Model Input Based on Latency Profiles for AI / ML

Abstract

This disclosure describes methods and systems for using delay profiles as inputs for AI / ML models. This disclosure includes solutions that use only delay profiles (DPs) as AI / ML model inputs, where the DP consists only of timing values ​​from multiple receive paths. Compared to receiving power profiles (PDPs), received power values ​​are not required as model inputs. Evaluation results show that the performance of the AI / ML model is maintained with only slight degradation when compared to using PDPs as model inputs. The significantly reduced model input size offers significant advantages to AI / ML models under this disclosure. For example, the measurement burden is greatly reduced. This advantage exists regardless of whether the measurements need to be signaled from one entity to another. Furthermore, when the measurements of the model input need to be transmitted from one entity to another, the signaling overhead is greatly reduced.

Description

[Technical Field] 【0001】 Cross-referencing of related information This application claims the advantages of U.S. priority application No. 63 / 457,975, filed on April 7, 2023, entitled "Delay Profile Based Model Input for AI / ML". 【0002】 This disclosure generally relates to a system and method for using delayed profile data as input for AI / ML models. [Background technology] 【0003】 Overview of AI / ML for Wireless Communication Artificial intelligence (AI) and machine learning (ML) have been studied in both academia and industry as promising tools for optimizing the design of air interfaces in wireless communication networks. Exemplary use cases include using autoencoders for CSI (Channel State Information) compression to reduce feedback overhead and improve channel prediction accuracy; using deep neural networks to classify LOS (Line of Sight) and NLOS (Non-Line of Sight) conditions to improve positioning accuracy; using reinforcement learning (RL) for beam selection on the network side and / or UE (User Equipment) side to reduce signaling overhead and beam alignment latency; and using deep reinforcement learning to learn optimal precoding policies for complex MIMO (Multiple Input Multiple Output) precoding problems. 【0004】 The 3GPP NR standardization work includes a new release of 18 research items on AI / ML for NR (New Radio) air interfaces, which began in May 2022. These research items explore the benefits of extending air interfaces with features that enable improved support for AI / ML-based algorithms for performance improvements and / or reduced complexity / overhead. By studying several selected use cases (CSI feedback, beam management, and positioning), this SI (System Information) aims to lay the foundation for future air interface use cases that leverage AI / ML techniques. 【0005】 Building an AI / ML model involves several development steps, and the actual training of the AI ​​model is just one step in the training pipeline. A crucial part of AI / ML development is the lifecycle management of the AI / ML model. This is illustrated in Figure 1. Figure 1 is a diagram of the training and inference pipelines, as well as their interactions within the model lifecycle management procedure. AI model lifecycle management typically includes the following: ● Training (Retraining) Pipeline: This involves data ingestion, which refers to collecting raw (training) data from data storage. After data ingestion, there may be steps to control the validity of the collected data. Data preprocessing refers to some feature engineering applied to the collected data, which may include, for example, data standardization and, in some cases, data transformations required for input data to the AI / ML model. This is followed by the actual model training steps outlined earlier, and model evaluation refers to benchmarking performance against several baselines. The iterative steps of model training and model evaluation continue until an acceptable level of performance is achieved (as illustrated earlier), and model registration refers to registering the AI / ML model, which may include any corresponding AI / ML metadata that provides information about how the AI / ML model was developed, and, in some cases, the AI / ML model evaluation performance results. ● The deployment phase to integrate trained (or retrained) AI / ML models into the inference pipeline. ● Inference Pipeline: This includes data ingestion, which involves collecting raw (inference) data from data storage; a data preprocessing stage, which is usually the same as the corresponding process performed in the training pipeline; model operation, which involves using a model that has been trained and deployed in an operating mode; data and model monitoring, which involves verifying that the inference data comes from a distribution that closely matches the training data; and monitoring the model output to detect any performance or operation drift. ● A drift detection stage that notifies about any drift in the model's operation. 【0006】 In the first scenario, we can assume that an AI / ML model operating on an existing standard air interface is located on the UE side. The UE uses the AI / ML model to generate outputs that are reported to a central node in the network to determine the UE's location. 【0007】 In the second scenario, we can assume that AI / ML models operating on existing standard air interfaces are deployed at various transmit / receive points (TRPs). The TRPs use the AI / ML models to generate outputs that are reported to a central node in the network to position the UE location. 【0008】 AI / ML for positioning One AI / ML use case is the positioning of target UEs. The following cases are under consideration. ●Case 1: UE-side model, UE-based positioning using direct AI / ML or AI / ML-assisted positioning. ●Case 2a: UE-assisted / LMF-based positioning using the UE-side model, AI / ML-assisted positioning. ●Case 2b: UE support by LMF-side model / LMF (position management function) based positioning, direct AI / ML positioning. ●Case 3a: NG-RAN node-assisted positioning and AI / ML-assisted positioning using the gNB side model. ● Case 3b: NG-RAN node assisted positioning and direct AI / ML positioning using the LMF side model. 【0009】 Most companies in 3GPP use the time-domain channel impulse response (CIR) or power delay profile (PDP) on a regular sampling grid as input to the AI / ML model. The CIR and PDP are usually truncated after a certain number of taps suitable for the specific sampling rate used and the intended use case. If the TRP has multiple RX (receive) antenna ports, the CIR and PDP may take on additional dimensions. This is illustrated in the following agreement in 3GPP RAN1. Agreement For the model input used in the evaluation of AI / ML-based positioning, if the time-domain channel impulse response (CIR) or power delay profile (PDP) is used as the model input in the evaluation, the company shall report the input dimension N TRP *N port *N t where N TRP is the number of TRPs, N port is the number of transmit / receive antenna port pairs, and N t is the number of time-domain samples. ● Note: The CIR and PDP can have various dimensions. ● Note: The company shall provide details regarding the company's assumptions on how the PDP is constructed and how (if applicable) the PDP is mapped to Nt samples. 【0010】 In addition to the above agreement, note that since both the I branch and Q branch are required, the time-domain CIR is represented by complex numbers. Therefore, the CIR samples require N TRP *N port *N t *2*B CIR,real bits, where BCIR,real is the number of bits for the real-valued part. 【0011】 Currently, several challenges exist. Existing AI / ML models for positioning use either CIR or Power Delay Profile (PDP) as model input. The use of CIR and PDP results in large model inputs (e.g., large vectors or large matrices). Consequently, obtaining CIR and PDP measurements is highly required. Signaling overhead is also significant when CIR or PDP measurements as model input need to be transmitted from one entity to another. [Overview of the project] 【0012】 One embodiment under this disclosure includes a method performed by a UE to use DP as input to an AI or ML model. The method includes receiving one or more delayed profile data for a training phase, preprocessing one or more delayed profile data for the training phase, and training an AI or ML model with the preprocessed one or more delayed profile data. 【0013】 Another embodiment under this disclosure is a method performed by a UE to use DP as input to an AI model or an ML model. This method includes receiving one or more DP data for an inference phase, preprocessing one or more DP data for an inference phase, and using one or more preprocessed DP data as input to an AI or ML model. 【0014】 Another embodiment under this disclosure is a method performed by a UE to provide a DP as an AI model input or an ML model input. This method includes receiving one or more radio signals from one or more radio nodes, generating one or more delay profile data based at least partially on one or more radio signals, and transmitting one or more delay profile data to a network node. 【0015】 Another embodiment under this disclosure is a method performed by a network node to use DP as an AI model input or an ML model input. This method includes receiving one or more delayed profile data, preprocessing one or more delayed profile data, sending the preprocessed one or more delayed profile data to the UE for use as one or more inputs to an AI or ML model, and receiving one or more outputs of the AI ​​or ML model from the UE. 【0016】 Another embodiment of the present disclosure is a method performed by a network node to use DP as an AI model input or an ML model input. The method includes receiving one or more delayed profile data, preprocessing one or more delayed profile data, using one or more preprocessed delayed profile data as one or more inputs for an AI or ML model, and obtaining one or more outputs of the AI ​​or ML model. 【0017】 Another embodiment under this disclosure includes a method performed by a network node to provide DP as an AI model input or an ML model input. This method includes receiving one or more radio signals from one or more radio nodes, generating one or more delay profile data based at least partially on one or more radio signals, and transmitting one or more delay profile data to a second network node for use in one or more inputs of an AI or ML model. 【0018】 This summary of the present invention is provided to introduce in a simplified form a selection of concepts that will be further described in the embodiments for carrying out the invention described below. This summary of the present invention is not intended to identify any major or essential features of the claimed subject matter, nor to be used as an indication of the scope of the claimed subject matter. 【0019】 For a more complete understanding of this disclosure, the following description, along with the attached drawings, is to be referenced. [Brief explanation of the drawing] 【0020】 [Figure 1] This diagram shows the training and inference pipelines, as well as their interactions within the model lifecycle management procedures. [Figure 2] This figure shows an example of a PDP (Photographic Display Panel). [Figure 3] This figure shows the complete PDP of the line-of-sight channel. [Figure 4] This figure shows the complete PDP for a non-line-of-sight channel. [Figure 5] This is a diagram of DP. [Figure 6] This figure shows an example of how UE reports timing information for RSTD and additional paths. [Figure 7] This table shows the reduction in model input size by using only path-by-path timing information compared to using both path-by-path timing information and power information. [Figure 8] This diagram shows the savings as a percentage of n, where n is the total number of passes. [Figure 9] This diagram shows the savings as a percentage of n, where n is the total number of passes. [Figure 10] This figure shows an example of AI / ML direct positioning where the LMF functions as a central node, processing DPs corresponding to numerous TRPs to generate an estimate of the target UE's position. [Figure 11] This figure shows an example of AI / ML-assisted positioning where a gNB acts as a central node, processing DPs corresponding to numerous TRPs to generate estimates of unobserved direct paths to arrival (ToA). [Figure 12] This figure shows an example of typical AI / ML-assisted positioning where multiple nodes process DP corresponding to numerous TRPs to generate estimates of unobserved direct paths to arrival (ToA). [Figure 13]This figure shows an exemplary 3GPP indoor factory (InF) ​​model with 18 TRPs deployed at known locations in the network. [Figure 14] This is a table of UE 2D positioning errors [in meters] at different percentiles. [Figure 15] This is a flowchart of a method embodiment under the present disclosure. [Figure 16] This is a flowchart of a method embodiment under the present disclosure. [Figure 17] This is a flowchart of a method embodiment under the present disclosure. [Figure 18] This is a flowchart of a method embodiment under the present disclosure. [Figure 19] This is a flowchart of a method embodiment under the present disclosure. [Figure 20] This is a flowchart of a method embodiment under the present disclosure. [Figure 21] This is a schematic diagram of an embodiment of a communication system under the present disclosure. [Figure 22] This is a schematic diagram of a user device embodiment under the present disclosure. [Figure 23] This is a schematic diagram of a network node embodiment under the present disclosure. [Figure 24] This is a schematic diagram of a host embodiment under the present disclosure. [Figure 25] This is a schematic diagram of an embodiment of a virtualization environment under this disclosure. [Figure 26] This figure shows a schematic representation of one embodiment of communication between a node, a host, and user equipment under this disclosure. [Modes for carrying out the invention] 【0021】 Before describing in detail the various embodiments of this disclosure, it should be understood that this disclosure is not limited to the parameters of the systems, methods, apparatus, products, processes, and / or kits illustrated in detail, and that such parameters may, of course, vary. Therefore, while some embodiments of this disclosure are described in detail with respect to specific settings, parameters, components, elements, etc., the descriptions are illustrative and should not be construed as limiting the scope of the claimed embodiments. Furthermore, the terms used herein are for the purpose of describing embodiments and are not necessarily intended to limit the scope of the claimed embodiments. Some of the embodiments considered herein are then described more fully with reference to the accompanying drawings. The embodiments are provided as examples to convey the scope of the subject matter to those skilled in the art. 【0022】 In some embodiments, the general term “network node” is used, which can refer to any type of radio network node or any network node that communicates with the UE and / or another network node. Examples of network nodes include NodeB, eNodeB, gNodeB (or gNB), gNB-DU, gNB-CU, MeNB, SeNB, network nodes belonging to MCG or SCG, base stations (BS), multi-standard radio (MSR) radio nodes such as MSR BS, network controllers, radio network controllers (RNC), base station controllers (BSC), relays, D2D UE-network relays, donor node control relays, base station transceiver stations (BTS), access points (AP), transmit points (TP), transmit / receive points (TRP), transmit nodes, RRU, RRH, nodes of distributed antenna systems (DAS), core network nodes (e.g., MSC, MME, etc.), O&M, OSS, SON, positioning nodes (e.g., E-SMLC), location servers, location management entities (LMF), MDT, etc. In some embodiments, the non-limiting terms UE or wireless device are used interchangeably. UE as used herein can be any type of wireless device capable of communicating with a network node or another UE via radio signals. UE can also be a wireless communication device, a target device, a D2D (device-to-device) UE, a machine-type UE or a machine-to-machine communication (M2M) capable UE, a low-cost and / or low-complexity UE, a sensor equipped with a UE, a tablet, a mobile terminal, a smartphone, a laptop embedded equipment (LEE), a laptop mounted equipment (LME), a USB dongle, customer premises equipment (CPE), an Internet of Things (IoT) device, or a narrowband IoT (NB-IoT) device. 【0023】 As mentioned above, conventional technologies currently have several challenges, and the use of CIR and PDP results in large model inputs (e.g., large vectors or large matrices). Consequently, there is a strong demand for obtaining CIR and PDP measurements. When CIR or PDP measurements as model inputs need to be transmitted from one entity to another, the signaling overhead is also significant. 【0024】 Certain aspects of this disclosure and their embodiments can provide solutions to these or other challenges. This disclosure includes a solution that uses only delayed profiles (DPs) as AI / ML model inputs, for example, in which the DPs include only timing values ​​for multiple receive paths. Compared to PDPs, received power values ​​are not required as model inputs. Evaluation results show that the performance of the AI / ML model is maintained with only slight degradation. Other embodiments include various encoding and storage methods for reducing dataset size. 【0025】 The significantly reduced model input size offers a major advantage to AI / ML models. For example, the measurement burden is greatly reduced. This advantage exists regardless of whether the measurements need to be signaled from one entity to another. Furthermore, when the measurements in the model input need to be sent from one entity to another, the signaling overhead is greatly reduced. 【0026】 Certain embodiments may provide one or more of the following technical advantages. Certain disclosed embodiments offer significant advantages to AI / ML models because the model input size is greatly reduced. The advantages include a greatly reduced measurement burden. This advantage exists regardless of whether the measurements need to be signaled from one entity to another. Another advantage is a greatly reduced signaling overhead when measurements of the model input need to be sent from one entity to another. Such signaling includes at least the following: ● If the measurement entity and the training data collection entity are different during the training data collection phase, signaling is provided to send the measured values ​​of the model input from the measurement entity to the training data collection entity. ● When the measurement entity and the model inference entity are different during the model inference phase, signaling is provided for measuring the model input from the measurement entity to the model inference entity. ● In the model monitoring phase, if the measurement entity and the model monitoring entity are different, and the model monitoring method requires observation of model inputs, signaling is provided from the measurement entity to the model monitoring entity for the measurement of model inputs. Another advantage is that, for training data collection, the significantly reduced model input size means a reduced memory size for storing the model input measurements. Also, for model training, using DP instead of PDP reduces the number of features in the model input by half, resulting in a smaller model and enabling more efficient hardware implementation. In addition, for model monitoring, halving the number of features in the model input makes it easier to analyze the model input distribution and generate model monitoring metrics. 【0027】 The following describes how to use DP as an AI / ML model input. An AI / ML-based positioning use case is used to perform specific signaling analysis and performance evaluation. While this is a typical use case for AI / ML in wireless communications, it is a non-limiting example and is used only to illustrate the method. Those skilled in the art should understand that the same method and design principles can be extended to many other use cases where DP has features useful for inclusion in AI / ML model inputs. 【0028】 It should also be noted that the disclosed methods and design principles can be applied to AI / ML models deployed in various entities within a wireless communication network, including, but not limited to, UEs, network nodes, gNBs, DUs (distributed units), RUs (radio units), and LMFs (location and management functions). 【0029】 DP model input Existing methods often use two types of model inputs as AI / ML model inputs: CIR and PDP. CIR provides rich information, but the model input size is extremely large. As a smaller alternative, PDP can be used instead of CIR. When PDP is used for one pair of TRP-UEs, for a time window containing a total of N samples and n receive paths, the following two values ​​are measured and reported for each of the n receive paths: ● Arrival times of the i-th pass {t0, t1, t2, ..., t (n-1)} and ●Power of the i-th path, {P0, P1, P2, ..., P (n-1) i=0,1,.,n-1. 【0030】 A PDP is shown in Figure 2. Because both timing and power measurements need to be taken, PDPs tend to be more compact than CIRs but have a larger set of information. 【0031】 Considering a typical radio propagation environment, it can be expected that strong CIR or PDP taps will indicate clustered locations. Figure 3 is a diagram of a complete PDP for a line-of-sight channel. Figure 4 is a diagram of a complete PDP for a non-line-of-sight channel. In the example of the line-of-sight channel shown in Figure 3, strong taps clustered around tap #20 can be observed. In the example of the non-line-of-sight channel shown in Figure 4, strong taps concentrated in two clusters can also be observed. Using such radio knowledge, a second exemplary embodiment is to encode the locations of non-zero taps using run-length coding. 【0032】 To reduce the size of the information required to describe the observed paths, according to certain embodiments of this disclosure, high performance can be achieved by using only DP as the input to the AI / ML model. That is, the power aspect of each path can be ignored, and the timing information of each path, i.e., t0, t1, t2, ..., t (n-1) Only is retained. It can also be seen as a simplified version of PDP, where power is always set to a known fixed value (e.g., 1), i.e., no information about power aspects is provided. This is shown in Figure 5, which shows an example of DP that can be compared to the downsampled PDP shown in Figure 2. Note that DP can be written in various forms, which will be described below. 【0033】 In one method (direct instruction method), the arrival times of n paths are given by n timing values ​​{t0, t1, t2, ..., t (n-1) Provided via}, t0 <t1<t (n-1) Generally, timing values ​​can be represented by floating-point or fixed-point values. When timing information needs to be signaled via an interface, timing values ​​are usually quantized to integer values, where the value range is [t min ,t max The ] and resolution (i.e., quantization step size) are defined for the mapping between floating-point values ​​and integer values. 【0034】 In one variant, the reference time t ref (Seconds) are given as absolute values, but other timing values ​​are given as relative values ​​t i -t ref It is provided as follows. Absolute and relative values ​​can be expressed as floating-point or fixed-point values. If timing information needs to be signaled via the interface, absolute and relative values ​​are usually quantized to integer values. Arrival time t of one pass, for example, the first pass. ref If t0 is taken as the reference time, there is no need to send the reference path, for example, the relative timing value of t0. In this case, the timing value is (a) reference time t ref (b) the absolute value of t0, and the relative timing value t i -t ref Provided by [the formula], i = 1, 2, ..., n-1. 【0035】 In another method (bitmap encoding), a measurement time window T (seconds) and a sampling period T_s (seconds) are defined, resulting in a total of N samples within the time window, where N = T / T s Next, a binary vector (i.e., a bitmap) of length N is used to provide the arrival times of the paths, where {t0, t1, t2, ..., t (n-1) In the sample corresponding to}, the vector element has the value "1", but the vector element has the value "0" for the other Nn samples. Path arrival time t i The corresponding sample index is round(t i / T s It can be calculated as follows: 【0036】 Another method (reference time + bitmap method) provides a bitmap that gives path timing values ​​along with the reference time t ref (Seconds) is shown directly. The time window for bitmaps is t ref from t ref Up to +T. The time window contains a total of N samples, where N = T / T. sAnd here, the sampling period T s (seconds) is defined for a bitmap. A binary vector of length N (i.e., a bitmap) is t ref It is configured to provide the arrival times of paths to {t0, t1, t2, ..., t}, where {t0, t1, t2, ..., t}. (n-1) In the sample corresponding to}, the vector element has the value "1", but in the other Nn samples, the vector element has the value "0". Path arrival time t i The corresponding sample index is round(t i -t ref / T s It can be calculated as follows: In one variant, the arrival time t of one path, for example, the first path. ref =t0 is taken as the reference time. In this case, the bitmap does not need to include bits for the reference path, e.g., t0. 【0037】 There are several further embodiments that use DP as input to an AI / ML model, including embodiments A to D, which are further described below. Embodiment A uses delay profiles of the first detected path and additional paths. Embodiment B uses delay profiles of the n strongest paths. Embodiment C uses delay profiles along with signal quality indicators. Embodiment D uses delay profiles along with a small set of per-path indicators. These embodiments are given as examples and do not limit the teachings disclosed herein. 【0038】 Embodiment A: DP of the initially detected path and additional paths In this embodiment, the delay profile includes the first detected path regardless of the intensity of the first detected path; that is, t0 always indicates the arrival time of the first detected path. The remaining delay profile {t1, t2, ..., t (n-1)This provides information about (n-1) additional paths that come after the initially detected path. Typically, the (n-1) additional paths are selected to represent the strongest paths, e.g., the (n-1) paths with the highest received path power that come after the initially detected path. 【0039】 For uplinks, the UL relative arrival time (for the first pass) is T UL-RTOA :UL relative arrival time and its relative delay are given by path delay {t0, t1, t2, ..., t (n-1) This is an NG-RAN (Next Generation Wireless Access Network) measurement for}. UL-RTOA To perform the measurement, UL relative arrival time (T UL-RTOA ) is the start of subframe i containing the SRS (Sounding Reference Signal) received at the receiving point (RP)j, relative to the RTOA reference time. 【0040】 For downlinks, timing information for n paths can be provided in several ways. In one example, the DL Reference Signal Time Difference (DL RSTD) is obtained for a reference path, where RSTD provides the relative timing difference between the j-th neighboring TRP and the PRS (Positioning Reference Signal) reference TRP. In addition, additional detected path timing values ​​for the TRP or PRS resource are provided, related to the path timing used to determine the RSTD value. This is shown in Figure 6, which illustrates the reporting of RSTD and additional path timing information by the UE. In another example, the UE Rx-Tx Time Difference (UE-RxTxTimeDiff) is obtained for a reference path. The UE Rx-Tx Time Difference is T UE_RX -T UE_TX Defined as, where T UE-RX This is the UE reception timing of downlink subframe #i from the transmit point (TP), defined by the first path detected in time, and T UE-TXThis is the UE transmit timing of the uplink subframe #j that is temporally closest to subframe #i received from the TP. In addition, additional detected path timing values ​​for the TRP or PRS resource are provided, which are related to the path timing used to determine the UE-RxTxTimeDiff value. 【0041】 Embodiment B: DP of n strongest paths In this embodiment, the delay profile represents the n strongest paths. For example, in a time window T, the n paths with the highest received path power are selected. In this case, the first detected path may or may not be included, depending on whether the first path's received power belongs to the nth highest. On the other hand, this embodiment is the same as Embodiment A if the first path in the measurement report is simply considered to be the first detected path. That is, no additional effort is expended to search for the first detectable path, and the first detected path is found in the same way as the other paths, for example, by checking the highest set of path powers. 【0042】 Embodiment C: DP with signal quality indicator In this embodiment, the DP and signal quality indicator are used in combination for the AI / ML model input. That is, the DP provides pass-by-pass timing information, while the signal quality indicator provides overall received signal quality (i.e., not pass-by-pass). An example of a signal quality indicator is the received signal strength indicator (RSSI) (e.g., the UE measurement of the RSSI of an OFDM symbol carrying a PRS resource). Another example is the reference signal received power (RSRP), including DL PRS reference signal received power (DL PRS-RSRP) and UL SRS reference signal received power (UL SRS-RSRP). Another example is the reference signal received quality (RSRQ), which is the ratio between RSRP and RSSI. 【0043】 Embodiment D: DP with a small set of indicators per path In this embodiment, a small set of delay profiles and per-path indicators are provided as AI / ML model inputs. For example, the delay profiles of n detected paths are used as model inputs together with m received path powers (RSRPPs), where m < n. For example, a delay profile of n = 16 timing values is used together with m = 2 RSRPP values. The m = 2 RSRPP values can be provided for the two most representative paths. For example, (1) the first detected path and the path with the highest received power, (2) the two paths with the highest received path power. 【0044】 Analysis of Model Input Size Reduction Using only DP as a model input has significant advantages compared to using PDP or CIR. This advantage is due to at least two factors. One is that the number of features in the model input is significantly reduced. For a given number of paths (n), only DP uses n features (i.e., n path timings), PDP uses 2×n features (i.e., n path timings and n power values), and CIR uses 3×n features (i.e., n path timings, and 2×n values of CIR since the CIR value is complex). The other is that the model input size in terms of bits is significantly reduced. Below, as an example, the positioning use case is used to analyze the reduction in model input size (in bits) by using DP instead of PDP. In some examples, the analysis also includes CIR. 【0045】 Model Input Analysis Based on New Signaling Formats When CIR is used, the channel impulse response of Nt time-domain samples is used as input, and the size (number of bits) of the CIR for one pair of TRP-UE is N port *N t *B CIR has the size of, B CIRis the number of bits required to represent a single complex value of CIR at sample time. The CIR value at sample time should be represented by two floating-point values ​​of either {real, imaginary} or {magnitude, phase}. Therefore, B CIR =2*B CIR,real And here, B CIR,real n is the number of bits required to represent a single real value of the CIR. When CIRs related to NTRP are taken as model inputs, the total size of the CIR inputs is N TRP Multiplied by, i.e., N TRP *N port *N t *2*B CIRreal (bits). The CIR input is expected to make the model input size extremely large. 【0046】 Alternatively, PDP can be used instead of CIR to reduce the model input size. In 3GPP, the PDP input to the AI / ML model is dimensional N. TRP *N port *N t It is agreed that this is the case, but further improvements can be made. Here, N TRP This is the number of TRPs, N port is the number of transmit / receive antenna port pairs, N t n is the number of time-domain samples. For a TRP with multiple RX antenna ports, it is not actually necessary to retain the port dimension of the PDP, because the actual PDP of the channel is identical for all RX ports. Instead, according to one embodiment, the PDP should be averaged across all RX ports as an additional average over fast fading. This average across RX ports is N TRP *1*N t This provides more precise dimensions. Note that PDP is expressed as a real number. Power value B PDP When expressed in bits, the total size of the PDP in the model input is N TRP *1*N t *B PDP It is (bit). 【0047】 In summary, without downsampling, the training dataset contains N CIR or PDP samples. samples To store this information, the dataset occupies the following: ● About CIR: N samples *N TRP *N port *N t *2*B CIR,real (bit) ●About PDP: N samples *N TRP *1*N t *B PDP (bit) 【0048】 To further reduce the size of the dataset, downsampling of time-domain taps may be considered. In one exemplary embodiment, downsampling is performed to reduce the number of active (non-zero) time-domain taps while retaining as much wireless environment information as possible. In a non-limiting embodiment, the number of active non-zero time-domain taps is such that N taps have higher power than the remaining taps. t By holding only a few taps, N t Individual taps are selected. In the case of CIR, such tap selection is determined by averaging the power across the RX port. This downsampling (or subsampling) procedure attempts to retain only the most prominent channel information while discarding weaker, noisier information. 【0049】 To store such subsampled time-domain taps, an efficient representation is required to minimize the dataset size. In one exemplary embodiment, such a subsampled CIR or PDP has each sample representing two pieces of information, namely, the position of a non-zero tap for a TRP link, of length N. t The bitmap and the non-zero tap values ​​are stored. Therefore, in downsampling, the N of the downsampled CIR samples samplesTo store, the dataset uses the following. For the bitmap, N samples *N TRP *N t bits, and for the non-zero taps, N samples *N TRP *N port *N t ’*2*B CIR,real bits. This gives the total number of bits for the CIR as follows. ● = N samples *N TRP *N t +N samples *N TRP *N port *N t ’*2*B CIR,real ● = N samples *N TRP *(N t +N t ’*2*B CIR,real ) 【0050】 Similarly, for downsampling, to store the N samples downsampled PDP samples, the dataset uses the following. ● For the bitmap, N samples *N TRP *N t bits, and ● For the power values at the non-zero taps, N samples *N TRP *1*N t ’*B PDP bits. 【0051】 This gives the total number of bits for the PDP as follows. ● = N samples *N TRP *N t +N samples *N TRP *1*N t ’*B PDP ● = N samples *N TRP *(N t +N t ’*B PDP) 【0052】 Similarly, using a bitmap coding technique, N out of DP samples are downsampled. samples To store the data, the dataset is the total number of bits related to DP, i.e., N related to the bitmap. samples *N TRP *N t It uses bits, which is a significant reduction compared to the size of CIR or PDP inputs. 【0053】 To make a rough comparison between the three types of model inputs, the following specific values ​​are used as examples of high-resolution training datasets. TIFF2026519929000002.tif17170 【0054】 The resulting dataset size is as follows: ●2.949GB related to all 256 tap CIR inputs ● 737.3MB related to all 256 taps PDP input ● 391.7MB related to 32-tap CIR input ● 115.2MB related to 32-tap PDP input ● 23.04MB related to 32-tap DP input 【0055】 In the above example, it is assumed that time-domain taps (or paths) are represented via bitmaps. Bitmaps for timing indication are acceptable when the timing window T of the path is short. Otherwise, a large timing window T means a long bitmap, which is costly for signaling. In the above example, the number of bits required to represent real values ​​is the same regardless of whether it is CIR or PDP (i.e., B CIR,real =B PDP ) is also assumed. Implementation limitations, such as the achievable timing detection accuracy for a wideband reference signal versus a narrowband reference signal, are not considered. 【0056】 Model input analysis based on existing signaling formats The following sections provide an alternative form of analysis for representing PDP and DP values ​​as model inputs, where the timing of the received paths is given by (a) the absolute value of the reference time and (b) the relative value of the additional paths. 【0057】 The analysis assumes that measurement reports for model inputs need to be signaled from measurement entities (e.g., UE or NG-RAN) to model inference entities (e.g., LMF) via a standardized interface (e.g., LPP or NRPPa). Therefore, the number of bits required to signal a single measured model input value is specified. Downlink and uplink scenarios are considered. 【0058】 For positioning purposes, during the downlink, the UE performs measurements to the PRS from one or more TRPs. For the fifth path, the measurements include: ● First path timing information, which may be DL-RSTD or UE-RxTxTimeDiff. The range of the reported value for the first path timing information is a function of the integer k (see columns 2 and 3 in Figure 7). ●First-path power information, i.e., DL PRS-RSRPP of the first path. Up to Rel-17, the reported range of PRS-RSRPP is 0..126, which is the same as the range of PRS-RSRP. Therefore, 7 bits are required to report the measured value of DL PRS-RSRPP. 【0059】 For (n-1) additional paths, the measurements include the following: ● Per-path timing information, which is relative path delay and therefore has a smaller value range for reporting. The range of reported values ​​for additional path timing information is a function of the integer k (see columns 4 and 5 in Figure 7). ● Power information per path, i.e., DL PRS-RSRPP for each of the additional paths. Similarly, 7 bits are required to report the DL PRS-RSRPP measurements. 【0060】 For positioning, on the uplink, the NR-RAN performs measurements against the SRS transmitted by the target UE. For the first pass, the measurements include: ● The first pass timing information is UL RTOA. The range of the reported value for the first pass timing information is a function of the integer k (see columns 2 and 3 in Figure 7). ●First-pass power information, i.e., UL SRS-RSRPP for the first pass. Up to Rel-17, the reporting range for SRS-RSRPP is 0..126, which is the same as the range for SRS-RSRP. Therefore, 7 bits are required to report the UL SRS-RSRPP measurement. 【0061】 For (n-1) additional paths, the measurements include the following: ● Per-path timing information, which is relative path delay and therefore has a smaller value range for reporting. The range of reported values ​​for additional path timing information is a function of the integer k (see columns 2 and 3 in Figure 7). ● Power information per path, i.e., UL SRS-RSRPP for each of the additional paths. Similarly, 7 bits are required to report the UL SRS-RSRPP measurements. 【0062】 Since the number of bits for reporting timing and power information is the same for DL ​​and UL, the same table can be used to provide the model input size corresponding to the measurements. In Figure 7, the model input size for using only per-pass timing information (i.e., DP) can be calculated compared to using both per-pass timing and power information (i.e., PDP). Note that in Figure 7, the size calculation is for only one pair of TRP-UE. TRP When a pair is used to position a target UE, the full model input size for these measurements is N TRPIt is twice as large. Similarly, if we consider collecting N samples in the training dataset, N samples The multiplier also needs to be applied. The value of k is an integer and can be set. 【0063】 In Figure 7, the range of values ​​specified in NR Rel-16 / 17 is reused. The range of timing values ​​for the first pass is 2 k ×T c -985024×T resolution step c From 985024×T c The range of relative path delay is 2 k ×T c -8175×T resolution step c From 8175×T c Up to this point. The reported range is the integer value after quantization (i.e., mapping floating-point values ​​to integer values). Here, T c T is the basic timing unit used in NR. c = 1 / (Δf max ·N f ) = 0.51ns, where Δf max =480·10 3 It is Hz, N f = 4096. 【0064】 Figure 7 shows the reduction in model input size by using only per-path timing information compared to using both per-path timing and power information. The total number of paths is n. The same calculation applies to both DL and UL. The size is calculated for only one pair of TRP-UE. 【0065】 Corresponding to Figure 7, Figure 8 shows the savings as a percentage of n as a function of n, where n is the total number of observed paths. For n=9, the savings range from 32% to 42% for k=0 to 5, and for n=128, the savings range from 33% to 43% for k=0 to 5. Overall, the savings range is 32% to 43%. Figure 8 assumes the Rel-16 / 17 timing value range. 【0066】 The range of timing values ​​used in Figures 7 and 8 is extremely large, allowing it to support the maximum cell radius (e.g., in rural areas), for example, a rural macrocell with an inter-site distance (ISD) of 5000m. However, the AI / ML model is designed for small cells with typical ISD values ​​of 20m to 50m, such as indoor office or factory floors. Therefore, the range of path timing values ​​can be significantly reduced in the case of the AI / ML model. 【0067】 An example is shown. The reduced range of the timing value for the first pass is -7700xT c From 7700xT c This represents a 128x reduction compared to Rel-16 / 17, thus saving 7 bits when reporting absolute timing values ​​for the first pass. The reduced range of relative path delay is -255xT. c From 255xT c This represents a 32-fold reduction compared to Rel-16 / 17, thus saving 5 bits when reporting relative timing values ​​for additional passes. When using the reduced value range, further savings are achieved by using a delay profile instead of a power delay profile, as shown in Figure 9. As can be observed in Figure 9, the savings range is 42% to 63%. Figure 9 shows the savings as a percentage as a function of n, where n is the total number of passes. For AI / ML models deployed for smaller cells (including indoor factory cells), the reduced timing value range from Rel-16 / 17 is assumed. 【0068】 Since AI / ML models are being developed for smaller cells where conventional positioning methods fail (e.g., indoor factories, urban valleys), this demonstrates the advantages of using delay profiles as input to the AI / ML models. 【0069】 The impact of AI / ML models on the LCM (Life Cycle Management) stage. Generally, feature engineering is a crucial step in identifying the most important features to include in a model input, while removing redundant, irrelevant, or unimportant information. The goal is to keep the number of features in the model input small without significantly sacrificing model performance. This is achieved precisely by using DP only, rather than PDP or CIR. 【0070】 Reducing the number of features in a model input significantly impacts model training, model structure, and model size. Generally, smaller model input sizes lead to faster model training because the model needs to learn fewer features. Smaller model input sizes also allow for simpler model structures and / or smaller model sizes, which in turn makes it easier to implement the model in hardware for model inference. 【0071】 The reduced number of features for model input also significantly facilitates generating measurements, sending measurement reports (if necessary), storing measurements (e.g., for training data collection), and analyzing measurements (e.g., for model monitoring). 【0072】 Compared to PDP, for a given number of paths, DP has half the number of features, namely, comparisons of path timing only and both path timing and path power. Therefore, using DP as a model input is a significant advantage if it can be used instead of PDP without significantly degrading model performance. This impacts every stage of AI / ML model lifecycle management (LCM), as detailed below. 【0073】 Model training and generation: When using dynamic programming (DP), the number of features in the model input is significantly reduced, allowing for simpler model structures and / or smaller model sizes. Faster model training is also achievable, which is particularly important when on-device training is desired. 【0074】 Measuring model inputs is simpler when only timing information for the received path needs to be measured, i.e., when it is not necessary to measure power per path. Therefore, when using DP alone, the measurement burden is significantly reduced. This advantage exists regardless of whether the measured values ​​need to be signaled from one entity to another. 【0075】 Signaling of measurement reports for model input: The signaling overhead for sending measurement reports for model input is significantly reduced when using DP instead of PDP. In addition, the signaling overhead is significantly reduced when measurement values ​​for model input need to be sent from one entity to another. Such signaling includes at least the following: ● For all positioning cases 1 / 2a / 2b / 3a / 3b: In the training data collection phase, if the measurement entity and the training data collection entity are different, signaling is performed to send measured values ​​of the model input from the measurement entity to the training data collection entity. In the model monitoring phase, if the measurement entity and the model monitoring entity are different and the model monitoring method requires observation of model input statistics, signaling is performed to measure the model input from the measurement entity to the model monitoring entity. ● For positioning cases 2b and 3b: Signaling for measuring model input from measurement entities (UE in case 2b, NG-RAN node in case 3b) to model inference entities (LMF in case 2b / 3b) during the model inference phase. In cases 1 / 2a / 3a, when considering the model inference stage, it should be noted that the measurements do not need to be sent from one measuring entity to another. However, in cases 2b and 3b, the entity that performs the measurements for the model input is different from the entity that performs the model inference. Therefore, the measurement reports for the model input need to be sent from one entity to another for model inference. 【0076】 Memory of collected training datasets: When collecting training data, the significant reduction in the number of features and the model input size (in bits) also has the advantage of reducing the memory size required to store the model input measurements. 【0077】 Model monitoring: In model monitoring methods that use statistical analysis of model inputs, reducing the number of features in the model input makes it easier to analyze the model input data and generate model monitoring metrics. In addition, considering that model input values ​​within a sliding observation window are typically stored for model monitoring purposes, the storage size for model inputs is also reduced. 【0078】 In summary, for positioning cases 1 / 2a / 2b / 3a / 3c, when the number of features in the model input is reduced, as achieved by DP-only model input, there are advantages in model training and generation, measurement for model input, storage of collected training datasets, and model monitoring. Regarding signaling of measurement reports for model input, there are advantages when measurements need to be sent from one entity to another. 【0079】 Examples of application to positioning cases In a direct positioning AI / ML model, DPs corresponding to numerous TRPs are collected at an appropriate central node (such as an LMF) and used as model input to estimate the location of the UE. This is illustrated in Figure 10. DPs can be measured and signaled by the NG-RAN relative to the target UE, or measured and signaled by the target UE. Figure 10 illustrates AI / ML direct positioning where the LMF acts as a central node, processing DPs corresponding to numerous TRPs to generate an estimate of the target UE's location. 【0080】 If the model input is a measured value of the UL reference signal acquired by NG-RAN: At the i-th TRP, the delay profile of the {i-th TRP, target UE} pair can be determined from the UL signal (such as a sounding reference signal) transmitted by the target UE and received at the i-th TRP. The i-th TRP forwards or reports its DP measurement to the central node. Multiple TRPs transmit measured values ​​of delay profiles associated with the same target UE. Thus, an AI / ML model at the central node can take DPs from multiple TRPs as input and determine the location of the target UE as the model output. 【0081】 If the model input is a measurement of the DL reference signal acquired by the target UE: For the i-th TRP, the delay profile of the {i-th TRP, target UE} pair can be determined from the DL signal (such as the positioning reference signal) transmitted by the i-th TRP and received by the target UE. The UE reports its DP measurement associated with the i-th TRP to the central node. After the UE reports its DP measurement associated with multiple TRPs, the AI / ML model at the central node can take them as model inputs and determine the location of the target UE as the model output. 【0082】 In a centralized AI / ML-assisted positioning model, delay profiles corresponding to numerous transmit / receive points (TRPs) are collected at appropriate centralized nodes (such as gNBs or LMFs) and used as model input to estimate the direct path arrival time (ToA) between these numerous TRPs and the target UE. The estimated ToA is then forwarded to the LMF to determine the UE's position. 【0083】 The delay profile of the i-th TRP can be determined from the UL signal (such as a sounding reference signal) transmitted by the target UE and received at the i-th TRP. The i-th TRP transmits its DP measurement to a central node (e.g., a gNB associated with the i-th TRP). In the case of a central node connected to multiple TRPs, each TRP can transmit its DP measurement to a central node associated with a given TRP. This allows an AI / ML model at the central node to take multiple DPs from different TRPs as input and generate a ToA estimate as the model output. This is illustrated in Figure 11. Figure 11 illustrates AI / ML-assisted positioning where a gNB acts as a central node, processing DPs corresponding to multiple TRPs to generate an estimate of an unobserved direct path ToA, which is then further processed by a conventional positioning algorithm to position the target UE. 【0084】 The delay profile of the i-th TRP can be determined from the DL signal (such as a positioning reference signal) transmitted by the i-th TRP and received by the target UE. The AI / ML model on the UE side takes multiple DP measurements as model input and estimates ToA as model output. 【0085】 More generally, two or more nodes can process two or more DPs to generate direct path-to-a, which are then forwarded to the LMF to determine the UE location. This is illustrated in Figure 12. Figure 12 illustrates typical AI / ML-assisted positioning, where multiple nodes process DPs corresponding to numerous TRPs to generate estimates of unobserved direct path-to-a, which are then further processed by conventional positioning algorithms to determine the location of the target UE. 【0086】 Evaluation results of positioning cases Next, the 3GPP indoor factory (InF) ​​model shown in Figure 13 is used as a non-limiting example of a known deployment. In this scenario, 18 TRPs are deployed within the factory, and the locations of the TRPs are known in the network. With a clutter density of 60% and clutter heights and widths of 6m and 2m, respectively, this indoor factory scenario has a less than 1% Loss of Service (LoS) probability from the UE to any TRP. 【0087】 In this example, a concentrated ToA estimation model was used to determine the UE location, as shown in Figure 11. Two types of inputs, PDP and DP, were compared. In the two cases of PDP, a complete PDP (with 256 taps) and the strongest 9-tap PDP were considered. In DP, the number of taps considered were 128, 64, 32, 16, and 9. The AI / ML model was N samples The model is trained on a dataset of 20,000 PDP or DP samples. The trained model is then tested on a separate dataset of 4,000 PDP or DP samples. 【0088】 Figure 14 shows the 2D position errors at different percentiles. The following analysis uses the 90th percentile 2D positioning error as a reference point for comparison. If the 90th percentile 2D positioning error is E, it means that the UE 2D positioning error is less than E for 90% of the time. It can be expected that a PDP input with more taps will achieve better performance. Therefore, the x-tap PDP performance is expected to be between the 256-tap PDP performance and the 9-tap PDP performance in the table below, i.e., 0.668 to 1.044m. For DP-only inputs, there is actually an optimal number of tap settings. This is expected. Too few taps and the DP will not capture all useful information. Too many taps and useful information will be lost because both strong and weak taps will be represented by the same value of 1. 【0089】 It can be observed that a 16-tap DP input is 10% better than a 9-tap PDP. Furthermore, the best performance is achieved with a 32-tap DP input, which is 15cm worse than a 256-tap PDP but 22cm better than a 9-tap PDP. 【0090】 Additional Embodiments Another possible method embodiment under this disclosure is shown in Figure 15. Method 1000 includes a method performed by the UE to use DP as input to an AI or ML model. Step 1010 is to receive one or more delay profile data for the training phase. Step 1020 is to preprocess one or more delay profile data for the training phase. Step 1030 is to train the AI ​​or ML model with the preprocessed one or more delay profile data. Method 1000 may include a plurality of variations and embodiments, as well as / or additional and / or alternative steps. 【0091】 Another possible method embodiment under this disclosure is shown in Figure 16. Method 1200 includes a method performed by a UE to use DP as input to an AI or ML model. Step 1210 is to receive one or more DP data for the inference phase. Step 1220 is to preprocess one or more DP data for the inference phase. Step 1230 is to use one or more preprocessed DP data as one or more inputs for an AI or ML model. Method 1200 may include a plurality of variations and embodiments, as well as / or additional and / or alternative steps. 【0092】 Another possible method embodiment under this disclosure is shown in Figure 17. Method 1400 includes a method performed by a UE to provide a DP as input to an AI or ML model. Step 1410 is to receive one or more radio signals from one or more radio nodes. Step 1420 is to generate one or more delay profile data based at least in part on one or more radio signals. Step 1430 is to transmit one or more delay profile data to a network node. Method 1400 may include a plurality of variations and embodiments, as well as additional and / or alternative steps. 【0093】 Another possible method embodiment under this disclosure is shown in Figure 18. Method 1600 includes a method performed by a network node to use DP as input to an AI or ML model. Step 1610 is to receive one or more delay profile data. Step 1620 is to preprocess one or more delay profile data. Step 1630 is to send one or more preprocessed delay profile data to the UE for use as one or more inputs to an AI or ML model. Step 1640 is to receive one or more outputs of the AI ​​or ML model from the UE. Method 1600 may include a plurality of variations and embodiments, as well as / or additional and / or alternative steps. 【0094】 Another possible method embodiment under this disclosure is shown in Figure 19. Method 1800 includes a method performed by a network node to use DP as input to an AI or ML model. Step 1810 is to receive one or more delay profile data. Step 1820 is to preprocess one or more delay profile data. Step 1830 is to use one or more preprocessed delay profile data as one or more inputs for an AI or ML model. Step 1840 is to obtain one or more outputs of the AI ​​or ML model. Method 1800 may include a plurality of variations and embodiments, as well as / or additional and / or alternative steps. 【0095】 Another possible method embodiment under this disclosure is shown in Figure 20. Method 1900 includes a method performed by a first network node to provide a DP as input to an AI or ML model. Step 1910 is to receive one or more radio signals from one or more radio nodes. Step 1920 is to generate one or more delay profile data based at least in part on one or more radio signals. Step 1930 is to transmit one or more delay profile data to a second network node for use in one or more inputs to an AI or ML model. Method 1900 may include a plurality of variations and embodiments, as well as additional and / or alternative steps. 【0096】 Figure 21 shows examples of communication systems 2100 according to several embodiments. In this example, communication system 2100 includes a communication network 2102 which includes an access network 2104 such as a RAN and a core network 2106 which includes one or more core network nodes 2108. The access network 2104 includes one or more access network nodes (one or more of which may generally be referred to as network nodes 2110), such as network nodes 2110a and 2110b, or any other similar Third Generation Partnership Project (3GPP) access nodes or non-3GPP access points. Network nodes 2110 facilitate direct or indirect connectivity of UEs, such as by connecting UEs 2112a, 2112b, 2112c, and 2112d (one or more of which may generally be referred to as UE2112) to the core network 2106 over one or more wireless connections. 【0097】 Exemplary wireless communication on a wireless connection includes transmitting and / or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and / or other types of signals suitable for transmitting information, without using wires, cables, or other material conductors. Furthermore, in different embodiments, the communication system 1100 may include any number of wired or wireless networks, network nodes, UEs, and / or any other components or systems that may facilitate or participate in the communication of data and / or signals, whether via wired or wireless connections. The communication system 2100 may include and / or interface with any type of communication, telecommunications, data, cellular, wireless network, and / or other similar types of systems. 【0098】 UE2112 may be any of a wide variety of communication devices, including wireless devices that are positioned, configured, and / or operable to communicate wirelessly with network node 2110 and other communication devices. Similarly, network node 2110 may be configured, capable, set up, and / or operable to communicate directly or indirectly with UE2112 and / or other network nodes or devices in communication network 2102 in order to enable and / or provide network access, such as wireless network access, and / or to perform other functions, such as administration in communication network 2102. 【0099】 In the illustrated example, the core network 2106 connects network node 2110 to one or more hosts, such as host 2116. These connections may be direct or indirect, via one or more intermediate networks or devices. In other examples, network nodes may be directly coupled to hosts. The core network 2106 includes another core network node (e.g., core network node 2108) structured with hardware and software components. Since the characteristics of these components may be substantially similar to those described for the UE, network nodes, and / or hosts, their descriptions are generally applicable to the corresponding components of core network node 2108. An exemplary core network node includes one or more functions from among the following: Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (AUSF), Subscriber Identifier Decryption Function (SIDF), Unified Data Management (UDM), Security Edge Protected Proxy (SEPP), Network Exposure Function (NEF), and / or User Plane Function (UPF). 【0100】 Host 2116 may be owned or controlled by a service provider other than the operator or provider of the access network 2104 and / or the communication network 2102, and may be operated by or on behalf of the service provider. Host 2116 may host a variety of applications to provide one or more services. Examples of such applications include live and pre-recorded audio / video content, data collection services such as extracting and compiling data on various ambient conditions detected by multiple UEs, analytical functions, social media, functions for controlling or, optionally, interacting with remote devices, functions for alarms and surveillance centers, or any other such functions performed by the server. 【0101】 Overall, the communication system 2100 in Figure 21 enables connectivity between UEs, network nodes, and hosts. In that sense, the communication system may be configured to operate according to predefined rules or procedures, including but not limited to, certain standards: Global System for Mobile Communications (GSM), Universal Mobile Telecommunications System (UMTS), Long Term Evolution (LTE), and / or other suitable 2G, 3G, 4G, 5G, or any applicable next-generation standard (e.g., 6G); Wireless Local Area Network (WLAN) standards such as the IEEE 802.11 standard (WiFi); and / or any other suitable wireless communication standards such as Global Interoperability for Microwave Access (WiMAX), Bluetooth, Z-wave, Near Field Communication (NFC) ZigBee, LiFi, and / or any Low Power Wide Area Network (LPWAN) standards such as LoRa and Sigfox. 【0102】 In some examples, the communication network 2102 is a cellular network implementing 3GPP standardization features. Therefore, the communication network 2102 can support network slicing to provide different logical networks to different devices connected to the communication network 2102. For example, the communication network 2102 can provide ultra-high reliability low latency communication (URLLC) services to several UEs while providing extended mobile broadband (eMBB) services to other UEs and / or massive machine-type communications (mMTC) / massive IoT services to even further UEs. 【0103】 In some examples, UE2112 is configured to transmit and / or receive information without direct human interaction. For example, the UE may be designed to transmit information to access network 2104 on a predetermined schedule when triggered by an internal or external event, or in response to a request from access network 2104. Furthermore, the UE may be configured to operate in single, multi-RAT, or multi-standard modes. For example, the UE may operate with one or a combination of Wi-Fi, NR (New Radio), and LTE, i.e., it may be configured for multi-radio dual connectivity (MR-DC), such as E-UTRAN (Enhanced UMTS Terrestrial Radio Access Network) New Radio Dual Connectivity (EN-DC). 【0104】 In this example, hub 2114 communicates with access network 2104 to facilitate indirect communication between one or more UEs (e.g., UE2112c and / or 2112d) and a network node (e.g., network node 2110b). In some examples, hub 2114 may be a controller, router, content source and content analysis, or any other communication device described herein with respect to the UE. For example, hub 2114 may be a broadband router that enables access to the core network 2106 for the UE. In another example, hub 2114 may be a controller that sends commands or instructions to one or more actuators within the UE. Commands or instructions may be received from the UE, network node 2110, or by executable code, scripts, processes, or other instructions in hub 2114. In yet another example, hub 2114 may be a data collector acting as temporary storage for UE data, and in some embodiments may perform data analysis or other processing. In yet another example, hub 2114 may be a content source. For example, with respect to a UE, such as a VR headset, display, loudspeaker, or other media distribution device, the hub 2114 may retrieve VR assets, video, audio, or other media or data related to sensory information via network nodes, which the hub 2114 then provides to the UE either directly, after performing local processing, and / or after adding additional local content. In yet another example, the hub 2114 may function as a proxy server or orchestrator for the UE, particularly if one or more of the UEs are low-energy IoT devices. 【0105】 Hub 2114 can have a permanent / persistent or intermittent connection to network node 2110b. Hub 2114 can also enable different communication methods and / or schedules between Hub 2114 and UEs (e.g., UE2112c and / or 2112d), and between Hub 2114 and the core network 2106. In other examples, Hub 2114 connects to the core network 2106 and / or one or more UEs via a wired connection. Furthermore, Hub 2114 may be configured to connect to an M2M service provider on the access network 1104 and / or another UE via a direct connection. In some scenarios, a UE may establish a wireless connection with network node 2110 while still connected via Hub 2114 via a wired or wireless connection. In some embodiments, Hub 2114 may be a dedicated hub, i.e., a hub whose primary function is to route communications to and from UEs to and from network node 2110b. In other embodiments, the hub 2114 may be a non-dedicated hub, i.e., a device capable of routing communication between the UE and the network node 2110b, but also capable of acting as a communication start and / or end point for a specific data channel. 【0106】 Figure 22 shows a UE2200 in some embodiments. As used herein, UE refers to a device that is capable of, configured, and / or operable of communicating wirelessly with network nodes and / or other UEs. Examples of UEs include, but are not limited to, smartphones, mobile phones, cell phones, voice over IP (VoIP) phones, wireless local loop phones, desktop computers, personal digital assistants (PDAs), wireless cameras, game consoles or devices, music storage devices, playback devices, wearable terminal devices, wireless endpoints, mobile stations, tablets, laptop computers, laptop embedded equipment (LEE), laptop computer-equipped equipment (LME), smart devices, wireless customer premises equipment (CPE), vehicle-mounted or vehicle-embedded / integrated wireless devices, etc. Other examples include any UE identified by the Third Generation Partnership Project (3GPP), including narrowband Internet of Things (NB-IoT) UEs, machine-type communications (MTC) UEs, and / or enhanced MTC (eMTC) UEs. 【0107】 A UE can support device-to-device (D2D) communication, for example, by implementing 3GPP standards for side-link communication, dedicated short-range communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-board (V2I), or vehicle-to-all (V2X). In other examples, a UE does not necessarily have a user in the sense of a human user who owns and / or operates the associated device. Instead, a UE may represent a device (e.g., a smart sprinkler controller) that is intended for sale to or operation by a human user, but may not be associated with a specific human user, or may not be initially associated with a specific human user. Alternatively, a UE may represent a device (e.g., a smart electricity meter) that is not intended for sale to or operation by an end user, but may be associated with a user or operated for the user's benefit. 【0108】 The UE2200 includes processing circuitry 2202 operably coupled via bus 2204 to input / output interface 2206, power supply 2208, memory 2210, communication interface 2212, and / or any other components, or any combination thereof. Several UEs may utilize all or a subset of the components shown in Figure 10. The level of integration between components may vary from one UE to another. Furthermore, some UEs may include multiple instances of components, such as multiple processors, memories, transceivers, transmitters, and receivers. 【0109】 The processing circuit 2202 is configured to process instructions and data and may be configured to implement any sequential state machine capable of executing instructions stored in memory 2210 as a machine-readable computer program. The processing circuit 2202 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), etc.), programmable logic with appropriate firmware, a microprocessor or digital signal processor (DSP) with appropriate software, one or more stored computer programs, a general-purpose processor, or any combination of the above. For example, the processing circuit 2202 may include multiple central processing units (CPUs). 【0110】 In this example, the input / output interface 2206 may be configured to provide one or more interfaces to input devices, output devices, or one or more input and / or output devices. Examples of output devices include speakers, sound cards, video cards, displays, monitors, printers, actuators, emitters, smart cards, other output devices, or any combination thereof. Input devices can allow users to take in information to the UE2200. Examples of input devices include touch-sensitive or presence-sensitive displays, cameras (e.g., digital cameras, digital video cameras, webcams, etc.), microphones, sensors, mice, trackballs, directional pads, trackpads, scroll wheels, smart cards, etc. Presence-sensitive displays may include capacitive or resistive touch sensors to detect user input. Sensors may include, for example, accelerometers, gyroscopes, tilt sensors, force sensors, magnetometers, light sensors, proximity sensors, biosensors, or any combination thereof. Output devices can use the same type of interface ports as input devices. For example, a Universal Serial Bus (USB) port may be used to provide input and output devices. 【0111】 In some embodiments, the power supply 2208 is constructed as a battery or battery pack. Other types of power sources may be used, such as an external power source (e.g., an electrical outlet), a photovoltaic device, or a battery. The power supply 2208 may further include a power circuit for distributing power from the power supply 2208 itself and / or from an external power source via an interface such as an input circuit or power cable. Distributing power may, for example, be for charging the power supply 2208. The power circuit may perform any formatting, converting, or other modifications to the power from the power supply 2208 to make that power suitable for each component of the UE2200 to which it is supplied. 【0112】 Memory 2210 is or may be configured to include random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic disks, optical disks, hard disks, removable cartridges, flash drives, and other types of memory. For example, memory 2210 may include one or more application programs 2214, such as an operating system, a web browser application, a widget, a gadget engine, or other application, and corresponding data 2216. Memory 2210 can store a wide variety of operating systems or combinations of operating systems for use by the UE2200. 【0113】 Memory 2210 may be configured to include several physical drive units such as a redundant array of independent disks (RAID), flash memory, USB flash drives, external hard disk drives, thumb drives, pen drives, key drives, high-density digital versatile disk (HD-DVD) optical disc drives, internal hard disk drives, Blu-ray optical disc drives, holographic digital data storage (HDDS) optical disc drives, external mini dual in-line memory modules (DIMMs), synchronous dynamic random access memory (SDRAM), external microDIMM SDRAM, smart card memory such as a tamper-resistant module in the form of a universal integrated circuit card (UICC) containing one or more subscriber identification modules (SIMs) such as USIM and / or ISIM, other memory, or any combination thereof. The UICC may be, for example, an embedded UICC (eUICC), an integrated UICC (iUICC), or a removable UICC commonly known as a "SIM card". Memory 2210 can enable the UE2200 to access instructions, application programs, etc., stored on temporary or non-temporary memory media, offload data, or upload data. Products that utilize communication systems, such as manufactured goods, may be tangibly embodied as memory 2210 or within memory 2210, and memory 2210 is or may comprise a device-readable storage medium. 【0114】 The processing circuit 2202 may be configured to communicate with an access network or other networks using a communication interface 2212. The communication interface 2212 may comprise one or more communication subsystems and may include or be communicatively coupled to an antenna 2222. The communication interface 2212 may include one or more transceivers used to communicate, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or network node in the access network). Each transceiver may include a transmitter 2218 and / or receiver 2220 suitable for providing network communication (e.g., optical, electrical, frequency-allocated, etc.). Furthermore, the transmitter 2218 and receiver 2220 may be coupled to one or more antennas (e.g., antenna 2222) and may share circuit components, software or firmware, or alternatively, be implemented separately. 【0115】 In the illustrated embodiment, the communication functions of the communication interface 2212 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communication such as Bluetooth, near-field communication, location-based communication such as the use of the Global Positioning System (GPS) for determining location, other similar communication functions, or any combination thereof. The communication may be implemented in accordance with one or more communication protocols and / or standards such as IEEE 802.11, Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMAX, Ethernet, Transmission Control Protocol / Internet Protocol (TCP / IP), Synchronous Optical Network (SONET), Asynchronous Transfer Mode (ATM), QUIC, and Hypertext Transfer Protocol (HTTP). 【0116】 Regardless of the sensor type, the UE can provide an output of the data captured by its sensor through its communication interface 2212 via a wireless connection to a network node. The data captured by the UE's sensor can be communicated via another UE through a wireless connection to a network node. The output may be periodic (e.g., once every 15 minutes if it reports the temperature it detected), random (e.g., to equalize the load from reports from several sensors), responsive to a triggering event (e.g., when moisture is detected and an alert is sent), responsive to a request (e.g., a user-initiated request), or a continuous stream (e.g., a live video feed of a patient). 【0117】 As another example, the UE may include actuators, motors, or switches related to a communication interface configured to receive radio input from a network node via a wireless connection. The state of the actuators, motors, or switches may change in response to the received radio input. For example, the UE may include a motor that adjusts the control surface or rotors of a drone in flight according to the received input, or a robotic arm that performs a medical procedure according to the received input. 【0118】 When a UE takes the form of an Internet of Things (IoT) device, it may be a device for use in one or more application areas, which include, but are not limited to, urban wearable technology, augmented industrial applications, and healthcare. Non-exclusive examples of such IoT devices are devices that are connected refrigerators or freezers, TVs, connected lighting devices, electricity meters, robotic vacuum cleaners, voice-controlled smart speakers, home security cameras, motion detectors, thermostats, smoke detectors, door / window sensors, flood / humidity sensors, electric door locks, connected doorbells, air conditioning systems such as heat pumps, autonomous vehicles, surveillance systems, weather monitoring devices, vehicle parking monitoring devices, electric vehicle charging stations, smartwatches, fitness trackers, head-mounted displays for augmented reality (AR) or virtual reality (VR), wearables for haptic augmentation or perceptual augmentation, water sprinklers, animal or product tracking devices, sensors for monitoring plants or animals, industrial robots, unmanned aerial vehicles (UAVs), and any kind of medical device such as a heart rate monitor or remotely controlled surgical robot, or devices incorporated into them. The UE in the form of an IoT device comprises circuitry and / or software depending on the intended application of the IoT device, in addition to the other components described with respect to the UE2200 shown in Figure 22. 【0119】 In another specific example, in an IoT scenario, a UE may represent a machine or other device that performs monitoring and / or measurement and transmits the results of such monitoring and / or measurement to another UE and / or network node. In this case, the UE may be an M2M device, which may be called an MTC device in the context of 3GPP. In one specific example, the UE may implement the 3GPP NB-IoT standard. In other scenarios, the UE may represent a vehicle such as a car, bus, truck, ship, and airplane, or other equipment capable of monitoring its operating status and / or reporting on its operating status, or other functions associated with its operation. 【0120】 In practice, any number of UEs can be used together for a single use case. For example, the first UE may be the drone itself, or integrated within the drone, providing speed information of the drone (acquired through a speed sensor) to the second UE, which is a remote controller operating the drone. When the user makes changes from the remote controller, the first UE can adjust the throttle on the drone (for example, by controlling an actuator) to increase or decrease the drone's speed. The first and / or second UEs can also include two or more of the functions described above. For example, the UE may be equipped with sensors and actuators and be able to handle data communication about both the speed sensor and the actuator. 【0121】 Figure 23 shows a network node 3300 according to one embodiment. As used herein, a network node refers to a device that is configured, set up, and / or operable to communicate directly or indirectly with UEs in a communication network and / or with other network nodes or devices. Examples of network nodes include, but are not limited to, access points (APs) (e.g., radio access points) and base stations (BSs) (e.g., radio base stations, node Bs, evolved node Bs (eNBs), and NR node Bs (gNBs)). 【0122】 Base stations may be classified based on the amount of coverage they provide (or, in other words, the base station's transmit power level), and therefore may be called femto base stations, pico base stations, micro base stations, or macro base stations in response to the amount of coverage they provide. A base station may also be a relay node or relay donor node that controls relays. A network node may also include one or more (or all) parts of a distributed radio base station, such as a centralized digital unit and / or a remote radio unit (RRU), which may be called a remote radio head (RRH). Such a remote radio unit may or may not be integrated with an antenna as an antenna-integrated radio. Parts of a distributed radio base station may also be called nodes in a distributed antenna system (DAS). 【0123】 Other examples of network nodes include multiple transmit point (multi-TRP) 5G access nodes, MSR equipment such as multi-standard radio (MSR) BS, network controllers such as radio network controllers (RNCs) or base station controllers (BSCs), base station transceiver stations (BTSs), transmit points, transmit nodes, multi-cell / multicast coordination entities (MCEs), operation and maintenance (O&M) nodes, operation support system (OSS) nodes, self-organizing network (SON) nodes, positioning nodes (e.g., evolved serving mobile location centers (E-SMLCs)), and / or drive test minimization (MDTs). 【0124】 Network node 3300 includes processing circuitry 3302, memory 3304, communication interface 3306, and power supply 3308. Network node 3300 may consist of multiple physically separate components (e.g., node B components and RNC components, or BTS components and BSC components), each of which may have its own separate components. In certain scenarios where network node 3300 has multiple separate components (e.g., BTS components and BSC components), one or more of the separate components may be shared among several network nodes. For example, a single RNC may control multiple node Bs. In such a scenario, each unique node B-RNC pair may, in some instances, be considered a single separate network node. In some embodiments, network node 1300 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be duplicated (e.g., separate memory 3304 for different RATs), and some components may be reused (e.g., the same antenna 3310 may be shared by different RATs). Network node 3300 may also include multiple sets of various indicated components for different wireless technologies, such as GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, radio frequency identification (RFID), or Bluetooth wireless technologies, which are integrated into network node 1300. These wireless technologies may be integrated into the same or different chips or sets of chips and other components within network node 1300. 【0125】 The processing circuit 3302 may include one or more combinations of microprocessors, controllers, microcontrollers, central processing units, digital signal processors, application-specific integrated circuits, field-programmable gate arrays, or any other suitable computing devices, resources, or combinations of hardware, software, and / or encoded logic, which are capable of operating alone or in conjunction with other network node 3300 components such as memory 3304 to provide network node 3300 functionality. 【0126】 In some embodiments, the processing circuit 3302 includes a system-on-a-chip (SOC). In some embodiments, the processing circuit 3302 includes one or more of the radio frequency (RF) transceiver circuit 3312 and the baseband processing circuit 3314. In some embodiments, the radio frequency (RF) transceiver circuit 3312 and the baseband processing circuit 3314 may be on separate chips (or sets of chips), boards, or units such as radio and digital units. In alternative embodiments, some or all of the RF transceiver circuit 3312 and the baseband processing circuit 3314 may be on the same chip or set of chips, board, or unit. 【0127】 Memory 3304 may include, but is not limited to, any form of volatile or non-volatile computer-readable and / or computer-executable memory device that stores information, data, and / or instructions that can be used by processing circuit 3302, including persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random-access memory (RAM), read-only memory (ROM), mass storage media (e.g., hard disk), removable storage media (e.g., flash drive, compact disc (CD), or digital video disc (DVD)), and / or any other volatile or non-volatile non-temporary device-readable and / or computer-executable memory device. Memory 3304 can store any appropriate instructions, data, or information, including computer programs, software, and applications that include one or more logic, rules, codes, tables, and / or other instructions that can be executed by processing circuit 3302 and made available to network node 3300. Memory 3304 may be used to store calculations performed by processing circuit 3302 and / or data received via communication interface 3306. In some embodiments, the processing circuit 3302 and the memory 3304 are integrated. 【0128】 The communication interface 3306 is used in wired or wireless communication of signaling and / or data between network nodes, access networks, and / or UEs. As illustrated, the communication interface 3306 includes, for example, a port / terminal 3316 for sending and receiving data to and from the network over a wired connection. The communication interface 3306 also includes a wireless front-end circuit 3318, which is coupled to or, in some embodiments, may be part of the antenna 3310. The wireless front-end circuit 3318 includes a filter 3320 and an amplifier 3322. The wireless front-end circuit 3318 may be connected to the antenna 3310 and the processing circuit 3302. The wireless front-end circuit may be configured to adjust signals communicated between the antenna 3310 and the processing circuit 3302. The wireless front-end circuit 3318 may receive digital data transmitted to other network nodes or UEs via the wireless connection. The wireless front-end circuit 3318 can convert digital data into a radio signal with appropriate channel and bandwidth parameters using a combination of filter 3320 and / or amplifier 3322. The radio signal may then be transmitted via antenna 3310. Similarly, when receiving data, antenna 3310 can collect a radio signal, which is then converted into digital data by the wireless front-end circuit 3318. The digital data may then be passed to processing circuit 3302. In other embodiments, the communication interface may include different components and / or different combinations of components. 【0129】 In certain alternative embodiments, the network node 3300 does not include a separate radio front-end circuit 3318; instead, the processing circuit 3302 includes the radio front-end circuit and is connected to the antenna 3310. Similarly, in some embodiments, all or part of the RF transceiver circuit 3312 is part of the communication interface 3306. In yet another embodiment, the communication interface 3306, as part of a radio unit (not shown), includes one or more ports or terminals 3316, a radio front-end circuit 3318, and an RF transceiver circuit 3312, and the communication interface 3306 communicates with a baseband processing circuit 3314, which is part of a digital unit (not shown). 【0130】 Antenna 3310 may include one or more antennas or antenna arrays configured to transmit and / or receive radio signals. Antenna 3310 may be coupled to the radio front-end circuit 3318 and may be any type of antenna capable of wirelessly transmitting and receiving data and / or signals. In certain embodiments, antenna 3310 may be isolated from the network node 3300 and connectable to the network node 3300 via an interface or port. 【0131】 The antenna 3310, the communication interface 3306, and / or the processing circuit 3302 may be configured to perform any receiving operations and / or some acquiring operations as described herein as being performed by a network node. Any information, data, and / or signals may be received from the UE, another network node, and / or any other network equipment. Similarly, the antenna 3310, the communication interface 3306, and / or the processing circuit 3302 may be configured to perform any transmitting operations as described herein as being performed by a network node. Any information, data, and / or signals may be transmitted to the UE, another network node, and / or any other network equipment. 【0132】 Power supply 3308 supplies power to the various components of network node 3300 in a form appropriate to each component (for example, at the voltage and current levels required for each component). Power supply 3308 may further include, or be coupled to, a power management circuit for supplying power to the components of network node 3300 to perform the functions described herein. For example, network node 3300 may be connectable to an external power source (e.g., a power grid, an electrical outlet) via an interface such as an input circuit or electrical cable, thereby allowing the external power source to power the power circuit of power supply 3308. As a further example, power supply 3308 may include a power source in the form of a battery or battery pack connected to or integrated into the power circuit. The battery can provide backup power in the event of an external power failure. 【0133】 Embodiments of the network node 3300 may include additional components other than those shown in Figure 23 to provide some aspects of the network node's functionality, including any of the functions described herein and / or functions necessary to support the subject matter described herein. For example, the network node 3300 may include user interface equipment for enabling information input to and output from the network node 3300. This may enable a user to perform diagnostic, maintenance, repair, and other administrative functions for the network node 3300. 【0134】 Figure 24 is a block diagram of host 4400, which may be an embodiment of host 2116 in Figure 21, according to various aspects described herein. Host 4400 as used herein may be a variety of combinations of hardware and / or software, or comprise a variety of combinations of hardware and / or software, including standalone servers, blade servers, cloud implementation servers, distributed servers, virtual machines, containers, or processing resources in a server farm. Host 4400 can provide one or more services to one or more UEs. 【0135】 The host 4400 includes an input / output interface 4406, a network interface 4408, a power supply 4410, and a processing circuit 4402 operably coupled via a bus 4404 to a memory 4412. Other embodiments may include other components. The characteristics of these components may be substantially the same as those described with respect to the devices in previous figures, such as Figures 22 and 23, and thereafter, their descriptions are generally applicable to the corresponding components of the host 4400. 【0136】 Memory 4412 may include one or more computer programs, each containing one or more host application programs 4414 and data 4416, the data 4416 of which may include user data, e.g., data generated by the UE for host 4400, or data generated by host 4400 for the UE. Embodiments of host 4400 may utilize only a subset or all of the illustrated components. Host application programs 4414 may be implemented in a container-based architecture and may provide support for video codecs (e.g., Multipurpose Video Coding (VVC), High Efficiency Video Coding (HEVC), Advanced Video Coding (AVC), MPEG, VP9) and audio codecs (e.g., FLAC, Advanced Audio Coding (AAC), MPEG, G.711), including transcoding for multiple different classes, types, or implementation forms of UEs (e.g., handsets, desktop computers, wearable display systems, head-up display systems). The host application program 4414 can also provide user authentication and licensing checks and periodically report health, route, and content availability to central nodes such as devices within or on the edge of the core network. Thus, host 4400 can select and / or direct different hosts for over-the-top services for the UE. The host application program 4414 can support various protocols such as HTTP Live Streaming (HLS), Real-Time Messaging Protocol (RTMP), Real-Time Streaming Protocol (RTSP), and Dynamic Adaptive Streaming over HTTP (MPEG-DASH). 【0137】 Figure 25 is a block diagram showing a virtualization environment 5500 in which functions implemented by some embodiments may be virtualized. In this context, virtualization means creating a virtual version of an apparatus or device, which may include virtualizing hardware platforms, storage devices, and networking resources. The virtualization used herein may apply to any device or its components described herein and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functions described herein may be implemented as virtual components run by one or more virtual machines (VMs) implemented within one or more virtualization environments 5500, which are hosted by one or more hardware nodes, such as network nodes, UEs, core network nodes, or hardware computing devices acting as hosts. Furthermore, in embodiments in which the virtual nodes do not require wireless connectivity (e.g., core network nodes or hosts), the nodes may be fully virtualized. 【0138】 Application 5502 (which may alternatively be referred to as a software instance, virtual appliance, network function, virtual node, virtual network function, etc.) runs within a virtualized environment 5500 to realize some of the features, functions, and / or benefits of some of the embodiments disclosed herein. 【0139】 Hardware 5504 includes processing circuits, memory for storing software and / or instructions executable by the hardware processing circuits, and / or other hardware devices described herein, such as network interfaces and input / output interfaces. The software is executed by the processing circuits to instantiate one or more virtualization layers 5506 (also called hypervisors or virtual machine monitors (VMMs)), providing VM5508a and 5508b (one or more of which may commonly be referred to as VM5508), and / or may implement any of the functions, features, and / or benefits described with respect to some embodiments described herein. The virtualization layer 5506 may present a virtual operating platform that appears to the VM5508 as networking hardware. 【0140】 VM5508 may include virtual processing, virtual memory, virtual networking or interfaces, and virtual storage, and may be run by the corresponding virtualization layer 5506. Different embodiments of the virtual appliance 5502 example may be implemented in one or more VM5508s, and the implementation may be carried out in different ways. Hardware virtualization is referred to as network function virtualization (NFV) in some contexts. NFV may be used to aggregate many types of network equipment on industry-standard high-volume server hardware, physical switches, and physical storage that can be located in data centers and customer premises equipment. 【0141】 In the context of NFV, VM5508 may be a software implementation of a physical machine that runs programs as if they were running on a physical, non-virtualized machine. Each VM5508 and the portion of hardware 5504 running its VM form a separate virtual network element, whether it is dedicated hardware for the VM and / or hardware shared between that virtual machine and other VMs. Furthermore, in the context of NFV, the virtual network function is responsible for handling specific network functions running on one or more VM5508s on hardware 5504, and corresponds to application 5502. 【0142】 Hardware 5504 may be implemented in a standalone network node with general or specific components. Hardware 5504 can achieve some functions through virtualization. Alternatively, hardware 5504 may be part of a larger cluster of hardware (such as in a data center or CPE) where many hardware nodes cooperate and are managed via management and orchestration 5510, which oversees the lifecycle management of applications 5502. In some embodiments, hardware 5504 is coupled to one or more radio units, each including one or more transmitters and one or more receivers, which may be coupled to one or more antennas. The radio units may communicate directly with other hardware nodes via one or more suitable network interfaces and may be used in combination with virtual components to provide a virtual node with radio capabilities, such as a radio access node or base station. In some embodiments, some signaling may be provided by using a control system 5512, which may alternatively be used for communication between hardware nodes and radio units. 【0143】 Figure 26 shows a communication diagram of host 6602 communicating with UE 6606 via network node 6604 through a partial wireless connection, according to one embodiment. Next, exemplary implementations of various embodiments of the UEs (such as UE 2112a in Figure 21 and / or UE 2200 in Figure 22), network nodes (such as network node 2110a in Figure 21 and / or network node 3300 in Figure 23), and hosts (such as host 2116 in Figure 21 and / or host 4400 in Figure 24), as discussed in the previous paragraph, will be described with reference to Figure 26. 【0144】 Like host 4400, an embodiment of host 6602 includes hardware such as a communication interface, processing circuitry, and memory. Host 6602 also includes software that is stored in or accessible by host 6602 and executable by the processing circuitry. The software may include a host application that can operate to serve remote users, such as UE6606, connected via an over-the-top (OTT) connection 6650 extending between UE6606 and host 6602. When serving remote users, the host application may provide user data transmitted using the OTT connection 6650. 【0145】 Network node 6604 includes hardware that enables network node 6604 to communicate with host 6602 and UE6606. The connection 6660 may be direct or pass through a core network (similar to core network 2106 in Figure 21) and / or one or more other intermediate networks, such as one or more public networks, private networks, or hosted networks. For example, the intermediate network could be a backbone network or the internet. 【0146】 The UE6606 includes hardware and software that is stored in or accessible by the UE6606 and executable by the UE's processing circuitry. The software includes client applications, such as a web browser or operator-specific “app,” which may be capable of operating to serve human or non-human users via the UE6606 with the support of host 6602. On host 6602, a running host application can communicate with a running client application via an OTT connection 6650 that terminates on UE6606 and host 6602. When serving a user, the UE's client application can receive request data from the host's host application and provide user data in response to the request data. The OTT connection 6650 can transfer both the request data and the user data. The UE's client application can interact with the user and generate user data to provide to the host application via the OTT connection 6650. 【0147】 The OTT connection 6650 can provide connectivity between host 6602 and UE6606 by extending through connection 6660 between host 6602 and network node 6604, and through wireless connection 6670 between network node 6604 and UE6606. Connections 6660 and wireless connection 6670, which may be provided by the OTT connection 6650, are depicted abstractly to illustrate communication between host 6602 and UE6606 via network node 6604 without explicitly referring to any intermediate devices and the exact routing of messages through these devices. 【0148】 As an example of transmitting data via the OTT connection 6650, in step 6608, host 6602 provides user data, which may be done by running a host application. In some embodiments, the user data is associated with a specific human user interacting with UE6606. In other embodiments, the user data is associated with UE6606 sharing data with host 6602 without explicit human interaction. In step 6610, host 6602 initiates a transmission carrying user data toward UE6606. Host 6602 may initiate a transmission in response to a request sent by UE6606. The request may be triggered by human interaction with UE6606 or by the operation of a client application running on UE6606. The transmission may pass through network node 6604 in accordance with the teachings of embodiments described throughout this disclosure. Thus, in step 6612, network node 6604 transmits the user data carried in the transmission initiated by host 6602 toward UE6606 in accordance with the teachings of embodiments described throughout this disclosure. In step 6614, UE6606 receives the user data carried in the transmission, which may be executed by a client application running on UE6606 associated with a host application run by host 6602. 【0149】 In some examples, UE6606 runs a client application that provides user data to host 6602. User data may be provided in response to or in reaction to data received from host 6602. Thus, in step 6616, UE6606 may provide user data, which may be done by running a client application. When providing user data, the client application may further consider user input received from the user via the input / output interface of UE6606. Regardless of the particular manner in which the user data is provided, in step 6618, UE6606 initiates a transmission of the user data toward host 6602 via network node 6604. In step 6620, in accordance with the teachings of embodiments described throughout this disclosure, network node 6604 receives user data from UE6606 and initiates a transmission of the received user data toward host 6602. In step 6622, host 6602 receives the user data carried in the transmission initiated by UE6606. 【0150】 One or more of the various embodiments improve the implementation of OTT services provided to UE6606 by using an OTT connection 6650 in which the wireless connection 6670 forms the final segment. More precisely, the teachings of these embodiments may improve data rate, latency, and / or power consumption, thereby providing benefits such as reduced user latency, relaxed file size limitations, improved content resolution, enhanced responsiveness, and / or extended battery life. 【0151】 In an exemplary scenario, factory status information may be collected and analyzed by host 6602. As another example, host 6602 may process audio and video data that may have been extracted from the UE for use in creating maps. As yet another example, host 6602 may collect and analyze real-time data to assist in controlling traffic congestion (e.g., controlling traffic signals). As yet another example, host 6602 may store surveillance video uploaded by the UE. As yet another example, host 6602 may store or control access to media content, such as video, audio, VR, or AR, which host 6602 can broadcast, multicast, or unicast to the UE. As yet another example, host 6602 may be used for energy pricing, remote control of non-time-critical electrical loads to balance power generation demand, location services, presentation services (such as accumulating diagrams, etc., from data collected from remote devices), or any other function that collects, retrieves, stores, analyzes, and / or transmits data. 【0152】 In some embodiments, measurement procedures may be provided for the purpose of monitoring data rate, latency, and other factors, which are improved by one or more embodiments. Furthermore, optional network functions may exist for reconfiguring the OTT connection 6650 between host 6602 and UE6606 in response to variations in measurement results. Measurement procedures and / or network functions for reconfiguring the OTT connection may be implemented in the software and hardware of host 6602 and / or UE6606. In some embodiments, sensors (not shown) may be deployed in or in relation to other devices through which the OTT connection 6650 passes, and the sensors may participate in the measurement procedure by supplying values ​​for the monitored quantities exemplified above, or values ​​for other physical quantities that the software can calculate or estimate the monitored quantities for. Reconfiguring the OTT connection 6650 may include message formatting, retransmission settings, preferred routing, etc., and the reconfiguration does not need to directly change the operation of network node 6604. Such procedures and functions are known and practiced in the art. In certain embodiments, the measurements may involve proprietary UE signaling that facilitates measurements such as throughput, propagation time, and latency by the host 6602. The measurements may be implemented in such a way that software uses an OTT connection 6650 to ensure that messages, particularly empty or "dummy" messages, are sent while monitoring propagation time, errors, etc. 【0153】 The computing devices described herein (e.g., UEs, network nodes, hosts) may include the shown combinations of hardware components, but other embodiments may comprise computing devices having different combinations of components. It should be understood that these computing devices may comprise any suitable combination of hardware and / or software required to perform the tasks, features, functions, and methods disclosed herein. The determining, calculating, acquiring, or similar operations described herein may be performed by processing circuits, which may process information by, for example, converting acquired information to other information, comparing acquired or converted information with information stored in a network node, and / or performing one or more operations based on the acquired or converted information and as a result of the processing making a decision. Furthermore, although components are shown as a single box located within a larger box, or as a single box nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that constitute a single shown component, and functions may be separated between distinct components. For example, a communication interface may be configured to include any of the components described herein, and / or the functions of the components may be separated between the processing circuit and the communication interface. In another example, the non-computation-intensive functions of any of such components may be implemented in software or firmware, while the computation-intensive functions may be implemented in hardware. 【0154】 In certain embodiments, some or all of the functions described herein may be provided by a processing circuit that executes instructions stored in memory, which in certain embodiments may be a computer program product in the form of a non-temporary computer-readable storage medium. In alternative embodiments, some or all of the functions may be provided by a processing circuit without executing instructions stored in a separate or individual device-readable storage medium, such as in a hardwired manner. In any of those particular embodiments, whether or not it executes instructions stored in a non-temporary computer-readable storage medium, the processing circuit may be configured to perform the functions described. The benefits provided by such functions are enjoyed by the computing device as a whole, and / or generally by the end user and the wireless network, not limited to the processing circuit alone or other components of the computing device. 【0155】 It will be understood that computer systems take on an increasingly diverse range of forms. In this description and in the claims, the terms “controller,” “computer system,” or “computing system” are broadly defined to include any device or system, or any combination thereof, comprising at least one physical and tangible processor, and physical and tangible memory capable of having computer-executable instructions that can be executed by the processor. Not limited to examples, the terms “computer system” or “computing system” as used herein are intended to include personal computers, desktop computers, laptop computers, tablets, handheld devices (e.g., mobile phones, PDAs, pagers), microprocessor-based or programmable consumer electronics, minicomputers, mainframe computers, multiprocessor systems, network PCs, distributed computing systems, data centers, message processors, routers, switches, and even devices not previously considered computing systems, such as wearables (e.g., eyeglasses). 【0156】 A computing system also has several structures on which it rests, often referred to as “executable components.” For example, the memory of a computing system may contain executable components. The term “executable component” is a name for a structure that is well understood by those skilled in the art of computing as a structure that may be software, hardware, or a combination thereof. For example, when implemented in software, those skilled in the art will understand that the structure of an executable component may include software objects, routines, methods, etc., that can be executed by one or more processors on the computing system, regardless of whether such executable components reside in the heap of the computing system or on a computer-readable storage medium. The structure of an executable component resides on a computer-readable medium in such a form that, when executed by one or more processors of the computing system, it is operable to cause the computing system to perform one or more functions, such as the functions and methods described herein. Such a structure may be directly computer-readable by the processor, as is the case when the executable component is binary. Alternatively, the structure may be configured and / or compiled in an interpretable manner, whether in a single step or multiple steps, to produce a binary number that is directly interpretable by the processor. 【0157】 The terms “components,” “services,” “engines,” “modules,” “controls,” and “generators” may also be used in this description. These terms, as used in this description and in this context, are intended to be synonymous with the term “executable components,” whether expressed with or without modifying clauses, thereby having the same structure as will be well understood by those skilled in computing. 【0158】 In terms of computer implementations, a computer is generally understood to comprise one or more processors or one or more controllers, and the terms computer, processor, and controller may be used interchangeably. When provided by a computer, processor, or controller, the functionality may be provided by a single dedicated computer or processor or controller, a single shared computer or processor or controller, or by multiple individual computers or processors or controllers, some of which may be shared or distributed. Furthermore, the terms “processor” or “controller” may also refer to other hardware capable of performing such functionality and / or running software, such as the exemplary hardware described above. 【0159】 In general, various exemplary embodiments may be implemented in hardware or dedicated chips, circuits, software, logic, or any combination thereof. For example, some embodiments may be implemented in hardware, while others may be implemented in firmware or software that can be executed by a controller, microprocessor, or other computing device, but this disclosure is not limited thereto. Various embodiments of the exemplary embodiments of this disclosure may be shown and described as block diagrams, flowcharts, or using any other graphical representation, but it should be understood that these blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuits or logic, general-purpose hardware or controllers or other computing devices, or any combination thereof, as non-limiting examples. 【0160】 Not all computing systems require a user interface, but in some embodiments, a computing system includes a user interface for use in communicating information with a user. The user interface may include output and input mechanisms. The principles described herein are not limited to strict output or input mechanisms and therefore depend on the nature of the device. However, output mechanisms may include, for example, speakers, displays, haptic outputs, projections, holograms, etc. Examples of input mechanisms may include, for example, microphones, touchscreens, projections, holograms, cameras, keyboards, styluses, mice, or other pointer inputs, any type of sensor, etc. 【0161】 Abbreviations and defined terms To aid in understanding the scope and content of this specification and the appended claims, several selected terms are defined below directly. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this disclosure relates. 【0162】 As used herein, the terms “approximately,” “about,” and “substantially” refer to an amount or condition that is close to a specific stated amount or condition that still performs the desired function or achieves the desired result. For example, the terms “approximately,” “about,” and “substantially” may refer to an amount or condition that deviates by less than 10%, or less than 5%, or less than 1%, or less than 0.1%, or less than 0.01%, from the specifically stated amount or condition. 【0163】 Various aspects of the Disclosure, including devices, systems, and methods, may be shown with respect to one or more embodiments or implementations that are essentially illustrative. As used herein, the term “exemplary” means “acting as an example, case, or illustration” and should not necessarily be construed as being preferable or advantageous to other embodiments disclosed herein. Furthermore, references to “implementations” of the Disclosure or embodiments include specific references to one or more embodiments thereof, and vice versa, and are intended to provide illustrative examples without limiting the scope of the Disclosure as directed not by this Specified Publication but by the appended claims. 【0164】 Unless implicitly or explicitly understood or stated otherwise, words appearing in the singular form as used herein include their plural equivalents, and words appearing in the plural form include their singular equivalents. Therefore, note that the singular forms “a,” “an,” and “the” as used herein and in the appended claims include plural referents unless the context explicitly specifies otherwise. For example, a reference to a singular referent (e.g., “a widget”) includes one, two, or more referents unless implicitly or explicitly understood or stated otherwise. Similarly, a reference to multiple referents should be interpreted as including one and / or multiple referents unless the content and / or context explicitly specifies otherwise. For example, a reference to a plural form of a referent (e.g., “widgets”) does not necessarily require multiple such referents. Instead, unless otherwise stated, it should be understood that one or more referents are intended herein, regardless of the presumed number of referents. 【0165】 References herein to “one embodiment,” “an embodiment,” and “exemplary embodiment” indicate that the embodiments described may include certain features, structures, or characteristics, but not all embodiments necessarily include such features, structures, or characteristics. Furthermore, such phrases do not necessarily refer to the same embodiment. Moreover, when certain features, structures, or characteristics are described in relation to an embodiment, it is known to those skilled in the art that such features, structures, or characteristics will be affected in relation to other embodiments, whether or not they are explicitly described. 【0166】 Terms such as “first” and “second” may be used herein to describe various elements, but it should be understood that these elements should not be limited by these terms. These terms are used merely to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, the first element may be called the second element, and similarly, the second element may be called the first element. The term “and / or” as used herein includes any and all combinations of one or more of the relevant enumerated terms. 【0167】 As used herein, the terms “comprises,” “comprising,” “has,” “having,” “includes,” and / or “including” specify the presence of the described features, elements, and / or components, but do not exclude the presence or addition of one or more other features, elements, components, and / or combinations thereof. 【0168】 conclusion This disclosure includes any novel features or combinations of features disclosed herein, whether express or generalized. Various modifications and adaptations to the exemplary embodiments of this disclosure may become apparent to those skilled in the art in light of the above description when read together with the accompanying drawings. However, any and all modifications still fall within the scope of the non-limiting and exemplary embodiments of this disclosure. 【0169】 With respect to any given component or embodiment described herein, it should be understood that any of the possible candidate or alternative forms listed for that component may be used individually or in combination with each other, unless implicitly or explicitly understood otherwise or otherwise stated. Furthermore, it should be understood that the list of such candidate or alternative forms is illustrative and not limiting, unless implicitly or explicitly understood otherwise or otherwise stated. 【0170】 Furthermore, unless otherwise indicated, numbers representing quantities, components, distances, or other measurements used herein and in the claims should be understood to be modified by the term “approximately” when the term is defined herein. Therefore, unless otherwise indicated, the numerical parameters described herein and in the appended claims are approximations that may vary depending on the desired properties to be obtained by the subject matter presented herein. At a minimum, and not as an attempt to limit the application of the doctrine of equivalents to the claims, each numerical parameter should be interpreted at least in light of the number of significant figures reported and by applying ordinary rounding techniques. While the numerical ranges and parameters describing a wide range of the subject matter presented herein are approximations, the numbers described in specific examples are reported as accurately as possible. However, any numerical value inherently contains some error, which inevitably arises from the standard deviation found in their respective test measurements. 【0171】 Any headings and subheadings used herein are for organizational purposes only and are not intended to limit the scope of this specification or the claims. The terms and expressions used herein are descriptive, not restrictive, and in the use of such terms and expressions, no equivalents or parts thereof of the features shown and described are excluded, and various modifications are possible within the scope of this disclosure. Therefore, although this disclosure is specifically disclosed in part by some embodiments and optional features, modifications and variations of the concepts disclosed herein may be used by those skilled in the art, and such modifications and variations are considered to be within the scope of this specification. 【0172】 It will be understood that some embodiments of the present disclosure may include, incorporate, or otherwise possess properties or features (e.g., components, members, elements, parts, and / or portions) described in other embodiments disclosed and / or described herein. Accordingly, various features of some embodiments may be compatible with, combined with, included in, and / or incorporated into other embodiments of the present disclosure. Accordingly, the disclosure of some features for a particular embodiment of the present disclosure should not be construed as limiting the application or inclusion of such features to that particular embodiment. Rather, it will be understood that other embodiments may also include such features, members, elements, parts, and / or portions without necessarily departing from the scope of the present disclosure. 【0173】 Furthermore, unless a feature is described as requiring another feature to be combined with it, any feature herein may be combined with any other feature of the same or different embodiments disclosed herein. Moreover, various well-known embodiments, such as exemplary systems, methods, and apparatus, are not described herein in particular detail in order to avoid obscuring the embodiments of the exemplary models. However, such embodiments are also contemplated herein. 【0174】 It will be apparent to those skilled in the art that methods, devices, device elements, materials, procedures, and techniques other than those specifically described herein can be applied to the practices of the embodiments described herein as being broadly disclosed herein, without relying on excessive experimentation. All technically known functional equivalents of the methods, devices, device elements, materials, procedures, and techniques specifically described herein are intended to be encompassed by this disclosure. 【0175】 When a group of materials, compositions, components, or compounds is disclosed herein, it should be understood that all individual members of those groups and all of their subgroups are disclosed separately. When a Markush group or other grouping is used herein, all individual members of that group, as well as all possible combinations and partial combinations of that group, are intended to be individually included in this disclosure. 【0176】 The embodiments described above are merely examples. Modifications, alterations, and variations of specific embodiments can be made by those skilled in the art without departing from the scope of this description, as defined solely by the appended claims.

Claims

[Claim 1] A method performed by a user device (UE) (2100) for using a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, Receiving one or more delay profile data for the training phase (1010), Preprocessing one or more delay profile data for the training phase (1020), Training an AI or ML model using one or more of the pre-processed delay profile data (1030) A method that includes this. [Claim 2] A method performed by a user device (UE) (2100) for using a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, Receiving one or more delayed profile data for the inference phase (1210), Preprocessing one or more of the delayed profile data for the inference phase (1220), Using the pre-processed one or more delay profile data as one or more inputs for an AI or ML model (1230) Methods that include... [Claim 3] A method performed by a user device (UE) (2100) to provide a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, Receiving one or more radio signals from one or more radio nodes (3300) (1410), (1420) generating one or more delay profile data based at least partially on one or more of the aforementioned wireless signals, Transmitting one or more of the aforementioned delay profile data to a network node (3300) (1430) A method that includes this. [Claim 4] The method according to claim 3, wherein the one or more delayed profile data are transmitted for use as one or more inputs for or related to an AI or ML model. [Claim 5] The method according to claim 3 or 4, wherein the network node includes at least one of gNB and location management functions. [Claim 6] The method according to claim 1, further comprising the steps described in claim 2. [Claim 7] The method according to claim 2, further comprising updating the AI ​​or ML model based at least partially on one or more of the aforementioned inputs. [Claim 8] The method according to any one of claims 1 to 7, further comprising obtaining the output of the AI ​​or ML model. [Claim 9] The method according to claim 8, further comprising reporting the output to a network node. [Claim 10] The method according to claim 8 or 9, wherein the output relates to the position of the UE. [Claim 11] The method according to claim 2 or 4, wherein only the one or more delay profile data is used as the one or more inputs. [Claim 12] The method according to any one of claims 1 to 11, wherein the one or more delay profile data includes only timing value data of a plurality of receiving paths. [Claim 13] The method according to claim 12, wherein the plurality of receiving paths are selected based on the strongest receiving path power. [Claim 14] The method according to claim 12, wherein the plurality of receiving paths are selected based on the earliest received path timing. [Claim 15] The method according to claim 14, wherein the earliest received pass timing includes the first detected pass and / or additional passes. [Claim 16] The method according to claim 12, 14, or 15, wherein the one or more delay profile data does not include a received power value for each of the plurality of receiving paths. [Claim 17] The method according to any one of claims 12 to 16, wherein, among the power values ​​for each of the plurality of receiving paths, there is no power value that includes the one or more inputs. [Claim 18] The method according to any one of claims 1 to 17, wherein the one or more delay profile data and / or timing value data include at least one of the following formats: bitmap, binary vector. [Claim 19] The method according to any one of claims 1 to 18, wherein the one or more delay profile data includes at least one of timing values, floating-point values, fixed-point values, reference time and other timing values ​​referencing the reference time, the number of samples in a measurement time window, a bitmap providing reference time and path timing values, delay profiles of the first detected path and additional paths, delay profiles of the n strongest paths, a delay profile with a signal quality indicator, and a delay profile with a small set of indicators per path. [Claim 20] The method according to any one of claims 1 to 19, further comprising measuring a wireless signal, generating a power value from the wireless signal, and transmitting the power value to the network node. [Claim 21] The method according to claim 20, wherein the power value includes at least one of the following: reference signal received power (RSRP), received signal strength indicator (RSSI), reference signal received quality (RSRQ), downlink positioning reference signal (PRS) (RSRP) (DL PRS-RSRP), and uplink sounding reference signal (SRS) (RSRP) (UL SRS-RSRP). [Claim 22] The method according to any one of claims 1 to 21, further comprising detecting one or more power delay profile data and extracting one or more delay profile data from the one or more power delay profile data. [Claim 23] A method performed by a network node (3300) for using a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, Receiving one or more delay profile data (1610), Preprocessing one or more of the aforementioned delay profile data (1620), Transmitting the pre-processed one or more delay profile data to a user device (UE) (2200) (1630) for use as one or more inputs to the AI ​​or ML model, Receiving one or more outputs of the AI ​​or ML model from the UE (1640) A method that includes this. [Claim 24] A method performed by a network node (3300) for using a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, Receiving one or more delay profile data (1810), Preprocessing one or more of the aforementioned delay profile data (1820), Using the pre-processed one or more delay profile data as one or more inputs for the AI ​​or ML model (1830), (1840) Obtaining one or more outputs of the aforementioned AI or ML model A method that includes this. [Claim 25] The method according to claim 23 or 24, wherein the network node includes at least one of gNB and location management functions. [Claim 26] The method according to any one of claims 23 to 25, wherein the one or more delay profile data are received from at least one of a gNB, UE, location management function (LMF), or next-generation radio access network (NG-RAN) node. [Claim 27] The method according to any one of claims 23 to 26, wherein the preprocessing includes combining one or more signal power measurements with one or more delay profile data. [Claim 28] A method performed by a first network node (3300) to provide a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, Receiving one or more radio signals from one or more radio nodes (3300) (1910), (1920) generating one or more delay profile data based at least partially on one or more of the aforementioned wireless signals, To transmit the one or more delay profile data to a second network node (3300) for use in one or more inputs of an AI or ML model (1930) A method that includes this. [Claim 29] The method according to claim 28, wherein the network node includes at least one of gNB and location management functions. [Claim 30] The method according to claim 28 or 29, further comprising receiving one or more outputs of the AI ​​or ML model. [Claim 31] The method according to any one of claims 23 to 30, wherein at least one of the one or more outputs is related to the position of the UE. [Claim 32] The method according to any one of claims 23 to 31, wherein only the one or more delay profile data is used as the one or more inputs. [Claim 33] The method according to any one of claims 23 to 32, wherein the one or more delay profile data includes only timing value data of a plurality of receiving paths. [Claim 34] The method according to any one of claims 23 to 33, further comprising transmitting the one or more outputs to another network node. [Claim 35] The method according to any one of claims 28 to 30, wherein the first network node includes a gNB and the second network node includes a next-generation radio access network (NG-RAN) node. [Claim 36] The method according to any one of claims 28 to 30, wherein the first network node includes a location management function (LMF), and the second network node includes a next-generation wireless access network (NG-RAN) node. [Claim 37] A user device (2200) for using a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, A processing circuit (2202) configured to perform any of the steps described in any one of claims 1 to 22, A power supply circuit (2208) configured to supply power to the processing circuit and User equipment (2200) equipped with [this]. [Claim 38] A network node (3300) for using or providing a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, A processing circuit (3302) configured to perform any of the steps described in any one of claims 23 to 36, A power supply circuit (3308) configured to supply power to the processing circuit and A network node (3300) equipped with [this feature]. [Claim 39] A user device (UE) (2200) for using a delay profile as input to an artificial intelligence (AI) or machine learning (ML) model, An antenna (2222) configured to transmit and receive wireless signals, A wireless front-end circuit (2212) connected to the antenna and processing circuit (3302) and configured to adjust signals communicated between the antenna and the processing circuit, wherein the processing circuit is configured to perform any of the steps described in any one of claims 1 to 22, An input interface (2206) connected to the processing circuit and configured to enable the input of information to the UE to be processed by the processing circuit, An output interface (2206) connected to the processing circuit and configured to output information processed by the processing circuit from the UE, A battery (2208) connected to the processing circuit and configured to supply power to the UE, User equipment (UE) (2200) equipped with the following.

Images