-RS can be transmitted in the first one or two symbols of a slot. The PUCCH DM-RS can be transmitted in the last symbol of a slot, depending on the PUCCH format. The UE can transmit a sounding reference signal (SRS). The SRS can be transmitted in the last symbol of a slot. The SRS can have a comb-2 structure, with the UE transmitting comb-wise in the odd and even symbols of a slot. The SRS can be used by a base station for channel quality feedback, to enable frequency-dependent scheduling on the UL.

[0051] Figure 2D An example of various UL channels is shown. The PUCCH can be positioned as indicated in one configuration. The PUCCH carries uplink control information (UCI), such as scheduling requests, a channel quality indicator (CQI), a precoding matrix indicator (PMI), a rank indicator (RI), and hybrid automatic repeat request (HARQ) acknowledgment (ACK) (HARQ-ACK) information (ACK / negative ACK (NACK) feedback). The PUSCH carries data, and can additionally be used to carry a buffer status report (BSR), a power headroom report (PHR), and / or UCI.

[0052] Figure 3 is a block diagram of the base station 310 communicating with the UE 350 in an access network. In the DL, IP packets from the EPC 160 can be provided to a controller / processor 375. The controller / processor 375 implements layer 3 and layer 2 functionality. Layer 3 includes a radio resource control (RRC) layer, and layer 2 includes a service data adaptation protocol (SDAP) layer, a packet data convergence protocol (PDCP) layer, a radio link control (RLC) layer, and a medium access control (MAC) layer. The controller / processor 375 provides RRC layer functionality associated with, e.g., broadcasting of system information (e.g., MIB, SIBs), RRC connection control (e.g., RRC connection paging, RRC connection establishment, RRC connection modification, and RRC connection release), inter radio access technology (RAT) mobility, and measurement configuration for UE measurement reporting; PDCP layer functionality associated with, e.g., header compression / decompression, security (ciphering, deciphering, integrity protection, integrity verification), and handover support functions; RLC layer functionality associated with, e.g., transfer of upper layer

[0053] The transmit (TX) processor 316 and the receive (RX) processor 370 implement layer 1 functionality associated with various signal processing functions. Layer 1, which includes a physical (PHY) layer, can include error detection on the transport channels, forward error correction (FEC) coding / decoding of the transport channels, interleaving, rate matching, mapping to physical channels, modulation / demodulation of physical channels, and MIMO antenna processing. The TX processor 316 handles mapping to signal constellations based on various modulation schemes (e.g., binary phase-shift keying (BPSK), quadrature phase-shift keying (QPSK), M-phase-shift keying (M-PSK), M-quadrature amplitude modulation (M-QAM)). The coded and modulated symbols are then split into parallel streams. Each stream can then be mapped to a OFDM subcarrier, multiplexed with a reference signal (e.g., pilot) in the time and / or frequency domain, and then combined together using an inverse fast Fourier transform (IFFT) to produce a physical channel carrying a time domain OFDM symbol stream. The OFDM stream is spatially precoded to produce multiple spatial streams. Channel estimates from a channel estimator 374 can be used to determine the coding and

[0054] At the UE 350, each receiver 354RX receives a signal through its respective antenna 352. Each receiver 354RX recovers information modulated onto an RF carrier and provides the information to the receive (RX) processor 356. The TX processor 368 and the RX processor 356 implement layer 1 functionality associated with various signal processing functions. The RX processor 356 can perform spatial processing on the information to recover any spatial streams destined for the UE 350. If multiple spatial streams are destined for the UE 350, they can be combined by the RX processor 356 into a single OFDM symbol stream. The RX processor 356 then converts the OFDM symbol stream from the time-domain to the frequency domain using a fast Fourier transform (FFT). The frequency domain signal comprises a separate OFDM symbol stream for each subcarrier of the OFDM signal. The symbols on each subcarrier, and the reference signal, are recovered and demodulated by determining the most likely signal constellation points transmitted by the base station 310. These soft decisions can be based on channel estimates computed by the channel estimator 358. The soft decisions are then decoded and deinterleaved to recover the data and control signals that were originally transmitted by the base station 310 on the physical channel. The data and control signals are then provided to the controller / processor 359, which implements layer 3 and layer 2 functionality.

[0055] The controller / processor 359 can be associated with a memory 360 that stores program codes and data. The memory 360 can be referred to as a computer-readable medium. In the UL, the controller / processor 359 provides demultiplexing between transport and logical channels, packet reassembly, deciphering, header decompression, and control signal processing to recover IP packets from the EPC 160. The controller / processor 359 is also responsible for error detection using an ACK and / or NACK protocol to support HARQ operations.

[0056] Similar to the functionality described in connection with the DL transmission by the base station 310, the controller / processor 359 provides RRC layer functionality associated with system information (e.g., MIB, SIBs) acquisition, RRC connections, and measurement reporting; PDCP layer functionality associated with header compression / decompression, and security (ciphering, deciphering, integrity protection, integrity verification); RLC layer functionality associated with the transfer of upper layer PDUs, error correction through ARQ, concatenation, segmentation, and reassembly of RLC SDUs, re-segmentation of RLC data PDUs, and reordering of RLC data PDUs; and MAC layer functionality associated with mapping between logical channels and transport channels, multiplexing of MAC SDUs onto TBs, demultiplexing of MAC SDUs from TBs, scheduling information reporting, error correction through HARQ, priority handling, and logical channel prioritization.

[0057] Channel estimates derived by the channel estimator 358 from a reference signal or feedback transmitted by the base station 310 can be used by the TX processor 368 to select the appropriate coding and modulation schemes, and to facilitate spatial processing. The spatial streams generated by the TX processor 368 can be provided to different antenna 352 via separate transmitters 354TX. Each transmitter 354TX can modulate an RF carrier with a respective spatial stream for transmission.

[0058] The UL transmission is processed at the base station 310 in a manner similar to that described in connection with the receiver function at the UE 350. Each receiver 318RX receives a signal through its respective antenna 320. Each receiver 318RX recovers information modulated onto an RF carrier and provides the information to a RX processor 370.

[0059] The controller / processor 375 can be associated with a memory 376 that stores program codes and data. The memory 376 can be referred to as a computer-readable medium. In the UL, the controller / processor 375 provides demultiplexing between transport and logical channels, packet reassembly, deciphering, header decompression, control signal processing to recover IP packets from the UE 350. IP packets from the controller / processor 375 can be provided to the EPC 160. The controller / processor 375 is also responsible for error detection using an ACK and / or NACK protocol to support HARQ operations.

[0060] At least one of the TX processor 368, the RX processor 356, and the controller / processor 359 can be configured to perform aspects of the methods 200, 198 in FIGS. Figure 1

[0061] At least one of the TX processor 316, the RX processor 370, and the controller / processor 375 can be configured to perform aspects of the methods 199 in FIG. Figure 1

[0062] At least one of the TX processor 316, the RX processor 370, and the controller / processor 375 can be configured to perform aspects of the methods 191 in FIG. Figure 1

[0063] Wireless communication systems can be configured to share available system resources and provide various telecommunication services (e.g., telephony, video, data, messaging, broadcasts, etc.) based on multiple access technologies that enable communication with multiple users (such as CDMA systems, TDMA systems, FDMA systems, OFDMA systems, SC-FDMA systems, TD-SCDMA systems, etc.). In many instances, common protocols that facilitate communication with wireless devices are employed in various telecommunication standards. For example, communication methods associated with eMBB, mMTC, and URLLC can be incorporated into 5G NR telecommunication standards, while other aspects can be incorporated into 4G LTE standards. As mobile broadband technology is part of an ongoing evolution, further improvements to mobile broadband are still useful to continue the advancement of such technology.

[0064] Figure 4 A diagram 400 illustrating a network architecture for machine learning (ML) and neural network (NN) model discovery and management techniques is shown. Performance of a ML / NN model can be based on multiple criteria, such as feature selection, model selection, sample quantity, etc. Feature selection can correspond to input parameters of a ML / NN model used to train and test the ML / NN model. Model selection can correspond to determining a model to perform from a plurality of models (e.g., based on model complexity or optimization parameters). Sample quantity can correspond to a number of observations of one or more features. ​​​

[0065] The selected models and features can have an impact on the performance of the ML / NN models and the overall network architecture and system performance. The UE 402 and the base station 404 can perform procedures in the PHY layer, MAC layer, and upper layers, where the ML / NN models can provide performance enhancements. Different ML / NN models can be used for different procedures of the UE 402 and the base station 404 for different technologies in the PHY layer, MAC layer, and the like. For example, a first ML model can be used for a first procedure and a second ML model can be used for a second procedure based on one or more model selection techniques. If the UE 402 is associated with reduced power and / or computational complexity, the UE 402 can be configured with low computational complexity ML / NN models or the UE 402 can not be configured to perform ML / NN model and feature selection. Thus, the network (e.g., the base station 404 and an operations and management (OAM) core network 406) can assist the UE 402 with ML / NN model and feature selection for different technologies in the PHY layer, MAC layer, and upper layers.

[0066] The OAM core network 406 can include a plurality of network entities and interfaces that allow for the execution of ML / NN procedures at the UE 402, the base station 404, and / or the OAM core network 406. The UE 402 can be configured to transmit a ML / NN model and feature request message to the base station 404, which can also be configured to transmit a model and feature request message to a model data access coordinator (MDAC) 408 of the OAM core network 406 to update the ML / NN models. After the base station 404 receives an update to the ML / NN models based on the ML / NN model and feature request transmitted to the OAM core network 406, the base station 404 can transmit a ML / NN model and feature response message to the UE 402 based on the ML / NN model and feature request message received from the UE 402 and the update to the ML / NN models received from the OAM core network 406.

[0067] The MDAC 408 can communicate with multiple network entities, such as the ML / NN database 410, the data lake 412, the ML / NN server 414, and the like. In a configuration, the ML / NN server 414 can be located within the OAM core network 406 (e.g., it can be an Internet Service Provider (ISP) Mobile Edge Computing (MEC) core network) or the ML / NN server 414 can be located outside of such a network and hosted by a different entity. A standalone ML / NN ranker can be configured to rank the ML / NN models within the ML / NN server 414. The UE 402 and the base station 404 can utilize ML / NN models registered at the ML / NN database 410, while the OAM core network 406 can perform authentication / verification protocols for generated ML / NN models. The MDAC 408 can determine whether to utilize the authenticated / verified models or whether to select different models for a particular procedure.

[0068] The ML / NN database 410 can include ML / NN models for performing multiple operations, which can be based on updates to the ML / NN models. That is, the ML / NN database 410 can store different ML / NN models and updates. The granularity of the storage of the ML / NN models can be per network slice, per cell, per RAT, per target area (TA), per RAN notification area (RNA), per PLMN, and the like. The data lake 412 can be used to store datasets and perform principal component analysis (PCA) for different ML / NN procedures. For example, the data lake 412 can store data for feature selection by the OAM core network 406 based on requests from the UE 402 and / or the base station 404. The granularity of the storage of the features of the ML / NN models can be per network slice, per cell, per RAT, per TA, per RNA, per PLMN, and the like. The ML / NN server 414 can be configured to assist the UE 402, the RAN, or other ML / NN entities with model selection (e.g., outside of the OAM core network 406). The model selection can be discoverable per network slice, per cell, per RAT, per TA, per RNA, per PLMN. The MDAC 408 can be configured to interface the ML / NN database 410, the data lake 412, the ML / NN server 414, and the base station 404. For example, the MDAC 408 can be configured to be a coordinator entity that provides an interface between the other network entities, provides the ML / NN models to the UE 402 and the base station 404, and performs authentication / verification of the ML / NN models.

[0069] The MDAC 408 can be a logical entity for data discovery as well as ML / NN model and analytics discovery. The protocol of the MDAC 408 can be associated with a domain name system (DNS), a relational database, or a hypertext transfer protocol (HTTP) such as JavaScript Object Notation (JSON) or Extensible Markup Language (XML). The discovery process can be based on a sub-process for a ML training / inference host (e.g., the UE 402, the base station 404, or the OAM core network 406) that sends data and model requests to the MDAC 408 for coordination. The MDAC 408 can query the data lake / pool 412 and the ML / NN server 414 for a uniform resource identifier (URI) of the data and model. The MDAC 408 can signal the ML / NN database 410 for an update to the ML / NN model. The MDAC 408 can also request the data lake / pool 412 to initiate data collection to signal the base station 404 or the UE 402 for corresponding information if the data for the process is not current or unavailable. In aspects, the MDAC 408 can request the ML / NN server 414 to generate a ML / NN model for a different ML / NN process if the model is not available for such a process. The MDAC 408 can update the ML / NN database 410 with the updated ML / NN model by pushing the updated ML / NN model to the ML / NN database 410. The MDAC 408 can send a response to the ML host with the URI of the requested data and ML model, and the ML host can fetch the data (e.g., via a HTTP Get URI (GET URI) command).

[0070] The search and discovery process for the ML / NN model can be based on any one of one or more input parameters, one or more output parameters, a name, an identifier (ID), a keyword, or a hypertext. However, the ML / NN model can have to first register with the ML / NN database 410. Thus, an unregistered ML / NN model that is not stored in the ML / NN database 410 can not be utilized. The UE 402, the base station 404, or the OAM core network 406 can be configured to initiate a model process, and the OAM core network 406 can be configured to generate or update a data set (e.g., based on a PHY layer, a MAC layer, or an upper layer) for a ML / NN model to perform a different ML / NN process.

[0071] Request and response information signaled for ML / NN model and feature selection can be associated with ML / NN procedures of different entities of the network architecture. For example, ML / NN model selection for inference and / or training can be network initiated (e.g., initiated by the OAM core network 406 or base station 404) or UE initiated (e.g., initiated by the UE 402). Request and response information signaled for ML / NN model and feature selection can also be authenticated based on the ML / NN model.

[0072] Figures 5A-5B Call flow diagrams 500-550 are shown for ML / NN procedures based on signaling initiated by the OAM core network 506 / 556. The OAM core network 506 / 556 can initiate ML / NN model selection for inference and / or training procedures. The OAM initiated ML / NN procedures can be signaling based or management based. The signaling based procedures can correspond to initiation of ML / NN procedures associated with a particular UE 552, while the OAM management based procedures can correspond to initiation of ML / NN procedures associated with a particular TA. The OAM core network 506 / 556 can initiate the ML / NN procedures via a ML / NN model or model ID associated with the selected features. Upon receiving an indication of the OAM initiated ML / NN procedures, the base station 554 can also indicate to the UE 552 the ML / NN model or model ID to be used for the ML / NN procedures. The ML / NN model can be sent to the UE 552 based on RRC signaling, or the UE 552 can download the ML / NN model and features based on a HTTP GET URI request after receiving an indication of the model ID and features in an RRC message. That is, if the base station 554 sends the model directly to the UE 552, the UE 552 can use the sent model / features. If the base station 554 sends the model ID, the UE 552 can request and download the ML / NN model and features using a HTTP GET URI request. Thus, the base station 554 can provide the model and features to the UE 552, or the base station 554 can indicate the model ID to the UE 552 so that the UE 552 independently downloads the model and features. If the UE 552 is handed over to a second base station, the ML / NN model and features can be transferred / communicated to the second base station.

[0073] Referring to call flow diagram 500, OAM core network 506 can be configured to determine ML / NN models and features to send to a UE or base station to perform a corresponding ML / NN procedure. OAM core network 506 can initiate the ML / NN procedure at 516 via MDAC 508 based on a model and feature selection request. For example, MDAC 508 can receive an indication of the ML / NN procedure at 516 and send a feature selection request to data lake / pool 512 at 518. Additionally or alternatively, MDAC 508 can send an ML / NN model selection request to ML / NN database 510 at 520. If ML / NN database 510 includes a threshold amount of data for the ML / NN model, ML / NN database 510 can send an ML / NN model selection response to MDAC 508 at 528 indicating the ML / NN model. Similarly, data lake / pool 512 can send a feature selection response to MDAC 508 at 526 based on the feature selection request received at 518.

[0074] If ML / NN database 510 does not include a threshold amount of data for the ML / NN model, ML / NN database 510 can send an ML / NN model update request to MDAC 508 at 522a and also send the ML / NN model update request from MDAC 508 to ML / NN server 514 at 522b. ML / NN server 514 can send an ML / NN model update response to MDAC 508 at 524a indicating an update to the ML / NN model and also send the ML / NN model update response from MDAC 508 to ML / NN database 510 at 524b. Based on the update to the ML / NN model, ML / NN database 510 can send an ML / NN model selection response to MDAC 508 at 528 based on the ML / NN model selection request received at 520. MDAC 508 can indicate the ML / NN model and features to OAM core network 506 at 530, and OAM core network 506 initiates the operations of call flow diagram 550 based on the indicated ML / NN model and features.

[0075] Referring to call flow diagram 550, OAM core network 556 can initiate ML / NN procedures (e.g., via signaling-based ML / NN techniques or management-based ML / NN techniques) at 566a / 568 for different network procedures (e.g., at base station 554), UE procedures, or both. For a signaled request, OAM core network 556 can select UE 552 for the ML / NN procedure and indicate the request to AMF 555 at 566a. In aspects, AMF 555 can relay the request to base station 554 at 566b, which can also relay the request to UE 552 (e.g., at 574). For a management request, OAM core network 556 can send the request to a TA (e.g., which includes base station 554) at 568. Base station 554 can also send the request to UE 552 (e.g., at 574) if UE 552 is to perform the ML / NN procedure. Some ML / NN procedures can be independent of base station 554, independent of UE 552, or corresponding to both base station 554 and UE 552.

[0076] Based on the signaling-based request or the management-based request received from OAM core network 556 at 566b / 568, base station 554 can download ML / NN models and features from OAM core network 556 at 570. At 572, base station 554 can determine to inform UE 552 of the ML / NN procedure and / or a corresponding model ID. At 574, base station 554 can send the ML / NN models and features to UE 552 to perform the ML / NN procedure, or base station 554 can send the ML / NN model ID to UE 552 for UE 552 to independently download the ML / NN models.

[0077] Figure 6 Call flow diagram 600 is shown for a base station-initiated signaling based ML / NN procedure. Base station 604 can be configured to initiate ML / NN model selection techniques for inference and / or training procedures. In aspects, base station 604 can initiate the ML / NN procedure at 616 based on initialization or based on performance falling below a threshold. For example, if the ML / NN procedure is used to encode / decode data transmitted between base station 604 and UE 602, base station 604 can determine to initiate the ML / NN model selection techniques at 616 to improve performance. Base station 604 can also switch the ML / NN procedure based on switching to different models and features for performing the ML / NN procedure.

[0078] At 618, the base station 604 can send the ML / NN model and feature selection request to the OAM core network 606. When receiving the ML / NN model and feature selection response to the ML / NN model and feature selection request from the OAM core network 606 at 620, the base station 604 can send an RRC message to the UE 602 to initiate the ML / NN procedure. The RRC message can be an RRCReconfiguration message, an RRCSetup message, an RRCResume message, an RRCReestablishment message, and the like, which can indicate the model ID, the ML / NN procedure ID, and / or the feature metrics to the UE 602. The UE can utilize the information included in the RRC message, such as the model ID, the feature list, and the like, to perform the ML / NN procedure.

[0079] In a configuration, the base station 604 can indicate to the UE 602 to perform the ML / NN model for procedures such as cell reselection, logged measurements, early measurements, and other idle or inactive mode procedures. Separately from the connected mode procedures, the base station 604 can download the ML / NN model and signal the configuration (e.g., based on an RRCReconfiguration message) to the UE 602 for performing the idle or inactive mode procedures. The base station 604 can provide the ML / NN model or model ID and features to the UE 602 in an RRCRelease message.

[0080] The UE 602 can be configured to determine whether to perform the ML / NN inference and / or training procedure based on conditions of the UE 602. For example, the UE 602 can be configured to accept the ML / NN model, reject the ML / NN model, or propose an alternative ML / NN model. The UE 602 can send a request for ML / NN inference reduction based on conditions of the UE 602, such as overheating, limited processing (e.g., microprocessor without interlocked pipeline stages (MIPS), instruction per second limit), battery status, and the like. To send the message / request, the UE 602 can utilize UE assistance information or an RRC message to suspend ML / NN inference and training and / or reduce ML / NN complexity. In response, the network (e.g., the base station 604 and / or the OAM core network 606) can signal an alternative model as a fallback (e.g., if the UE 602 sends a request for reduced complexity), or the network can signal complexity reduction parameters that the UE 602 can use to reduce the ML / NN complexity. If the UE 602 is operating based on an increased number of features, the network can signal the UE 602 to stop performing a subset of the features (e.g., features 1 to k), but perform the remaining subset of the features to provide reduced ML / NN complexity.

[0081] Additionally or alternatively, for ML / NN inference and / or training procedures, the network can request multiple models and associated features. Thus, the network can provide multiple ML / NN models and feature selections to the UE 602 at the same time, rather than sending a request and receiving a response for each ML / NN inference and training procedure. The UE 602 can be configured to alternate between high performance and model simplification based on CPU metrics, power consumption, power state, etc. of the UE 602.

[0082] Figure 7 A call flow diagram 700 is shown for UE-initiated model and feature signaling. The UE 702 can determine to switch ML / NN models to a particular ML / NN procedure at 716. Alternatively, the UE 702 can initiate a procedure for downloading ML / NN models and features at 718 upon initialization of a ML / NN procedure. For example, if the performance of the UE 702 falls below a threshold, the UE 702 can send an ML / NN model and feature request to the base station 704 at 720 to switch to a ML / NN procedure of the UE 702 and improve the performance of the UE 702, such as for ML / NN procedures associated with encoding and decoding operations. An ML / NN model and feature request can also be sent from the base station 704 to the OAM core network 706 at 722, which can send an ML / NN model and feature response to the base station 704 based on the received ML / NN model and feature request at 724. The base station 704 can likewise relay the ML / NN model and feature response for the ML / NN procedure to the UE 702 based on the ML / NN model and feature request received at 720 and the ML / NN model and feature response received at 724 at 726.

[0083] In aspects, the UE 702 can provide a UEAssistanceInformation message or other dedicated RRC signaling for initialization / switching requests for ML / NN model and feature selection. Upon initiation of a ML / NN procedure, the UE 702 can determine to download an associated ML / NN model or send a request to a serving base station 704. For example, upon activation of a network (e.g., RAN slice), the UE 702 can determine to use a ML / NN model, which can request the ML / NN model from the base station 704. The UE 702 can request the ML / NN model with a UEAssistanceInformation message or other dedicated RRC signaling.

[0084] Authentication techniques can be performed prior to use of the ML / NN model. For example, use of the ML / NN model by the UE 702 can depend on a threshold security level of the ML / NN model. Thus, the ML / NN model and features can first be verified by the network. Different security aspects can be associated with downloading the ML / NN model. For a trusted model, each ML / NN model can be stored based on a signature issued by a trusted entity (e.g., similar to an SSL certificate). The UE 702 can verify the signature / certificate prior to use of the ML / NN model. For a trusted user, each ML / NN model can include a list of authorized users. The authority of each user can be verified via a download request.

[0085] Figure 8 is a flow diagram of a method of wireless communication. The method can be performed by a core network (e.g., core network 190; OAM core network 406, 506, 556, 606, 706; apparatus 1102, etc.), which can include memory 376 and can be the entire core network 190, 406, 506, 556, 606, 706 or a component of the core network 190, 406, 506, 556, 606, 706, such as the TX processor 316, the RX processor 370, and / or the controller / processor 375.

[0086] At 802, the core network can transmit, to an interface of the core network, a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure. For example, with reference to Figures 4-5A The core network 506 can indicate the ML / NN procedure to the MDAC 508 at 516. The interface of the core network can be the MDAC 408 / 508 interfacing with at least one of the database 410 / 510, the data lake 412 / 512, the data pool 412 / 512, the server 414 / 514, or the base station 404. The transmission of the request for at least one of a model or a feature (e.g., at 516) can be initiated based on an indication from at least one of the core network 406 / 506, the base station 404 / 604, or the UE 402 / 702. In aspects, the core network (e.g., core network 190) can correspond to an OAM entity (e.g., 406, 506, 556, 606, 706).

[0087] At 804, the core network can query at least one of a database for a URI of the model or at least one of a data lake or a data pool for the feature. For example, with reference to Figure 5A The MDAC 508 of the OAM core network 506 can transmit a ML / NN model selection request to the ML / NN database 510 at 520 and a feature selection request to the data lake / pool 512 at 518.

[0088] At 806, the core network can determine, based on the query, a status of at least one of a URI or a feature of the model. For example, referring to Figure 5A , the MDAC 508 of the OAM core network 506 can determine a status of the ML / NN model based on the ML / NN model selection response received at 528 from the ML / NN database 510 and / or based on the feature selection response received at 526 from the data lake / pool 512.

[0089] At 808, the core network can transmit, based on the status of the URI of the model, a model update request to the server. For example, referring to Figure 5A , the MDAC 508 of the OAM core network 506 can transmit, at 522b, an ML / NN model update request to the ML / NN server 514 based on the indication received at 522a from the ML / NN database 510.

[0090] At 810, the core network can receive, based on the model update request, an updated model from the server. For example, referring to Figure 5A , the MDAC 508 of the OAM core network 506 can receive, at 524a, an ML / NN model update response from the ML / NN server 514 based on the ML / NN model update request transmitted at 522b to the ML / NN server 514. The updated model can be registered in the ML / NN database 510 based on an authentication procedure for the updated model.

[0091] At 812, the core network can determine, based on the request, at least one of a model or a feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature, via an interface of the core network. For example, referring to Figure 5A , the MDAC 508 of the OAM core network 506 can determine the ML / NN model based on the ML / NN model selection response received at 528 from the ML / NN database 510. In aspects, the ML / NN model selection response can be based on the ML / NN model update response transmitted at 524a from the ML / NN server 514. The MDAC 508 of the core network 506 can additionally or alternatively determine the feature based on the feature selection response received at 526 from the data lake / pool 512, which can be based on the updated data.

[0092] At 814, the core network can receive, from the interface of the core network, a response to the request for the at least one of the model or the feature, the response to the request indicating the latest update to the at least one of the model or the feature. For example, referring to Figure 5AAt 530, the core network 506 can receive the ML / NN model and the features from the MDAC 508. The ML / NN model and the features received at 530 can be indicative of the ML / NN model selection response sent from the ML / NN database 510 at 528 and / or the features selection response sent from the data lake / pit 512 at 526.

[0093] At 816, the core network can transmit the at least one of the model or the features to at least one of the first base station or the AMF of the core network based on the latest update to the at least one of the model or the features. For example, with reference to Figures 5A-5B At 566a, the OAM 506 / 556 can transmit the ML / NN model and the features to the AMF via the signaling-based ML / NN technique, or at 568, the OAM 506 / 556 can transmit the ML / NN model and the features to the base station 554 via the management-based ML / NN technique. In aspects, the transmission at 566a / 568 can be based on the ML / NN model and the features received at 530. The transmission at 568 to at least one of the first base station (e.g., 554) or the AMF 555 of the core network (e.g., the OAM core network 556) can be configured to initiate at least one of an ML procedure or a NN procedure, the at least one of the ML procedure or the NN procedure being at least one of a signaling procedure or a management procedure.

[0094] At 818, the core network can transmit the at least one of the model or the features to the second base station based on a handover of the UE to the second base station. For example, with reference to Figure 1 At 818, the core network can transmit the at least one of the model or the features to the second base station based on a handover of the UE to the second base station. For example, with reference to

[0095] Figure 9 is a flow diagram of a method of wireless communication. The method can be performed by a base station (e.g., the base station 102, 180, 404, 554, 604, 704; the apparatus 1202, etc.), which can include the memory 376 and can be the entire base station 102, 180, 404, 554, 604, 704 or a component of the base station 102, 180, 404, 554, 604, 704, such as the TX processor 316, the RX processor 370, and / or the controller / processor 375.

[0096] At 902, the base station can determine to initiate a request for at least one of a model or features associated with at least one of an ML procedure or a NN procedure. For example, with reference to Figure 6At 616, the base station 604 can determine to initiate the ML / NN procedure. The determination at 616 regarding the initiation request can be based on at least one of an initialization of the ML procedure, an initialization of the NN procedure, or a performance degradation of the base station 604.

[0097] At 904, the base station can transmit the request to the core network. For example, with reference to Figure 6 At 618, the base station 604 can transmit the ML / NN model and feature selection request to the OAM core network 606.

[0098] At 906, the base station can receive, from the core network based on the request, at least one of a model or a feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature. For example, with reference to Figure 6 At 620, the base station 604 can receive, from the OAM core network 606 based on the ML / NN model and feature selection request transmitted at 618, a ML / NN model and feature selection response.

[0099] At 908, the base station can transmit, to the UE, an RRC message indicating the at least one of the model or the feature. For example, with reference to Figure 6 At 622, the base station 604 can transmit, to the UE 602, an RRC reconfiguration message indicating the model ID and the features. In aspects, the RRC message can correspond to at least one of an RRC setup message, an RRC resume message, or an RRC reestablishment message associated with at least one of an ID, a procedure ID, or a feature metric. In further aspects, the RRC message can correspond to an RRC release message associated with at least one of a cell reselection procedure, logged measurements, predicted measurements, or an idle mode procedure.

[0100] At 910, the base station can receive, from the UE based on the RRC message, a download request for the at least one of the model or the feature. For example, with reference to Figure 7 At 720, the base station 704 can receive, from the UE 702, a ML / NN model and feature request based on the download of the ML / NN model and features initiated by the UE 702 at 718.

[0101] At 912, the base station can transmit, to the UE based on the download request, the at least one of the model or the feature. For example, with reference to Figure 7 At 726, the base station 704 can transmit, to the UE 702, a ML / NN model and feature response based on the ML / NN model and feature request received from the UE 702 at 720.

[0102] Figure 10is a flowchart 1000 of a method of wireless communication. The method can be performed by a UE (e.g., the UE 104, 402, 552, 602, 702; the apparatus 1302, etc.) that can include the memory 360 and that can be the entire UE 104, 402, 552, 602, 702 or a component of the UE 104, 402, 552, 602, 702, such as the TX processor 368, the RX processor 356, and / or the controller / processor 359.

[0103] At 1002, the UE can determine to initiate a request for at least one of a model or a feature, where the request is transmitted to a base station based on a determination to initiate the request. For example, with reference to Figure 7 , the UE 702 can determine to initiate a ML / NN model and feature request to the base station 704. The determination to initiate the request transmitted at 720 can be based on a switch at 716 from at least one of a ML process or a NN process to at least one of a second ML process or a second NN process. Additionally or alternatively, the determination to initiate the request transmitted at 720 can be based on an initialization of at least one of a ML process or a NN process. The determination to initiate the request transmitted at 720 can also be based on a performance degradation of the UE 702.

[0104] At 1004, the UE can transmit, to a base station, a request for at least one of a model or a feature, the request associated with at least one of a ML process or a NN process. For example, with reference to Figure 7 , the UE 702 can transmit, at 720, a ML / NN model and feature request for a ML / NN process to the base station 704.

[0105] At 1006, the UE can receive, from the base station based on the request, at least one of a model or a feature, the at least one of a model or a feature corresponding to a latest update to the at least one of a model or a feature. For example, with reference to Figure 7 , the UE 702 can receive, at 726, a ML / NN model and feature response for a ML / NN process from the base station 704, which can be based on an update to the ML / NN model and feature via a request and response (e.g., received and transmitted at 722 and 724) to / from the OAM core network 706.

[0106] At 1008, the UE can determine whether to use the at least one of a model or a feature received from the base station. For example, with reference to Figure 7UE 702 can determine whether to use the ML / NN model and features indicated via the ML / NN model and features response from base station 704 at 726 for the ML / NN process. For example, UE 702 can determine to use different ML / NN model and features than the ML / NN model and features indicated via the ML / NN model and features response from base station 704.

[0107] Figure 11 FIG. 1100 is a diagram 1100 illustrating an example of a hardware implementation for the apparatus 1102. The apparatus 1102 is a BS and includes a baseband unit 1104. The baseband unit 1104 can communicate with the UE 104 through a cellular RF transceiver 1122. The baseband unit 1104 can include a computer- readable medium / memory. The baseband unit 1104 is responsible for general processing, including the execution of software stored on the computer-readable medium / memory. The software, when executed by the baseband unit 1104, causes the baseband unit 1104 to perform the various functions described supra. The computer-readable medium / memory can also be used for storing data that is manipulated by the baseband unit 1104 when executing software. The baseband unit 1104 further includes a reception component 1130, a communication manager 1132, and a transmission component 1134. The communication manager 1132 includes the one or more illustrated components. The components of the communication manager 1132 can be stored in the computer-readable medium / memory and / or configured as hardware within the baseband unit 1104. The baseband unit 1104 can be a component of the BS 310 and can include at least one of the TX processor 316, the RX processor 370, and the controller / processor 375, and / or the memory 376.

[0108] For example, as described in conjunction with 810 and 814, receiving component 1130 is configured to: receive an updated model from a server based on a model update request; and receive a response from an interface of the core network to a request for at least one of the models or features, the response indicating a recent update to at least one of the models or features. Communication manager 1132 includes query component 1140, which, for example, as described in conjunction with 804, is configured to: query at least one in a database for a model's URI or query at least one in a data lake or data pool for a feature. Communication manager 1132 also includes determining component 1142, which, for example, as described in conjunction with 806 and 812, is configured to: determine the status of at least one of the model's URI or feature based on the query; and determine, via an interface of the core network, at least one of the models or features based on the request, the at least one of the models or features corresponding to a recent update to at least one of the models or features. For example, as described in conjunction with 802, 808, 816, and 818, the transmitting component 1134 is configured to: transmit a request to an interface of the core network for at least one of a model or feature associated with at least one of the ML or NN processes; transmit a model update request to a server based on the state of the model's URI; transmit at least one of the model or feature to at least one of the first base station or AMF of the core network based on the latest update of at least one of the model or feature; and transmit at least one of the model or feature to a second base station based on a handover from the UE to the second base station.

[0109] The device may include the ability to perform the above-described actions. Figure 8 The flowchart shows the algorithm's additional components in each box. Therefore, the above can be performed by these components. Figure 8 Each box in the flowchart, and the apparatus may include one or more of those components. A component may be one or more hardware components specifically configured to perform the process / algorithm, implemented by a processor configured to perform the process / algorithm, stored in a computer-readable medium for implementation by a processor, or some combination thereof.

[0110] In one configuration, the apparatus 1102 (and in particular the baseband unit 1104) includes means for transmitting, to an interface of a core network, a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure; means for determining, via the interface of the core network, based on the request, at least one of the model or the feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature; and means for receiving, from the interface of the core network, a response to the request for the at least one of the model or the feature, the response to the request indicating the latest update to the at least one of the model or the feature. The apparatus 1102 also includes means for querying at least one of a database for a URI of the model or at least one of a data lake or a data pool for the feature; and means for determining, based on the querying, a status of at least one of the URI of the model or the feature. The apparatus 1102 also includes means for transmitting, based on the status of the URI of the model, a model update request to a server; and means for receiving, based on the model update request, an updated model from the server. The apparatus 1102 also includes means for transmitting, based on the latest update to the at least one of the model or the feature, at least one of the model or the feature to at least one of a first base station or an AMF of the core network. The apparatus 1102 also includes means for transmitting, based on a handover of the UE to a second base station, at least one of the model or the feature to the second base station. The aforementioned means can be one or more of the aforementioned components of the apparatus 1102 configured to perform the functions recited by the aforementioned means. As described above, the apparatus 1102 can include the TX processor 316, the RX processor 370, and the controller / processor 375. As such, in one configuration, the aforementioned means can be the TX processor 316, the RX processor 370, and the controller / processor 375 configured to perform the functions recited by the aforementioned means.

[0111] Figure 12Figure 1200 illustrates an example of a hardware implementation for device 1202. Device 1202 is a BS and includes a baseband unit 1204. Baseband unit 1204 can communicate with UE 104 via cellular RF transceiver 1222. Baseband unit 1204 may include computer-readable medium / memory. Baseband unit 1204 is responsible for general processing, including executing software stored on computer-readable medium / memory. When executed by baseband unit 1204, the software causes baseband unit 1204 to perform the various functions described above. Computer-readable medium / memory can also be used to store data manipulated by baseband unit 1204 during software execution. Baseband unit 1204 also includes a receiving component 1230, a communication manager 1232, and a transmitting component 1234. Communication manager 1232 includes one or more components shown. Components within communication manager 1232 may be stored in computer-readable medium / memory and / or configured as hardware within baseband unit 1204. The baseband unit 1204 may be a component of the BS 310 and may include at least one of the TX processor 316, the RX processor 370 and the controller / processor 375 and / or the memory 376.

[0112] For example, as described in conjunction with 906 and 910, receiving component 1230 is configured to: receive at least one of a model or feature from the core network based on a request, the at least one of the model or feature corresponding to a latest update of the at least one of the model or feature; and receive a download request for the at least one of the model or feature from the UE based on an RRC message. Communication manager 1232 includes determining component 1240, which, for example, as described in conjunction with 902, is configured to: determine to initiate a request for at least one of a model or feature associated with at least one of the ML or NN processes. For example, as described in conjunction with 904, 908, and 912, transmitting component 1234 is configured to: transmit the request to the core network; transmit an RRC message to the UE indicating the at least one of the model or feature; and transmit the at least one of the model or feature to the UE based on the download request.

[0113] The device may include the ability to perform the above-described actions. Figure 9 The flowchart shows the algorithm's additional components in each box. Therefore, the above can be performed by these components. Figure 9 Each box in the flowchart, and the apparatus may include one or more of those components. A component may be one or more hardware components specifically configured to perform the process / algorithm, implemented by a processor configured to perform the process / algorithm, stored in a computer-readable medium for implementation by a processor, or some combination thereof.

[0114] In one configuration, the apparatus 1202 (and in particular the baseband unit 1204) includes means for determining to initiate a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure, means for transmitting the request to a core network, and means for receiving at least one of a model or a feature from the core network based on the request, the at least one of a model or a feature corresponding to a latest update to the at least one of a model or a feature. The apparatus 1202 further includes means for transmitting an RRC message to a UE indicating the at least one of a model or a feature. The apparatus 1202 further includes means for receiving a download request for the at least one of a model or a feature from the UE based on the RRC message, and means for transmitting the at least one of a model or a feature to the UE based on the download request. The aforementioned means can be one or more of the aforementioned components of the apparatus 1202 configured to perform the functions recited by the aforementioned means. As described above, the apparatus 1202 can include the TX processor 316, the RX processor 370, and the controller / processor 375. As such, in one configuration, the aforementioned means can be the TX processor 316, the RX processor 370, and the controller / processor 375 configured to perform the functions recited by the aforementioned means.

[0115] Figure 13is a diagram 1300 showing an example of a hardware implementation for the apparatus 1302. The apparatus 1302 is a UE and includes a cellular baseband processor 1304 (also referred to as a modem) coupled with a cellular RF transceiver 1322 and one or more subscriber identity modules (SIM) cards 1320, an application processor 1306 coupled with a secure digital (SD) card 1308 and a screen 1310, a Bluetooth module 1312, a wireless local area network (WLAN) module 1314, a Global Positioning System (GPS) module 1316, and a power supply 1318. The cellular baseband processor 1304 communicates with the UE 104 and / or BS 102 / 180 by the cellular RF transceiver 1322. The cellular baseband processor 1304 can include a computer-readable medium / memory. The computer-readable medium / memory can be non-transitory. The cellular baseband processor 1304 is responsible for general processing, including the execution of software stored on the computer-readable medium / memory. The software, when executed by the cellular baseband processor 1304, causes the cellular baseband processor 1304 to perform the various functions described supra. The computer-readable medium / memory can also be used for storing data that is manipulated by the cellular baseband processor 1304 when executing software. The cellular baseband processor 1304 further includes a reception component 1330, a communication manager 1332, and a transmission component 1334. The communication manager 1332 includes the one or more illustrated components. The components of the communication manager 1332 can be stored in the computer-readable medium / memory and / or configured as hardware within the cellular baseband processor 1304. The cellular baseband processor 1304 can be a component of the UE 350 and can include at least one of the TX processor 368, the RX processor 356, and the controller / processor 359, and / or the memory 360. In one configuration, the device 1302 can be a modem chip and include only the baseband processor 1304, and in another configuration, the apparatus 1302 can be an entire UE (e.g., see 350 of FIG. 13A) and include the aforementioned additional modules of the apparatus 1302. Figure 3

[0116] ​For example, as described in conjunction with 1006, receiving component 1330 is configured to receive at least one of a model or feature from a base station based on a request, the at least one of the model or feature corresponding to a recent update of the at least one of the model or feature. Communication manager 1332 includes determining component 1340, which, for example, as described in conjunction with 1002 and 1008, is configured to: determine whether to initiate a request for at least one of the model or feature, wherein the request is sent to the base station based on the determination of initiating the request; and determine whether to use the at least one of the model or feature received from the base station. For example, as described in conjunction with 1004, sending component 1334 is configured to: send a request to the base station for at least one of the model or feature, the request being associated with at least one of the ML or NN processes.

[0117] The device may include the ability to perform the above-described actions. Figure 10 The flowchart shows the algorithm's additional components in each box. Therefore, the above can be performed by these components. Figure 10 Each box in the flowchart, and the apparatus may include one or more of those components. A component may be one or more hardware components specifically configured to perform the process / algorithm, implemented by a processor configured to perform the process / algorithm, stored in a computer-readable medium for implementation by a processor, or some combination thereof.

[0118] In one configuration, apparatus 1302 (and particularly cellular baseband processor 1304) includes: a unit for sending a request to a base station for at least one of a model or feature associated with at least one of the ML or NN processes; and a unit for receiving, based on the request, at least one of the model or feature from the base station, the at least one of the model or feature corresponding to a recent update of the at least one of the model or feature. Apparatus 1302 further includes: a unit for determining whether to initiate a request for at least one of the model or feature, wherein the request is sent to the base station based on the determination of initiating the request. Apparatus 1302 further includes: a unit for determining whether to use at least one of the model or feature received from the base station. The aforementioned units may be one or more components of apparatus 1302 configured to perform the functions described by the aforementioned units. As described above, apparatus 1302 may include TX processor 368, RX processor 356, and controller / processor 359. Therefore, in one configuration, the aforementioned units may be TX processor 368, RX processor 356, and controller / processor 359 configured to perform the functions described by the aforementioned units.

[0119] It is to be understood that the specific order or hierarchy of blocks in the disclosed process / flow diagrams is an illustration of exemplary processes. Based upon design preferences, it is understood that the specific order or hierarchy of blocks in the processes / flow diagrams can be re-arranged. Further, some blocks can be combined or omitted. The accompanying method claims give expression to the elements of the various blocks in the exemplary order presented but are not meant to be limited to the specific order or hierarchy presented.

[0120] The foregoing description is provided to enable any person skilled in the art to implement the various aspects described herein. Various modifications to these aspects will be apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects. Therefore, the claims are not intended to be limited to the aspects shown herein, but are given the full scope consistent with the textual claims, wherein reference to the singular form of an element, unless expressly stated otherwise, is not intended to mean “one and only one,” but rather “one or more.” Terms such as “if,” “when,” and “while,” should be interpreted as “under the condition of,” rather than implying a direct temporal relationship or reaction. That is, these phrases (e.g., “when”) do not imply an immediate action in response to the occurrence of an action or during the occurrence of such action, but merely that the action will occur if the condition is met, without requiring a specific or immediate temporal constraint on the occurrence of the action. The term “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any aspect described herein as “exemplary” is not necessarily to be construed as preferred over or superior to other aspects. Unless expressly stated otherwise, the term “some” refers to one or more. Combinations such as "at least one of A, B, or C", "one or more of A, B, or C", "at least one of A, B, and C", "one or more of A, B, and C", and "A, B, C, or any combination thereof" include any combination of A, B, and / or C, and may include multiples of A, multiples of B, or multiples of C. Specifically, combinations such as "at least one of A, B, or C", "one or more of A, B, or C", "at least one of A, B, and C", "one or more of A, B, and C", and "A, B, C, or any combination thereof" may be only A, only B, only C, A and B, A and C, B and C, or A and B and C, wherein any such combination may contain one or more members of A, B, or C. All structural and functional equivalents of the elements throughout the various aspects described in this disclosure are expressly incorporated herein by reference and intended to be included by the claims, and such structural and functional equivalents are known to or will be known later to those skilled in the art. Furthermore, nothing disclosed herein is intended to be offered to the public, whether or not such disclosure is explicitly stated in the claims. Terms such as “module,” “mechanism,” “element,” “device,” etc., are not necessarily substitutes for the term “unit.” Therefore, no claim element should be interpreted as a unit plus a function unless the element is explicitly stated using the phrase “unit for…”.

[0121] The following aspects are illustrative only and may be combined with, but not limited to, other aspects or teachings described herein.

[0122] Aspect 1 is a method of wireless communication of a core network, featuring: transmitting, to an interface of the core network, a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure; determining, via the interface of the core network, the at least one of the model or the feature based on the request, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature; and receiving, from the interface of the core network, a response to the request for the at least one of the model or the feature, the response to the request indicating the latest update to the at least one of the model or the feature.

[0123] Aspect 2 can be in combination with Aspect 1, and features that the interface of the core network is a MDAC interfacing with at least one of a database, a data lake, a data pool, a server, or a base station.

[0124] Aspect 3 can be in combination with any of Aspects 1-2, and features further that querying at least one of the database for a URI of the model or at least one of the data lake or the data pool for the feature; and determining a status of at least one of the URI of the model or the feature based on the querying.

[0125] Aspect 4 can be in combination with any of Aspects 1-3, and features further that transmitting, to the server, a model update request based on the status of the URI of the model; and receiving, from the server, an updated model based on the model update request.

[0126] Aspect 5 can be in combination with any of Aspects 1-4, and features that the updated model is registered in the database based on an authentication procedure for the updated model.

[0127] Aspect 6 can be in combination with any of Aspects 1-5, and features further that transmitting, to at least one of a first base station or an AMF of the core network, the at least one of the model or the feature based on the latest update to the at least one of the model or the feature.

[0128] Aspect 7 can be in combination with any of Aspects 1-6, and features that the transmission to the at least one of the first base station or the AMF of the core network is configured to initiate the at least one of the ML procedure or the NN procedure, the at least one of the ML procedure or the NN procedure being at least one of a signaling procedure or a management procedure.

[0129] Aspect 8 can be combined with any of aspects 1-7 and characterized further by transmitting the model or the at least one of the features to a second base station based on a handover of the UE to the second base station.

[0130] Aspect 9 can be combined with any of aspects 1-8 and characterized by the transmission of the request for the at least one of the model or the features being initiated based on an indication from at least one of the core network, a base station, or a UE.

[0131] Aspect 10 can be combined with any of aspects 1-9 and characterized by the core network corresponding to an OAM entity.

[0132] Aspect 11 is a method of wireless communication of a base station, characterized by determining to initiate a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure; transmitting the request to a core network; and receiving, from the core network based on the request, the at least one of the model or the feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature.

[0133] Aspect 12 can be combined with aspect 11 and characterized by the determination to initiate the request being based on at least one of an initialization of the ML procedure, an initialization of the NN procedure, or a degradation in performance of the base station.

[0134] Aspect 13 can be combined with any of aspects 11-12 and characterized further by transmitting, to a UE, an RRC message indicating the at least one of the model or the feature.

[0135] Aspect 14 can be combined with any of aspects 11-13 and characterized further by receiving, from the UE based on the RRC message, a download request for the at least one of the model or the feature; and transmitting, to the UE based on the download request, the at least one of the model or the feature.

[0136] Aspect 15 can be combined with any of aspects 11-14 and characterized by the RRC message corresponding to at least one of an RRC setup message, an RRC resume message, or an RRC reestablishment message associated with at least one of a model ID, a procedure ID, or a feature metric.

[0137] Aspect 16 can be combined with any of aspects 11-15 and characterized by the RRC message corresponding to an RRC release message associated with at least one of a cell reselection procedure, logged measurements, predicted measurements, or an idle mode procedure.

[0138] Aspect 17 is a method of wireless communication of a UE, featuring: transmitting, to the base station, a request for at least one of a model or a feature associated with at least one of a ML procedure or a NN procedure; and receiving, from the base station based on the request, the at least one of the model or the feature, the at least one of the model or the feature corresponding to a latest update to the at least one of the model or the feature.

[0139] Aspect 18 can be combined with aspect 17 and feature further that determining the request for the at least one of the model or the feature is based on a determination to initiate the request.

[0140] Aspect 19 can be combined with any of aspects 17-18 and feature that the determination to initiate the request is based on a switch from the at least one of the ML procedure or the NN procedure to at least one of a second ML procedure or a second NN procedure.

[0141] Aspect 20 can be combined with any of aspects 17-19 and feature that the determination to initiate the request is based on an initialization of the at least one of the ML procedure or the NN procedure.

[0142] Aspect 21 can be combined with any of aspects 17-20 and feature that the determination to initiate the request is based on a performance degradation of the UE.

[0143] Aspect 22 can be combined with any of aspects 17-21 and feature further that determining whether to use the at least one of the model or the feature received from the base station.

[0144] Aspect 23 is an apparatus for wireless communication, comprising at least one processor coupled to a memory and configured to implement a method as in any of aspects 1-22.

[0145] Aspect 24 is an apparatus for wireless communication, comprising means for implementing a method as in any of aspects 1-22.

[0146] Aspect 25 is a non-transitory computer-readable medium storing computer-executable code which, when executed by at least one processor, causes the at least one processor to implement a method as in any of aspects 1-22.

Claims

1. A method for wireless communication of a core network, comprising: The core network receives a request for at least one of the models or features corresponding to the input parameters of the model, wherein the model or feature is associated with at least one of a machine learning (ML) process or a neural network (NN) process, and wherein the MDAC is a centralized interface between at least one of a database, a data lake or a data pool, a server and a base station. At the MDAC in the core network, perform at least one of the following: query the database for the Uniform Resource Indicator (URI) of the model, or query at least one of the data lake or the data pool for the feature, wherein the database is part of the core network; At the MDAC in the core network, the state of at least one of the URIs or features of the model is determined based on the query; At the MDAC in the core network, a model update request is sent to the server based on the state of the URI of the model, wherein the server is not part of the core network; At the MDAC in the core network, the updated model is received from the server based on the model update request; The updated model received from the server is sent to the database at the MDAC in the core network. At the MDAC in the core network, a model selection response is received from the database based on the updated model; At the MDAC in the core network, at least one of the model or the feature is determined based on the request and the model selection response received from the database; and At the MDAC in the core network, a response to the request for at least one of the models or features is sent, the response to the request indicating the model selection response received from the database.

2. The method according to claim 1, wherein, The updated model is registered in the database based on the authentication process used for the updated model.

3. The method according to claim 1, further comprising: The model or at least one of the features is sent to at least one of the first base station or the Access and Mobility Management Function (AMF) of the core network based on the latest update of the model or the feature.

4. The method according to claim 3, wherein, The transmission to the first base station of the core network or at least one of the AMFs is configured to initiate at least one of the ML process or the NN process, wherein the ML process or the at least one of the NN process is at least one of the signaling process or the management process.

5. The method according to claim 3, further comprising: At least one of the models or features is sent to the second base station based on the handover from the user equipment (UE) to the second base station.

6. The method according to claim 1, wherein, The transmission of the request for at least one of the models or features is initiated based on an instruction from at least one of the core network, base station, or user equipment (UE).

7. The method according to claim 1, wherein, The core network corresponds to the Operations and Administration (OAM) entity.

8. A method for wireless communication of a base station, comprising: Determine to initiate a request for at least one of the models or features associated with at least one of the machine learning (ML) process or neural network (NN) process; Send the request to the core network; and Based on the request, at least one of the model or the feature is received from the core network, the at least one of the model or the feature corresponding to the latest update of the model or the at least one of the features; The method further includes sending a radio resource control (RRC) message to a user equipment (UE) indicating at least one of the models or features.

9. The method according to claim 8, wherein, The determination of initiating the request is based on at least one of the initialization of the ML process, the initialization of the NN process, or the base station performance degradation.

10. The method of claim 8, further comprising: Based on the RRC message, receive a download request from the UE for at least one of the models or features; as well as Based on the download request, send at least one of the model or the feature to the UE.

11. The method according to claim 8, wherein, The RRC message corresponds to at least one of the following: an RRC establishment message, an RRC recovery message, or an RRC re-establishment message, which is associated with at least one of the following: a model identifier (ID), a process ID, or a feature metric.

12. The method according to claim 8, wherein, The RRC message corresponds to an RRC release message associated with at least one of the following: a cell reselection process, a recorded measurement, a predicted measurement, or an idle mode process.

13. A method for wireless communication of a user equipment (UE), comprising: Sending a request to the base station for at least one of the models or features associated with at least one of the machine learning (ML) process or neural network (NN) process; and Based on the request, at least one of the model or the feature is received from the base station, the at least one of the model or the feature corresponding to the latest update of the model or the at least one feature; The method further includes receiving a radio resource control (RRC) message from a base station (BS) indicating at least one of the models or features.

14. The method of claim 13, further comprising: Determine to initiate a request for at least one of the models or features, wherein the request is sent to the base station based on the determination to initiate the request.

15. The method according to claim 14, wherein, The determination of initiating the request is based on switching from at least one of the ML process or the NN process to at least one of the second ML process or the second NN process.

16. The method of claim 14, wherein, The determination of initiating the request is based on the initialization of at least one of the ML process or the NN process.

17. The method of claim 14, wherein, The determination that the request was initiated is based on the performance degradation of the UE.

18. The method of claim 13, further comprising: Determine whether to use at least one of the models or features received from the base station.

19. An apparatus for wireless communication in a core network, comprising: Memory; as well as At least one processor, coupled to the memory, is configured to: The core network receives a request for at least one of the models or features corresponding to the input parameters of the model, wherein the model or feature is associated with at least one of a machine learning (ML) process or a neural network (NN) process, and wherein the MDAC is a centralized interface between at least one of a database, a data lake or a data pool, a server and a base station. At the MDAC in the core network, perform at least one of the following: query the database for the Uniform Resource Indicator (URI) of the model, or query at least one of the data lake or the data pool for the feature, wherein the database is part of the core network; At the MDAC in the core network, the state of at least one of the URIs or features of the model is determined based on the query; At the MDAC in the core network, a model update request is sent to the server based on the state of the URI of the model, wherein the server is not part of the core network; At the MDAC in the core network, the updated model is received from the server based on the model update request; The updated model received from the server is sent to the database at the MDAC in the core network. At the MDAC in the core network, a model selection response is received from the database based on the updated model; At the MDAC in the core network, at least one of the model or the feature is determined based on the request and the model selection response received from the database; and At the MDAC in the core network, a response to the request for at least one of the models or features is sent, the response to the request indicating the model selection response received from the database.

20. The apparatus according to claim 19, wherein, The updated model is registered in the database based on the authentication process used for the updated model.

21. The apparatus according to claim 19, wherein, The at least one processor is further configured to send the at least one of the model or the at least one of the features to at least one of the first base station or the Access and Mobility Management Function (AMF) of the core network based on the latest update of the at least one of the model or the features.

22. The apparatus according to claim 21, wherein, The transmission to the first base station of the core network or at least one of the AMFs is configured to initiate at least one of the ML process or the NN process, wherein the ML process or the at least one of the NN process is at least one of the signaling process or the management process.

23. The apparatus according to claim 21, wherein, The at least one processor is further configured to send at least one of the models or features to the second base station based on a handover from a user equipment (UE) to the second base station.

24. The apparatus according to claim 19, wherein, The transmission of the request for at least one of the models or features is initiated based on an instruction from at least one of the core network, base station, or user equipment (UE).

25. The apparatus according to claim 19, wherein, The core network corresponds to the Operations and Administration (OAM) entity.

26. An apparatus for wireless communication for a base station, comprising: Memory; as well as At least one processor, coupled to the memory, is configured to: Determine to initiate a request for at least one of the models or features associated with at least one of the machine learning (ML) process or neural network (NN) process; Send the request to the core network; and Based on the request, at least one of the model or the feature is received from the core network, the at least one of the model or the feature corresponding to the latest update of the model or the at least one of the features; The at least one processor is further configured to send a radio resource control (RRC) message to a user equipment (UE) indicating at least one of the models or features.

27. The apparatus according to claim 26, wherein, The determination of initiating the request is based on at least one of the initialization of the ML process, the initialization of the NN process, or the base station performance degradation.

28. The apparatus according to claim 26, wherein, The at least one processor is further configured to: Based on the RRC message, receive from the UE a download request for at least one of the models or features; and Based on the download request, send at least one of the model or the feature to the UE.

29. The apparatus according to claim 26, wherein, The RRC message corresponds to at least one of the following: an RRC establishment message, an RRC recovery message, or an RRC re-establishment message, which is associated with at least one of the following: a model identifier (ID), a process ID, or a feature metric.

30. The apparatus according to claim 26, wherein, The RRC message corresponds to an RRC release message associated with at least one of the following: a cell reselection process, a recorded measurement, a predicted measurement, or an idle mode process.

31. An apparatus for wireless communication for a user equipment (UE), comprising: Memory; as well as At least one processor, coupled to the memory, is configured to: Sending a request to the base station for at least one of the models or features associated with at least one of the machine learning (ML) process or neural network (NN) process; and Based on the request, at least one of the model or the feature is received from the base station, the at least one of the model or the feature corresponding to the latest update of the model or the at least one feature; The at least one processor is further configured to receive from a base station (BS) a radio resource control (RRC) message indicating at least one of the models or features.

32. The apparatus according to claim 31, wherein, The at least one processor is further configured to: determine to initiate the request for at least one of the model or the feature, wherein the request is sent to the base station based on the determination to initiate the request.

33. The apparatus according to claim 32, wherein, The determination of initiating the request is based on switching from at least one of the ML process or the NN process to at least one of the second ML process or the second NN process.

34. The apparatus according to claim 32, wherein, The determination of initiating the request is based on the initialization of at least one of the ML process or the NN process.

35. The apparatus according to claim 32, wherein, The determination that the request was initiated is based on the performance degradation of the UE.

36. The apparatus according to claim 31, wherein, The at least one processor is further configured to: determine whether to use at least one of the models or features received from the base station.

37. An apparatus for wireless communication in a core network, comprising: A unit for receiving, at the Model and Data Access Coordinator (MDAC) of the core network, a request for at least one of the models or features corresponding to the input parameters of the model, wherein the model or feature is associated with at least one of a machine learning (ML) process or a neural network (NN) process, and wherein the MDAC is a centralized interface between at least one of a database, a data lake or a data pool, a server and a base station. A unit for performing at least one of the following at the MDAC in the core network: querying the database for a Uniform Resource Indicator (URI) of the model, or querying at least one of the data lake or the data pool for the feature, wherein the database is part of the core network; A unit for determining the state of at least one of the URIs or features of the model based on the query at the MDAC in the core network; A unit for sending a model update request to the server at the MDAC of the core network based on the state of the URI of the model, wherein the server is not part of the core network; A unit for receiving an updated model from the server at the MDAC in the core network based on the model update request; A unit for sending the updated model received from the server to the database at the MDAC in the core network; A unit for receiving a model selection response from the database based on the updated model at the MDAC in the core network; A unit for determining, at the MDAC in the core network, at the request and the model selection response received from the database, at least one of the model or the feature; and A unit for sending a response at the MDAC of the core network to a request for at least one of the models or features, wherein the response to the request indicates a model selection response received from the database.

38. An apparatus for wireless communication for a base station, comprising: A unit for determining a request to initiate a request for at least one of a model or feature associated with at least one of a machine learning (ML) process or a neural network (NN) process; A unit used to send the request to the core network; A unit for receiving at least one of the models or features from the core network based on the request, wherein the at least one of the models or features corresponds to the latest update of the at least one of the models or features; as well as A unit for sending a radio resource control (RRC) message to a user equipment (UE) indicating at least one of the models or features.

39. An apparatus for wireless communication for a user equipment (UE), comprising: A unit for sending a request to a base station for at least one of a model or feature associated with at least one of a machine learning (ML) process or a neural network (NN) process; A unit for receiving at least one of the model or the feature from the base station based on the request, wherein the at least one of the model or the feature corresponds to the latest update of the model or the at least one of the features; as well as A unit for receiving from a base station (BS) a radio resource control (RRC) message indicating at least one of the models or features.

40. A computer-readable medium storing computer-executable code, said code, when executed by at least one processor, causing said at least one processor to perform the following operations: At the Model and Data Access Coordinator (MDAC) of the core network, a request is received for at least one of the models or features corresponding to the input parameters of the model, wherein... The model or feature is associated with at least one of a machine learning (ML) process or a neural network (NN) process, and wherein the MDAC is a centralized interface between at least one of a database, a data lake or a data pool, a server and a base station. At the MDAC in the core network, perform at least one of the following: query the database for the Uniform Resource Indicator (URI) of the model, or query at least one of the data lake or the data pool for the feature, wherein the database is part of the core network; At the MDAC in the core network, the state of at least one of the URIs or features of the model is determined based on the query; At the MDAC in the core network, a model update request is sent to the server based on the state of the URI of the model, wherein the server is not part of the core network; At the MDAC in the core network, the updated model is received from the server based on the model update request; The updated model received from the server is sent to the database at the MDAC in the core network. At the MDAC in the core network, a model selection response is received from the database based on the updated model; At the MDAC in the core network, at least one of the model or the feature is determined based on the request and the model selection response received from the database; and At the MDAC in the core network, a response to the request for at least one of the models or features is sent, the response to the request indicating the model selection response received from the database.

41. A computer-readable medium storing computer-executable code, said code, when executed by at least one processor, causing said at least one processor to perform the following operations: Determine to initiate a request for at least one of the models or features associated with at least one of the machine learning (ML) process or neural network (NN) process; Send the request to the core network; and Based on the request, at least one of the model or the feature is received from the core network, the at least one of the model or the feature corresponding to the latest update of the model or the at least one of the features; in, When executed by the at least one processor, the code also causes the at least one processor to perform the following operation: send a radio resource control (RRC) message to the user equipment (UE) indicating at least one of the models or features.

42. A computer-readable medium storing computer-executable code, said code, when executed by at least one processor, causing said at least one processor to perform the following operations: Sending a request to the base station for at least one of the models or features associated with at least one of the machine learning (ML) process or neural network (NN) process; and Based on the request, at least one of the model or the feature is received from the base station, the at least one of the model or the feature corresponding to the latest update of the model or the at least one feature; in, When executed by the at least one processor, the code also causes the at least one processor to perform the following operation: receive a radio resource control (RRC) message from a base station (BS) indicating at least one of the models or features.

Images