Artificial intelligence model training for idle mode assistance

Abstract

One or more radio access network nodes may determine learning model configuration information for training a learning model corresponding to a user equipment in an idle mode. The node may broadcast a training configuration resource indication in the information block, the training configuration resource indication indicating resources available for broadcasting the learning model training configuration or indicating resources available for broadcasting the training results. While idle, the user equipment may decode the training configuration according to the training configuration resource indication and perform a training action indicated in the training configuration. The learning model may be trained based on the training action while the user device is idle. While idle, the user equipment may estimate radio parameters using a model trained while the user equipment is idle, and transmit the estimated radio parameters to the node for establishing a connection with the node.

Description

Cross Reference to Related Applications

[0001] This application claims priority to U.S. Non-Provisional Patent Application 18 / 303,375, filed April 19, 2023, entitled “ARTIFICIAL INTELLIGENCE MODEL TRAINING FOR IDLE MODE ASSISTANCE.” The entirety of this priority application is hereby incorporated by reference herein. BACKGROUND

[0002] The term “New Radio” (NR) in relation to the fifth generation of mobile wireless communication systems (“5G”) refers to technical aspects in a wireless Radio Access Network (“RAN”) for wireless, including several Quality-of-Service Classes (QoS), including Ultrareliable and Low Latency Communications (“URLLC”), Enhanced Mobile Broadband (“eMBB”), and Massive Machine Type Communication (“mMTC”). The URLLC QoS class is associated with strict latency requirements (e.g., low latency or low signal / message delay) and high reliability of radio performance, while traditional eMBB use cases can be associated with high-capacity wireless communications, which can allow for less strict latency requirements (e.g., higher latency than URLLC) and lower reliability of radio performance than URLLC. The performance requirements of mMTC can be lower than those of eMBB use cases. Some application scenarios involving mobile devices or mobile user equipment, such as smartphones, wireless tablets, smartwatches, and similar devices, can impose different loads or demands on a given RAN resource. SUMMARY

[0003] The following presents a simplified summary of the disclosed subject matter to provide a basic understanding of some embodiments of various embodiments. This summary is not an extensive overview of the various embodiments. It is not intended to identify key or critical elements of the various embodiments nor to delineate the scope of the various embodiments. Its sole purpose is to present some concepts of the disclosure in a simplified form as a prelude to the more detailed description that is presented later.

[0004] In an example embodiment, a method can include the following method, comprising: broadcasting, by a radio access network node comprising a processor, a learning model configuration information block message, the learning model configuration information block message comprising a training configuration resource indication, the training configuration resource indication indicating a training configuration resource, the training configuration resource being usable by the radio access network node for broadcasting a learning model training configuration, and broadcasting, by the radio access network node, the learning model training configuration according to the training configuration resource. The learning model training configuration can comprise a training action indication, the training action indication indicating a training action, the training action being executable by an idle user equipment.

[0005] In an embodiment, the learning model training configuration can comprise a training action indication, the training action indication indicating a training action, wherein the training action can correspond to a radio function learning model or a radio function learning model feature. Executing the training action can result in a determined radio function parameter value corresponding to the radio function learning model / feature. Examples of training actions can include: transmitting a timing advance preamble by the user equipment; receiving a reference signal identifier that can be used to determine an optimal beam relative to the user equipment; or transmitting a sounding reference signal group identifier. The learning model configuration information block message can further comprise a training result resource indication, the training result resource indication indicating a training result resource that can be used to transmit the determined radio function parameter value. The example method can further comprise: transmitting, by the radio access network node, the determined radio function parameter value to the idle user equipment according to the training result resource.

[0006] The training action can correspond to a radio function learning model, wherein executing the training action will result in a determined radio function parameter value corresponding to the radio function learning model. The example method can further comprise: receiving, by the radio access network node, a radio resource control signal message comprising the determined radio function parameter value from the idle user equipment. The example method can further comprise: establishing, by the radio access network node, a connection with the idle user equipment using the determined radio function parameter value received in the radio resource control signal message from the idle user equipment, whereby the idle user equipment becomes a connected user equipment. In an embodiment, the determined radio function parameter value can comprise a timing advance, the timing advance corresponding to a timing advance relative to the radio access network node corresponding to the idle user equipment. In an embodiment, the determined radio function parameter value can comprise an optimal serving beam indication, the optimal serving beam indication corresponding to a beam associated with the radio access network node, the beam having a signal strength corresponding to the radio access network node relative to the user equipment that is higher than other signal strengths associated with other beams than the beam.

[0007] In embodiments, the learning model training configuration can include a training resource indication indicating, to the idle user equipment, training resources available for performing the training action by the idle user equipment. In embodiments, the radio access network node can be a first radio access network node, and the example method can further include receiving, by the first radio access network node from a second radio access network node, a non-training resource indication, the second radio access network node being a neighboring radio access network node relative to the first radio access network node, the non-training resource indication indicating, to the first radio access network node, non-training resources to be reserved by the second radio access network node and available for the second radio access network node for non-training operations. The example method can further include scheduling, by the first radio access network node, the training resources to avoid overlap of the training resources corresponding to the first radio access network node with the non-training resources corresponding to the second radio access network node.

[0008] In embodiments of the example method, the radio access network node can be a first radio access network node, and the learning model training configuration can include a training action indication for indicating a training action to be performed by at least one idle user equipment relative to the first radio access network node to result in a first determined learning model parameter value. The example method can further include receiving, by the first radio access network node from the at least one idle user equipment, the first determined learning model parameter value. The example method can further include receiving, by the first radio access network node from a second radio access network node, a second determined learning model parameter value, the second radio access network node being a neighboring radio access network node relative to the first radio access network node, wherein the training action is performed by at least one of the at least one idle user equipment relative to the second radio access network node to result in the second determined learning model parameter value. The example method can further include determining, by the first radio access network node, a composite determined learning model parameter value based on the first determined learning model parameter value and based on the second determined learning model parameter value, and broadcasting, by the first radio access network node, the composite determined learning model parameter value to the at least one idle user equipment via a composite result information block message. The composite determined learning model parameter value can be based on information sent by the user equipment to the first radio access network node and sent by the user equipment or by another user equipment to the second radio access network node. The composite determined learning model parameter value can be based on information sent by the user equipment or by another user equipment to the first radio access network node and sent by another user equipment to the second radio access network node.

[0009] In an embodiment, the training action can correspond to a radio function learning model, and the composite determination learning model parameter value can be used by the at least one idle user equipment to train the radio function learning model to result in a trained learning model at the at least one idle user equipment. The example method can further include receiving, by the first radio access network node from the at least one idle user equipment, a connection request message including a performance indicator estimated by the at least one idle user equipment using the trained learning model to result in an estimated performance indicator. Based on the estimated performance indicator, the method can further include establishing a connection with the idle user equipment, whereby the idle user equipment becomes a connected user equipment with respect to the first radio access network node.

[0010] In one embodiment, the learning model configuration information block message can be a system information block message. In another embodiment, the learning model configuration information block message can be a master information block message. Accordingly, the idle mode user equipment can obtain the configuration resource for the result resource information from an information block that the user equipment is configured to decode in idle.

[0011] In another example embodiment, the first radio access network node can include a processor configured to receive, from a second radio access network node that is a neighboring radio access network node with respect to the first radio access network node, a non-training resource indication indicating to the first radio access network node non-training resources to be used by the second radio access network node to perform non-training operations, and schedule training resources to be used by the at least one idle mode user equipment to perform training actions with respect to the first radio access network node, whereby the training resources do not overlap the non-training resources. The processor can be further configured to broadcast a master information block message including a training configuration resource indication to indicate training configuration resources to be used by the first radio access network node to broadcast learning model training configurations. The processor can be further configured to broadcast the learning model training configurations in accordance with the training configuration resources.

[0012] The learning model training configurations can include a training resource indication indicating to the at least one idle mode user equipment training resources to be used by the at least one idle mode user equipment to perform the training actions. The master information block message can further include a training result resource indication to indicate training result resources to be used by the at least one idle mode user equipment to receive training results from the first radio access network node, the training results resulting from the at least one idle mode user equipment performing the training actions.

[0013] In an embodiment, the processor can be further configured to: determine a training result resulting from the at least one idle-mode user equipment performing the training action; and establish a connection with the at least one idle-mode user equipment based on the training result, whereby the at least one idle-mode user equipment becomes at least one connected-mode user equipment.

[0014] In another example embodiment, a non-transitory machine-readable medium can include executable instructions that, when executed by a processor of a first radio access network node, facilitate performance of operations comprising broadcasting a first information block message including a training configuration resource indication indicating a training configuration resource. The operations can further comprise broadcasting a learning model training configuration in accordance with the training configuration resource, wherein the learning model training configuration includes a training action indication for indicating a training action to be performed by a first idle user equipment of the at least one idle user equipment with respect to the first radio access network node to result in a first determined learning model parameter value. The operations can further comprise receiving the first determined learning model parameter value from the first idle user equipment of the at least one idle user equipment and receiving a second determined learning model parameter value from a second radio access network node that is a neighboring radio access network node with respect to the first radio access network node, wherein the training action is performed by at least a second idle user equipment of the at least one idle user equipment with respect to the second radio access network node to result in the second determined learning model parameter value. The operations can further comprise determining an updated learning model based on the first determined learning model parameter value and based on the second determined learning model parameter value and broadcasting the updated learning model to the first idle user equipment of the at least one idle user equipment via a second information block message. In an embodiment, the training action can have been performed by the first idle user equipment of the at least one idle user equipment with respect to the second radio access network node to result in the second determined learning model parameter value. In an embodiment, the operations can further comprise sending the updated learning model to the second radio access network node via a backhaul link.

[0015] In another example embodiment, a method can include receiving, by a user equipment comprising a processor, a learning model configuration information block message from a first radio access network node, the learning model configuration information block message comprising a training configuration resource indication for indicating a training configuration resource available for broadcasting, by the first radio access network node, a learning model training configuration. The method can further include receiving, by the user equipment, the learning model training configuration according to the training configuration resource. The method can further include decoding, by the user equipment, the learning model training configuration. The decoding of the learning model training configuration comprises blind decoding. The learning model training configuration can comprise a training action indication for indicating a training action to be performed by the user equipment.

[0016] In an embodiment, the learning model training configuration can comprise at least one timing advance preamble corresponding to a second radio access network node, the second radio access network node being a neighboring radio access network node relative to the first radio access network node, and the training action can comprise transmitting, by the user equipment, one of the at least one timing advance preamble to the second radio access network node. The method can further include transmitting, by the user equipment, one of the at least one timing advance preamble corresponding to the second radio access network node to the second radio access network node, wherein the one of the at least one timing advance preamble corresponding to the second radio access network node can be used by the second radio access network node to derive at least one updated timing advance learning model parameter corresponding to a timing advance learning model.

[0017] The learning model configuration information block message can comprise a training result resource indication for indicating a training result resource available for receiving, by the user equipment, the at least one updated timing advance learning model parameter. The method can further include receiving, by the user equipment, the at least one updated timing advance learning model parameter via the training result resource. Based on the at least one updated timing advance learning model parameter, the method can further include updating, by the user equipment, the timing advance learning model to derive an updated timing advance learning model.

[0018] In an embodiment, based on the updated timing advance learning model, the method can further include determining, by the user equipment, a timing advance corresponding to the user equipment relative to the first radio access network node. The method can further include transmitting, by the user equipment, a connection establishment request message to the first radio access network node, the connection establishment request message comprising the timing advance, and establishing, by the user equipment, a communication connection with the first radio access network node based on the connection establishment request message, whereby the user equipment is in a connected mode or becomes connected relative to the first radio access network node.

[0019] In an embodiment, based on the updated timing advance learning model, the method can further comprise determining, by the user equipment, a timing advance relative to the second radio access network node corresponding to the user equipment. The method can further comprise transmitting, by the user equipment, a connection establishment request message including the timing advance to the second radio access network node. Based on the connection establishment request message, the method can further comprise establishing, by the user equipment, a communication connection with the second radio access network node, whereby the user equipment is in a connected mode relative to the second radio access network node.

[0020] In an example embodiment, a user equipment can comprise a processor configured to receive a learning model configuration information block message from a radio access network node, wherein the learning model configuration information block message can comprise a training configuration resource indication indicating a training configuration resource available for receiving a learning model training configuration from the radio access network node. The processor can be further configured to receive the learning model training configuration according to the training configuration resource and decode the learning model training configuration, wherein the learning model training configuration comprises a training action indication indicating a training action to be performed by the user equipment. The processor can be further configured to perform the training action to obtain a training action result and transmit the training action result to the radio access network node. In an embodiment, the training action comprises generating a sounding reference signal, such that the training action result is the generated sounding reference signal, and wherein the generated sounding reference signal is transmitted to the radio access network node, which can be used by the radio access network node to train an uplink resource grant learning model to obtain a trained uplink resource grant learning model.

[0021] In an embodiment, the processor can be further configured to establish a communication connection with the radio access network node, wherein the communication connection comprises at least one uplink resource granted by the radio access network node based on the trained uplink resource grant learning model. The grant of the at least one uplink resource by the radio access network node can be based on excluding, by the user equipment, transmission of the sounding reference signal after transmitting the generated sounding reference signal by the user equipment. In other words, the at least one uplink resource can be granted by the radio access network node based on the prospectively transmitted sounding reference signal by the user equipment when the user equipment is idle.

[0022] In embodiments, the radio access network node can be a first radio access network node, the user equipment can perform the training action with respect to the first radio access network node to result in a first training action result, and the user equipment can transmit the first training action result to the first radio access network node. The processor can be further configured to perform the training action with respect to a second radio access network node to result in a second training action result, the second radio access network node being a neighboring radio access network node with respect to the first radio access network node. The processor can be further configured to transmit the second training action result to the second radio access network node.

[0023] In yet another example embodiment, a non-transitory machine-readable medium can include executable instructions that, when executed by a processor of a user equipment, facilitate performance of operations comprising: receiving, from a first radio access network node when the user equipment is idle, a learning model configuration information block message, the learning model configuration information block message including a training result resource, or including an indication of a training result resource from which the user equipment can receive a training result from the first radio access network node; and receiving, from the first radio access network node when the user equipment is idle, a learning model training configuration, the learning model training configuration including a training action indication to indicate a training action that can be performed by the user equipment with respect to at least the first radio access network node. The operations can further include performing, when the user equipment is idle, the training action with respect to the first radio access network node to result in a first training action result.

[0024] In embodiments, the operations can further include receiving the first training action result from the first radio access network node. The first training action result can be received in accordance with the training result resource indicated in the learning model configuration information block message.

[0025] In embodiments, the operations can further include performing, when the user equipment is idle, the training action with respect to a second radio access network node to result in a second training action result, the second radio access network node being a neighboring radio access network node with respect to the first radio access network node; and receiving the second training action result from the second radio access network node.

[0026] In embodiments, the first training action result can be used by the first radio access network node to update the learning model.

[0027] In an embodiment, the first training action result can be used by the user equipment to update the learning model to obtain an updated learning model for use by the user equipment. The learning model is a beam selection learning model, and wherein the updated learning model is an updated beam selection learning model.

[0028] In an embodiment, the operations can further include determining, when the user equipment is idle, a determined preferred serving beam corresponding to the first radio access network node using the updated beam selection learning model, the preferred serving beam to be used during connection setup with the first radio access network node. The operations can further include sending, when the user equipment is idle, a connection setup message to the first radio access network node, the connection setup message including a preferred serving beam indication indicating to the first radio access network node the determined preferred serving beam to be used to establish a connection with the first radio access network node. The operations can further include establishing the connection with the first radio access network node, wherein the connection includes the determined preferred serving beam, and wherein the establishment of the connection with the first radio access network node excludes beam sweeping to determine an optimal beam corresponding to the user equipment.

[0029] In an embodiment, the learning model can include a timing advance learning model. The updated learning model can include an updated timing advance learning model. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 A wireless communication system environment is shown.

[0031] Figure 2 An example environment of radio functions implemented in conjunction with corresponding learning models is shown.

[0032] Figure 3 An example environment of artificial intelligence machine learning model information exchanged between radio access network nodes and used to update artificial intelligence machine learning models at idle mode user equipment is shown.

[0033] Figure 4 An example radio resource control signal learning model configuration information block message is shown.

[0034] Figure 5A An example learning model training configuration to be used to train a learning model at idle mode user equipment is shown.

[0035] Figure 5B An example learning model training configuration to be used to train a timing advance learning model at idle mode user equipment is shown.

[0036] Figure 5CAn example learning model training configuration is shown that is to be used to train a channel state information learning model at an idle mode user equipment.

[0037] Figure 5D An example learning model training configuration is shown that is to be used to train an uplink beam selection model learning model at a radio access network node.

[0038] Figure 6 An example training result information block is shown that is to send results of training actions.

[0039] Figure 7 An example training result information block is shown that is to send timing advance results.

[0040] Figure 8 An example radio resource control signal message connection setup message is shown that includes a preferred beam determined by a user equipment.

[0041] Figure 9 A timing diagram of an example method for training a user equipment in idle mode is shown.

[0042] Figure 10 A flowchart of an example method for training a user equipment in idle mode is shown.

[0043] Figure 11 A block diagram of an example method is shown.

[0044] Figure 12 A block diagram of an example first radio access network node is shown.

[0045] Figure 13 A block diagram of an example non-transitory machine-readable medium is shown.

[0046] Figure 14 A block diagram of an example method is shown.

[0047] Figure 15 A block diagram of an example user equipment is shown.

[0048] Figure 16 A block diagram of an example non-transitory machine-readable medium is shown.

[0049] Figure 17 An example computer environment is shown.

[0050] Figure 18 A block diagram of an example wireless user equipment is shown. DETAILED DESCRIPTION

[0051] As a first matter, those skilled in the art will appreciate that the embodiments of the present application are susceptible to broad utility and application. Numerous methods, embodiments and adaptations thereof, and numerous variations, modifications and equivalent arrangements will be readily apparent to those skilled in the art from the teachings herein without departing from the scope or spirit of the present application.

[0052] Accordingly, although specific embodiments have been illustrated and described herein, it will be appreciated that the disclosure is intended to be illustrative of one or more concepts expressed within the various examples embodiments, and that the purpose of the disclosure is merely to provide a complete and enabling disclosure. The following disclosure is not intended or should not be construed to limit the present application in any way, nor to otherwise exclude any other embodiments, adaptations, variations, modifications and equivalent arrangements, the present application being limited only by the claims and equivalents thereof.

[0053] As used in this disclosure, the terms "component," "system" and the like are intended to refer to or comprise a computer-related entity or an entity related to an operational apparatus with one or more specific functions, wherein the entity can be either hardware, a combination of hardware and software, software, or an entity running software, as examples. As an example, a component can be, but is not limited to being, a process running on a processor, a processor, an object, an executable, a thread of execution, a computer-executable instruction, a program, and / or a computer. By way of illustration, both an application running on a server and the server can be a component. One or more components can reside within a process and / or thread of execution, and a component can be localized, co-resident, and / or distributed, among one computer and / or across multiple computers.

[0054] One or more components can reside within a process and / or thread of execution, and a component can be localized, co-resident, and / or distributed, among one computer and / or across multiple computers. Moreover, these components can execute from various computer readable media having various data structures stored thereon. The components can communicate via local and / or remote processes such as in accordance with a signal having one or more packets of data, e.g., data from components interacting with one another over a network, over a local network, and / or over the Internet. As another example, a component can be an apparatus with specific functionality provided by mechanical parts operated by electric or electronic circuitry, which is operated by a software application or firmware application executed by a processor, wherein the processor can be located either in the apparatus or outside of the apparatus, and executes at least a part of the software or firmware application. As yet another example, a component can be an apparatus that provides specific functionality through electronic components without mechanical parts, such as a software application or firmware application executed by a processor, wherein the processor can be located either in the apparatus or outside of the apparatus, and executes at least a part of the software or firmware application. While various components have been shown as separate components, it will be understood that multiple components can be implemented as a single component, or a single component can be implemented as multiple components, without departing from example embodiments.

[0055] The term "facilitating," as used herein in the context of a system, device, or component "facilitating" one or more actions or operations, is in keeping with the nature of complex computing environments in which multiple components and / or multiple devices can participate in some computing operation. Non-limiting examples of actions that can or can not involve multiple components and / or multiple devices include: sending or receiving data, establishing a connection between devices, determining intermediate results toward obtaining a result, etc. In this regard, a computing device or component can facilitate an operation by playing any role in the completion of the operation. Thus, when operations of a component are described herein, it is understood that where the operation is described as being facilitated by the component, the operation can optionally be completed in cooperation with one or more other computing devices or components, such as but not limited to sensors, antennas, audio and / or video output devices, other devices, etc.

[0056] Furthermore, various embodiments can be implemented as a method, apparatus, or article of manufacture using standard programming and / or engineering techniques to produce software, firmware, hardware, or any combination thereof to control a computer to implement the disclosed subject matter. The term "article of manufacture" as used herein is intended to encompass a computer program accessible from any computer-readable (or machine-readable) device or computer- readable (or machine-readable) storage / communication media. A computer-readable storage medium can include, but is not limited to, magnetic storage devices (e.g., hard disk, floppy disk, magnetic strips), optical disks (e.g., compact disk (CD), digital versatile disk (DVD)), smart cards, and flash memory devices (e.g., card, stick, key drive). Of course, those skilled in the art will recognize many modifications can be made to this configuration without departing from the scope or spirit of the various embodiments.

[0057] Artificial intelligence ("AI") and machine learning ("ML") models can facilitate performance and operational functionality and improvements in 5G implementations, such as, for example, network automation, optimization of signaling overhead, device energy saving, and traffic capacity maximization. Artificial intelligence machine learning model ("AI / ML model") functionality can be implemented and built in a variety of different forms and various vendor proprietary designs. A 5G radio access network node ("RAN") in a network to which user devices can attach or register can manage or control real-time AI / ML model performance at different user device apparatuses for various radio functions.

[0058] The network RAN can dynamically control the activation, deactivation, trigger model retraining (possibly specific to a radio function) or update the learning model according to the monitoring and analysis of defined real-time performance indicators corresponding to the learning model being implemented at the user equipment. It should be understood that even if the learning model can be implementing a specific radio function, the monitored or analyzed indicators can be learning model indicators and not necessarily radio function indicators (e.g. mathematical / statistical indicators, not necessarily radio function indicators such as e.g. signal strength).

[0059] Turning now to the drawings, Figure 1 An example of a wireless communications system 100 that supports blind decoding of PDCCH candidates or search spaces according to aspects of the present disclosure is illustrated. The wireless communications system 100 can include one or more base stations 105, one or more UEs 115, and a core network 130. In some examples, the wireless communications system 100 can be a Long Term Evolution (LTE) network, an LTE-Advanced (LTE-A) network, an LTE-A Pro network, or a New Radio (NR) network. In some examples, the wireless communications system 100 can support enhanced broadband communications, ultra-reliable (e.g., mission critical) communications, low latency communications, communications with low-cost and low-complexity devices, or any combination thereof. As illustrated, examples of UEs 115 can include a smartphone, a car or other transportation vehicle, or a drone or other flying vehicle. Another example of a UE can be a virtual reality device 117, such as smart glasses, a virtual reality headset, an augmented reality headset, and other similar devices, which can provide images, video, audio, haptics, taste, or smell to a wearer. A UE, such as the VR device 117, can transmit or receive wireless signals via a long-range wireless link 125 with a RAN base station 105, or the UE / VR device can receive or transmit wireless signals via a short-range wireless link 137, which can include a wireless link with a UE device 115 (such as a Bluetooth link, a Wi-Fi link, etc.). A UE, such as the device 117, can communicate via multiple wireless links simultaneously, such as communicating over a link 125 with a base station 105 and over a short-range wireless link. The VR device 117 can also communicate with a wireless UE via a cable or other wired connection. The RAN or components thereof can be implemented by one or more computer components, which can be described with reference to the computer system 600. Figure 12

[0060] Continuing with the description of Figure 1 ​The base stations 105 can be dispersed throughout the geographic area 100 and can be geographic distributed according to the coverage area 110 of these base stations 105. These base stations 105 can be heterogeneous in terms of coverage area 110, supported communications standards, the transmit power each base station 105 can employ, or the like. The base stations 105 can also be referred to as, and can include some or all, of an access point, a broadcast transmitter, a network node, or other suitable terminology.

[0061] The UEs 115 can be dispersed throughout the coverage areas 110 of the wireless communications system 100, and each UE 115 can be stationary, or mobile, or both at different times. The UEs 115 can be devices in different forms or have different capabilities. Figure 1 Some example UEs 115 are illustrated in FIG. 1. The UEs 115 described herein can be able to communicate with various types of devices, such as other UEs 115, base stations 105, or network equipment (e.g., core network nodes, relay devices, integrated access and backhaul (IAB) nodes, or other network equipment), as shown in FIG. 1. Figure 1

[0062] The base stations 105 can communicate with the core network 130, or with one another, or both. For example, the base stations 105 can interface with the core network 130 through one or more backhaul links 120 (e.g., via an SI, N2, N3, or other interface). The base stations 105 can communicate with one another over the backhaul links 120 (e.g., via an X2, Xn, or other interface) directly (e.g., direct

[0063] One or more of the base stations 105 described herein can include or can be referred to by a person of ordinary skill in the art as a base transceiver station, a radio base station, an access point, a radio transceiver, a NodeB, an eNodeB (eNB), a next-generation NodeB or giga-NodeB (either of which can be referred to as a gNB), a Home NodeB, a Home eNodeB, or other suitable terminology.

[0064] ​A UE 115 can include or can be referred to as a mobile device, a wireless device, a remote device, a handheld device, or a subscriber device, or some other suitable terminology, where the “device” can also be referred to as a unit, a station, a terminal, or a client, among other examples. A UE 115 can also include or can be referred to as a personal electronic device such as a cellular phone, a personal digital assistant (PDA), a tablet computer, a laptop computer, a personal computer, or a router, among other examples. In some examples, a UE 115 can include or be referred to as a wireless local loop (WLL) station, an Internet of Things (IoT) device, an Internet of Everything (IoE) device, or a machine type communications (MTC) device, among other examples, which can be implemented in various objects such as appliances, vehicles, or smart meters, among other examples.

[0065] A UE 115 can be able to communicate with various types of devices, such as other UEs 115 that can sometimes act as relays for the Figure 1 UE 115 and a base station 105, as well as network equipment including macro eNBs or gNBs, small cell eNBs or gNBs, or relay base stations, among other examples, as shown in FIG. 1.

[0066] The UEs 115 and the base stations 105 can wirelessly communicate with one another via one or more communication links 125 over one or more carriers. The term “carrier” can refer to a set of radio frequency spectrum resources with a defined physical layer structure

[0067] In some examples (e.g., in carrier aggregation configurations), a carrier can also have acquisition signaling or control signaling for other-carrier coordination operations. A carrier can be associated with a frequency channel (e.g., an evolved universal mobile telecommunication system terrestrial radio access (E-UTRA) absolute radio frequency channel number (EARFCN)) and can be positioned according to a channel raster for discovery by the UEs 115. Carriers can be operated in an independent mode, where initial acquisition and connection can be conducted by the UE 115 via the carrier; or can be operated in an non-independent mode, where connection is anchored using a different carrier (e.g., of same or different radio access technologies).

[0068] The communication links 125 shown in wireless communication system 100 can include uplink transmissions from a UE 115 to a base station 105, or downlink transmissions from a base station 105 to a UE 115. Carriers can carry downlink or uplink communications (e.g., in an FDD mode) or can be configured to carry downlink and uplink communications (e.g., in a TDD mode).

[0069] A carrier can be associated with a particular bandwidth of radio frequency spectrum, and in some examples the carrier bandwidth can be referred to as a “system bandwidth” of the carrier or wireless communications system 100. For example, the carrier bandwidth can be one of a number of predetermined bandwidths for carriers of a particular radio access technology (e.g., 1.4, 3, 5, 10, 15, 20, 40, or 80 megahertz (MHz)). Devices of wireless communications system 100 (e.g., base stations 105, UEs 115, or both) can have hardware configurations that support communications over a particular carrier bandwidth, or can be configurable to support communications over one of a set of carrier bandwidths. In some examples, wireless communications system 100 can include base stations 105 or UEs 115 that support communications via carriers associated with multiple carrier bandwidths. In some examples, each served UE 115 can be configured for operating over portions (e.g., a sub-band, a BWP) or all of a carrier bandwidth.

[0070] Signal waveforms transmitted over a carrier can be composed of multiple subcarriers (e.g., using multi-carrier modulation (MCM) techniques such as orthogonal frequency division multiplexing (OFDM) or discrete Fourier transform spread orthogonal frequency division multiplexing (DFT-S-OFDM). In a system employing MCM techniques, a resource element can consist of one symbol period (e.g., a duration of one modulation symbol) and one subcarrier, where the symbol period and subcarrier spacing are inversely related. The number of bits carried by each resource element can depend on the modulation scheme (e.g., the order of the modulation scheme, the coding rate of the modulation scheme, or both). Thus, the more resource elements that a UE 115 receives and the higher the order of the modulation scheme, the higher the data rates for the UE can be. A wireless communications resource can refer to a combination of a radio frequency spectrum resource, a time resource, (e.g., search space) or a spatial resource (e.g., spatial layers or beams), with multiple spatial layers further increasing the data rate or data integrity for communications with a UE 115.

[0071] One or more parameter sets, which can include a subcarrier spacing (Af) and a cyclic prefix, can be supported for a carrier. A carrier can be partitioned into one or more BWPs with the same or different numerologies. In some examples, a UE 115 can be configured with multiple BWPs. In some examples, a single BWP for a carrier can be active at a given time and communications for the UE 115 can be restricted to one or more active BWPs.

[0072] Time intervals for the base station 105 or the UE 115 can be expressed in multiples of a basic time unit, which may, for example, refer to a sampling period of Ts= 1 / (A s f max ·N f ) seconds, where Af max may represent the maximum supported subcarrier spacing, and N f may represent the maximum supported discrete Fourier transform (DFT) size. Time intervals of a communications resource can be organized as radio frames, each

[0073] Each frame can include a plurality of sequentially numbered subframes or slots, and each subframe or slot can have the same duration. In some examples, a frame can be partitioned (e.g., in the time domain) into subframes, and each subframe can be further partitioned into a number of slots. Alternatively, each frame can include a variable number of slots, and the number of slots can depend on the subcarrier spacing. Each slot can include a number of symbol periods (e.g., depending on the length of the cyclic prefix prepended to each symbol period). In some wireless communication systems 100, a slot can be further partitioned into a plurality of mini-slots containing one or more symbols. In addition to the cyclic prefix, each symbol period can contain one or more (e.g., N f The duration of a symbol period can depend on the subcarrier spacing or the operating band.

[0074] A subframe, a slot, a mini-slot, or a symbol can be the smallest scheduling unit (e.g., in the time domain) of the wireless communications system 100 and can be referred to as a transmission time interval (TTI). In some examples, the TTI duration (e.g., the number of symbol periods in a TTI) can be variable. Additionally or alternatively, the smallest scheduling unit of the wireless communications system 100 can be dynamically selected (e.g., in a burst of shortened TTIs (sTTIs)).

[0075] Physical channels can be multiplexed on a carrier according to various techniques. A physical control channel and a physical data channel can be multiplexed on a downlink carrier (e.g., using one or more of time division multiplexing (TDM) techniques, frequency division multiplexing (FDM) techniques, or hybrid TDM-FDM techniques). A control region (e.g., a control resource set (CORESET)) for a physical control channel can be defined by a number of symbol periods and can extend across the system bandwidth or a subset of the system bandwidth of the carrier. One or more control regions (e.g., CORESETs) can be configured for a set of UEs 115. For example, one or more of the UEs 115 can monitor or search control regions or spaces for control information according to one or more search space sets, and each search space set can include one or more control channel candidates, in one or more aggregation levels, arranged in a cascaded manner. An aggregation level for a control channel candidate can refer to a number of control channel resources (e.g., control channel elements (CCEs)) associated with encoded information for a control information format having a given payload size. Search space sets can include common search space sets for sending control information to multiple UEs 115 and UE-specific search space sets for sending control information to a specific UE 115. Other search spaces and configurations for monitoring and decoding them are disclosed herein, which are novel and unorthodox.

[0076] The base stations 105 can provide communication coverage for a respective geographic area via one or more cells (e.g., macro cells, small cells, hot spots, or other types of cells, or any combination thereof). The term “cell” can refer to a logical communication entity used for communication with a base station 105 (e.g., on a carrier or aggregation of carriers) and can be associated with a identifier, such as a physical cell identifier (PCID), a virtual cell identifier (VCID), or other identifiers. In some examples, a cell can also refer to a geographic coverage area of a logical communication entity or a portion of a geographic coverage area of a logical communication entity (e.g., a sector). Depending on various factors, such as capacity of the base station 105, the geographic coverage area for a cell can extend a relatively small area (e.g., a building, a subset of a building, or other example) to a relatively large area. For example, a cell can be or include a building, a subset of a building, or an outdoor space between or overlapping with geographic coverage areas 110, among other examples.

[0077] A macro cell can generally cover a relatively large geographic area (e.g., a radius of several kilometers) and can allow unrestricted access by UEs 115 with service agreements with a network provider supporting the macro cell. A small cell can be associated with a lower-powered base station 105 and can provide restricted or no access by UEs 115 with service agreements that do not include access to the small cell or other restrictions. A base station 105 can support one or multiple cells, and can communicate with UEs 115 on one or more carriers.

[0078] In some examples, a carrier can support multiple cells, and different cells can be configured according to different protocol types (e.g., MTC, narrowband IoT (NB-IoT), enhanced mobile broadband (eMBB)) to provide access to devices of different types.

[0079] In some examples, a base station 105 can be movable and therefore provide communication coverage for a moving geographic coverage area 110. In some examples, different geographic coverage areas 110 associated with different technologies can overlap, but the different geographic coverage areas 110 can be supported by the same base station 105. In other examples, the overlapping geographic coverage areas 110 associated with different technologies can be supported by different base stations 105. The wireless communications system 100 can include, for example, a heterogeneous network in which different types of base stations 105 provide coverage for various geographic coverage areas 110 using the same or different radio access technologies.

[0080] The wireless communications system 100 can support synchronous or asynchronous operation. For synchronous operation, the base stations 105 can have similar frame timings, and transmissions from different base stations 105 can be approximately aligned in time. For asynchronous operation, the base stations 105 can have different frame timings, and transmissions from different base stations 105 can not be aligned in time. The techniques described herein can be used for either synchronous or asynchronous operations.

[0081] Some UEs 115, such as MTC or IoT devices, can be low cost or low complexity devices, and can provide for automated communication between machines (e.g., via machine-to-machine (M2M) communication). M2M communication or MTC can refer to data communication technologies that allow devices to communicate with one another or a base station 105 without human intervention. In some examples, M2M communication or MTC can include communications from devices that integrate sensors or meters to measure or capture information and relay such information to a central server or application program, which can make use of the information or present the information to humans in interaction with the application program. Some UEs 115 can be designed to collect information or enable automated behavior of machines or other devices. Examples of applications for MTC devices include smart metering, inventory monitoring, water level monitoring, equipment monitoring, healthcare monitoring, wildlife monitoring, weather and geological event monitoring, fleet management and tracking, remote security sensing, physical access control, and transaction-based business charging.

[0082] Some UEs 115 can be configured to employ operating modes that reduce power consumption, such as a half-duplex communications (e.g., a mode of operation in which a UE 115 can support either transmission or reception, but not simultaneously). In some examples, half-duplex communications can be performed at a reduced peak rate. Other power conservation techniques for UEs 115 include entering a power saving deep sleep mode when not engaging in active communications, operating over a limited bandwidth (e.g., according to a narrowband

[0083] The wireless communications system 100 can be configured to support ultra-reliable communications or low-latency communications, or various combinations thereof. For example, the wireless communications system 100 can be configured to support ultra-reliable low-latency communications (URLLC) or mission critical communications. UEs 115 can be designed to support ultra-reliable, low-latency, or critical functions (e.g., mission critical functions). Ultra-reliable communications can include dedicated communications or group communications, and can be supported by one or more mission critical services such as mission critical push-to-talk (MCPTT), mission critical video (MCVideo), or mission critical data (MCData). Support for mission critical functions can include service priority scheduling, and mission critical services can be used for public safety or general commercial applications. The terms “ultra-reliable,” “low-latency,” “mission critical,” and “ultra-reliable low- latency” can be used interchangeably herein.

[0084] In some examples, UEs 115 can also communicate directly with other UEs 115 using a device-to-device (D2D) communication link 135 (e.g., using a peer-to-peer (P2P) or D2D protocol). The communication link 135 can include a sidelink communication link. One or more UEs 115 utilizing D2D communications can be within the geographic coverage area 110 of a base station 105. Other UEs 115 in such a group can be outside the geographic coverage area 110 of a base station 105, or be otherwise unable to receive transmissions from a base station 105. In some examples, groups of the UEs 115 communicating via D2D communications can utilize a one-to-many (1:M) system in which each UE 115 transmits to other UEs 115 in the group. In some examples, a base station 105 facilitates the scheduling of resources for D2D communications. In other cases, D2D communications are carried out between the UEs 115 without the involvement of a base station 105.

[0085] In some systems, the D2D communication link 135 can be an example of a communication channel, such as a sidelink communication channel, between vehicles (e.g., UEs 115). In some examples, vehicles can communicate using vehicle-to-everything (V2X) communications, car-to-car (V2V) communications, or a combination of these. Vehicles can signal information related to traffic conditions, signal scheduling, weather, safety, emergencies, or any other information related to V2X systems. In some examples, vehicles in a V2X system can communicate with roadside infrastructure, such as roadside units, or with the network via one or more RAN network nodes (e.g., base stations 105) using vehicle-to-network (V2N) communications, or both.

[0086] The core network 130 can provide user authentication, access authorization, tracking, Internet Protocol (IP) connectivity, and other access, routing, or mobility functions. The core network 130 can be an evolved packet core (EPC) or 5G core (5GC), which can include at least one control plane entity that manages access and mobility (e.g., a mobility management entity (MME), an access and mobility management function (AMF)) and at least one user plane entity that routes packets or interconnects with external networks (e.g., a serving gateway (S-GW), a packet data network (PDN) gateway (P-GW), or a user plane function (UPF)). The control plane entity can manage non-access stratum (NAS) functions such as mobility, authentication, and bearer management for UEs 115 served by base stations 105 associated with the core network 130. User IP packets can be transferred through the user plane entity, which can provide IP address allocation as well as other functions. The user plane entity can be connected to the IP services 150 of the Internet 155, an intranet, an IP multimedia subsystem (IMS), or a packet- switched streaming service.

[0087] Some of the network devices, such as a base station 105, can include subcomponents such as an access network entity 140, which can be an example of an access node controller (ANC). Each access network entity 140 can communicate with UEs 115 through one or more other access network transmission entities 145, which can be referred to as radio heads, smart radio heads, or transmission / reception points (TRPs). Each access network transmission entity 145 can include one or more antenna panels. In some configurations, various functions of each access network entity 140 or base station 105 can be distributed across various network devices (e.g., radio heads and ANCs) or consolidated into a single network device (e.g., a base station 105).

[0088] The wireless communications system 100 can operate using one or more frequency bands, typically in the range of 300 megahertz (MHz) to 300 gigahertz (GHz). Generally, the region from 300 MHz to 3 GHz is known as the ultra-high frequency (UHF) region or decimeter band, since the wavelengths range from approximately one decimeter to one meter in length. UHF waves can be blocked or redirected by buildings and environmental features, but the waves can penetrate structures sufficiently for a macro cell to provide service to the UEs 115 located indoors. The transmission of UHF waves can be associated with smaller antennas and shorter ranges (e.g., less than 100 kilometers) compared to transmission using the low frequency (LF) or very high frequency (VHF) parts of the spectrum below 300 MHz.

[0089] The wireless communications system 100 can also operate in a super high frequency (SHF) region, using frequency bands from 3 GHz to 30 GHz, also known as centimeter wave, or in an extremely high frequency (EHF) region, for example, from 30 GHz to 300 GHz, also known as millimeter wave. In some examples, the wireless communications system 100 can support millimeter wave (mmW) communications between the UEs 115 and the base stations 105, and EHF antennas of the respective devices can be smaller and more closely spaced than UHF antennas. In some examples, this can facilitate using antenna arrays within a device. However, the propagation of EHF transmissions can be subject to even greater atmospheric attenuation than SHF or UHF transmissions, and EHF transmissions can therefore have a shorter range than SHF or UHF transmissions. The techniques disclosed herein can be employed across transmissions that use one or more different frequency regions, and designated use of bands across these frequency regions can differ by country or regulatory organization.

[0090] The wireless communications system 100 can utilize both licensed and unlicensed radio frequency spectrum bands. For example, the wireless communications system 100 can employ License Assisted Access (LAA), LTE-Unlicensed (LTE-U) radio access technology, or NR technology in an unlicensed

[0091] The base stations 105 or UEs 115 can be equipped with multiple antennas, which can be used to employ techniques such as transmit diversity, receive diversity, multiple-input multiple-output (MIMO) communications, or beamforming. The antennas of a base station 105 or a UE 115 can be co-located within one or more antenna arrays or antenna panels, which can support MIMO operations or transmit or receive beamforming. For example, one or more base station antennas or antenna arrays can be co-located at an antenna assembly, such as an antenna tower. In some examples, antennas or antenna arrays associated with a base station 105 can be located in different geographic locations. A base station 105 can have antenna arrays with a number of rows and columns of antenna ports that can be used to support beamforming of communications signals to UEs 115. Similarly, a UE 115 can have one or more antenna arrays that can support various MIMO or beamforming operations. Additionally, or alternatively, antenna panels can support radio frequency beamforming of signals transmitted via the antenna ports.

[0092] The base stations 105 or the UEs 115 can use MIMO communications to exploit multipath signal propagation and increase the spectral efficiency. Such techniques can be referred to as spatial multiplexing. The multiple signals may, for example, be transmitted by the transmitting device via different antennas or different combinations of antennas. Similarly, the multiple signals can be received by the receiving device via different antennas or different combinations of antennas. Each of the multiple signals can be referred to as a spatial stream, and can carry bits associated with the same data stream (e.g., a same codeword) or different data streams (e.g., different codewords). Different spatial layers can be associated with different antenna ports used for channel measurement and reporting. MIMO techniques include single-user MIMO (SU-MIMO), where multiple spatial layers are transmitted to the same receiving device, and multiple-user MIMO (MU-MIMO), where multiple spatial layers are transmitted to multiple devices.

[0093] Beamforming, which can also be referred to as spatial filtering, directional transmission, or directional reception, is a signal processing technique that can be used at a transmitting device or a receiving device (e.g., a base station 105, a UE 115) to shape or steer a beam of energy in a specific direction. Beamforming can be achieved by combining the signals communicated by antennas of a base station 105 or a UE 115 in a way that causes the overall signal energy to be concentrated in a specific direction. This can be achieved by the transmitting device or receiving device applying amplitude and phase offsets to signals carried on each of its antennas. The amplitude and phase offsets used can depend on the desired shape and / or direction of the resulting beam.

[0094] The base stations 105 or the UEs 115 can use beamforming techniques to increase the spectral efficiency of communication between the base stations 105 and the UEs 115, particularly at high frequencies where a signal can be less likely to bounce around and cause interference between transmission and reception. Some signals can be transmitted by a base station 105 using multiple beams simultaneously. In this case, the base station 105 can transmit different layers of data by applying beamforming to different spatial layers or different antenna subarrays. The different spatial layers or different antenna subarrays can be transmitted in different directions. The base station 105 can transmit different beams of a single signal in different directions to different UEs 115 within the same transmission.

[0095] Some signals, such as data signals, can be transmitted by a base station 105 in a single beam direction (e.g., associated with a receiving device, such as a UE 115). In some examples, the beam direction associated with transmissions along a single beam direction can be determined based on a signal that is transmitted in one or more beam directions. For example, a UE 115 can receive one or more of the signals transmitted by the base station 105 in different directions and can report to the base station an indication of the signal that the UE 115 received with a highest signal quality, or other acceptable signal quality.

[0096] In some examples, transmissions by a device (e.g., base station 105 or UE 115) can be performed using multiple beam directions, and the device can use a combination of digital precoding or radio frequency beamforming to generate a combined beam for transmissions (e.g., from a base station 105 to a UE 115). The UE 115 can report feedback that indicates precoding weights for one or more beam directions, and the feedback can correspond to a number of beams configured across a system bandwidth or one or more sub-bands. The base station 105 can transmit reference signals (e.g., cell-specific reference signals (CRS), channel state information reference signals (CSI-RS)) that can be precoded or unprecoded. The UE 115 can provide feedback for beam selection, which can be a precoding matrix indicator (PMI) or codebook-based feedback (e.g., a multi-panel type codebook, a linear combination type codebook, a port selection type codebook). Although these techniques are described with reference to signals transmitted by a base station 105 in one or more directions, a UE 115 can employ similar techniques for transmitting signals multiple times in different directions (e.g., for identifying beam directions for subsequent transmissions or receptions by the UE 115) or for transmitting a signal in a single direction (e.g., for transmitting data to a receiving device).

[0097] A receiving device (e.g., a UE 115) can try multiple receive configurations (e.g., directional listening) when receiving various signals from base stations 105, such as synchronization signals, reference signals, beam selection signals, or other control signals. For example, a receiving device can try multiple receive directions by receiving via different antenna subarrays, by processing received signals according to different antenna subarrays, by receiving according to different receive beamforming weight sets applied individually to antenna array elements or antenna subarrays, by processing received signals according to different antenna subarrays, by moving antenna elements in different directions, or by using combinations of the above. Any of these can be indicated by a single or double arrow in the network diagram. For example, a receiving device can try multiple receive directions while receiving a signal. The signal can be associated with initial access to a base station, or other purposes. The receiving device can determine a receive beam that provides the highest post-processing SNR or other indication of signal quality. The receiving device can then perform tasks associated with the signal (e.g., receiving system information, accessing a paging channel, or other tasks) using the receive beam.

[0098] Wireless communications system 100 can be a packet-based network that operates according to a layered protocol stack. In the user plane, communications at the bearer or Packet Data Convergence Protocol (PDCP) layer can be IP -based. A Radio Link Control (RLC) layer can perform packet segmentation and reassembly to communicate over logical channels. A Medium Access Control (MAC) layer can perform priority handling and multiplexing of logical channels into transport channels. The MAC layer can also use error detection techniques, error correction techniques, or both, to support retransmissions at the MAC layer, e.g., using an appropriate hybrid automatic repeat request (HARQ) technique, to improve link efficiency. In the control plane, the Radio Resource Control (RRC) protocol layer can provide establishment, configuration, and maintenance of an RRC connection between a UE 115 and a base station 105 or core network 130, which can support radio bearers for the user plane data. At the physical layer, the transport channels can be mapped to physical channels.

[0099] The UEs 115 and the base stations 105 can support data retransmission to increase the likelihood that data is received successfully. Hybrid automatic repeat request (HARQ) feedback is one technique used to increase the likelihood that data is received correctly over the communication link 125. HARQ can include a combination of error detection (e.g., using a cyclic redundancy check (CRC)), forward error correction (FEC), and retransmission (e.g., automatic repeat request (ARQ)). HARQ can improve throughput at the MAC layer in poor radio conditions (e.g., low signal-to-noise conditions). In some examples, a device can support same-slot HARQ feedback, where the device can provide HARQ feedback in a specific time slot for data received in a previous time slot during the same frame. In other cases, the device can provide HARQ feedback in a subsequent time slot or according to other time intervals.

[0100] Conventional rule-based models can be implemented in user equipment to perform various radio frequency (“RF”) or signal processing functions, such as beamforming, channel estimation, demodulation, and decoding, and can be based on well-refined system models. As long as the model closely follows the actual behavior of the radio network system in which the user equipment is operating, satisfactory performance can be achieved. However, the performance of conventional models can not reach optimal performance. AI / ML-based models are generally superior to their conventional counterparts; unlike conventional rule-based models, AI / ML-based models can be based on data, rather than pre-determined rules of conventional models. Thus, the output or result of a conventional rule-based model can be considered “deterministic” in that the input is applied to static rules that result in a “determined” output, whereas the output or result of an AI / ML model can be considered probabilistic in that the learned model generally infers a likely output based on coefficients, factors, functions, or other variables that can have been derived based on previous inputs to the model.

[0101] Using AI / ML models can facilitate improved user equipment performance compared to using conventional rule-based models. Multiple AI / ML driven use cases can include AI / ML timing advance acquisition / prediction, AI / ML channel state information (“CSI”) acquisition / prediction, AI / ML radio positioning, and AI / ML beam management. While AI / ML based models trained using data from actual real-world operations can outperform conventional rule-based models, learning models can not be robust enough to provide less than ideal results in cases where changes can have occurred in the radio system / environment that were not experienced or “seen” during learning model training, and thus the learning model can extrapolate less than ideal outputs in cases that are “unknown” to the learning model. For example, such problematic cases can be caused by particular network / user equipment conditions or configurations, or by the architecture of the AI / ML learning model, or a combination thereof. Accordingly, it is desirable to implement a procedure that enables the network RAN to update the AI / ML learning model.

[0102] For AI / ML learning model implementations for radio functions at user equipment, the user equipment or gNB / RAN can predict a modulation and coding scheme (“MCS”), and for which a given number of channel state information reporting instances can be used. The modulation and coding scheme can be referred to as a format. The format or scheme can be associated with a quality of service. Channel conditions or interference conditions that were not present during training or the model can systematically result in less than optimal MCS selection, which in turn can result in violation of minimum device performance targets.

[0103] Turning now to Figure 2 the figure illustrates a system 200 that includes a RAN node 105 that communicates with a user equipment 115 via a wireless link 125. The UE 115 can perform various radio functions 205A through 205n, which can be facilitated by corresponding machine learning models 215A through 215n, respectively. During UE 115 wireless operation and communication with the RAN 105, the UE can transmit parameter indicator reports 220A through 220n, which can include one or more learning model parameter indicators (corresponding to 215A through 215n, respectively). The reports 220A through 220n can include one or more control action requests, e.g., requesting deactivation or retraining of one or more of the models 215A through 215n. The RAN 105 can transmit radio resource control messages 225 corresponding to the learning model information 215 to the UE 115.

[0104] AI / ML learning models deployed at UE devices 115, such as Figure 2The illustrated model 215) can be implementation specific (e.g., vendor proprietary learning model). (Examples of vendors that can provide proprietary learning models can include a user equipment manufacturer or a provider of an application for a user equipment, a network equipment provider or a provider of an application for a network equipment, or a mobile network operator or a provider of an application for a mobile network operator). The network RAN can determine the overall performance of the learning model deployed at the UE to facilitate the minimum device performance requirements. The dynamic reporting procedure can facilitate the user equipment to compile and report indications that can be configured or preconfigured to reflect or indicate the model performance of the corresponding learning model.

[0105] A particular user equipment device can employ several different AI / ML learning model implementations for running, executing, or otherwise facilitating different radio functions. Different learning model parameter indicators can indicate the performance of different learning models. The user equipment can compile and report one or more different learning model performance indication parameter indicators or indications for each learning model. Different learning model indicators can be associated with different corresponding filtering or time resolution configurations. Thus, such customized indicator reporting for a given learning model can facilitate the optimized tracking and reporting of each active learning model for each user equipment device 115 (that can be served by the RAN 105), as illustrated. Figure 1 or Figure 2 illustrated. Accordingly, the network RAN 105 can acquire and use the real-time performance of each learning model active at the UE 115 to facilitate the best performance of the learning model and the inferences it can generate. Moreover, several reporting variants can be customized to accommodate the implementation or purpose of various AI / ML learning models, e.g., accurate absolute, accurate relative, quantized, or time (e.g., historical) indicator reporting. The network node RAN 105 can dynamically trade-off the AI / ML learning model reporting overhead for the accuracy in acquiring the AI / ML model performance indicators.

[0106] For AI / ML learning model performance, various parameters and their corresponding indicators can be considered, analyzed, or evaluated depending on the nature of the problem being solved and the corresponding learning model function (e.g., regression or classification) or radio function that the learning model is executing or facilitating. For example, for radio functions such as channel estimation or channel state information (“CSI”) compression, a regression function can be used in the learning model and the following parameters or their corresponding indicators can be evaluated: mean squared error (“MSE”); root mean squared error (“RMSE”); normalized mean squared error (“NMSE”); mean absolute error (“MAE”); R-squared; generalized cosine similarity (“GCS”); or squared generalized cosine similarity (“SGCS”). Table 1 illustrates example functions that define the corresponding learning model parameters that can be monitored and evaluated with their corresponding indicators as listed above. Table 1

[0107] For classification problems such as beam index prediction, accuracy parameter metrics can be analyzed to determine the performance of learning models that facilitate beam index prediction. Other example learning model parameter metrics that can indicate the performance of learning models that solve classification problems can include, but are not limited to, the absolute number of true negatives, true positives, false negatives, and false positives; precision and recall; or F1 score. The F1 score can include an evaluation metric used to express the performance of a machine learning model or classifier and provides combined information about the precision and recall of a learning model. A higher F1 score metric generally indicates high values for both the recall and precision metrics.

[0108] As described above, AI / ML learning model implementations at different devices can be vendor specific and transparent to network nodes (e.g., a RAN serving a UE can not have access to the specific functionality and programming of a given learning model deployed in a UE that facilitates radio functions). To manage and facilitate UE device implementation performance targets, a RAN node can be aware of the functionality of a UE device and overall AI / ML learning model performance. Accordingly, an active UE device, upon first connecting to a serving network RAN, can transmit device specific AI / ML capability information that includes the following information elements (“IEs”): the type of AI / ML supported algorithms, including supervised learning, unsupervised learning, and reinforcement learning; a list of AI / ML supported radio functions; a list of supported AI / ML model specific metrics for estimation and reporting; a model bank size per radio function, e.g., the number of models that can be stored for each radio function; or an indication of model classification (small / medium / large), which can facilitate a network RAN to define or determine data sets to be used by learning models. For example, for a large number of neurons (e.g., nodes of a learning model neural network), a determination of a commensurate number of information samples can be used to avoid overfitting of a learning model. The AI / ML capability information elements can be transmitted based on subsequent radio resource control (“RRC”) signaling or based on dynamic scheduling uplink control information (“UCI”) and become part of device capability signaling. Accordingly, a network RAN can determine updates to one or more learning models and communicate the updated models or their corresponding coefficients to a user device.

[0109] The prior art optimizes, updates, and delivers AI / ML models, model information, and model parameter values only for connected mode devices (e.g., user equipment devices having a communication connection established with a serving radio access network node). Using conventional techniques, channel conditions, control channels, data channels, reference signals, AI / ML capabilities, and supported AI / ML radio characteristics corresponding to a connected mode user equipment can be identified and established. However, conventional techniques do not facilitate AI / ML training and assistance support for idle mode user equipment devices. Idle mode user equipment devices are “unknown” to a radio access network node because the user equipment is not connected to the node. As a result, AI / ML capabilities, supported AI / ML driven characteristics, and location of the user equipment are unknown to the network.

[0110] AI / ML model training facilitates efficient deployment of AI / ML algorithms within a cellular wireless communication network. A trained AI / ML model can provide predictive output based on actual data that provides results sufficient to support satisfactory performance. Without training, AI / ML models used at user equipment devices or at RAN nodes can output erroneous or inaccurate results that result in inappropriate actions or predictive estimations. With a large, diverse set of training samples, a model can be trained that results in an AI / ML model that is able to identify a variety of radio conditions and is able to responsively optimize the output of the model based on detected radio conditions. However, AI / ML model training is a time consuming, signaling overhead intensive, and processing heavy operation that is typically adjusted according to the needs of a particular user equipment, AI / ML model capabilities of the user equipment, or radio functionality corresponding to the model being trained. For connected mode user equipment devices, channel conditions, data channels, control channels, serving radio access network nodes, or AI / ML capabilities are “known” and established at the network, and AI / ML model training can be fully adjusted according to the needs and radio conditions of each connected mode device. AI / ML model training information can be sent or communicated via device specific or device common established control channels or data channels for connected mode user equipment. Training of AI / ML learning models with respect to user equipment in connected mode can be accomplished using conventional techniques by sending or downloading model information to connected user equipment for execution using real-time data. As a result, sending of learning model information for connected mode devices can be exchanged via control channel variations or communicated via data channels because data and control channels are already identified, adapted to device channel conditions, and otherwise established for connected mode devices.

[0111] However, during typical usage conditions, user equipment devices are in an idle state (e.g., the UE has not established a connection with a radio access network node) for a majority of the time that the device is on. In the idle state, no data channel, control channel, and serving RAN are established, or even known. Thus, AI / ML model training for idle mode user equipment is not performed using conventional techniques.

[0112] Since most cellular user equipment devices are in an idle state for a majority of the time that the user equipment is on, excluding AI / ML model training support for idle mode devices severely constrains or limits the true potential of AI / ML deployment. Moreover, training an AI / ML model using current radio condition information as one or more inputs to the learning model during an extended idle mode period, during which radio conditions can change (e.g., the user equipment is moving in the idle state, or in some mode), can result in a faster, more reliable transition to an active connected state when the device attempts to establish a network session, compared to a user equipment using an AI / ML model that is not trained based on current conditions. For example, a user equipment with a well-trained AI / ML model (e.g., a model trained using current radio condition information) driving a downlink beam prediction function during idle mode time can use a learning model trained based on recent radio conditions to proactively determine and identify to a radio access network node a predicted optimal downlink beam, which the user equipment device expects will result in sufficient coverage when the user equipment transitions to a connected mode with the radio access network node. Thus, by proactively training a learning model using a beam selection learning model (e.g., training the model while the UE is idle) to determine, predict, or estimate an optimal beam when the user equipment device connects to a radio access network node, the node can avoid or exclude sending reference signals that will be used to optimize downlink beam determination corresponding to the device, thereby potentially reducing access latency and signaling overhead.

[0113] Embodiments disclosed herein facilitate AI / ML model assistance and training for multiple radio functions for idle mode user equipment devices. Examples include AI / ML assisted timing advance (“TA”) acquisition, AI / ML assisted downlink reference signal and beam estimation, or AI / ML assisted uplink reference signal estimation. Embodiments disclosed herein facilitate training of AI / ML models implementing various radio functions when the user equipment is in an idle mode, idle state, inactive mode, or inactive state (which can be collectively referred to herein as “idle” or “idle mode”). Training AI / ML models while the user equipment is idle can facilitate a shortened time to transition to connected mode, and can facilitate a more reliable, more efficient transition from idle to connected than if the AI / ML models were trained only after the user equipment is connected to a serving radio access network node. For example, when an idle mode device is sufficiently trained for TA acquisition and beam estimation while idle, the user equipment can proactively and prospectively efficiently perform TA estimation (while the user equipment is in idle mode), and predict / estimate one or more radio function parameter values, and transmit the one or more parameter values to the RAN node that the user equipment attempts to connect to (e.g., establish a data / call session). Thus, the RAN node that the UE attempts to connect to can avoid performing a procedure to determine or acquire radio parameter values, as the parameter values can be proactively provided by the UE using a trained AI / ML model (trained while the UE is in idle mode), or prospectively acquired by the RAN node, thus transitioning to connected mode and establishing a data / call session more quickly, more efficiently than if the parameter values were determined after the UE begins establishing a connection.

[0114] Embodiments disclosed herein can implement inter-cell / RAN coordination procedures via backhaul links to exchange coordinated AI / ML assisted training data, parameter values, updated models, samples, or other model information to provide to idle mode user equipment devices. Embodiments described herein can implement a number of signaling procedures to deliver AI / ML assisted idle mode configuration and result feedback for two idle mode devices and to coordinate RAN nodes to achieve the required AI / ML model training accuracy corresponding to radio functions. In embodiments, a new radio application programming interface (“API”) can facilitate AI / ML capable devices reporting AI / ML model predicted radio key performance indicators (“KPIs”) based on estimated parameter values generated by AI / ML models trained while the UE is in idle mode or values derived therefrom when transitioning from idle mode to connected mode. Accordingly, not only can AI / ML models at user equipment be trained while the user equipment is in idle mode; idle mode user equipment devices can use AI / ML training moments of various radio functions to provide proactive / forward looking predictive intelligence when attempting to connect to a radio access network node (e.g., when transitioning to connected mode).

[0115] Dynamic AI / ML training assistance for idle mode devices.

[0116] Turning now to Figure 3The actions shown in the environment 300 facilitate idle mode AI / ML training assistance. At action 1, a RAN node in a coordinating group of RAN nodes (which can include neighboring RAN nodes / cells 105A and 105B) can coordinate, via backhaul link 320, in terms of idle mode AI / ML training assistance, including coordination regarding support for radio function AI / ML training while a user equipment is in idle mode. The RAN nodes 105A and 105B can coordinate expected operation of a user equipment device UE 115A or 115B and coordinate actions of the RAN nodes (e.g., the RAN nodes can coordinate transmission of reference signals or transmission of preambles or reference signals by the user equipment), as well as AI / ML training resources associated with each RAN node (e.g., the RAN nodes coordinate to avoid scheduling of resources for training purposes to overlap with scheduling of resources for non-AI idle mode operations, such as paging). At action 2A or 2B, the coordinating RAN node 105A or 105B can respectively communicate (e.g., via broadcast signaling) an AI / ML configuration information block signal message 500A or 500B that can be detected by idle mode user equipment devices 115A or 115B in coverage of the coordinating RAN node. The message 500A or 500B can include or indicate configuration information corresponding to currently available AI / ML learning model idle mode training assistance and support for idle mode radio functions (e.g., learning models for facilitating radio functions that can be trained while the UE is in idle mode).

[0117] At action 3A or 3B, the idle mode user equipment device apparatus 115A or 115B with AI / ML capability can perform the AI / ML training actions indicated in the configuration message 500A or 500B or indicated by the configuration message 500A or 500B (e.g., measuring AI / ML training reference signals transmitted from the RAN 105A or RAN 105B, transmitting uplink preambles and / or reference signals to the RAN 105A or RAN 105B, etc.) for radio functions that can be supported or potentially of interest (e.g., radio functions of interest to the RAN or UE). At action 4, in case all or a subset of the AI / ML models trained in the idle mode need to feed back the training results to the user equipment apparatus, the coordinating RAN nodes 105A and 105B can exchange AI / ML training results 340A or 340B (e.g., via the backhaul interface link 320 (e.g., the results exchanged at action 4 can include, for example, uplink coverage received from reference signals transmitted by the UE device, estimated timing advance received from the UE device, etc.)) in case all or a subset of the AI / ML models trained in the idle mode need to feed back the training results to the user equipment apparatus, to facilitate coordination of the learning models corresponding to the radio functions at one RAN to be consistent with the same learning models at another RAN. The RAN node 105A or 105B can transmit AI / ML training assistance results information block at action 5A or action 5B, respectively, as broadcast signal message 350A or 350B, containing the result(s) of inter-cell AI / ML training collected from neighboring cells / RAN nodes of the same AI / ML coordination group, which can include 105A or 105B.

[0118] In embodiments, at the time of establishing a connection with the RAN 105A or 105B, the new radio signaling API can facilitate the user equipment device 115A or 115B indicating to the RAN node 105A or 105B a proactively determined radio access key performance indicator (“KPI”). This indicator can be estimated at the UE 115A or 115B based on an AI / ML model corresponding to a radio function associated with the KPI while the UE 115A or 115B is in idle mode. Thus, the user equipment device 115A or 115B can train the AI / ML model during an idle mode period, run various radio functions based on the trained model(s) to estimate radio function information / KPIs, and subsequently provide this estimated / predicted intelligence to the RAN node via the new radio API at the time of establishing a connection with the RAN 105A or 105B, resulting in faster, more reliable network access. For example, when the user equipment device proactively provides the RAN node with the expected optimal set of downlink beams (e.g., determined with a learned model trained while the UE is idle) at the time the user equipment device establishes a connection with the RAN node, the RAN node can minimize the delay and use of overhead resources to send reference signals to the user equipment device that the UE can use to determine the determined optimal downlink beams for the UE by using the proactively determined optimal beams. As the UE has proactively provided this KPI intelligence, access delay and use of downlink / uplink signaling overhead can be reduced compared to what would otherwise be possible if the indication of the optimal beams predicted by an AI / ML learned model trained while the UE is idle were not sent to the RAN at the time of connection establishment.

[0119] Turning now to Figure 4 , the figure illustrates a learned model configuration information block message 410, which can include a training configuration resource indication 405 including or indicating training configuration resources to be used or available for broadcasting learned model training configurations by a radio access network node, such as can be described in reference to Figure 3 sent at action 2A or 2B of the described actions. In embodiments, Figure 4The illustrated training resource configuration resource indication 405 can include a training result resource indication that can include or indicate a training result resource that can be used to transmit determined radio function parameter values that can be determined by the RAN that has transmitted the information block message 410. The training configuration resource indication 405 can indicate a training configuration resource that will be used or that can be used to broadcast an AI / ML training configuration information block message 415 that can include or can indicate a learning model training configuration. In embodiments, the configuration information block message 415 can indicate one or more resources that the radio access network node will use or that can be used to transmit a learning model training configuration.

[0120] The training configuration resource indication 405 can indicate one or more training result resources that will be used or that can be used to broadcast an AI / ML training result information block message 420 that can indicate a learning model training result or that can indicate resources that the radio access network node will use or that can be used to transmit or broadcast a learning model training result, such as the result message 350A or 350B transmitted at action 5 illustrated in Figure 3 . It should be understood that the message 350A can include different information than the message 350B.

[0121] Continuing with the description of Figure 4 , as part of a broadcast master information block (“MIB”) 410 or a basic system information block (“SIB”) 410 (the information block 410 can be a MIB or a SIB), the RAN can indicate (via the indication 405) the presence of an AI / ML idle mode model training assistance configuration information block 415 or the presence of an AI / ML result information block 420. The RAN can transmit a new presence indication 405 (as part of a MIB or SIB) to indicate the availability of a new system information block 415 or its corresponding resources that can carry idle mode AI / ML training assistance configuration or that can indicate resources that can carry idle mode AI / ML training assistance configuration that can be embodied in one or more example configurations described with reference to Figure 5A , Figure 5B , Figure 5C or Figure 5D . The indication 405 can indicate resources that can carry a new result system information block 420 that can carry idle mode AI / ML training assistance sampling results or that can indicate resources that can carry idle mode AI / ML training assistance sampling results that can be embodied in one or more example configurations described with reference to Figure 6 or Figure 7In the described examples. When an idle mode user equipment device determines the presence of an AI / ML SIB and corresponding resource(s) via indication 405, the user equipment can decode (may include blind decoding) one or more AI / ML configuration blocks 415 or one or more AI / ML result blocks 420, which can include information corresponding to one or more idle mode indicated radio functions, such as a learning model training configuration or determined radio function parameter values.

[0122] Turning now to Figure 5A , an idle mode AI / ML configuration 500 SIB is illustrated. Configuration 500 can be contained in a reference Figure 4 described information block 415 or can be indicated by an indication contained in information block 415 (e.g., information block 415 can include configuration 500 or can indicate resources available for a user equipment to receive configuration 500). Configuration 500 can include an identifier field 502 indicating an AI-ML learning model, a radio function implementable by the AI / ML learning model, or an AI / ML learning model feature. One or more coordinating RAN node identifiers in field 504 can be associated in configuration 500 with each defined radio feature 502A to 502n, which can be indicated by a feature identifier defined or determined in field 502. Configuration 500 can associate a training action 506 for an uplink direction or a downlink direction with a training feature 502 and a coordinating RAN associated in a coordinating group of RANs. Configuration 500 can associate training resources 508 available or to be used to perform corresponding training action(s) 506 with corresponding RAN(s) 504 relative to feature 502. Thus, configuration 500 can be used to indicate one or more AI / ML models or features to idle mode user equipment, which can be trained for one or more radio functions. Configuration 500 can not necessarily indicate or command a user equipment to perform a training action 506 for a feature 502 for RAN(s) indicated in RAN field 504.

[0123] In Figure 5BIn the illustrated example, the example AI / ML configuration contained in or indicated by the configuration information block 510 can facilitate random access channel (“RACH”) timing advance (“TA”) training of a TA learning model or TA learning model feature via a feature corresponding to a feature identifier contained in the feature identifier field 512. Adjacent RAN nodes in a RAN coordination group (which can include two or more RAN nodes that can be adjacent or proximate) can coordinate on RACH TA training resources 518A, 518B, 518n that can be used by user devices to transmit a corresponding defined TA preamble. The TA preamble can be contained in or as part of a defined preamble group indicated in the preamble group field 516. The user device can transmit the preamble indicated in field 516 in the uplink direction toward the coordinating RAN identified in the corresponding RAN field 514. The preamble group can include a set of defined preambles that have been configured to be associated with a given RAN 514 such that the RAN, upon receiving via resources 518 corresponding to the preamble group 516 of the given RAN, can indicate to the receiving RAN that the RAN is to determine a TA with respect to a user device that can have transmitted the preamble. In response to receiving a preamble of a TA preamble group or TA preamble pool, the receiving RAN can train or update a learning model or learning model feature corresponding to TA acquisition. The training / update results of the RAN can be transmitted to user devices via a results SIB. Thus, upon decoding the AI / ML configuration SIB 510, if an idle mode device is interested in or supports the AI / ML driven TA acquisition 512, an indicated uplink TA preamble can be transmitted that can be randomly selected by the idle mode user device from the configured TA preamble group 516B (e.g., via corresponding resources 518B toward the associated target RAN node 514B). Adjacent RAN nodes can coordinate with one another such that the resources indicated in field 518 that are to be used with respect to the RANs identified in field 514 as corresponding to the resources do not overlap (e.g., in time or frequency) with resources that can be used by the adjacent RANs for non-AI / ML training purposes.

[0124] In another embodiment as shown in FIG. 5, for example, an AI / ML configuration information block 520 can facilitate training of channel state information (“CSI”) features 522. Coordinating RAN nodes indicated in field 524 can transmit a set of CSI beams indicated in field 526 via resources indicated in field 528. An idle mode user device can receive and detect received coverage levels corresponding to beams via corresponding resources indicated in field 528, and can subsequently train a CSI AI / ML model at the user device. The user device can transmit a determined optimal beam result indication corresponding to the RAN identified in field 524 that transmitted beams in the uplink direction via small data transmission (e.g., according to existing technologies that facilitate fast, small payload transmissions by idle mode devices), or the idle mode user device apparatus can transition to connected mode and transmit the optimal beam indication as data transmission before resuming to idle mode.

[0125] In Figure 5D another embodiment as shown, an example AI / ML configuration block 530 can facilitate training of a sounding reference signal (“SRS”) learning model feature 532 at a user device. A training action indication 536 in configuration 530 can include an indication to transmit a configured SRS reference signal pattern by a device in use via resources 538 toward RAN nodes (indicated in field 534) of an AI / ML RAN coordination group while idle. SRS training can facilitate RAN and user device optimization of uplink decoding performance. Training of RAN nodes relative to user devices on uplink channel conditions while user devices are idle can facilitate providing dynamic uplink transmission configurations to user devices almost immediately when the user devices connect to the RAN nodes and establish uplink sessions. By training SRS learning model features while user devices are in idle mode, RAN nodes can avoid requesting additional SRS transmissions from transitioning user devices. Thus, based on the SRS model features 532 trained while user devices are idle, less time is spent on optimizing uplink channel quality determinations before establishing uplink sessions, which can result in faster uplink session establishment.

[0126] Turning now to Figure 6 , an example SIB 600 for carrying AI / ML result samples is shown. SIB 600 can be indicated by an indication 405 as shown, for example (e.g., SIB 600 can be the same as SIB 420 as shown). In embodiments, Figure 4 Figure 4 Figure 6 ​​The SIB 600 shown may include indications of training result resources that can be used or will be used by the RAN to send AI / ML model training information / results, such as determined radio function parameter values, to the UE.

[0127] In one example, training result 604A corresponding to the identified learning model feature 602 can be transmitted or broadcast to the user equipment via SIB 600. For example, one or more RAN nodes belonging to the AI / ML training coordination group can transmit or broadcast SIB 600, which contains AI / ML training samples that may have been determined at the RAN nodes and are available for use by idle-mode user equipment. Therefore, for the radio feature ID corresponding to the AI / ML model 602 trained at the RAN, the RAN can transmit the training result or training result sample 604 to the user equipment via SIB 600.

[0128] exist Figure 7 In the illustrated embodiment, the AI / ML training result SIB 700 can facilitate the training of the RACH TA learning model or the training of RACH TA learning model features 702. For the RACH TA learning model features, multiple information objects 720 can be indicated as corresponding to radio access network nodes (such as…) Figure 3 As shown in RAN 105A or RAN 105B, where (multiple) RANs may belong to or participate in the AI / ML coordination group. Therefore, the RAN corresponding to the preamble indicated by indication 720 can report the estimated TA level or TA parameter value to the UE via result SIB 700. For example, if Figure 3 The UE shown sends a message to RAN 105A from... Figure 5B The preamble selected from the preamble group 516A shown (assuming RAN 105A is identified in) Figure 5B As shown in the example in field 514A of configuration 510, RAN 105A can be identified in field 720 of SIB 700. RAN 105A can return a TA result 720A-1 to the UE, which indicates the TA generated by RAN 105A based on the AI / ML training operation performed on the preamble y1 received from the UE while the UE is in idle mode. If the UE transmits preamble y2 or y3 respectively in idle mode... i Then RAN 105A can also return results 720A-2 or 720A-iI. Accordingly, user equipment devices with AI / ML capabilities can, in idle mode, according to, for example, MIB 410 (such as... Figure 4The resource indicated by instruction 405 (as shown) is received and decoded, and the result SIB 700 is extracted from it, along with one or more preambles 516 that the UE may have sent to it (as shown). Figure 5B (as shown) the TA level corresponding to one or more RAN nodes.

[0129] Now go to Figure 8 The figure illustrates a radio API signaling message environment 800. Message 820 can output radio estimation results from User Equipment (UE) device 115 to RAN node 105, carrying an actively / proactively determined AI / ML model, in a new message section 825. For example, when UE device 115 transitions from idle mode to connected mode with RAN 105, the UE can proactively provide one or more radio KPI estimates to the RAN node. One or more KPIs can be generated by an AI / ML model trained when UE 115 is idle. The trained model can be a model implementing one or more radio functions. Figure 8 In the example shown, User Equipment (UE) device 115 can indicate to RAN 105 via the new API portion 825 of RRC message 820 the optimal downlink CSI beam estimated by AI / ML, which has a beam index 802 corresponding to a trained beam feature 801, via which the UE expects (e.g., based on a trained AI / MNL model) to receive the optimal coverage level relative to other downlink beams corresponding to RAN 105. UE 115 can also transmit an expected / predicted uplink timing advance 806 corresponding to the trained feature 805 in the new API portion 825. The indication of the optimal beam index 802 can correspond to the predicted beam prediction accuracy 804 (determined by the AI / ML model trained by the UE when idle), and the indication of TA 806 can correspond to the predicted TA accuracy 808 (determined by the AI / ML model trained by the UE when idle).

[0130] Accordingly, the session establishment or connection establishment between the UE 115 and the RAN 105 can be based on the information and KPI elements that are determined proactively / forward looking (e.g., determined when the UE is idle), and thus the RAN can avoid triggering CSI beam optimization procedures or uplink timing advance acquisition procedures during session establishment, as the UE device has proactively provided the predicted KPIs to the RAN. As part of the AI KPI information block 825 that can be transmitted during the RRC connection request signaling 820, the UE 115 can signal the achievable prediction accuracy 804 or 808 of the corresponding AI / ML model of the generated respective predicted KPI 802 or 806. Since the AI / ML model can have been trained while the device was in idle mode, during which time the RAN 105 typically has no knowledge of the performance of the AI / ML model, the RAN can use the accuracy prediction 804 or 808 to determine whether to use the estimated KPI 802 or 806, respectively. Accordingly, the new KPI information elements 802 and 806 and the corresponding accuracy information elements 804 and 808 can facilitate the RAN to determine whether the AI model available at the UE 115 to generate the predictive KPI information elements is operating satisfactorily. If the reported prediction accuracy level 804 or 806 is low (e.g., below a threshold configured at the RAN 105), the RAN can disregard the corresponding KPI 802 or 806 received from the UE 115 and instead trigger regular radio procedures (e.g., determine beam selection via beam sweeping or TA acquisition during RRC connection establishment) for establishing the device connection. As an example, if the respective accuracy 804-1 or 804-2 does not satisfy a configured threshold, the RAN 105 can override and disregard the signaled optimal beam indication 802-1 or 802-2 and trigger the beam optimization procedure as regular during connection establishment.

[0131] Turning now to Figure 9 FIG. 13 illustrates a timing diagram of an example method 1300. At action 1305, the neighboring RAN nodes 105A and 105B can exchange supported AI / ML model training features via backhaul / Xn interface, which can be trained when the UE 115 is idle. The information exchanged at action 1305 can include specific receive and transmit signal information corresponding to each coordinating RAN 105A or 105B. The information exchanged at action 1305 can include timing, frequency, and periodicity information corresponding to the RAN 105A or 105B. At action 1310A or 1310B, the RAN node 105A or 105B can transmit a new AI / ML configuration system information block scheduling indication (e.g., indication 405 shown in FIG. 4) as part of a synchronization signal block (SSB) and / or as part of a master information block (e.g., indication 405 shown in FIG. 4) to the UE 115, respectively. Figure 4 Figure 4 ​part of the information block 410 shown in FIG. 13). The new indication sent at action 910A or 910B can include an idle mode inter-cell training configuration indication. In Figure 9 At actions 915A and 915B shown in FIG. 13, the RAN nodes 105A and 105B can send an AI / ML idle mode training configuration to the UE 115 via the first scheduled configuration AI / ML SIB resources (e.g., corresponding to the information block 415 shown in FIG. 13). Figure 4 At actions 915A and 915B shown in FIG. 13, the RAN nodes 105A and 105B can send an AI / ML idle mode training configuration to the UE 115 via the first scheduled configuration AI / ML SIB resources (e.g., corresponding to the information block 415 shown in FIG. 13). Figure 5A At actions 920A or 920B, the UE 115 can receive, decode, or train an AI / ML model according to the configuration received at actions 915A or 915B.

[0132] If the configured AI / ML idle mode training period for a certain AI / ML feature to be trained during the idle mode of the UE 115 expires, the RAN nodes 105A or 105B can receive one or more AI / ML idle mode training reports from each other via the backhaul link at action 925. At actions 930A or 930B, the RAN nodes 105A or 105B can send one or more new AI / ML result system information block scheduling indications to the UE 115 as part of the synchronization signal block and / or the master information block, which can indicate resources scheduled to send AI / ML idle mode inter-cell training result samples (e.g., the information exchanged at action 925) to the UE 115. The indication mentioned at actions 930A or 930B can correspond to the indication 405 shown in FIG. 13, or can be the indication. In Figure 4 At actions 935A or 935B shown in FIG. 13, the RAN nodes 105A or 105B can send one or more AI / ML training result samples via the second scheduled AI / ML training result sample SIB resources (e.g., corresponding to the resources of the training result information block 420 shown in FIG. 13). Figure 9 At actions 935A or 935B shown in FIG. 13, the RAN nodes 105A or 105B can send one or more AI / ML training result samples via the second scheduled AI / ML training result sample SIB resources (e.g., corresponding to the resources of the training result information block 420 shown in FIG. 13). Figure 4 At actions 935A or 935B shown in FIG. 13, the RAN nodes 105A or 105B can send one or more AI / ML training result samples via the second scheduled AI / ML training result sample SIB resources (e.g., corresponding to the resources of the training result information block 420 shown in FIG. 13). Figure 9 At action 940 shown in FIG. 13, the UE 115 can use the AI / ML performance indicators determined prospectively while the UE is in the idle mode (e.g., the TA determined prospectively and the corresponding accuracy indicator) to establish a connection with the RAN 105A or the RAN 105B.

[0133] Turning now to Figure 10The figure illustrates a flowchart of an example method 1000. The method 1000 begins at act 1005. At act 1010, a radio access network node can determine learning model configuration information, or a set of neighboring radio access network nodes can coordinate and exchange learning model configuration information with each other. The determined or exchanged learning model configuration information can include an indication of a radio function that can be implemented or facilitated by one or more learning models at the radio access network node or at a user equipment. The determined or exchanged learning model configuration information can correspond to resources that a coordinating radio access network node can use to broadcast learning model configuration or learning model results for idle mode user equipment to receive. The determined or exchanged learning model configuration information can correspond to a radio function learning model that can be trained using information resulting from training actions performed while a user equipment is idle. The training actions can include one or more actions that are executable by a user equipment while the user equipment is idle. The training actions can include one or more actions that are executable by a radio access network node while the user equipment is idle. Examples of training actions can include transmitting a reference Figure 5B described preamble 516, receiving a reference Figure 5C described CSI reference signal identifier 526, or transmitting a reference Figure 5D described SRS group identifier 536.

[0134] Continuing with the description of Figure 10 at act 1015, the radio access network node that determines the configuration information or the radio access network node that coordinates with another radio access network node to determine the configuration information at act 1010 can broadcast a training configuration information block message resource indication, such as reference Figure 4 described indication 405 in the MIB 410. Continuing with the description of Figure 10 at act 1020, the radio access network node can broadcast a learning model configuration in a training configuration information block (such as reference Figure 4 described information block 415) via the resource indication indicated in the indication 405. The learning model configuration can be according to the configuration 500 described with reference Figure 5A

[0135] Continuing with the description of Figure 10 at act 1025, a user equipment can receive and detect a configuration information block according to the training configuration resource indicated at act 1015 (such as according to the training configuration resource indicated in the indication 405 described with reference Figure 4 Figure 10 ​​In accordance with the description of the actions 1010-1045, at action 1035 the user equipment can perform the training actions indicated in the learning model configuration broadcast in the learning model configuration information block. The learning model configuration information block or the learning model configuration contained therein can include an indication of more than one training action corresponding to more than one learning model, or an indication of learning model features corresponding to radio functions implementable by the learning model. At action 1035, the user equipment can decide to perform zero, one, more or all of the training actions indicated in the learning model configuration. At action 1040, based on or in response to the user equipment performing the training actions indicated in the learning model configuration, the radio access network node that transmitted the learning model configuration or another radio access network node that can have cooperated with the radio access network node that broadcast the learning model configuration at action 1010 can determine a training result.

[0136] At action 1045, the radio access network node that determined the training result at action 1040 can determine whether the determined training result should be broadcast, transmitted or otherwise provided to the user equipment that performed the training actions at action 1035. The determination made at action 1045 can include a determination of whether the training result determined at action 1040 should be broadcast, transmitted or otherwise provided to user equipment other than the user equipment that performed the training actions at action 1035. If it is determined at action 1045 that the training result determined at 1040 need not be provided to the user equipment, the method 1000 can proceed to action 1070. At action 1070, the radio access network node and the user equipment that performed the training actions at action 1035 can establish a communication connection in accordance with a learning model that can have been trained based on the training actions performed by the user equipment at action 1035. It will be appreciated that the user equipment can be in idle mode or can otherwise not have established a connection with the radio access network node during the performance of at least action 1015 through the performance of action 1045, and the establishment of a connection with the radio access network node can be initiated at action 1070. Accordingly, the establishment of a connection at action 1070 can include the transmission and reception of radio resource control signaling messages. After the connection between the user equipment that performed the training actions at action 1035 in idle mode and the radio access network node is established at action 1070, the method 1000 proceeds to action 1075 and ends.

[0137] Returning to the description at action 1045, if the radio access network node that determined the training result at 1040 determines that the user equipment may need to receive the training result determined at action 1040, then method 1000 proceeds to action 1050. At action 1050, the radio access network node that determined the training result at action 1040 can broadcast the training result determined at action 1045, based on the training result resources indicated in the training configuration information block message broadcast at action 1015, for the user equipment performing the training action at action 1035 to receive or detect. Therefore, in addition to the training model configuration information block message resource indication sent or broadcast at action 1015 for indicating resources available for sending or broadcasting the configuration information block at action 1020 (e.g., besides referencing...), Figure 4 The described instruction 405 can be used via reference Figure 4 In addition to the training configuration information block 415 described (referring to the resources of configuration 500 described in Figure 5), the training configuration information block message resource indication (e.g., referring to...) sent at action 1015 is also included. Figure 4 The described instruction 405 may also include an indication of training resources available or to be used by a radio access network node, which determines the training results at action 1040 to broadcast the training results at 1050. At action 1055, the user equipment can, based on the training configuration information block message resource indication broadcast at action 1015, specify the training result resources (e.g., according to the reference) that are already indicated. Figure 4 The resource indicated in instruction 405 may receive one or more training results broadcast at action 1050. The user equipment may use the training results received at 1055 to update or train the learning model or learning model features corresponding to the training action performed at action 1035.

[0138] At act 1060, in embodiments, the user device can determine one or more key performance indicators prospectively using an updated or trained learning model that can have been trained or updated based on, from, or in response to the training actions performed at act 1035, or that can have been trained or updated based on, from, or in response to the training results received at act 1055. For example, the training actions performed at act 1035 can include the user device sending a preamble that will be used by the radio access network node to train the timing advance machine learning model or the timing advance machine learning model feature. The results of training the machine learning model or the machine learning model feature associated with determining timing advance can result in the model parameter results determined by the radio access network node at act 1045, which will be provided to the user device at act 1050. After updating the timing advance machine learning model or the timing advance machine learning model feature, the user device can use the updated / trained timing advance machine learning model or the updated / trained timing advance machine learning model feature to prospectively determine the timing advance with respect to the radio access network node that can have determined the learning model configuration information, or that can have cooperatively determined the learning model configuration information (act 1010), at act 1063. The term “prospectively” (the term “proactively” is used interchangeably herein) means that the user device prospectively determining the timing advance can be in an idle mode and can determine the timing advance prior to initiating or attempting to establish a connection with the radio access network node that can have determined or cooperatively determined the machine learning model configuration information at act 1010.

[0139] At act 1065, the user device that prospectively determined one or more key performance indicators at act 1063 can transmit the one or more prospectively determined KPIs to the radio access network node prior to or during the connection establishment procedure performed at act 1070, thereby avoiding determining the one or more key performance indicators, such as, for example, the timing advance or the optimal beam selection, during establishing the connection at act 1070. After establishing the connection at act 1070, the method 1000 proceeds to act 1075 and ends.

[0140] Now turning to Figure 11The figure illustrates an example embodiment method 1100, the method 1100 comprising: at block 1105, broadcasting, by a radio access network node comprising a processor, a learning model configuration information block message, the learning model configuration information block message comprising a training configuration resource indication, the training configuration resource indication for indicating a training configuration resource usable for broadcasting a learning model training configuration by the radio access network node; at block 1110, broadcasting, by the radio access network node, a learning model training configuration according to the training configuration resource; at block 1115, wherein the learning model training configuration comprises a training action indication, the training action indication for indicating a training action performable by an idle user equipment; at block 1120, wherein the training action corresponds to a radio function learning model, wherein performing the training action is to result in a determined radio function parameter value corresponding to the radio function learning model, the method further comprising: receiving, by the radio access network node, a radio resource control signal message comprising the determined radio function parameter value from the idle user equipment; and at block 1125, establishing, by the radio access network node, a connection with the idle user equipment using the determined radio function parameter value received in the radio resource control signal message from the idle user equipment, thereby causing the idle user equipment to become a connected user equipment.

[0141] Turning now to Figure 12 The figure illustrates a first radio access network node 1200, at block 1205, comprising a processor configured to receive, from a second radio access network node being a neighboring radio access network node relative to the first radio access network node, a non-training resource indication, the non-training resource indication indicating to the first radio access network node non-training resources to be used by the second radio access network node for non-training operations; at block 1210, schedule training resources, the training resources to be used by at least one idle mode user equipment for performing a training action relative to the first radio access network node, whereby the training resources and the non-training resources do not overlap; at block 1215, broadcast a master information block message comprising a training configuration resource indication, the training configuration resource indication for indicating training configuration resources to be used by the first radio access network node for broadcasting a learning model training configuration; and at block 1220, broadcast the learning model training configuration according to the training configuration resources.

[0142] Turning now to Figure 13The figure illustrates a non-transitory machine-readable medium 1300 comprising executable instructions that, when executed by a processor of a first radio access network node, facilitate performing the following operations: broadcasting a first information block message comprising a training configuration resource indication, the training configuration resource indication being for indicating a training configuration resource; 1310, in accordance with the training configuration resource, broadcasting a learning model training configuration, wherein the learning model training configuration comprises a training action indication, the training action indication being for indicating a training action to be performed by a first idle user equipment of at least one idle user equipment relative to the first radio access network node to result in a first determined learning model parameter value; at block 1315, receiving the first determined learning model parameter value from the first idle user equipment of the at least one idle user equipment; at block 1320, receiving a second determined learning model parameter value from a second radio access network node that is a neighboring radio access network node relative to the first radio access network node, wherein the training action is performed by at least a second idle user equipment of the at least one idle user equipment relative to the second radio access network node to result in the second determined learning model parameter value; at block 1325, determining an updated learning model based on the first determined learning model parameter value and based on the second determined learning model parameter value; and at block 1330, broadcasting the updated learning model to the first idle user equipment of the at least one idle user equipment via a second information block message.

[0143] Turning now to Figure 14 The figure illustrates an example embodiment method 1400 comprising: at block 1405, receiving, by a user equipment comprising a processor, a learning model configuration information block message from a first radio access network node, the learning model configuration information block message comprising a training configuration resource indication, the training configuration resource indication being for indicating a training configuration resource usable for broadcasting, by the first radio access network node, a learning model training configuration; at block 1410, receiving, by the user equipment, the learning model training configuration in accordance with the training configuration resource; at block 1415, decoding, by the user equipment, the learning model training configuration; at block 1420, wherein the decoding of the learning model training configuration comprises blind decoding; and at block 1425, wherein the learning model training configuration comprises a training action indication, the training action indication being for indicating a training action to be performed by the user equipment.

[0144] Turning now to Figure 15The figure illustrates an example user equipment 1500, at block 1505, comprising a processor configured to receive a learning model configuration information block message from a radio access network node, wherein the learning model configuration information block message comprises a training configuration resource indication for indicating a training configuration resource usable for receiving a learning model training configuration from the radio access network node; at block 1510, receive the learning model training configuration according to the training configuration resource; at block 1515, decode the learning model training configuration, wherein the learning model training configuration comprises a training action indication for indicating a training action to be performed by the user equipment; at block 1520, perform the training action to result in a training action result; at block 1525, send the training action result to the radio access network node; at block 1530, wherein the training action comprises generating a sounding reference signal, such that the training action result is the generated sounding reference signal, and wherein the generated sounding reference signal is sent to the radio access network node, which can use the generated sounding reference signal to train an uplink resource grant learning model to result in a trained uplink resource grant learning model; at block 1535, establish a communication connection with the radio access network node, wherein the communication connection comprises at least one uplink resource granted by the radio access network node based on the trained uplink resource grant learning model; and at block 1540, the grant of the at least one uplink resource by the radio access network node is based on the user equipment excluding transmission of a sounding reference signal after the user equipment sends the generated sounding reference signal.

[0145] Turning now to Figure 16The figure illustrates a non-transitory machine-readable medium 1600 that includes, at block 1605, executable instructions that, when executed by a processor of a user equipment, facilitate performance of operations comprising: receiving, from a first radio access network node, a learning model configuration information block message while the user equipment is idle, the learning model configuration information block message comprising a training result resource usable by the user equipment to receive training results from the first radio access network node; receiving, from the first radio access network node, a learning model training configuration while the user equipment is idle, the learning model training configuration comprising a training action indication to indicate a training action performable by the user equipment with respect to at least the first radio access network node; performing, at block 1615, the training action with respect to the first radio access network node while the user equipment is idle to result in a first training action result; performing, at block 1620, the training action with respect to a second radio access network node while the user equipment is idle to result in a second training action result, wherein the second radio access network node is a neighboring radio access network node with respect to the first radio access network node; and receiving, at block 1625, the second training action result from the second radio access network node.

[0146] To provide additional context for various embodiments described herein, Figure 17 And the following discussion is intended to provide a brief overview of a suitable computing environment 1700 in which various embodiments described herein can be implemented. Although the embodiments have been described above in the general context of computer-executable instructions of a program that runs on a computer, those skilled in the art will recognize that the embodiments also can be implemented in combination with other program modules and / or as a combination of hardware and software.

[0147] Generally, program modules include routines, programs, components, data structures, etc., that perform particular tasks or implement particular abstract data types. Moreover, those skilled in the art will appreciate that the methods can be practiced with other computer system configurations, including single-processor or multiprocessor computer systems, minicomputers, mainframe computers, Internet of Things devices, distributed computing systems, as well as personal computers, hand-held computing devices, microprocessor-based or programmable consumer electronics, and the like, each of which can be operatively coupled to one or more related devices.

[0148] The embodiments described herein also can be implemented in a distributed computing environment, where certain tasks are performed by remote processing devices that are linked through a communications network. In this distributed computing environment, program modules can be located in both local and remote memory storage devices.

[0149] Computing devices typically include a variety of media, which can include computer-readable storage media, machine-readable storage media, and / or communications media, as these terms are used herein, in different contexts and with different applications, are intended to have the same general meaning. Computer-readable storage media or machine-readable storage media can be any available storage media that can be accessed by a computer, including both volatile and nonvolatile media, removable and non-removable media. By way of example, and not limitation, computer-readable storage media or machine-readable storage media can be implemented in connection with any method or technology for storage of information (e.g., computer- readable or machine-readable instructions, program modules, structured data, or unstructured data).

[0150] Computer-readable storage media can include, but are not limited to, random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD), Blu-ray disc (BD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, solid state drives or other solid state storage devices, or other tangible and / or non-transitory media which can be used to store desired information. In this context, the term “tangible” or “non-transitory” as applied to a memory, memory, or computer-readable media, should be understood only to exclude propagating transitory signals per se as modifiers and does not detract from all standard memories, memories, or computer-readable media that are not propagating transitory signals per se.

[0151] Computer-readable storage media can be accessed by one or more local or remote computing devices, such as through access requests, queries, or other data retrieval protocols, to perform various operations in relation to the information stored by the media.

[0152] Communication media typically embody computer-readable instructions, data structures, program modules, or other structured or unstructured data in a data signal such as a modulated data signal, e.g., a carrier wave or other transport mechanism, and includes any information delivery or transport media. The term “modulated data signal” or signal refers to a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. By way of example, and not limitation, communication media includes wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, radio frequency, infrared, and other wireless media.

[0153] Referring again to Figure 17The example environment 1700 for implementing various embodiments of the aspects described herein includes a computer 1702, which includes a processing unit 1704, a system memory 1706, and a system bus 1708. The system bus 1708 couples system components including, but not limited to, the system memory 1706 to the processing unit 1704. The processing unit 1704 can be any of various commercially available processors, and can include

[0154] The system bus 1708 can be any of several types of bus structures including a memory bus with or without a memory controller, a peripheral bus, and a local bus using any of a variety of commercially available bus architectures. The system memory 1706 includes ROM 1710 and RAM 1712. A basic input / output system (BIOS) can be stored in non-volatile memory such as ROM, erasable programmable read-only memory (EPROM), EEPROM, flash memory or the like. The BIOS contains routines that help to transfer information between elements within the computer 1702, such as during startup. The RAM 1712 can also include high-speed RAM such as static RAM for caching data.

[0155] The computer 1702 further includes an internal hard disk drive (HDD) 1714, e.g., EIDE, SATA, one or more external storage devices 1716, e.g., a magnetic floppy or hard disk drive (FDD) 1716, a memory stick or flash drive reader, a memory card reader, etc., and an optical disk drive 1720, e.g., a CD-ROM, DVD, BD, etc. reader / writer. While a built-in HDD 1714 is shown in the figures as being within the computer 1702, the built-in HDD 1714 can also be configured for external use in an appropriate chassis (not shown). Additionally, while not shown in the environment 1700, a solid state drive (SSD) can be included in addition to or in place of the HDD 1714. The HDD 1714, external storage device(s) 1716, and optical disk drive 1720 can connect to the system bus 1708 via an HDD interface 1724, an external storage interface 1726, and an optical drive interface 1728, respectively. The interface 1724 for external drive implementations can include at least one or both Universal Serial Bus (USB) and Institute of Electrical and Electronics Engineers (IEEE) 1394 interface technologies. Other external drive connection technologies are within the contemplation of the embodiments described herein.

[0156] The drive and its associated computer-readable storage medium provide non-volatile storage of data, data structures, computer-executable instructions, etc. For computer 1702, the drive and storage medium can accommodate storage of any data in a suitable digital format. Although the above description of computer-readable storage media refers to a corresponding type of storage device, those skilled in the art will understand that other types of computer-readable storage media (whether existing or developed in the future) may also be used in the example operating environment, and furthermore, any such storage medium may contain computer-executable instructions for performing the methods described herein.

[0157] The drive and RAM 1712 can store multiple program modules, including an operating system 1730, one or more application programs 1732, other program modules 1734, and program data 1736. All or part of the operating system, applications, modules, and / or data can also be cached in RAM 1712. The systems and methods described herein can be implemented using various commercially available operating systems or combinations of operating systems.

[0158] Computer 1702 may optionally include emulation technology. For example, a hypervisor (not shown) or other intermediary may emulate a hardware environment for operating system 1730, and the emulated hardware may optionally be compatible with... Figure 17 The hardware shown is different. In this embodiment, the operating system 1730 may include one of a plurality of virtual machines (VMs) hosted on the computer 1702. Furthermore, the operating system 1730 may provide a runtime environment for the application 1732, such as the Java Runtime Environment or the .NET Framework. A runtime environment is a consistent execution environment that allows the application 1732 to run on any operating system that includes that runtime environment. Similarly, the operating system 1730 may support containers, and the application 1732 may be in the form of a container, which is a lightweight, standalone executable software package that includes, for example, code, runtime, system tools, system libraries, and settings for the application.

[0159] Furthermore, computer 1702 may include a security module, such as a Trusted Processing Module (TPM). For example, using a TPM, the boot component hashes the next boot component and waits for the result to match a security value before loading the next boot component. This process can be performed at any level of the computer 1702's code execution stack, such as at the application execution level or the operating system (OS) kernel level, thereby achieving security at any level of code execution.

[0160] A user can enter commands and information into the computer 1702 through one or more wire / wireless input devices, e.g., a keyboard 1738, a touch screen 1740, and a pointing device, such as a mouse 1742. Other input devices (not shown) can include a microphone, an infrared (IR) remote control, a radio frequency (RF) remote control, or other remote control, a joystick, a virtual reality controller and / or virtual reality headset, a gamepad, a stylus, an image input device, e.g., a camera(s), a gesture sensor input device, a visual motion sensor input device, an emotion or facial detection device, a biometric input device, e.g., a fingerprint or iris scanner, etc. These and other input devices are often connected to the processing unit 1704 through an input device interface 1744 that can be coupled to the system bus 1708, but can be connected by other interfaces such as a parallel port, an IEEE 1394 serial port, a game port, a USB port, an IR interface, a BLUETOOTH® interface, etc.

[0161] A monitor 1746 or other type of display device can also be connected to the system bus 1708 via an interface, such as a video adapter 1748. In addition to the monitor 1746, a computer typically includes other peripheral output devices (not shown), such as speakers, printers, etc.

[0162] The computer 1702 can operate in a networked environment using logical connections to one or more remote computers, such as a remote computer(s) 1750. The remote computer(s) 1750 can be a workstation, a server computer, a router, a personal computer, portable computer, microprocessor-based entertainment appliance, a peer device or other common network node, and typically includes many or all of the elements described relative to the computer 1702, although, for purposes of brevity, only a memory / storage device 1752 is illustrated. The logical connections depicted include wire / wireless connectivity to a local area network (LAN) 1754 and / or larger networks, e.g., a wide area network (WAN) 1756. Such LAN and WAN networking environments are commonplace in offices and companies and facilitate enterprise-wide computer networks, such as intranets, all of which can connect to a global communications network, e.g., the Internet.

[0163] When used in a LAN networking environment, the computer 1702 can be connected to the local network 1754 through a wire / wireless communication network interface or adapter 1758. The adapter 1758 can facilitate wire or wireless communication to the LAN 1754, which can also include a wireless access point (AP) disposed thereon for communicating in wireless mode with the adapter 1758.

[0164] When used in a WAN networking environment, computer 1702 may include modem 1760, or may be connected to a communication server on WAN 1756 via other means of establishing communication on WAN 1756, such as via the Internet. Modem 1760 may be built-in or external, and may be a wired or wireless device, connected to system bus 1708 via input device interface 1744. In a networking environment, program modules associated with computer 1702 or parts thereof may be stored in remote memory / storage device 1752. It should be understood that the network connection shown is an example, and other means of establishing communication links between computers may be used.

[0165] When used in a LAN or WAN networking environment, computer 1702 can access cloud storage systems or other network-based storage systems (as a supplement to or replacement for the external storage device 1716 described above). Typically, the connection between computer 1702 and the cloud storage system can be established on LAN 1754 or WAN 1756 (e.g., via adapter 1758 or modem 1760, respectively). After connecting computer 1702 to the associated cloud storage system, external storage interface 1726 can manage the storage provided by the cloud storage system, just like managing other types of external storage, using adapter 1758 and / or modem 1760. For example, external storage interface 1726 can be configured to provide access to cloud storage sources as if these sources were physically connected to computer 1702.

[0166] Computer 1702 can communicate with any wireless device or entity operatively arranged in wireless communication, such as printers, scanners, desktop and / or portable computers, portable data assistants, communication satellites, any device or location associated with a wirelessly detectable tag (e.g., self-service kiosks, newsstands, store shelves, etc.), and telephones. This can include Wi-Fi and BLUETOOTH® wireless technologies. Therefore, communication can be a predefined structure along with a conventional network, or simply self-organizing communication between at least two devices.

[0167] Go to Figure 18The figure shows a block diagram of example UE 1860. UE 1860 may include a smartphone, wireless tablet computer, wireless-enabled laptop computer, wearable device, machine equipment supporting vehicle telematics, tracking device, remote sensing device, etc. UE 1860 includes a first processor 1830, a second processor 1832, and shared memory 1834. UE 1860 includes a radio front-end circuitry 1862, which may be referred to herein as a transceiver, but should generally be understood to include transceiver circuitry, independent filters, and independent antennas for use on wireless links (such as...). Figure 1 Transceiver 1862 facilitates the transmission and reception of signals on one or more wireless links 125, 135, and 137 shown. Furthermore, transceiver 1862 may include multiple sets of circuitry or may be tuned to accommodate different frequency ranges, different modulation schemes, or different communication protocols to facilitate long-range wireless links (such as link 135), device-to-device links (such as link 135), and short-range wireless links (such as link 137).

[0168] Continue to Figure 18 As described, UE 1860 may also include SIM 1864 or SIM profile, which may include information stored in memory (memory 1834 or a separate memory section) for facilitating communication with... Figure 1 The wireless communication of RAN 105 or core network 130 shown. Figure 18 The SIM 1864 is presented as a single component within the form of a conventional SIM card; however, it should be understood that the SIM 1864 can represent multiple SIM cards, multiple SIM profiles, or multiple eSIMs, some or all of which can be implemented in hardware or software. It should be understood that a SIM profile may include information such as security credentials (e.g., encryption keys, values ​​that can be used to generate encryption keys), or information between the SIM 1864 and another device (which could be...). Figure 1 Shared values ​​shared between components of RAN 105 or core network 130 (as shown). SIM profile 1864 may also include SIM or SIM profile-specific identification information, such as, for example, the International Mobile Subscriber Identity (“IMSI”) or information constituting the IMSI.

[0169] The SIM 1864 is shown as coupled to both the first processor portion 1830 and the second processor portion 1832. An advantage of this implementation is that the first processor portion 1830 can not need to request or receive information or data from the SIM 1864 that the second processor 1832 can request, thus eliminating the use of the first processor as a "go-between" when the second processor uses information from the SIM to perform its functions and applications. The first processor 1830 (which can be a modem processor or baseband processor) is shown as less than the processor 1832 (which can be a more complex application processor) to intuitively indicate the relative level of complexity (i.e., processing power and performance) and corresponding relative level of operational power consumption between the two processor portions. When the UE 1860 does not need the second processor portion 1832 to perform applications and process data related to the applications, it can be advantageous to keep the second processor portion 1832 in a sleep / inactive / low power consumption state to reduce power consumption (when the UE only needs to monitor regular configured bearer management and mobility management / maintenance procedures using the first processor portion 1830 in a listening mode, or monitor a search space that the UE has been configured to monitor, while the second processor portion remains inactive / sleep state).

[0170] The UE 1860 can also include sensors 1866 such as, for example, temperature sensors, accelerometers, gyroscopes, barometers, humidity sensors, and the like (which can provide signals to the first processor 1830 or the second processor 1832). The output devices 1868 can include, for example, one or more visual displays (e.g., computer displays, VR devices, etc.), acoustic transducers such as speakers or microphones, vibrating components, and the like. The output devices 1868 can include software that interfaces with output devices external to the UE 1860 (e.g., visual displays, speakers, microphones, tactile devices, olfactory or gustatory devices, etc.).

[0171] The glossary of terms given in Table 2 below can apply to one or more of the descriptions of embodiments disclosed herein. Table 2

[0172] The above description includes non-limiting examples of various embodiments. Of course, for the description of the disclosed subject matter, it is not possible to describe all possible combinations and permutations of components or methods, and one skilled in the art can recognize that further combinations and permutations of the various embodiments are possible. The disclosed subject matter is intended to cover all such alterations, modifications, and variations as come within the spirit and scope of the appended claims.

[0173] With respect to various functions performed by the above-described components, devices, circuits, systems, and the like, terms used to describe such components (including references to "means") are also intended to refer to any structure for performing the specified function of that component, i.e., that is functionally equivalent, even though structural equivalents can not be identical to disclosed structures. Furthermore, although certain features of the disclosed subject matter can be disclosed with respect to only one or a few embodiments, such features can be combined with one or more other features of the same or different embodiments as can be desired and advantageous for any given or particular application.

[0174] The terms "exemplary" and / or "illustrative," as used herein, are intended to mean serving as an example, instance, or illustration. For the avoidance of doubt, the subject matter of the present disclosure is not limited by such examples. In addition, any aspect or design described herein as "exemplary" and / or "illustrative" is not necessarily to be construed as being superior to or superior to other aspects or designs, nor is it intended to exclude any equivalents or alternates. Moreover, the use of the terms "including," "containing," "comprising," "having," and other similar words in the detailed description section herein is intended to be broad and under the doctrine of equivalents— similar to the term "comprising" as an open transition word— not to exclude any additional or other elements.

[0175] The term "or," as used herein, is intended to mean an inclusive "or" rather than an exclusive "or." That is, unless specified otherwise, or clear from the context, the phrase "X employs A or B" is intended to mean that X employs A or B or both A and B. Additionally, the terms "a" and "an," as used herein, generally should be interpreted to mean "one or more" unless otherwise indicated or clear from the context to indicate a singular form.

[0176] The term "set," as used herein, does not include an empty set, i.e., a set having no elements. Thus, a "set" in the present disclosure contains one or more elements or entities. Likewise, the term "group," as used herein, refers to a collection of one or more entities.

[0177] The terms "first," "second," "third," etc. as used in the claims, unless otherwise expressly specified, are used for clarity only and do not otherwise limit the scope of the claims. For example, "a first determination," "a second determination," and "a third determination" do not mean that the first determination should be performed before the second determination, and the second determination should be performed before the third determination, and so on, unless otherwise expressly specified.

[0178] The description of the example embodiments of the subject disclosure provided herein, including the description in the Abstract, is not intended to be exhaustive or to limit the disclosed embodiments to the precise form disclosed. While specific embodiments and examples are described herein for illustrative purposes, various modifications are possible within the scope of the embodiments and examples as those skilled in the relevant art will appreciate. In this regard, while the subject matter has been described in connection with various embodiments and corresponding drawings (as applicable), it should be understood that other similar embodiments can be used or modifications and additions can be made to the described embodiments for performing the same, similar, alternative, or substitute function of the disclosed subject matter without deviating from the spirit of the disclosed subject matter. Accordingly, the disclosed subject matter should not be limited to any single embodiment, but instead has broad applicability to the use of other similar embodiments and, in fact, to all equivalent arrangements, modifications, and additions that are within the scope of the disclosed subject matter.

Claims

1. A method comprising: A user equipment including a processor receives a learning model configuration information block message from a first radio access network node. The learning model configuration information block message includes a training configuration resource indication, which indicates training configuration resources available for broadcast training configuration of the learning model by the first radio access network node. The user equipment receives the training configuration of the learning model according to the training configuration resources; as well as The user equipment decodes the training configuration of the learning model.

2. The method of claim 1, wherein decoding of the training configuration of the learning model includes blind decoding.

3. The method of claim 1, wherein the learning model training configuration includes a training action indication, the training action indication being used to indicate a training action to be performed by the user device.

4. The method of claim 1, wherein the learning model training configuration includes at least one timing advance preamble corresponding to a second radio access network node, the second radio access network node being a neighboring radio access network node relative to the first radio access network node, and wherein the training action includes sending one of the at least one timing advance preamble to the second radio access network node, the method further comprising: The user equipment sends one of the at least one timing advance preambles corresponding to the second radio access network node to the second radio access network node, wherein the timing advance preamble can be used by the second radio access network node to obtain at least one updated timing advance learning model parameter corresponding to the timing advance learning model.

5. The method of claim 4, wherein the learning model configuration information block message includes a training result resource indication, the training result resource indication being used by the user equipment to receive the at least one updated time-advance learning model parameter, the method further comprising: The user equipment receives the at least one updated time-advance learning model parameter via the training result resource; as well as Based on the at least one updated timing advance learning model parameter, the user equipment updates the timing advance learning model to obtain the updated timing advance learning model.

6. The method according to claim 5, further comprising: Based on the updated timing advance learning model, the user equipment determines the timing advance relative to the user equipment corresponding to the first radio access network node; The user equipment sends a connection establishment request message, including the timed advance, to the first radio access network node; as well as Based on the connection establishment request message, the user equipment establishes a communication connection with the first radio access network node, thereby placing the user equipment in a connection mode relative to the first radio access network node.

7. The method according to claim 5, further comprising: Based on the updated timing advance learning model, the user equipment determines the timing advance relative to the user equipment corresponding to the second radio access network node; The user equipment sends a connection establishment request message, including the timed advance, to the second radio access network node; as well as Based on the connection establishment request message, the user equipment establishes a communication connection with the second radio access network node, thereby placing the user equipment in a connection mode relative to the second radio access network node.

8. A user equipment, comprising: Processor, the processor being configured to: Receive a learning model configuration information block message from a radio access network node, wherein the learning model configuration information block message includes a training configuration resource indication, the training configuration resource indication being used to indicate training configuration resources available for receiving training configuration of the learning model from the radio access network node; Receive the training configuration of the learning model according to the training configuration resources; The learning model training configuration is decoded, wherein the learning model training configuration includes training action instructions, which are used to indicate training actions to be performed by the user device; Perform the training actions to obtain training results; and The training action results are sent to the radio access network node.

9. The user equipment of claim 8, wherein the training action includes generating a probe reference signal such that the result of the training action becomes the generated probe reference signal, and wherein the generated probe reference signal is sent to the radio access network node, the generated probe reference signal being used by the radio access network node to train an uplink resource grant learning model to obtain a trained uplink resource grant learning model.

10. The user equipment according to claim 9, wherein the processor is further configured to: Establish a communication connection with the radio access network node, wherein the communication connection includes at least one uplink resource authorized by the radio access network node based on the trained uplink resource authorization learning model.

11. The user equipment of claim 10, wherein the authorization of the radio access network node to the at least one uplink resource is based on: the user equipment excluding the transmission of the probe reference signal after the user equipment transmits the generated probe reference signal.

12. The user equipment of claim 8, wherein the radio access network node is a first radio access network node, wherein the user equipment performs the training action relative to the first radio access network node to make the training action result a first training action result, wherein the user equipment sends the first training action result to the first radio access network node; the processor is further configured to: The training action is performed relative to a second radio access network node to obtain the result of the second training action, wherein the second radio access network node is a neighboring radio access network node relative to the first radio access network node; and The results of the second training action are sent to the second radio access network node.

13. A non-transient machine-readable medium comprising executable instructions that, when executed by a processor of a user equipment, facilitate the execution of operations, said operations including: When the user equipment is idle, it receives a learning model configuration information block message from the first radio access network node. The learning model configuration information block message includes training result resources, which can be used by the user equipment to receive training results from the first radio access network node. When the user equipment is idle, it receives a learning model training configuration from a first radio access network node. The learning model training configuration includes a training action instruction, which indicates a training action that can be performed by the user equipment relative to at least the first radio access network node. as well as When the user equipment is idle, it performs the training action relative to the first radio access network node to obtain the first training action result.

14. The non-transient machine-readable medium of claim 13, further comprising: Receive the results of the first training action from the first radio access network node.

15. The non-transient machine-readable medium of claim 13, wherein the operation further comprises: When the user equipment is idle, the training action is performed relative to the second radio access network node to obtain the second training action result, wherein the second radio access network node is a neighboring radio access network node relative to the first radio access network node. as well as Receive the results of the second training action from the second radio access network node.

16. The non-transient machine-readable medium of claim 13, wherein the result of the first training action is used by the first radio access network node to update the learning model.

17. The non-transient machine-readable medium of claim 13, wherein the result of the first training action is used by the user equipment to update the learning model to obtain an updated learning model to be used by the user equipment.

18. The non-transient machine-readable medium of claim 17, wherein the learning model is a beamselection learning model, and wherein the updated learning model is an updated beamselection learning model.

19. The non-transient machine-readable medium of claim 18, further comprising: When the user equipment is idle, the updated beam selection learning model is used to determine a determined preferred service beam corresponding to the first radio access network node, and the determined preferred service beam will be used during the connection establishment corresponding to the first radio access network node. When the user equipment is idle, it sends a connection establishment message to the first radio access network node. The connection establishment message includes a preferred service beam indication, which indicates to the first radio access network node the determined preferred service beam that will be used to establish a connection with the first radio access network node. as well as The connection is established with the first radio access network node, wherein the connection includes the determined preferred serving beam, and wherein establishing the connection with the first radio access network node excludes beam scanning to determine the optimal beam corresponding to the user equipment.

20. The non-transient machine-readable medium of claim 17, wherein the learning model is a time-advance learning model, and wherein the updated learning model is an updated time-advance learning model.

Images