Wireless resource management using machine learning

Abstract

The wireless device (104) is configured to receive a set of sensor data (114) from one or more sensors of the wireless device, receive a set of wireless measurements (116) from the wireless interface of the wireless device, process the set of sensor data and the set of wireless measurements with the wireless resource management (RRM) neural network (112) of the wireless device to generate an output representing an RRM action (118), and then execute the RRM action. The wireless device is further configured to provide a representation of the sensor function (122) of the wireless device to the sensors for reception by infrastructure components (110) of a network infrastructure (102) wirelessly connected to the wireless device, and in response to providing the representation of the sensor function, receive a neural network architecture configuration (124) from the infrastructure components and be able to implement the neural network architecture configuration in the RRM neural network.

Description

[Background Technology]

[0001] Cellular networks and other wireless networks often use Radio Resource Management (RRM) to manage network capacity issues at a large-scale level (e.g., at the level of multiple users or multiple cells) rather than addressing point-to-point network capacity problems. Therefore, RRM utilizes various techniques to provide efficient overall network throughput while ensuring efficient power consumption across various networked components. These techniques may include, for example, those addressing power control, scheduling, cell discovery, cell re-selection, handover, radio link or connection monitoring, connection establishment / re-establishment, and interference management. User equipment (UEs) within a wireless network often play a crucial role in RRM by collecting various radio measurements and other system observations and reporting them to the network for use in implementing RRM actions on the network side, as well as in carrying out specific procedures based on these measurements. In traditional approaches, the service base station (BS) or other components of the network infrastructure dictate that the UE use static or algorithmic configurations for its role in the overall RRM process. This can be done, for example, by configuring the UE to use static schedules for scanning within, between, or between RATs (Radio Access Technologies) of a serving cell or adjacent cell, or by specifying a particular algorithm or fixed threshold set for the UE to use when making conditional handover decisions (CHOs) for switching between cells. This static approach to defining the UE's role in RRM often fails to consider the specific circumstances of the UE and can therefore lead to suboptimal RRM behavior at the UE. This often results in inefficient acquisition of RRM-related information at the UE, and thus can impair the overall RRM decision process that utilizes such information. Furthermore, a suboptimal RRM configuration at the UE can lead to unnecessary power consumption at the UE, as the UE performs various RRM actions that are irrelevant or out of time to the overall efficiency of the RRM process.

[0002] By referring to the accompanying drawings, the present disclosure may be better understood and numerous features and advantages will become apparent to those skilled in the art. The use of the same reference symbols in different drawings indicates similar or identical items.

Brief Description of the Drawings

[0003] [Figure 1] FIG. shows an exemplary wireless system that uses neural network-based RRM management in a wireless device according to some embodiments. [Figure 2] FIG. shows an exemplary configuration of a wireless device that implements a neural network for RRM management on the UE side according to some embodiments. [Figure 3] FIG. shows an exemplary configuration of an infrastructure component that implements a UE neural network selection and update scheme according to some embodiments. [Figure 4] FIG. shows a machine learning module that uses a neural network for use in a neural network-based sensor and transceiver fusion scheme according to some embodiments. [Figure 5] FIG. shows an exemplary operating environment of an RRM scheme in a wireless device according to some embodiments. [Figure 6] FIG. shows an exemplary operating environment of an RRM scheme in an infrastructure component according to some embodiments. [Figure 7] FIG. shows a flowchart of an exemplary method for the configuration and implementation of an RRM neural network in a wireless device according to some embodiments. [Figure 8] FIG. shows a flowchart of an exemplary method for the configuration and implementation of an RRM neural network in a wireless device according to some embodiments. [Figure 9] FIG. is a ladder diagram showing examples of the methods of FIGS. 7 and 8. [Figure 10]This flowchart shows alternative embodiments of the methods in Figures 7 and 8, according to several embodiments. [Figure 11] This flowchart illustrates exemplary methods for configuring and implementing a Conditional Handover (CHO) neural network in wireless devices for managing CHO decision-making in wireless devices, according to several embodiments. [Figure 12] This is a ladder diagram showing an example of the method in Figure 11 according to several embodiments. [Modes for carrying out the invention]

[0004] Static or algorithmic configurations of the UE for RRM operation, as seen in many conventional wireless systems, can lead to poor utilization efficiency of the UE for the overall RRM process, at the cost of excessive power consumption at the UE. Furthermore, configuring the UE to employ these static RRM configurations typically requires considerable design, testing, and implementation effort. As described below with reference to Figures 1-12, static or algorithmic approaches to RRM operation on the UE side can be replaced or complemented by a neural network (NN)-based approach that works to fuse sensor data from the UE's available sensors with radio measurements performed by the UE to arrive at one or more RRM actions used by the UE. Such actions may include, for example, setting the frequency or type of radio measurement to be performed, determining specific conditional RRM actions (such as a conditional handover (CHO) decision or conditional primary cell change (CPC)), performing specific types of measurements, or other RRM actions.

[0005] In at least one embodiment, a base station (BS) (or other infrastructure component) is provided with access to a set of neural network architecture configurations trained using various training datasets that reflect various sensor functions, various sensor data, various RRM-related measurements performed by a UE or other wireless device, etc. During or after the establishment of a wireless connection between the BS and the UE, the UE provides the BS with a representation of its sensor functions, or in particular, sensor functions related to the sensor type used to train neural network architecture configurations accessible to the BS. The BS then selects a neural network architecture configuration based on the UE's indicated sensor functions and instructs the UE to implement the selected neural network architecture configuration in its RRM neural network (e.g., a deep neural network (DNN)).

[0006] When the UE is configured in this way, it captures a series of one or more sensor measurements from a set of sensors and feeds them to the RRM neural network as a set of sensor data. Similarly, the UE captures a series of one or more radio measurements (also called radio parameters or radio metrics), such as signal power measurements from one or both of the serving cells or one or more target cells or neighboring cells, via a separate radio interface from the set of sensors, and feeds the measurements to the RRM neural network as a set of radio measurements. The RRM neural network then uses these inputs to generate outputs that represent one or more RRM actions that the UE will take in response to these input datasets. The RRM actions specified by the output of the RRM neural network may include instructions to take a specific RRM action, such as taking a specified RRM-related measurement, or instructions to take action on a CHO decision (e.g., to initiate a handover to a target cell, or conversely, to postpone any handover), or instructions to refrain from taking a specific RRM action. RRM actions may also include, or alternatively, settings for aspects of the RRM action, such as how often a particular RRM action is performed (e.g., how often data is checked when connected, or how often cell reselection is evaluated when idle), and settings for parameters used in the RRM action (e.g., specifying a particular frequency band used for inter-frequency measurements).

[0007] Furthermore, the UE can report to the BS one or both of the sensor data and / or radio measurements used by the RRM neural network in determining the resulting RRM action, and the BS can then use this feedback to retrain the model or copy of the RRM neural network and provide the UE with an updated version of the RRM neural network for future use. Alternatively, the UE can use this same data and other information to retrain or modify its local copy of the RRM neural network.

[0008] Integrating or fusing recent local sensor data of the UE with recent radio measurements from an RRM neural network facilitates the local RRM process used by the UE to adapt more readily to the UE's current transmission environment or other current circumstances than would be provided through the UE's static / algorithmic RRM configuration. For example, sensor data indicating the proximity of an object, building, or other interference source, as a result of training an implemented neural network architecture, could trigger the RRM neural network to instruct an RRM action that reduces the frequency with which a particular measurement is made using the UE's millimeter-wave (mmW) antenna (and thus saves power), or to use a lower threshold for signal power parameters measured via the mmW antenna before triggering a CHO, thus potentially eliminating an unnecessary handover process. Furthermore, by leveraging a neural network with an architectural configuration trained on sensor data that matches the UE's indicated sensor capabilities, the UE can effectively implement various UE-side RRM actions without requiring the substantial design, testing, and implementation attempts that would otherwise be necessary for implementing conventional static / algorithmic RRM configurations.

[0009] The neural network-based systems and technologies for RRM detailed herein utilize coordination between infrastructure components and wireless devices in a wireless network. For ease of explanation, these systems and technologies are described in the exemplary context of a cellular network in which base stations (BS) function as the aforementioned infrastructure components and UEs function as the aforementioned wireless devices. However, these systems and technologies are not limited to this exemplary embodiment. For example, in the same cellular context, the infrastructure components may be servers or other components separate from the BS, or may include multiple infrastructure components such as a coordinating server and a BS. As another example, in an embodiment of a wireless local area network (WLAN), the aforementioned infrastructure components may be wireless access points (APs), or other “upstream” components from the aforementioned wireless devices connected to the wireless APs.

[0010] Figure 1 shows an exemplary wireless communication network 100 using a neural network-based RRM scheme in several embodiments. In the illustrated example, the wireless communication network 100 is a cellular network including a network infrastructure 102 wirelessly connected to one or more wireless devices such as a UE 104. The network infrastructure 102 includes a core network 106 coupled to one or more wide area networks (WANs) 108 or other packet data networks (PDNs) such as the Internet. The core network 106 is further connected to one or more BS110s, such as BS110-1 and BS110-2. Each BS110 supports wireless communication with one or more wireless devices such as a UE 104 via radio frequency (RF) signaling using one or more applicable RATs as defined by one or more communication protocols or standards. Thus, each BS110 acts as a wireless interface between one or more wireless devices and various networks and services provided by the network infrastructure 102, such as packet-switched (PS) data services and circuit-switched (CS) services. Traditionally, signaling communication from BS110 to UE104 has been referred to as "downlink" or "DL," while signaling communication from UE104 to BS110 has been referred to as "uplink" or "UL."

[0011] Each BS110 can employ any of the following RATs: operating as a NodeB (or Base Transceiver Station (BTS)) of a Universal Mobile Telecommunications System (UMTS) RAT (also known as "3G"), operating as an Extended NodeB (eNodeB) of a 3G Partnership Project (3GPP®) Long-Term Evolution (LTE) RAT, or operating as a 5G NodeB ("gNB") of a 3GPP 5G New Radio (NR) RAT. UE104 then represents any of the various electronic devices capable of operating to communicate with the BS110 via the appropriate RAT, including, for example, a mobile phone, a cellular-enabled tablet or laptop computer, a desktop computer, a cellular-enabled video game system, a server, a cellular-enabled consumer electronics appliance, a cellular-enabled automotive communication system, a cellular-enabled smartwatch, or other wearable device.

[0012] For the purposes described below, in Figure 1, BS110-1 is currently wirelessly connected to UE104 and provides network services via the resulting wireless connection, and is therefore referred to herein as “serving” BS110-1 (also known in the art as “primary” BS). In this example, BS110-2 is not currently providing network services to UE104, but is available for handover and initiation of network service provision, and is therefore referred to as “target” BS110-1 (or known in the art as “adjacent” BS or “secondary” BS).

[0013] The exemplary wireless communication network 100 in Figure 1 shows only one UE 104 and two BS 110 for the sake of clarity, but in a real-world implementation, such a wireless system would have numerous BS and UE 104 operating in close geographical areas, thus increasing the opportunities for RF interference and competition over shared RF resources or network resources. Therefore, in at least one embodiment, the wireless communication network 100 utilizes an RRM scheme to manage network capacity and network resource competition at a large scale. This RRM scheme involves the UE 104 both in performing various measurements that can be used for various RRM processes, and in performing some or all of these RRM processes. As described above, conventional RRM schemes rely on algorithms or static approaches that utilize the UE for RRM-related measurements and operations.

[0014] In contrast, in at least one embodiment, the wireless communication network 100 utilizes a neural network-based RRM scheme in which the UE 104 is configured to adaptively and non-statically implement RRM actions using a trained RRM neural network (NN) 112 based on information about the current context of the UE 104. In detail, in at least one embodiment, the RRM NN 112 receives local sensor data 114 and radio measurement data 116 as separate inputs and provides an output indicating at least one RRM action 118 from at least these inputs.

[0015] Local sensor data 114 includes sensor data acquired from one or more sensors of the UE104's sensor set (see Figure 2), and may be provided as a single-point-of-time sample of sensor data / sensor status from the involved sensors, or as a time series of samples of sensor data from the involved sensors over a fixed or variable sliding time window. It will be recognized that the information captured by the sensors of the UE104's sensor set can reflect the current operating state of the UE104, both with respect to the UE104's physical local RF transmission environment and the UE104's internal operating status. For example, the current situation (and absence) of the UE104 that could impair RF signaling between the UE104 and the serving BS110-1 may be detectable (or otherwise represented) from sensor data generated by object detection sensors such as radar, lidar, or image sensors (e.g., imaging cameras), which generate sensor data reflecting the presence or absence of interfering objects in the line of sight (LOS) propagation path between the serving BS110-1 and the UE104. Similarly, positioning data from a Global Positioning System (GPS) sensor, gyroscope, accelerometer, or camera-based visual-odor sensor system, for example, can find one that positions and / or moves UE104 relative to Serving BS110-1, and thus represent the current RF signal propagation environment. As another example, an optical sensor, image sensor, or touch sensor can provide sensor data indicating the orientation of UE104 relative to the user's body, thereby serving as an indicator of the likely RF signal propagation environment in which UE104 is present. Regarding the current internal operating environment of UE104, a battery power sensor can indicate the remaining battery power and thus the UE's ability to continue performing various tasks without compromising the overall operation of UE104, while a thermal sensor of UE104 can indicate whether UE104 is close to its thermal limit and thus the extent to which UE104 can continue performing various RRM actions that may contribute to its thermal output.Therefore, inputting local sensor data 114 into the trained RRM NN112 facilitates the RRM NN112's decision-making process, allowing it to adapt the resulting input to reflect the current operating environment of the UE104.

[0016] With respect to the radio measurement data 116, this information can also be provided as a single-point-of-time measurement sample, as a time-series measurement sample across a fixed or variable sliding window, or a combination thereof. The radio measurement data 116 reflects measurements of various parameters or metrics of either or both of the received RF signaling or the transmitted RF signaling. Examples of such radio measurements may include reference signal received power (RSRP) measurements, reference signal received quality (RSRP) measurements, signal-to-noise ratio (SNR) measurements, signal-to-noise plus interference ratio (SNIR), signal-to-interference plus noise ratio (SINR) measurements, received signal strength indicator (RSSI) measurements, reference signal received quality (RSRQ) measurements, and thermal interference (IoT) ratio measurements.

[0017] As described above, the output of the RRM NN112 in response to the input of local sensor data 114 and wireless measurement data 116 represents one or more RRM actions 118 performed by the UE104. The RRM actions 118 can be classified into at least one of the following categories: (1) an instruction for the UE104 to perform or refrain from an RRM process, (2) an instruction to change or set parameters of an RRM process performed by the UE, or (3) a determination of a decision regarding a conditional RRM process for the UE104. The first category of RRM actions represents an RRM action 118 in which the UE104 responds by either performing an RRM process or refraining from an RRM process that the UE104 would otherwise perform. For example, based on local sensor data 114 indicating that UE104 is stationary and in a location with relatively few signal interferences, RRM NN112 may output RRM action 118 instructing UE104 to perform a single RRM measurement (e.g., a scan of a specific frequency band) and report the results to serving BS110-1. Conversely, based on local sensor data 114 indicating that UE104 is moving at a very high speed, or that UE104 is indoors and surrounded by substantial RF blocking objects, RRM NN112 may output RRM action 118 instructing UE104 to skip the next scheduled RRM measurement (and thus potentially save power and computing resources of the UE), reflecting the potential futility of attempting a scheduled RRM measurement in the UE's current operating environment.

[0018] A second category of RRM actions represents RRM actions 118 in which UE104 responds by modifying an RRM process that has already been scheduled or instructed to be performed. For example, UE104 may be configured to perform a particular RRM measurement according to a specific timing, and RRM action 118 may represent a change to this timing. For example, the 3GPP 5th Generation New Radio (5G NR) specification provides for comparing the relative signal strength of a serving cell (e.g., provided by serving BS110-1) with an adjacent cell (e.g., provided by target BS110-2) through a synchronous signal (SS) / physical broadcast channel (PBCH) block, or cell signal measurement process using SSB, where the timing, or periodicity, of each consecutive measurement using SSB is controlled via an SSB-based RRM measurement timing configuration (SMTC) window. Therefore, either or both of the input local sensor data 114 or the current wireless measurement data 116 may provide an output representing an RRM action 118 that triggers the trained RRM NN 112 to instruct the UE 104 to adjust the SMTC window, thereby increasing or decreasing the period of SSB-based cell measurements for the serving cell and adjacent cells (as adjusted).

[0019] A third category of RRM actions represents RRM action 118, which represents a decision made for one or more conditional RRM processes that can be performed by UE 104. For example, the 3GPP 5G NR Release 16 specification defines a conditional handover (CHO) RRM process, in which a handover command is sent to the UE along with one or more conditions that are monitored by the UE in relation to the handover command. Instead of immediately performing the handover, the handover command is stored and the UE monitors the specified one or more conditions. Once the monitored conditions are met, the UE then initiates the previously received handover. Thus, with respect to this example, RRM action 118 can mimic a CHO process without requiring that the specific handover conditions be defined statically or algorithmically. Rather, since the RRM NN112 is trained using a training dataset having similar training sensor data and similar training radio measurement data to make decisions like CHO, the RRM action 118 output by the RRM NN112 in the field may include a CHO decision to either initiate a handover or refrain from initiating a handover, based on the current network and UE context reflected in the input local sensor data 114 and radio measurement data 116. Other such conditional RRM process decisions that may be designated as RRM action 118 may include, for example, a conditional primary cell change (CPC), where a primary cell (Pcell) change is performed by the UE if one or more specified conditions are met.

[0020] The RRM action 118 may also be a combination of two or more of the above three categories. That is, the RRM action 118 may be a hybrid RRM action. By way of example, the output of the RRM NN112 may represent the RRM action 118, including activating a previously deactivated RRM measurement process and setting one or more parameters for that RRM measurement process (e.g., activating an inter-RAT scan and setting the frequencies to be scanned).

[0021] RRM NN112 may also receive other inputs that help generate an output representing RRM action 118. For example, the 5G NR specification is specific to a combination of the UE's RRC state (IDLE, INACTIVE, CONNECTED) and whether the UE's 5G NR RAT is in standalone (SA) mode or non-standalone (NSA) mode. For example, if the UE is in the RRC IDLE state and in SA mode, the UE is permitted to perform cell selection or cell reselection, but if the UE is in NSA mode while in the RRC IDLE state, the 5G NR specification means that the UE is intended to refrain from cell selection or cell reselection. Similarly, if the UE is in RRC CONNECTED, the UE can perform random access to primary cells regardless of whether the 5G NR RAT is in SA or NSA mode, but the 5G NR specification means that the UE is intended to perform handover only if it is in SA mode while in this RRC state. Therefore, UE104 can provide operational state data 120 as input to RRM NN112, which represents specific operational state parameters of UE104 and / or the RAT of UE104, such as the RRC state and SA / NSA mode of the 5G NR RAT of UE104. In this example, RRM NN112 can be trained using training data to avoid providing an output that triggers an RRM action 118 that represents an RRM process inconsistent with the limitations of the RRC state and SA / NSA mode for a particular RRM measurement.

[0022] For UE104 to utilize RRM NN112 for RRM actions, UE104 is initially configured with RRM NN112. Furthermore, in some embodiments, the underlying architectural configuration of RRM NN112 can be updated either via local updates or remote updates performed through reporting of relevant information by UE104. View 121 in Figure 1 provides an overall overview of the initial setup and update process for UE104 using RRM NN112. In at least one embodiment, the RRM NN112 used by UE104 is trained using training data similar to the data inputs expected to be provided by UE104. As described above, data input to RRM NN112 includes local sensor data 114 generated by a specific set of sensors available to UE104 and radio measurement data 116 representing specific types of radio measurements that UE104 can perform. Therefore, for an RRM NN112 adopted by UE104 that is trained using training data from similar sensors, it is advantageous to select an NN architectural configuration trained for similar types of radio measurements. For example, employing an NN architecture configuration of RRM NN112 trained with training data heavily influenced by radar sensor data would typically yield less effective results in a UE without a radar sensor. For this purpose, serving BS110-1 stores, or otherwise accesses, a set of various NN architecture configurations, each trained according to a specific sensor configuration and / or radio measurement configuration. During the initialization process between serving BS110-1 and UE104, UE104 provides serving BS110-1 with a representation of its functions (UE function message 122). The UE function message 122 includes representations of various functions of UE104, including either or both representations of UE104's sensor functions or representations of UE104's radio measurement functions.The representation of the sensor functions of UE104 may include, for example, the representation of the types of sensors included in the sensor set of UE104, the functions or other parameters of each of some or all of the sensors. For example, during the connection process, the serving BS110-1 may send a UE Capabilities Enquiry radio resource control (RRC) message to UE104, and UE104 may use one or more fields of the UECapabilitiesInformation RRC message, which includes data or other information representing the types, quantities, and parameters of the sensors in the sensor set of UE104, to respond with the UECapabilitiesInformation RRC message.

[0023] As shown in FIG. 1, the serving BS110-1 (or another infrastructure component such as a server in the core network 106) then selects a UE104 RRM NN architecture configuration 124 appropriate for the indicated sensor functions and / or radio measurement functions using the sensor and / or radio measurement functions of UE104 shown in the UE function message 122. This selection can be performed using any of a variety of techniques, such as using one or more lookup tables (LUTs) indexed by the indicated functions through a weighted evaluation that compares the functions indicated by the sensor type / parameters and / or radio measurement type used to train the corresponding candidate RRM NN architecture configurations, etc. Next, the serving BS110-1 signals to UE104 to use the selected RRM NN architecture configuration 124. For example, in some embodiments, the serving BS110-1 may send one or more messages to UE104 using data representing the actual RRM NN architecture configuration 124. In other embodiments, UE104 may be pre-provided with a set of stored RRM NN architecture configurations, and then the serving BS110-1 may send a message including an index or other identifier of the specific RRM NN architecture configuration utilized from its stored set by UE104.

[0024] In response to the identification or provision of an RRM NN architecture configuration 124, UE 104 configures RRM NN 112 to utilize the identified RRM NN architecture configuration 124 and begins operating RRM NN 112 as configured. During operation, and in accordance with many RRM schemes, UE 104 may provide and return various RRM reports 126 to service BS 110-1. This RRM report 126 may include a report of RRM measurements made by UE 104, including those directed by or configured by RRM NN 112. The RRM report 126 may further include a report of decisions or settings reflected in RRM actions 118 output by RRM NN 112. For example, UE 104 may report via RRM report 126 that, as a result of an RRM action 118 output by RRM NN 112, the UE changed the frequency at which it performs inter-RAT or intra-RAT scans. Next, Serving BS110-1 can take local action in response to this RRM report 126, or it can forward the RRM report 126 to the core network 106 for further consideration by the network infrastructure 102.

[0025] In some embodiments, the RRM NN architecture configuration used by RRM NN112 can be dynamically updated based on usage. In some embodiments, NN updates to RRM NN112 are provided by Serving BS110-1 (or other infrastructure components). In this method, UE104 can provide Serving BS110-1 with a copy 128 of local sensor data 114 and / or radio measurement data 116, and Serving BS110-1 or other components can use this data, for example, along with RRM reports 126, to retrain the NN architecture configuration 124 used by UE104 to generate an updated NN architecture configuration, and provide this updated NN architecture configuration to UE104 for implementation as NN update 130. In other embodiments, NN updates are performed by UE104 itself. For example, UE104 may be configured to periodically provide Serving BS110-1 with a copy, or "snapshot," of the current state of the NN architecture configuration of RRM NN112 as NN update 130. Next, the serving BS110-1 can update its local or accessible copy of the RRM NN architecture configuration 124 to reflect the supplied NN update 130, and further, it can redistribute this updated NN architecture configuration to other UEs employing the same NN architecture configuration 124.

[0026] Figure 2 shows exemplary hardware configurations of the UE104 (as a typical wireless) in several embodiments. Note that the hardware configurations shown represent the processing and communication components most directly related to the neural network-based processes described herein, and omit certain components that are well known to be frequently implemented in such electronic devices, such as displays, non-sensor peripherals, and external power supplies.

[0027] In the illustrated configuration, the UE104 includes an RF front-end 202 having one or more antennas 203, and a radio interface 204 including one or more modems supporting one or more RATs. The RF front-end 202 effectively acts as a physical (PHY) transmit / receive interface, conducting and processing signaling between one or more processors 206 of the UE104 and the antennas 203 to facilitate various types of wireless communication. The antennas 203 can be arranged in one or more arrays of multiple antennas configured similarly or differently from one another, and can be tuned to one or more frequency bands associated with the corresponding RATs. The one or more processors 206 may include, for example, one or more central processing units (CPUs), graphics processing units (GPUs), tensor processing units (TPUs), or other application-specific integrated circuits (ASICs). For example, the processors 206 may include application processors (APs) used by the UE104 to run the operating system and various user-level software applications, as well as one or more processors used by the modem or baseband processor of the radio interface 204.

[0028] UE104 further includes one or more computer-readable media 208, which include any of the various media used by electronic devices to store data and / or executable instructions, such as random access memory (RAM), read-only memory (ROM), cache, flash memory, solid-state drive (SSD), or other mass storage devices. For ease of explanation and brevity, computer-readable media 208 are referred to herein as “memory 208” considering that system memory or other memory frequently uses to store data and instructions for execution by the processor 206, but references to “memory 208” are understood to apply equally to other types of storage media unless otherwise noted.

[0029] In at least one embodiment, the UE104 is further separate from the radio interface 204 and includes a number of sensors collectively referred to herein as a sensor set 210, at least some of which are utilized in the neural network-based scheme described herein. Generally, the sensors of the sensor set 210 include sensors that sense some aspect of the external RF environment of the UE104 (i.e., sensors that may sense parameters that affect, or reflect, at least to some extent, the RF propagation path of the UE104 or RF transmission / reception performance), and sensors that sense several aspects of the current operating status of the UE104, such as battery status, thermal status, operating mode, and screen status. Thus, the sensors of the sensor set 210 may include one or more sensors 212 for object detection, such as radar sensors, lidar sensors, image sensors, structured light-based depth sensors, and proximity sensors. The sensor set 210 may also include one or more sensors 214 for determining the position, attitude, or velocity / speed of the UE104, such as satellite positioning sensors like GPS sensors, Global Navigation Satellite System (GNSS) sensors, Internal Measurement Unit (IMU) sensors, visual / odor sensors, accelerometers, gyroscopes, barometers, altimeters, tilt sensors or other inclinometers, and ultra-wideband (UWB) based sensors. Other examples of sensor types in the sensor set 210 may include sensors 216 for determining the current operating status of the UE104, such as battery level sensors, temperature sensors, and screen mode sensors. It should be noted that, although not shown, the UE104 may further include one or more batteries or other portable power sources, one or more user interface (UI) components such as touchscreens, user-operable input / output devices (e.g., "buttons" or keyboards), or other touch / contact sensors, microphones, or other audio sensors for capturing audio content.

[0030] One or more memories 208 of the UE104 are used to store one or more sets of executable software instructions and associated data for operating one or more processors 206 and other components of the UE104, and to perform various functions described herein and belonging to the UE104. Examples of sets of executable software instructions include operating systems (OS) and various drivers (not shown), as well as various software applications. The sets of executable software instructions further include one or more of the neural network management module 218, function management module 220, or RRM module 222. The neural network management module 218 implements one or more neural networks in the UE104, as will be described in detail below. The function management module 220 determines various functions of UE104 that may be related to the neural network architecture configuration or selection, reports such functions to Serving BS110-1 (e.g., in one or more UECapabilitiesInformation RRC messages), and monitors UE104 for changes in such functions, including changes in RF and processing functions, accessory availability or functionality, and manages the reporting of such functions to Serving BS110-1 and changes in functions. The RRM module 222 operates to execute an RRM process on UE104, including RRM measurement and reporting, and RRM action(s) 118 specified by the output of RRM NN112.

[0031] To facilitate the operation of the UE104 described herein, one or more memories 208 of the UE104 may further store data related to these operations. This data may include, for example, one or more neural network architecture configurations 224 (RRM NN architecture configuration 124, embodiment of Figure 1), and device data 226. Device data 226 may represent, for example, user data, multimedia data, beamforming codebooks, and software application configuration information. Device data 226 may further include functional information of the UE104, such as sensor function information regarding one or more sensors of the sensor set 210, including the presence or absence of a particular sensor or sensor type, and for present sensors, one or more representations of the corresponding type and function, such as the range and resolution of a LiDAR or radar sensor, or the image resolution and color depth of an imaging camera. Functional information may further include, for example, information regarding the function or status of the battery, and the function or status of the wireless interface 204 and antenna(s) 203 (e.g., frequency function, radio measurement function, etc.).

[0032] Each neural network architecture configuration 224 includes one or more data structures containing data and other information that represent the corresponding architecture and / or parameter configuration used by the neural network management module 218 to form the corresponding RRM neural network 112 of the UE 104. The information contained in the neural network architecture configuration 224 includes, for example, parameters that specify the total connected layer neural network architecture, the convolutional layer neural network architecture, the recurrent neural network layers, the number of connected hidden neural network layers, the input layer architecture, the output layer architecture, the number of nodes used by the neural network, the coefficients used by the neural network (such as weights and biases), kernel parameters, the number of filters used by the neural network, the stride / pooling configuration used by the neural network, the activation function of each neural network layer, the interconnections between neural network layers, and the neural network layers to skip. Therefore, the neural network architecture configuration 224 includes any combination of neural network forming components (e.g., architecture and / or parameter configurations) that can be used to create a neural network architecture configuration (e.g., a combination of one or more neural network forming components) that defines and / or forms a DNN or other neural network.

[0033] Figure 3 shows exemplary hardware configurations of BS110 (as a representative infrastructure component) in several embodiments. Note that the hardware configurations shown represent the processing and communication components most directly related to the neural network-based processes described herein, and omit certain components that are well known to be frequently implemented in such electronic devices, such as displays, non-sensor peripherals, and external power supplies. Furthermore, while the illustrated diagrams represent embodiments of BS110 as a single network node (e.g., 5G NR node B, i.e., "gNB"), note that the functions, and therefore the hardware components of BS110, may instead be distributed across multiple network nodes or devices in a manner that performs the functions described herein.

[0034] In the illustrated configuration, the BS110 includes an RF front-end 302 with one or more antennas 303 and a radio interface 304 with one or more modems supporting one or more RATs, which acts as a PHY transceiver interface to facilitate various types of wireless communication by performing and processing signaling between one or more processors 306 of the BS110 and the antennas 303. The antennas 303 can be arranged in one or more arrays of multiple antennas configured similarly or differently from one another and can be tuned to one or more frequency bands associated with the corresponding RATs. The one or more processors 306 can include, for example, one or more CPUs, GPUs, TPUs, or other ASICs. The BS110 further includes one or more computer-readable media 308 containing any of the various media used by electronic devices to store data and / or executable instructions, such as RAM, ROM, cache, flash memory, SSD, or other mass storage devices. Similar to memory 208 of UE104, for the sake of ease and brevity of explanation, computer-readable medium 308 is referred to herein as “memory 308” considering that it frequently uses system memory or other memory to store data and instructions for execution by processor 306, but references to “memory 308” are understood to apply equally to other types of storage medium unless otherwise noted.

[0035] One or more memories 308 of the BS110 are used to store one or more sets of executable software instructions and related data for operating one or more processors 306 and other components of the BS110, and to perform various functions described herein and attributed to the BS110. Examples of sets of executable software instructions include the OS and various drivers (not shown), as well as various software applications. The sets of executable software instructions further include one or more of the neural network management module 310, training module 312, and RRM module 314. One or more memories 308 further store one or more sets 322 of candidate neural network architecture configurations 324 (embodiments of RRM NN architecture configuration 124, Figure 1), as well as various information such as various BS data 326. The BS data 326 represents, for example, beamforming codebooks, software application configuration information, RRM scheme information, etc. The neural network architecture configuration 324 represents a trained neural network architecture configuration that can be used in the RRM NN112 of UE014 or other UEs. Therefore, similar to the neural network architecture configuration 224 in Figure 2, each candidate neural network architecture configuration 324 includes one or more data structures containing data and other information representing the corresponding architecture and / or parameter configuration, which the neural network management module 218 of a UE, such as UE104, uses to form the corresponding RRM NN112 in the UE. The neural network management module 310 manages the training, retraining, selection, and distribution of the neural network architecture configurations 324 to UE104 and other UEs. The training module 312 performs the actual training / retraining of the selected neural network architecture configuration 324 using sensor data, radio measurement data, and other feedback from one or more UEs. The RRM module 314 operates to perform various RRM processes performed by BS110.

[0036] Figure 4 shows exemplary machine learning (ML) modules 400 for implementing neural networks in several embodiments. As described herein, UE104 implements one or more DNNs or other neural networks as RRM NN112 to configure or control the RRM process in UE104. In connection therewith, BS110 trains, retrains, or otherwise updates copies of one or more DNNs or other neural networks used as RRM NN112 in one or more UEs. Thus, ML module 400 shows exemplary modules for implementing one or more of these neural networks.

[0037] In the illustrated example, the ML module 400 implements at least one deep neural network (DNN) 402 having groups of connected nodes (e.g., neurons and / or perceptrons) organized into three or more layers. The nodes between layers can be configured in various ways, such as a partially connected configuration where a first subset of nodes in the first layer is connected to a second subset of nodes in the second layer, or a fully connected configuration where each node in the first layer is connected to each node in the second layer. Neurons process input data and produce continuous output values, such as any real number between 0 and 1. In some cases, the output value indicates how close the input data is to a desired category. Perceptrons perform linear classification, such as binary classification, on the input data. Nodes, whether neurons or perceptrons, can use a variety of algorithms to generate output information based on adaptive learning. Using DNN402, ML Module 400 performs a variety of different types of analysis, including single linear regression, multiple linear regression, logistic regression, stepwise regression, binary classification, multi-class classification, multivariate adaptive regression splines, local estimation scatter plot smoothing, and more.

[0038] In some embodiments, the ML module 400 learns adaptively based on supervised learning. In supervised learning, the ML module 400 receives various types of input data as training data. The ML module 400 processes the training data to learn how to map inputs to desired outputs. As an example, the ML module 400 receives sequences of training data in the form of training sensor data and training radio measurement data, and actually learns how to map the input training data to desired outputs, i.e., RRM actions. In detail, during the training procedure, the ML module 400 uses labeled data or known data as input to the DNN 402. The DNN 402 uses nodes to analyze the input and generate corresponding outputs. The ML module 400 compares the corresponding outputs to true data and adapts the algorithms implemented by the nodes to improve the accuracy of the output data. The DNN 402 then applies the adapted algorithms to unlabeled input data to generate corresponding output data. The ML module 400 maps inputs to outputs using either or both statistical analysis and / or adaptive learning. For example, the ML module 400 uses characteristics learned from training data to correlate an unknown input to a statistically likely output within a threshold range or threshold. This allows the ML module 400 to receive complex inputs and identify the corresponding output. As mentioned above, several embodiments train the ML module 400 on characteristics of RRM decisions based on input sensor data and radiometric data. This enables the trained ML module 400 to receive a set of sensor data (either as sensor data from a single time slice or sensor data over a sequence of time slices) and a set of radiometric data (either from a single time slice or over a sequence of time slices), and from these inputs to generate an output representing the RRM action(s) to be performed.

[0039] In the described example, the DNN 402 includes an input layer 404, an output layer 406, and one or more hidden layers 408 positioned between the input layer 404 and the output layer 406. Each layer has any number of nodes, and the number of nodes between layers may be the same or different. That is, the input layer 404 may have the same and / or different number of nodes as the output layer 406, the output layer 406 may have the same and / or different number of nodes as the one or more hidden layers 408, and so on.

[0040] Node 410 corresponds to one of several nodes included in the input layer 404, and each node performs a separate, independent computation. As will be further explained, a node receives input data and processes it using one or more algorithms to produce output data. Typically, the algorithms include weights and / or coefficients that change based on adaptive learning. Thus, the weights and / or coefficients reflect the information learned by the neural network. Each node can, in some cases, decide whether to pass the processed input data to one or more subsequent nodes. For example, after processing the input data, node 410 can decide whether to pass the processed input data to one or both of nodes 412 and 414 in the hidden layer 408. Alternatively or additionally, node 410 passes the processed input data to nodes based on the layer-connected architecture. This process can be repeated across multiple layers until the DNN 402 produces an output using a node in the output layer 406 (e.g., node 416).

[0041] Neural networks can also use various architectures that determine which nodes in the neural network are connected, how data is advanced and / or held within the neural network, what weights and coefficients are used to process the input data, and how the data is processed. These various elements collectively describe neural network architecture configurations, such as the neural network architecture configurations briefly described above. For example, recurrent neural networks, such as Long Short-Term Memory (LSTM) neural networks, form cycles between node connections to hold information from the previous part of the input data sequence. The recurrent neural network then uses the held information for the subsequent part of the input data sequence. As another example, feedforward neural networks advance connections by transferring information without forming cycles to hold information. Although described in the context of node connections, it should be recognized that neural network architecture configurations can include various parameter configurations that affect how DNN402 or other neural networks process input data.

[0042] The neural network architecture configuration of a neural network can be characterized by various architectures and / or parameter configurations. To illustrate this, consider the example of DNN402, which implements a convolutional neural network (CNN). In general, a convolutional neural network corresponds to a type of DNN in which layers process data using convolutional operations and filter the input data. Therefore, a CNN architecture configuration can be characterized, for example, by pooling parameters(s), kernel parameters(s), weights, and / or layer parameters(s).

[0043] Pooling parameters correspond to parameters that specify pooling layers within a convolutional neural network, reducing the dimensionality of the input data. For example, a pooling layer can combine the outputs of nodes in a first layer into the inputs of nodes in a second layer. Alternatively or additionally, pooling parameters specify how and where the neural network pools data within the layers of data processing. For example, a pooling parameter indicating "maximum pooling" configures the neural network to pool data by selecting the maximum value from a group of data generated by the nodes of the first layer, and using that maximum value as the input to a single node in the second layer. A pooling parameter indicating "average pooling" configures the neural network to generate an average value from a group of data generated by the nodes of the first layer, and using that average value as the input to a single node in the second layer.

[0044] Kernel parameters indicate the filter size (e.g., width and height) used to process the input data. Alternatively or additionally, kernel parameters specify the type of kernel method used to filter and process the input data. Support vector machines, for example, correspond to kernel methods that use regression analysis to identify and / or classify data. Other types of kernel methods include Gaussian processes, canonical correlation analysis, and spectral clustering methods. Thus, kernel parameters can indicate the filter size and / or the type of kernel method to apply to the neural network. Weight parameters specify the weights and biases used by the algorithm within the node to classify the input data. In some embodiments, the weights and biases are learned parameter settings, such as parameter settings generated from training data. Layer parameters specify the layer connections and / or layer types, such as fully connected layer types, which indicate that all nodes of a first layer (e.g., output layer 406) are connected to all nodes of a second layer (e.g., hidden layer 408); partially connected layer types, which indicate which nodes of the first layer are disconnected from the second layer; and activated layer types, which indicate which filters and / or layers are activated within the neural network. Alternatively or additionally, layer parameters specify the type of node layer, such as normalization layer type, convolutional layer type, and pooling layer type.

[0045] While pooling parameters, kernel parameters, weight parameters, and layer parameters are discussed in this context, it should be noted that other parameter settings can be used to form a DNN that conforms to the guidelines provided herein. Thus, the neural network architecture configuration can include any appropriate type of configuration parameters applicable to the DNN that influence how the DNN processes input data and produces output data.

[0046] In some embodiments, the configuration of the ML module 400 is further based on the sensor and / or radio measurement capabilities of the UE implementing the ML module 400. Thus, the architectural configuration of the ML module 400 may also be based on the capabilities of the UE implementing the ML module 400. For example, UE 104 may have a significant imaging capability, thereby allowing the ML module 400 of UE 104 to be trained on image data as input, thereby facilitating, for example, the generation of RRM actions that are fairly suitable for RF transmission environments, depending on the presence or absence of objects or other interference sources represented in such image data. However, in the case of a UE without an imaging capability, using an ML module 400 configured via training extensively on training image data would generally not be very suitable for generating effective RRM actions based on input sensor data without imaging data. Thus, in some embodiments, the device implementing the ML module 400 may be configured to implement different neural network architecture configurations for different combinations of sensor capabilities, radio measurement capabilities, or both. For example, a device may have access to one or more neural network architecture configurations for use when a radar sensor is available to the device, and to one or more different sets of neural network architecture configurations for use when a radar sensor is unavailable, but the radar sensor is available.

[0047] In some embodiments, the UE 104 implementing the ML module 400 locally stores some or all of the set of candidate neural network architecture configurations that can be used for the ML module 400. For example, candidate neural network architecture configurations 224 may be indexed in the UE 104 by a lookup table (LUT) or other data structure that takes one or more parameters as input, such as one or more sensor function parameters or radio measurement functions, and outputs an identifier associated with the corresponding locally stored candidate neural network architecture configuration that is suitable for operation considering the input parameter(s). In other embodiments, the UE 104 provides a representation of its functions, and the BS 110 or other infrastructure component selects a neural network architecture configuration to be implemented in the UE 104 based on these functions. To facilitate the process of selecting an appropriate neural network architecture configuration, in at least one embodiment, the BS 110 or other infrastructure component trains different versions of the ML module 400 using a neural network management module 310 and a training module 312. For example, the training module 312 can mathematically generate training data, access files storing the training data, and acquire real-world communication data. Next, the neural network management module 310 extracts and stores various trained neural network architecture configurations for subsequent use. In some embodiments, input characteristics are stored for each neural network architecture configuration, thereby describing various sensor characteristics and / or wireless measurement characteristics.

[0048] As described above, wireless devices such as the UE104 can be configured to determine one or more RRM actions using one or more RRM DNNs, each RRM DNN complementing or replacing one or more functions that were conventionally implemented by one or more hardcoded or fixed design blocks. Furthermore, each DNN can further incorporate current sensor data from one or more sensors of the wireless device's sensor set to modify or otherwise adapt its operation to actually take into account the current RF signal propagation environment or the operating state of the wireless device reflected in the sensor data. For this purpose, Figure 5 shows an exemplary operating environment 500 for a DNN implementation in the UE104. In the illustrated example, the neural network management module 218 of the UE104 implements an RRM processing module 502 for a general RRM decision-making process. Furthermore, to illustrate a specific use case of neural network-based RRM management in UE, an exemplary embodiment of the neural network management module 218 is shown as additionally implementing a CHO processing module 504, which is a specific exemplary embodiment of the RRM processing module for the specific purpose of conditional RRM decision-making, which in this case is the CHO's decision-making. Thus, the CHO processing module 504 can be understood as a separate RRM processing module implemented specifically for the CHO's decision-making, or alternatively, as that part of the RRM processing module 502 that brings about the RRM actions that influence the CHO's decision-making.

[0049] As an overview of the general operation of UE104 in relation to the operating environment 500 depicted in Figure 5, the neural network management module 218 supplies the neural network architecture configuration 224 (Figure 2) to the RRM processing module 502 and the CHO processing module 504, respectively, based on the sensor functions of the sensors of the sensor set 210 and / or the radio measurement functions of the radio interface 204. According to their respective schedules, the sensor set 210 provides a recent local sensor dataset 514 (one embodiment of local sensor data 114, Figure 1) acquired from one or more sensors of the sensor set 210, and the radio interface 204 provides a recent radio measurement dataset 516 (one embodiment of radio measurement data 116, Figure 1) to the neural network management module 218, which then provides these datasets as input to the RRM processing module 502 and the CHO processing module 504, respectively. Other inputs, such as the current RRC state (IDLE, INACTIVE, CONNECTED) or RAT mode (e.g., standalone (SA) or non-standalone (NSA) of a 5G NR RAT), may also be provided to processing modules 502 and 504.

[0050] The RRM processing module 502 utilizes these inputs to generate at least one RRM action 518 (an embodiment of RRM action 118), which is provided to the RRM module 222 for implementation. As described above, the RRM action 518 can represent an RRM process to be performed, such as performing an RRM measurement and reporting it to the serving BS 110-1; an RRM process to be skipped, such as skipping a scheduled cell reselection; a modification made to one or more RRM processes, such as resetting the measurement cycle within the RAT; or a combination thereof. Thus, the RRM module 222 controls one or more parameters and other controls of the radio IF 204 in order to implement the RRM action 518. For example, the periodicity of inter-RAT scans may be implemented as a value stored in a register of the radio IF 204, and the RRM module 222 may implement an RRM action 518 that attempts to change this periodicity by overwriting the register with a new value. As another example, the wireless interface 204 may provide an application programming interface (API) or other interface, and the RRM module 222 may trigger the wireless interface 204 to perform an RRM measurement represented by the RRM action 518 by triggering the RRM measurement via the API or other interface.

[0051] Simultaneously (or as part of the RRM processing module 502), the CHO processing module 504 utilizes its inputs to generate outputs representing CHO actions 520, which are similarly provided to the RRM module 222 for implementation. CHO actions 520 may include, for example, CHO decisions such as a decision to initiate a handover based on handover conditions interpreted through the local sensor dataset 514 and the radio measurement dataset 516, or decisions to postpone the initiation of a handover based on handover conditions interpreted through their inputs. Additionally or alternatively, CHO actions 520 may include modifications to static CHO configurations. For example, BS110 may send a message to UE104 to implement a CHO in which a handover is initiated when a specified set of monitoring conditions is met, and CHO action 520 may include modifications to thresholds or other triggers represented by this set of monitored conditions. For example, BS110 may specify that a handover is initiated when the measured RSRP is below a specified threshold. However, the CHO processing module 504 may have been trained using training data that reflects a situation where the measured RSRP of the UE is low but the speed is high, and the measured RSRP may change rapidly as the position of the UE relative to the serving cell changes rapidly. Thus, the training results may configure the CHO processing module 504 to function in order to correct the RSRP threshold before a handover is triggered, if the input local sensor dataset 514 includes sensor data indicating that the UE 104 is moving rapidly, and the radio measurement dataset 516 indicates that the measured RSRP bounces near the original specified threshold. In either method, the RRM module 222 receives the CHO action 520 and coordinates with the radio interface 204 to carry out the process represented by the CHO action 520, such as correcting one or more monitored parameters of conditions that the radio interface 204 monitors for the purpose of determining whether to initiate a handover, refrain from initiating a handover, or whether the radio interface 204 should initiate a handover.

[0052] Figure 6 shows an exemplary operating environment 600 of BS110 (e.g., Serving BS110-1) for supporting neural network-based RRM in UE104 according to at least one embodiment. As previously stated, for the neural network-based RRM process, BS110 operates, in some embodiments, to provide UE104 with a neural network architecture configuration suitable for specific sensor functions and / or radio measurement functions of UE104, to train / retrain the neural network architecture configuration, or a combination of both. For the selection and provision of an initial DNN architecture configuration to UE104, BS110 either implements a neural network architecture configuration data store 602 or has networked access to the data store 602 via a server or other infrastructure component of the network infrastructure 102 (Figure 1). The data store 602 stores one or more candidate neural network architecture configurations 324 (Figure 3) that can be indexed or otherwise accessed based on corresponding sensor function attributes, radio measurement attributes, etc. Therefore, if UE104 supplies UE function information 604 to BS110 indicating one or both of its sensor function and / or radio measurement function, the RF front end 302 of BS110 passes the UE function information 604 to the RRM module 314, which uses the UE function information 604 to select a neural network architecture configuration 324 from the data store 602 that has been trained using training data having the same or similar sensor function and / or the same or similar radio measurement function. This selected neural network architecture configuration 324, or its identifier or other representation, is then transmitted to UE104 for implementation.

[0053] Furthermore, in response to providing the UE104 with a neural network architecture configuration 324, in some embodiments, the neural network management module 310 of the BS110 can instruct the training module 312 to instantiate a training processing module 606 that implements the DNN or other neural network initially configured with the selected neural network architecture configuration 324 provided to the UE104. That is, the training processing module 606 is in fact a copy of the RRM NN112 running concurrently on the BS110. The UE104 then provides feedback in the form of a copy of the local sensor dataset 514 input to the RRM NN112, a copy of the radio measurement dataset 516 input to the RRM NN112, and / or an RRM report 608 from the UE104, which is provided in response to the UE104 performing various RRM processes as a result of the RRM actions 118 generated by the RRM NN112 as a result of receiving the local sensor dataset 514 and the radio measurement dataset 516 as input. Therefore, this feedback from UE104 represents the input provided to RRM NN112. Thus, the training module 312 can use this feedback as input to the training processing module 606 to retrain or update the DNN or any other neural networks implemented therein. Periodically, or in response to some other trigger event, the modified neural network architecture configuration 624 of the training processing module 606 is extracted and provided to the RRM module 314, which then stores a representation of the modified neural network architecture configuration 624 in the data store 602 and / or sends the representation of the modified neural network architecture configuration 624 to UE104 for use when updating RRM NN112.

[0054] Referring here to Figures 7 and 8, a method 700 for implementing an RRM scheme in a wireless communication network 100 that utilizes a neural network-based RRM process in UE 104 is described according to at least one embodiment. Note that the order of operations described with reference to method 700 is for illustrative purposes only, and operations may be performed in a different order, and one or more operations may be omitted, or one or more additional operations may be included in the method shown. For ease of understanding, method 700 is described with reference to an exemplary embodiment of method 700 as reflected in ladder diagram 900 in Figure 9.

[0055] In some embodiments, the UE104 may have any of several combinations of sensor and radio measurement functions. For example, the UE's optical sensor may be turned on at times, while at other times the UE turns it off to conserve power. As another example, some UE104s may have satellite-based positioning sensors, while others may not. As yet another example, one UE104 may have radar or lidar functionality but not camera functionality, while another UE104 may have camera functionality but not radar or lidar functionality. Since the RRM neural networks implemented in the UE104 utilize sensor data and / or radio measurements as input to direct their operation, it will often be understood that a particular neural network architecture configuration implemented in the UE104 is based on specific sensors available to provide sensor data as input, and / or specific radio measurements that the UE can perform. In other words, the specific neural network architecture configuration implemented in RRM NN112 reflects, in combination with the type of sensor currently providing input to RRM NN112, and one or both of the types and parameters of the wireless measurement currently provided as input to RRM NN112.

[0056] Therefore, Method 700 begins in Block 702 by determining the sensor and radio measurement capabilities of one or more test UEs, which may include UE104 or utilize UEs other than UE104. It will be understood that “test UE” in this sense may not actually be a physical UE, but rather a software simulation of the operation of a test UE for the purpose of neural network training. Therefore, reference to “test UE” should be understood to include reference to such simulations. With respect to the training process, a training module (such as training module 312 of BS110, or a similar training module of a server or other infrastructure component) selects a specific sensor configuration and / or radio measurement configuration to train a candidate neural network architecture of the RRM NN112 of the test UE. In some embodiments, the training module may attempt to train all permutations of available sensors and all permutations of available radio measurement capabilities. However, in embodiments where the non-test UE has a relatively large and diverse range of suitable sensor or radio measurement capabilities, this attempt may not be feasible. Therefore, in other embodiments, the training module selects only a limited representative set of potential sensors and sensor configurations and wireless measurement capabilities. For example, if lidar information from different lidar modules manufactured by the same company is relatively consistent, and therefore, for example, UE104 can implement any of several lidar sensors from its manufacturer, the training module may choose to exclude some lidar sensors from the sensor configuration during training. As another example, a very small subset of UE104 may have the ability to measure SINR over a wider range or at a higher resolution, but the training module may limit training to a smaller SINR range or a lower SINR resolution that still covers most embodiments of UE104.In yet another embodiment, there may be a defined set of sensor configurations that the training module can select for training, and so the training module selects a sensor configuration from this defined set (and also avoids selecting a sensor configuration that depends on sensor functions not commonly supported by the associated device).

[0057] Once a sensor / wireless measurement configuration is selected, the training module identifies one or more sets of training data to use when training candidate neural network architecture configurations based on the selected configuration. Specifically, one or more training datasets contain, or represent, sensor data that can be generated by equivalent sensors in the selected sensor configuration, and represent wireless measurement data that can be generated in accordance with the selected wireless measurement capability, making them suitable for training candidate neural network architecture configurations to operate on sensor data provided by specific sensors represented by the selected sensor / wireless measurement configuration. The training data may also include training data related to the operating state of the test UE, such as RRC state or 5G NR SA or NSA mode. Once one or more training sets are acquired, the training module begins training the RRM NN on the test UE. This training typically involves initializing the bias weights and coefficients of various RRM NNs with initial values, which are generally chosen pseudo-randomly. Then, a training dataset (e.g., representing known sensor data from sensors in a selected sensor configuration and known radiometric data matching a selected radiometric function configuration) is input. The test RRM NN processes the input to generate an output, determines the error between the actual output and the expected output, backpropagates the error across the entire test RRM NN, and repeats this process for the next input dataset. This process is repeated until a specific number of training iterations are performed or a specific minimum error rate is achieved. As a result of training the RRM NN of the test UE, the resulting neural network has a specific neural network architecture configuration, or a DNN architecture configuration if the implemented neural network is a DNN, which characterizes the architecture and parameters of the corresponding RRM NN, such as the number of hidden layers, the number of nodes in each layer, the connections between each layer, and the weights, coefficients, and other bias values ​​implemented at each node.Therefore, once the training of the RRM NN for the selected sensor / wireless measurement configuration test UE is complete, the current neural network architecture configuration of the trained RRM NN is extracted and made available to one or more BS110s as the corresponding candidate neural network architecture configuration 324 (Figure 3) by saving a local copy of the candidate neural network architecture configuration 324 to the BS110 in block 704, or by storing a copy of the candidate neural network architecture configuration 324 in a remote data store accessible by the BS110. In at least one embodiment, the candidate neural network architecture configuration 324 can be generated at the end of training by extracting the architecture and parameters of the corresponding RRM NN, such as the number of hidden layers, the number of nodes, connections, coefficients, weights, and other bias values.

[0058] If there is one or more other untrained sensor / wireless measurement configurations, the training process is repeated, selecting the next sensor / wireless measurement configuration to be trained. Otherwise, if the candidate RRM NNs in UE104 have been trained for all intended sensor / wireless measurement configurations, network 100 can move on to supporting the neural network-based RRM process in UE104 using the trained DNN.

[0059] Therefore, as part of or following the initialization of the wireless connection between UE104 and serving BS110-1, in block 706, UE104 notifies serving BS110-1 of its sensor and / or wireless measurement capabilities by sending a sensor / measurement capabilities message 902 (Figure 9) to serving BS110-1. For example, during the connection procedure, serving BS110-1 may send a UECapabilitiesEnquiry RRC message to UE104, in which case UE104 sends a UECapabilitiesInformation RRC message to serving BS110-1 having one or more fields of the UECapabilitiesInformation RRC message (one embodiment of sensor / measurement capabilities message 902) containing data representing one or both of UE104's sensor and / or wireless measurement capabilities.

[0060] In block 708, the neural network management module 310 of serving BS110-1 selects a candidate neural network architecture configuration 324 for implementation in the RRM NN112 of UE104 from set 322, based on either or both of the sensor capabilities advertised by UE104 or the radio measurement capabilities advertised by UE104. For example, if UE104 has a radar sensor and imaging camera available for use and can perform RSRP measurements at a specific range and resolution, the neural network management module 310 can select a candidate neural network architecture configuration 324 for UE104 that has been trained for this specific sensor configuration and radio measurement capabilities. Once a suitable neural network architecture configuration is selected, in block 710, serving BS110-1 instructs UE104 to implement the selected neural network architecture configuration in the RRM NN112 by sending an NN configuration message 904 to UE104. The NN configuration message 904 may include one or more data structures that implement the selected neural network architecture configuration itself, or it may include an index or other identifier of the selected neural network architecture configuration, so that the UE104 can access the selected neural network architecture configuration from a local data store or from a remote server. In other embodiments, the neural network management module 218 of the UE104 selects a neural network architecture configuration to implement in the RRM NN112 based on the current sensor / measurement configuration of the UE104, regardless of the direction from the serving BS110-1.

[0061] In response to the NN configuration message 904, or in response to the selection of its own neural network architecture configuration, in block 712, UE104 implements the indicated neural network architecture configuration in UE104's RRM NN112. After RRM NN112 is initialized in this way, UE104 can continue RRM management based on RRM action 118 indicated by the output of RRM NN112. Therefore, in the case of an iteration of an input / output loop using the RRM NN112, in block 714, the sensor set 210 provides a sensor dataset 914 (one embodiment of local sensor data 114, Figure 1) representing sensor data captured by one or more sensors as input to the RRM NN112 of the UE 104, and simultaneously, in block 716, the RRM module 222 and the RF front end 202 of the UE 104 together acquire one or more radio measurements and provide these radio measurements as a radio measurement dataset 916 (one embodiment of radio measurement data 116, Figure 1) as another input to the RRM NN112. In some embodiments, these datasets 914 and 916 represent samples from a single time slice. For example, the UE 104 may provide separate sensor datasets 914 and radio measurement datasets 916 every 10 milliseconds (ms). In other embodiments, one or both of these datasets are sequences of sampled data across multiple time slices of a sliding window, such as a sequence of 10 separate datasets acquired every 100ms in the previous second.

[0062] In block 718, the RRM NN112 processes these inputs and, in some embodiments, processes additional inputs such as the current RRM state or the current 5G NR mode to generate outputs representing one or more RRM actions 918 (one embodiment of RRM action 118, Figure 1) based on the current neural network architecture configuration and the current state of the nodes in the RRM NN112. In block 720, the RRM module 222 of the UE104 initiates the execution of an RRM action(s) 918 by triggering the RF frontend 202 to perform an RRM measurement or other RRM process, instructing the RF frontend 202 to skip one or more scheduled RRM measurements or other RRM processes, reconfiguring one or more parameters of a scheduled RRM process, or a combination thereof.

[0063] In accordance with a typical RRM scheme, in at least one embodiment, UE104 contributes to the RRM scheme of wireless network 100 by providing an RRM report to serving BS110-1, which can then be used by serving BS110-1, core network 106, or other infrastructure components when making RRM decisions for the entire wireless system. This RRM report can be performed in response to conventional RRM processes carried out by UE104, such as embodiments of a static RRM measurement scheme separate from RRM NN112. This RRM report can also be performed in response to the execution of an RRM action(s) 918 by UE104. Therefore, continuing with method 700 in Figure 8, in block 722, UE104 provides an RRM report 920 to serving BS110-1 in response to one or more RRM processes being performed in UE104, either due to an RRM action 918 generated by RRM NN112 or due to a separate, parallel static RRM process RRM NN112 performed in UE104. Typically, the RRM report 920 includes an RRM measurement report containing one or more RRM measurements performed by UE104 in the preceding period. For example, if the RRM action 918 generated in block 718 represents an instruction to have UE104 perform an Authorization-Assisted Access (LAA) investigation, as part of this LAA investigation, UE104 may perform radio measurements of this subset of cells, such as identifying a subset of adjacent cells and then measuring the RSRP or SINR of each cell in the subset. Therefore, RRM report 920 in this example may include an RRM measurement report that includes the measured RSRP or measured SINR for each cell within this identified subset.

[0064] In response to information contained in RRM report 920, or to another trigger such as the expiration of a specified period since the last update, serving BS110-1 may determine that it is an appropriate point to update the underlying architectural configuration used in RRM NN112. This update process typically relies on utilizing the same real-world inputs provided to RRM NN112 in a test UE performed by serving BS110-1 (as described above), and then iteratively backpropagating any error between the actual results output by the test UE and the expected or intended results. Thus, in block 724, serving BS110-1 sends a sensor / radio request 922 (Figure 9) to UE104 requesting that UE104 provide RRM NN112 with some or all copies of the sensor dataset 914 and radio measurement dataset 916, for example, since the last update process. In response to the sensor / wireless request 922, in block 726, UE104 transmits the requested sensor / wireless measurement data 924 (Figure 9) to the serving BS110-1. In another embodiment, instead of buffering the sensor data and wireless measurement data to transmit to the serving BS110-1 on request, in a single burst, UE104 may transmit copies of each sensor dataset 914 and / or each wireless measurement dataset 916 to the serving BS110-1 as they are generated, in parallel with providing these datasets as input to the RRM NN112.

[0065] In block 728, the training module 312 of serving BS110-1 utilizes sensor data and / or radio measurement data received from UE104 to perform a neural network update process 926, providing this data as input to RRM NN112. The training module 312 then retrains or updates the test copy of RRM NN112 implemented as a test UE by reviewing the underlying neural network architecture configuration of the test copy based on a comparison of the resulting output with the expected or intended output. This training may further include an evaluation of RRM reports 920 enumerated for the period associated with the retraining data, assessing the accuracy of the operation of RRM NN112 in UE104, and consequently adjusting or reviewing the neural network architecture configuration of the test copy. Subsequently, the neural network management module 310 of serving BS110-1 extracts the updated neural network architecture configuration from the retrained test copy and provides the resulting updated neural network architecture configuration 324 to the RRM module 314 to replace or complement the previous version in the data store 602, and in block 730, can send an update message 928 containing a representation of the updated neural network architecture configuration 324 to the UE104. In block 732, the UE104 updates the RRM NN112 using the updated neural network architecture configuration 324 represented in the update message 928, either by replacing the "old" neural network architecture configuration of the RRM NN112 with the updated version, or by using the updated version to selectively modify or extend the previous version of the RRM NN112.

[0066] Referring to Figure 10, an alternative embodiment of Method 700, called Method 700-1, is shown. Method 700-1 proceeds similarly to Method 700 with respect to the processes in blocks 702 to 720 in Figure 7 and block 722 in Figure 8. However, following the RRM reporting process in block 722, in Method 700-1, UE 104 performs an NN update process and, rather than the other way around, reports the updated neural network architecture configuration to Serving BS 110-1. Thus, in Method 700-1, following the RRM reporting in block 722, in block 1024, the neural network management module 218 of UE 104 instantiates a copy of RRM NN 112 as a training copy and then evaluates some or all of the output RRM actions 118 generated from the input set of sensor data and radio measurement data against a set of RRM actions that would be expected or intended if the same inputs were given (these expected RRM actions are specified, for example, by training guidance provided by the network infrastructure 102). Next, the neural network management module 218 updates the underlying architecture configuration of the training copy of RRM NN112, for example, by iteratively backpropagating the error between the actual and expected RRM actions. Then, in block 1026, the neural network management module 218 extracts the modified neural network architecture configuration obtained as a result of the test copy of RRM NN112 as the updated neural network architecture configuration, and updates the actual RRM NN112 using the updated neural network architecture configuration. In block 1028, this update can also be sent to the serving BS110-1 for storage as a candidate neural network architecture configuration 324 and / or for propagation in one or more other UE104 using the same original neural network architecture configuration in their respective RRM neural networks.

[0067] As described above, the RRM NN112 can provide control for the network infrastructure 102 to implement various conditional RRM procedures, instructing the UE 104 to conditionally execute RRM-related processes depending on the status of one or more conditions monitored by the UE 104. CHO is an example of such a conditional RRM process, in which the UE is instructed to initiate a handover on the condition that one or more monitored conditions subsequently reach a corresponding threshold. In some embodiments, the conditional RRM process is represented by the same RRM NN112 that provides other types of RRM decisions at the UE 104. In other embodiments, one or more conditional RRM processes may be implemented as separate conditional RRM neural networks that operate in parallel with the RRM NN112 and utilize overlapping inputs. Figure 11 illustrates an exemplary method 1100 of the operation of a conditional RRM neural network, either as the RRM NN112 (or RRM processing module 502) or as separate neural networks such as the CHO processing module 504. It should be noted that the sequence of operations described with reference to Method 1100 is for illustrative purposes only, and operations may be performed in a different order, one or more operations may be omitted, or one or more additional operations may be included in the method shown. For ease of understanding, Method 1100 will be described with reference to an exemplary embodiment of Method 1100 as reflected in the ladder diagram 1200 of Figure 12.

[0068] Method 1100 begins with training a set of candidate CHO neural network architecture configurations using the training process described above. Thus, in block 1102, UE 104 notifies serving BS 110-1 of its sensor and / or radio measurement capabilities by sending a sensor / measurement capability message 1202 (Figure 12) to serving BS 110-1. This message 1202 may include, for example, a UECapabilitiesInformation RRC message provided in response to a UECapabilitiesEnquiry RRC message from serving BS 110-1. In block 1104, the neural network management module 310 of serving BS 110-1 selects a candidate CHO neural network architecture configuration (e.g., an example of neural network architecture configuration 324) from set 322 based on either or both of the sensor capabilities advertised by UE 104 or the radio measurement capabilities advertised by UE 104. Once an appropriate CHO neural network architecture configuration is selected, in block 1106, the serving BS110-1 instructs the UE104 to implement the selected CHO neural network architecture configuration in the CHO processing module 504 by sending a CHO configuration message 1204 (Figure 12) to the UE104. The CHO configuration message 1204 may include one or more data structures that implement the selected neural network architecture configuration itself, or it may include an index or other identifier of the selected neural network architecture configuration, so that the UE104 can access the selected neural network architecture configuration from a local data store or from a remote server. In other embodiments, the neural network management module 218 of the UE104 selects the CHO neural network architecture configuration to be implemented in the CHO processing module 504 based on the current sensor / measurement configuration of the UE104, regardless of the direction from the serving BS110-1.

[0069] In response to the CHO configuration message 1204, or in response to the selection of its own CHO neural network architecture configuration, in block 1108, the UE 104 implements the indicated CHO neural network architecture configuration in the CHO processing module 504, and thus completes the initialization of the CHO processing module 504.

[0070] The UE104 can then proceed with conditional RRM management based on the CHO action 520 indicated by the output of the CHO processing module 504. Thus, in the case of an iteration of the input / output loop using the CHO processing module 504, in block 1110, the sensor set 210 provides a sensor dataset 1214 (Figure 12) (one embodiment of local sensor data 114, Figure 1) representing sensor data captured by one or more sensors as input to the CHO processing module 504. As understood, the decision to initiate a handover as a result of a CHO instruction typically reflects the result of comparing the suitability of the current serving cell with one or more neighboring or target cells, where suitability is indicated by a comparison of various radio measurements of the serving cell and neighboring cells(s). Therefore, simultaneously with providing iterative sensor data, in block 1112, the RRM module 222 and the RF front end 202 of UE104 together acquire one or more radio measurements of serving BS110-1, and in block 1114, the RRM module 22 and the RF front end 202 acquire one or more radio measurements of each applicable neighboring BS110, such as target BS110-2, and provide these radio measurements as a radio measurement dataset 1216 (Figure 12) (one embodiment of radio measurement data 116, Figure 1) as another input to the CHO processing module 504. In some embodiments, these datasets 1214, 1216 represent samples from a single time slice, while in other embodiments, one or both of these datasets are sequences of sample data across multiple time slices in a sliding window.

[0071] In block 1116, the CHO processing module 504 processes these inputs and, in some embodiments, processes additional inputs such as the current RRC state or the current 5G NR mode to generate an output representing one or more CHO actions 1220 (Figure 12) (one embodiment of CHO action 520, Figure 5) based on the current neural network architecture configuration and the current state of the nodes of the CHO processing module 504. The RRM module 222 of UE 104 then initiates the execution of the CHO action(s) 1220. As described above, in some embodiments, the CHO action 1220 may include a process for adjusting some parameters of the CHO process, such as changing the threshold used when making a handover decision. However, in other embodiments, the CHO action 1220 represents the handover decision itself. In such embodiments, in block 1118, the RRM module 222 makes the handover decision represented by the CHO action 1220. If the handover decision is to postpone the start of the handover, method 1100 can return to blocks 1110, 1112, and 1114 for another iteration of the CHO action generation process using a new set of sensor and measurement data inputs to the CHO processing module 504.

[0072] Otherwise, if the handover decision is to initiate a handover, in block 1120, the RRM module 222 and RF front-end 202 are coordinated to proceed with a handover 1222 (Figure 12) to the indicated adjacent cell (represented in this example by target BS110-1). After the handover is performed, it will be understood that UE104 is now connected to a different cell which may have different RRM parameters compared to the previous cell associated with BS110-1. Thus, in at least one embodiment, at least one configuration of RRM NN112 (e.g., as RRM processing module 502) or CHO processing module 504 is replaced by a new configuration provided by BS110-2 as a new serving BS for UE104. Therefore, in block 1122, the same or similar neural network architecture configuration process described above with reference to blocks 1102-1108 may be performed again, but this time using a neural network architecture configuration(s) selected and provided by BS110-2 instead of BS110-1. Thus, UE104 can provide BS110-2 with a representation of its sensor and / or wireless measurement functions as a function message 1224 (Figure 12), and BS110-2 can select an appropriate neural network architecture configuration 1226 (Figure 12) based on these functions and instruct UE104 to implement the selected neural network architecture configuration 1226 in the CHO processing module 504.

[0073] Subsequently, UE104 can use the newly configured CHO processing module 504 to execute a neural network-based CHO decision-making process in the same manner as described above, which includes providing the sensor dataset 1234 (Figure 12) and the radio measurement dataset 1236 (Figure 12) acquired from the sensor set 210 as inputs to the CHO processing module 504. From these inputs, the CHO processing module 504 generates CHO actions 1240 suitable for the CHO-related RRM scheme implemented by BS110-2. UE104 can then execute one or more CHO processes represented by the CHO actions 1240 described herein.

[0074] Embodiments of this disclosure may also be better understood by considering the following non-limiting embodiments.

[0075] Example 1: A computer implementation method, in a wireless device, Receiving a first set of sensor data from one or more sensors of the wireless device, Receiving a first set of wireless measurement values ​​from a wireless interface of a wireless device different from the one or more sensors, The wireless resource management (RRM) neural network of the wireless device processes the first set of sensor data and the first set of wireless measurement values ​​to generate a first output representing a first RRM action, and Performing the first RRM action on the wireless device, Computer implementation methods, including those mentioned above.

[0076] Example 2: The wireless device is a user device, and the method is To provide at least one representation of the sensor function of the wireless device for reception by infrastructure components of a network infrastructure wirelessly connected to the wireless device, The method according to Embodiment 1, further comprising receiving a first neural network architecture configuration from the infrastructure component in response to providing a representation of the sensor function, and implementing the first neural network architecture configuration in the RRM neural network.

[0077] Example 3: The method according to Example 2, wherein the representation of the sensor function includes one or more fields of a UECapabilitiesInformation radio resource control (RRC) message.

[0078] Example 4: The method provides the infrastructure component with at least one of the first set of sensor data or the first set of wireless measurement values. The method further includes receiving a second neural network architecture configuration from the infrastructure component in response to providing at least one of the first set of sensor data or the first set of radio measurements, the second neural network architecture configuration representing a modification of the first neural network architecture configuration based on at least one of the first set of sensor data or the first set of radio measurements, and the method The RRM neural network implements the second neural network architecture configuration. The method according to Example 2 or 3, further comprising:

[0079] Example 5: Modifying the first neural network architecture configuration based on at least one of the first set of sensor data or the first set of wireless measurement values ​​to generate a second neural network architecture configuration, and The RRM neural network implements the second neural network architecture configuration. The method according to Example 2 or 3, further comprising:

[0080] Example 6: Further includes receiving a representation of the operating state of the wireless device, The method according to any one of Examples 1 to 5, wherein processing includes processing the first set of sensor data, the first set of wireless measurements, and the operating state in the RRM neural network in order to generate the first output.

[0081] Example 7: The method according to Example 6, wherein the operating state includes the radio resource control (RRC) state of the wireless device.

[0082] Example 8: The method according to any one of Examples 1 to 7, wherein the one or more sensors include at least one of a position sensor, an attitude sensor, an accelerometer, a pressure sensor, or a proximity sensor.

[0083] Example 9: The method according to Example 8, wherein the position sensor includes at least one of a satellite-based positioning sensor or a visual telemetry imaging sensor, and the attitude sensor includes at least one of a gyroscope or a visual telemetry imaging sensor.

[0084] Example 10: The method according to any one of Examples 1 to 9, wherein the first set of wireless measurements includes at least one signal power measurement of a serving cell or an adjacent cell.

[0085] Example 11: The method according to Example 10, wherein the signal power measurement includes at least one of signal-to-noise (SNR) measurement, signal-to-interference plus noise (SINR) measurement, reference signal received power (RSRP) measurement, or received signal strength indicator (RSSI) measurement.

[0086] Example 12: The method according to any one of Examples 1 to 11, wherein the RRM action includes at least one of performing RRM-related measurements by the wireless device, defining the characteristics of the RRM-related measurements performed by the wireless device, or performing an RRM reporting process in the wireless device.

[0087] Example 13: The method according to Example 12, wherein the RRM-related measurement includes at least one of inter-frequency measurement, intra-frequency measurement, or inter-radio access technology (RAT) measurement.

[0088] Example 14: The method according to Example 12 or 13, wherein the characteristics of the RRM-related measurement include at least one of the frequency or timing of the RRM-related measurement, or the frequency band or channel of the RRM-related measurement.

[0089] Example 15: The method according to Example 1, wherein the RRM action includes performing a conditional handover (CHO) decision or a conditional PSCell change (CPC) decision.

[0090] Example 16: In response to connecting to a cell as a result of a conditional handover decision, To provide a representation of the sensor function of the wireless device for reception by the infrastructure components of the cell, In response to providing a representation of the sensor function, the cell receives a first neural network architecture configuration from the infrastructure component, and Replacing the second neural network architecture configuration of the RRM neural network with the first neural network architecture configuration, The method according to Example 15, further comprising:

[0091] Example 17: Receiving a second set of sensor data from one or more sensors of the wireless device, Receiving a second set of wireless measurement values ​​from the wireless interface of the wireless device, The RRM neural network of the wireless device processes the second set of sensor data and the second set of wireless measurement values ​​to generate a second output representing a second RRM action, and Performing the second RRM action on the wireless device, The method according to Example 16, further comprising:

[0092] Example 18: The method according to any one of Examples 1 to 17, wherein the wireless device is a user device.

[0093] Example 19: The method according to Example 1, wherein the wireless device is a base station.

[0094] Example 20: The method according to Example 19, further comprising receiving an RRM report from a user device.

[0095] Example 21: Wireless interface, At least one processor coupled to the wireless interface, and A wireless device comprising at least one memory, the at least one memory being coupled to the at least one processor and storing executable instructions configured to operate the at least one processor to perform the method according to any one of Examples 1 to 20.

[0096] Example 22: A computer implementation method for an infrastructure component wirelessly connected to a wireless device, Receiving a representation of the sensor function of the wireless device from the wireless device, To determine the configuration of a first neural network architecture based on the representation of the aforementioned sensor function, and To determine the RRM actions performed by the wireless device, a representation of the first neural network architecture configuration is transmitted to the wireless device for implementation in the wireless resource management (RRM) neural network of the wireless device. Computer implementation methods, including those mentioned above.

[0097] Example 23: Receiving at least one of a set of sensor data or a set of wireless measurement values ​​from the wireless device. To modify the first neural network architecture configuration based on at least one of the set of sensor data or the set of wireless measurement values ​​to generate a second neural network architecture configuration, and For implementation in the RRM neural network of the wireless device, a representation of the second neural network architecture configuration is transmitted to the wireless device. The method according to Example 22, further comprising:

[0098] Example 24: The method according to Example 22 or 23, wherein the set of sensor data includes at least one of position data, orientation data, acceleration data, barometric pressure or altitude data, or proximity data.

[0099] Example 25: The method according to any one of Examples 22 to 24, wherein the set of wireless measurements includes at least one signal power measurement of a serving cell or an adjacent cell.

[0100] Example 26: The method according to any one of Examples 22 to 25, wherein the determined RRM action includes at least one of performing an RRM-related measurement by the wireless device or setting the characteristics of an RRM-related measurement performed by the wireless device.

[0101] Example 27: The method according to any one of Examples 22 to 26, wherein the determined RRM action includes performing a conditional handover decision.

[0102] Example 28: The method according to any one of Examples 22 to 26, wherein the infrastructure component includes a base station.

[0103] Example 29: The method according to any one of Examples 22 to 28, wherein the wireless device includes a user device.

[0104] Example 30: The method according to any one of Examples 22 to 29, wherein the representation of the sensor function includes one or more fields of a UECapabilitiesInformation radio resource control (RRC) message.

[0105] Example 31: A set of 1 or more sensors, A wireless interface different from the set of one or more sensors, At least one processor coupled to the wireless interface, and An infrastructure component comprising at least one memory, the at least one memory being coupled to the at least one processor and storing executable instructions configured to operate the at least one processor to perform the method described in any of embodiments 22 to 30.

[0106] In some embodiments, certain aspects of the techniques described above may be implemented by one or more processors of a processing system that executes the software. The software includes one or more sets of executable instructions, which are stored or otherwise tangibly embodied on a non-temporary computer-readable storage medium. The software may include instructions and certain data that, when executed by one or more processors, cause one or more processors to execute one or more aspects of the techniques described above. The non-temporary computer-readable storage medium may include, for example, magnetic or optical disk storage devices, and solid-state storage devices such as flash memory, cache, random-access memory (RAM), or a single or multiple non-volatile memory devices, and similar. The executable instructions stored on the non-temporary computer-readable storage medium may be source code, assembly language code, object code, or other instruction format that can be interpreted or otherwise executed by one or more processors.

[0107] Computer-readable storage media may include any storage media, or combinations of storage media, that are accessible by a computer system while in use to provide instructions and / or data to the computer system. Such storage media may include, but are not limited to, optical media (e.g., compact discs (CDs), digital versatile discs (DVDs), Blu-ray discs), magnetic media (e.g., floppy disks, magnetic tapes, or magnetic hard drives), volatile memory (e.g., random access memory (RAM) or cache), non-volatile memory (e.g., read-only memory (ROM) or flash memory), or microelectromechanical system (MEMS) based storage media. Computer-readable storage media may be embedded in a computing system (e.g., system RAM or ROM), fixedly mounted to a computing system (e.g., magnetic hard drives), removable mounted to a computing system (e.g., optical discs or Universal Serial Bus (USB) based flash memory), or connected to a computer system via a wired or wireless network (e.g., network-accessible storage (NAS)).

[0108] In addition to the foregoing, it should be noted that not all activities or elements described above are required, and that certain activities or parts of devices may not be required, or that one or more additional activities may be performed, or that one or more additional elements may be included. Furthermore, the order in which the activities are listed does not necessarily indicate the order in which they are performed. Also, concepts are described with reference to specific embodiments. However, those skilled in the art will understand that various modifications and variations can be made without departing from the scope of this disclosure, as described in the claims below. Accordingly, this specification and the drawings should be considered illustrative rather than restrictive, and all such variations are intended to be included within the scope of this disclosure.

[0109] Benefits, other advantages, and solutions to problems have been described above with respect to specific embodiments. However, benefits, advantages, and solutions to problems, as well as any features(if any) that may produce or make more prominent any benefit, advantage, or solution, should not be construed as material, necessary, or essential features of any or all claims. Furthermore, the subject matter of the disclosed invention can be modified and implemented in different but equivalent ways that will be apparent to those skilled in the art who have a benefit of teaching herein; therefore, the specific embodiments disclosed above are merely illustrative. No limitation is intended to any configuration or design details shown herein other than those described in the claims below. Accordingly, it will be apparent that the specific embodiments disclosed above can be modified or altered, and all such variations will be considered within the scope of the subject matter of the disclosed invention. Thus, the protection sought herein is as described in the claims below.

Claims

1. A method performed by a computer, In wireless devices, Receiving a first set of sensor data from one or more sensors of the wireless device, Receiving a first set of wireless measurement values ​​from the wireless interface of a wireless device different from the one or more sensors, The wireless resource management (RRM) neural network of the wireless device processes the set of first sensor data and the set of first wireless measurement values ​​to generate a first output representing a first RRM action, and The first RRM action is performed on the wireless device. Includes, The wireless device is a user device, and the method is To provide at least one representation of the sensor function of the wireless device for one or more sensors, for reception by infrastructure components of a network infrastructure wirelessly connected to the wireless device. In response to providing a representation of the sensor function, the infrastructure component receives a first neural network architecture configuration, and The RRM neural network implements the first neural network architecture configuration. Methods that are performed on a computer, including further details.

2. The method according to claim 1, wherein the representation of the sensor function includes one or more fields of a UECapabilitiesInformation radio resource control (RRC) message.

3. The aforementioned method, To provide the infrastructure component with at least one of the first set of sensor data or the first set of wireless measurement values, The method further includes receiving a second neural network architecture configuration from the infrastructure component in response to providing at least one of the first set of sensor data or the first set of radio measurements, the second neural network architecture configuration representing a modification of the first neural network architecture configuration based on at least one of the first set of sensor data or the first set of radio measurements, and the method The RRM neural network implements the second neural network architecture configuration. The method according to claim 1, further comprising:

4. To modify the first neural network architecture configuration based on at least one of the first set of sensor data or the first set of wireless measurement values ​​to generate a second neural network architecture configuration, and The RRM neural network implements the second neural network architecture configuration. The method according to claim 1, further comprising:

5. Further includes receiving a representation of the operating state of the wireless device, The method according to claim 1, wherein processing includes processing the first set of sensor data, the first set of wireless measurement values, and the operating state in the RRM neural network in order to generate the first output.

6. The method according to claim 5, wherein the operating state includes the radio resource control (RRC) state of the wireless device.

7. The method according to claim 1, wherein the one or more sensors include at least one of a position sensor, an attitude sensor, an accelerometer, a pressure sensor, or a proximity sensor.

8. The method according to claim 1, wherein the first set of wireless measurement values ​​includes at least one signal power measurement value of a serving cell or an adjacent cell.

9. The method according to claim 1, wherein the RRM action includes at least one of performing RRM-related measurements by the wireless device, setting the characteristics of the RRM-related measurements performed by the wireless device, or performing an RRM reporting process on the wireless device.

10. The method according to claim 9, wherein the characteristics of the RRM-related measurement include at least one of the frequency or timing of the RRM-related measurement, or the frequency band or channel of the RRM-related measurement.

11. The method according to claim 1, wherein the RRM action includes performing a conditional handover (CHO) decision or a conditional PSCell change (CPC) decision.

12. In response to connecting to a cell as a result of a conditional handover decision, To provide a representation of the sensor function of the wireless device for reception by the infrastructure components of the cell, In response to providing a representation of the sensor function, the first neural network architecture configuration is received from the infrastructure component of the cell, and Replacing the second neural network architecture configuration of the RRM neural network with the first neural network architecture configuration, The method according to claim 11, further comprising:

13. Receiving a second set of sensor data from one or more sensors of the wireless device, Receiving a second set of wireless measurement values ​​from the wireless interface of the wireless device, The RRM neural network of the wireless device processes the second set of sensor data and the second set of wireless measurement values ​​to generate a second output representing a second RRM action, and Performing the second RRM action on the wireless device, The method according to claim 12, further comprising:

14. The method according to claim 1, wherein the wireless device is a user device.

15. The method according to claim 1, wherein the wireless device is a base station.

16. The method according to claim 15, further comprising receiving an RRM report from a user device.

17. One or more sensors, A wireless interface different from the one or more sensors mentioned above, At least one processor coupled to the wireless interface, and A device comprising at least one memory, wherein the at least one memory is coupled to the at least one processor and stores executable instructions configured to operate the at least one processor to perform the method according to any one of claims 1 to 16. Wireless device.

18. A program that causes at least one processor to perform the method described in any one of claims 1 to 16.

Images