Ai / ML-based device positioning

Abstract

There are provided apparatuses, methods, systems, and / or the like for artificial intelligence (AI) and / or machine learning (ML) (AI / ML)-based device positioning. An exemplary apparatus may comprise at least one processor and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: (i) obtain, via running a first AI / ML model, based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); (ii) obtain, via running the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; and (iii) transmit, to a network entity, the one or more predicted ToA values and the one or more quality indicators for the one or more TRPs.

Description

AI / ML-BASED DEVICE POSITIONINGCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of US Provisional Application No. 63 / 717721, filed November 07, 2024. The entire content of the above-referenced application is hereby incorporated by reference.TECHNICAL FIELD

[0002] Various example embodiments relate generally to artificial intelligence (Al)- and / or machine learning (ML)-based device positioning, particularly to assessment of timing information quality in AI / ML- based device positioning.BACKGROUND

[0003] Some device positioning techniques, in some examples, may rely on assessment of the quality of timing information absent AI / ML models and / or techniques. Such device positioning techniques may not sufficiently capture timing measurement quality, for example, if timing information is estimated by AI / ML model (s).BRIEF DESCRIPTION

[0004] According to an exemplary embodiment, there is provided an apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: (i) obtain, via running a first artificial intelligence (Al) / machine learning (ML) model, based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); (ii) obtain, via running the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; and (iii) transmit, to a network entity, the one or more predicted ToA values and the one or more quality indicators for the one or more TRPs. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to predict, via running the first AI / ML model, and based on one or more input radio measurements, the one or more predicted ToA values. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to receive, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the network entity comprises a location management function (LMF). In some examples, the obtaining the one or more quality indicators is further based on one or more ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus. Insome examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to receive, from the network entity, the dataset. In some examples, the dataset comprises ground truth data, including at least one of the following: one or more ground truth ToA values; one or more input radio measurements; or one or more corresponding quality indicators of respective timing values. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to train the first AI / ML model, based on one or more ground truth ToA values, to obtain the one or more predicted ToA values and the one or more quality indicators. In some examples, the first AI / ML model is used for at least one of the following: training, or inference. In some examples, the first AI / ML model comprises a neural network based supervised learning model. In some examples, the apparatus comprises a terminal device or a network device.

[0005] According to an exemplary embodiment, there is provided a method, comprising: (i) obtaining, via running a first artificial intelligence (Al) / machine learning (ML) model, based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); (ii) obtaining, via running the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; and (iii) transmitting, to a network entity, the one or more predicted ToA values and the one or more quality indicators for the one or more TRPs. In some examples, the method further comprises predicting, via running the first AI / ML model, and based on one or more input radio measurements, the one or more predicted ToA values. In some examples, the method further comprises receiving, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the network entity comprises a location management function (LMF). In some examples, the obtaining the one or more quality indicators is further based on one or more ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus. In some examples, the method further comprises receiving, from the network entity, the dataset. In some examples, the dataset comprises ground truth data, including at least one of the following: one or more ground truth ToA values; one or more input radio measurements; or one or more corresponding quality indicators of respective timing values. In some examples, the method further comprises training the first AI / ML model, based on one or more ground truth ToA values, to obtain the one or more predicted ToA values and the one or more quality indicators. In some examples, the first AI / ML model is used for at least one of the following: training, or inference. In some examples, the first AI / ML model comprises a neural network based supervised learning model. In some examples, an apparatus performing the method comprises a terminal device or a network device.

[0006] According to an exemplary embodiment, there is provided a non-transitory computer readable medium comprising program instructions that, when executed by an apparatus, cause the apparatus to: (i) obtain, via running a first artificial intelligence (Al)Zmachine learning (ML) model, based on a dataset, one ormore predicted ToA values for one or more transmission and reception points (TRPs); (ii) obtain, via running the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; and (ill) transmit, to a network entity, the one or more predicted ToA values and the one or more quality indicators for the one or more TRPs. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus to predict, via running the first AI / ML model, and based on one or more input radio measurements, the one or more predicted ToA values. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus to receive, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the network entity comprises a location management function (LMF). In some examples, the obtaining the one or more quality indicators is further based on one or more ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus to receive, from the network entity, the dataset. In some examples, the dataset comprises ground truth data, including at least one of the following: one or more ground truth ToA values; one or more input radio measurements; or one or more corresponding quality indicators of respective timing values. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus to train the first AI / ML model, based on one or more ground truth ToA values, to obtain the one or more predicted ToA values and the one or more quality indicators. In some examples, the first AI / ML model is used for at least one of the following: training, or inference. In some examples, the first AI / ML model comprises a neural network based supervised learning model. In some examples, the apparatus comprises a terminal device or a network device.

[0007] According to an exemplary embodiment, there is provided an apparatus, comprising: (I) means for obtaining, via running a first artificial intelligence (Al) / machine learning (ML) model, based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); (II) means for obtaining, via running the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; and (ill) means for transmitting, to a network entity, the one or more predicted ToA values and the one or more quality indicators for the one or more TRPs. In some examples, the apparatus further comprises means for predicting, via running the first AI / ML model, and based on one or more input radio measurements, the one or more predicted ToA values. In some examples, the apparatus further comprises means for receiving, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the network entity comprises a location management function (LMF). In some examples, the obtaining the one or more quality indicators is further based on one or more ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus. In some examples, theapparatus further comprises means for receiving, from the network entity, the dataset. In some examples, the dataset comprises ground truth data, including at least one of the following: one or more ground truth ToA values; one or more input radio measurements; or one or more corresponding quality indicators of respective timing values. In some examples, the apparatus further comprises means for training the first AI / ML model, based on one or more ground truth ToA values, to obtain the one or more predicted ToA values and the one or more quality indicators. In some examples, the first AI / ML model is used for at least one of the following: training, or inference. In some examples, the first AI / ML model comprises a neural network based supervised learning model. In some examples, the apparatus comprises a terminal device or a network device.

[0008] According to an exemplary embodiment, there is provided an apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to receive, from a first apparatus, one or more predicted time of arrival (ToA) values for one or more transmission and reception points (TRPs) and one or more quality indicators for the one or more TRPs, wherein the one or more predicted ToA values and the one or more quality indicators are obtained by the first apparatus via a first artificial intelligence (Al) / machine learning (ML) model. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to transmit, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to obtain, via a second artificial intelligence (Al) / machine learning (ML) model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to: (I) request, from a positioning reference unit (PRU), ground truth data including one or more ground truth time ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; and (II) receive, from the PRU, the ground truth data. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to transmit, to the first apparatus, a dataset comprising the ground truth data.

[0009] According to an exemplary embodiment, there is provided a method, comprising receiving, from a first apparatus, one or more predicted time of arrival (ToA) values for one or more transmission and reception points (TRPs) and one or more quality indicators for the one or more TRPs, wherein the one or more predicted ToA values and the one or more quality indicators are obtained by the first apparatus via a first artificial intelligence (Al) / machine learning (ML) model. In some examples, the method further comprises transmitting, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the method further comprises obtaining,via a second artificial intelligence (Al) / machine learning (ML) model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus. In some examples, the method further comprises: (I) requesting, from a positioning reference unit (PRU), ground truth data including one or more ground truth time ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; and (ii) receiving, from the PRU, the ground truth data. In some examples, the method further comprises transmitting, to the first apparatus, a dataset comprising the ground truth data.

[0010] According to an exemplary embodiment, there is provided a non-transitory computer readable medium comprising program instructions that, when executed by an apparatus, cause the apparatus at least to receive, from a first apparatus, one or more predicted time of arrival (ToA) values for one or more transmission and reception points (TRPs) and one or more quality indicators for the one or more TRPs, wherein the one or more predicted ToA values and the one or more quality indicators are obtained by the first apparatus via a first artificial intelligence (Al) / machine learning (ML) model. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to transmit, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to obtain, via a second artificial intelligence (Al) / machine learning (ML) model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to: (I) request, from a positioning reference unit (PRU), ground truth data including one or more ground truth time ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; and (ii) receive, from the PRU, the ground truth data. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to transmit, to the first apparatus, a dataset comprising the ground truth data.

[0011] According to an exemplary embodiment, there is provided an apparatus, comprising means for receiving, from a first apparatus, one or more predicted time of arrival (ToA) values for one or more transmission and reception points (TRPs) and one or more quality indicators for the one or more TRPs, wherein the one or more predicted ToA values and the one or more quality indicators are obtained by the first apparatus via a first artificial intelligence (Al) / machine learning (ML) model. In some examples, the apparatus further comprises means for transmitting, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the apparatus further comprises means for obtaining, via a second artificial intelligence (Al) / machine learning (ML) model and based on the received one or more predicted ToA values and the one or more quality indicators, at leastone position of the first apparatus. In some examples, the apparatus further comprises: (I) means for requesting, from a positioning reference unit (PRU), ground truth data including one or more ground truth time ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; and (ii) means for receiving, from the PRU, the ground truth data. In some examples, the apparatus further comprises means for transmitting, to the first apparatus, a dataset comprising the ground truth data.

[0012] According to an exemplary embodiment, there is provided an apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: (I) receive, from a network entity, a reporting criterion; (ii) obtain, via running a first artificial intelligence (Al) / machine learning (ML) model and based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); (ill) obtain, via the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; (iv) determine which of the one or more quality indicators meet the reporting criterion; and (v) transmit, to the network entity, one or more predicted ToA values and one or more quality indicators which meet the reporting criterion for a respective subset of the one or more TRPs. In some examples, the reporting criterion comprises a threshold value. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to receive, from the network entity, the dataset. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to receive, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the obtaining the one or more quality indicators is further based on ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to train the first AI / ML model, based on the dataset, to obtain the one or more predicted ToA values and the one or more quality indicators. In some examples, the first AI / ML model further comprises a neural network based supervised learning model. In some examples, the apparatus comprises a terminal device.

[0013] According to an exemplary embodiment, there is provided a method, comprising: (I) receiving, from a network entity, a reporting criterion; (ii) obtaining, via running a first artificial intelligence (Al) / machine learning (ML) model and based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); (ill) obtaining, via the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; (iv) determining which of the one or more quality indicators meet the reporting criterion; and (v) transmitting, to the network entity, one or more predicted ToA values and one or more quality indicators which meet the reporting criterion for a respective subset of the one or more TRPs. In some examples, the reporting criterion comprises athreshold value. In some examples, the method further comprises receiving, from the network entity, the dataset. In some examples, the method further comprises receiving, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the obtaining the one or more quality indicators is further based on ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus. In some examples, the method further comprises training the first AI / ML model, based on the dataset, to obtain the one or more predicted ToA values and the one or more quality indicators. In some examples, the first AI / ML model further comprises a neural network based supervised learning model. In some examples, an apparatus performing the method comprises a terminal device.

[0014] According to an exemplary embodiment, there is provided a non-transitory computer readable medium comprising program instructions that, when executed by an apparatus, cause the apparatus at least to: (i) receive, from a network entity, a reporting criterion; (ii) obtain, via running a first artificial intelligence (Al) / machine learning (ML) model and based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); (iii) obtain, via the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; (iv) determine which of the one or more quality indicators meet the reporting criterion; and (v) transmit, to the network entity, one or more predicted ToA values and one or more quality indicators which meet the reporting criterion for a respective subset of the one or more TRPs. In some examples, the reporting criterion comprises a threshold value. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to receive, from the network entity, the dataset. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to receive, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the obtaining the one or more quality indicators is further based on ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to train the first AI / ML model, based on the dataset, to obtain the one or more predicted ToA values and the one or more quality indicators. In some examples, the first AI / ML model further comprises a neural network based supervised learning model. In some examples, the apparatus comprises a terminal device.

[0015] According to an exemplary embodiment, there is provided an apparatus, comprising: (i) means for receiving, from a network entity, a reporting criterion; (ii) means for obtaining, via running a first artificial intelligence (Al)Zmachine learning (ML) model and based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); (iii) means for obtaining, via the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToAvalues; (iv) means for determining which of the one or more quality indicators meet the reporting criterion; and (v) means for transmitting, to the network entity, one or more predicted ToA values and one or more quality indicators which meet the reporting criterion for a respective subset of the one or more TRPs. In some examples, the reporting criterion comprises a threshold value. In some examples, the apparatus further comprises means for receiving, from the network entity, the dataset. In some examples, the apparatus further comprises means for receiving, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the obtaining the one or more quality indicators is further based on ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus. In some examples, the apparatus further comprises means for training the first AI / ML model, based on the dataset, to obtain the one or more predicted ToA values and the one or more quality indicators. In some examples, the first AI / ML model further comprises a neural network based supervised learning model. In some examples, the apparatus comprises a terminal device.

[0016] According to an exemplary embodiment, there is provided an apparatus, comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to: (i) transmit, to a first apparatus, a reporting criterion; and (ii) receive, from the first apparatus, one or more predicted time of arrival (ToA) values and one or more quality indicators which meet the reporting criterion for a respective subset of one or more transmission and reception points (TRPs), wherein the one or more predicted ToA values for the respective subset of TRPs and the one or more quality indicators for the respective subset of TRPs are obtained by the first apparatus via a first artificial intelligence (Al)Zmachine learning (ML) model. In some examples, the reporting criterion comprises a threshold value. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to transmit, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to obtain, via a second AI / ML model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to: (i) request, from a positioning reference unit (PRU), ground truth data including one or more ground truth ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; and (ii) receive, from the PRU, the ground truth data. In some examples, the instructions, when executed by the at least one processor, further cause the apparatus at least to transmit, to the first apparatus, the dataset comprising the ground truth data.

[0017] According to an exemplary embodiment, there is provided a method, comprising: (i) transmitting, to a first apparatus, a reporting criterion; and (ii) receiving, from the first apparatus, one or more predicted time of arrival (ToA) values and one or more quality indicators which meet the reporting criterion for a respective subset of one or more transmission and reception points (TRPs), wherein the one or more predicted ToA values for the respective subset of TRPs and the one or more quality indicators for the respective subset of TRPs are obtained by the first apparatus via a first artificial intelligence (Al) / machine learning (ML) model. In some examples, the reporting criterion comprises a threshold value. In some examples, the method further comprises transmitting, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the method further comprises obtaining, via a second AI / ML model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus. In some examples, the method further comprises: (I) requesting, from a positioning reference unit (PRU), ground truth data including one or more ground truth ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; and (ii) receiving, from the PRU, the ground truth data. In some examples, the method further comprises transmitting, to the first apparatus, the dataset comprising the ground truth data.

[0018] According to an exemplary embodiment, there is provided a non-transitory computer readable medium comprising program instructions that, when executed by an apparatus, cause the apparatus at least to: (I) transmit, to a first apparatus, a reporting criterion; and (ii) receive, from the first apparatus, one or more predicted time of arrival (ToA) values and one or more quality indicators which meet the reporting criterion for a respective subset of one or more transmission and reception points (TRPs), wherein the one or more predicted ToA values for the respective subset of TRPs and the one or more quality indicators for the respective subset of TRPs are obtained by the first apparatus via a first artificial intelligence (Al) / machine learning (ML) model. In some examples, the reporting criterion comprises a threshold value. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to transmit, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to obtain, via a second AI / ML model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to: (I) request, from a positioning reference unit (PRU), ground truth data including one or more ground truth ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; and (ii) receive, from the PRU, the ground truthdata. In some examples, the program instructions, when executed by the apparatus, further cause the apparatus at least to transmit, to the first apparatus, the dataset comprising the ground truth data.

[0019] According to an exemplary embodiment, there is provided an apparatus, comprising: (i) means for transmitting, to a first apparatus, a reporting criterion; and (ii) means for receiving, from the first apparatus, one or more predicted time of arrival (ToA) values and one or more quality indicators which meet the reporting criterion for a respective subset of one or more transmission and reception points (TRPs), wherein the one or more predicted ToA values for the respective subset of TRPs and the one or more quality indicators for the respective subset of TRPs are obtained by the first apparatus via a first artificial intelligence (Al)Zmachine learning (ML) model. In some examples, the reporting criterion comprises a threshold value. In some examples, the apparatus further comprises means for transmitting, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model. In some examples, the apparatus further comprises means for obtaining, via a second AI / ML model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus. In some examples, the apparatus further comprises: (i) means for requesting, from a positioning reference unit (PRU), ground truth data including one or more ground truth ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; and (ii) means for receiving, from the PRU, the ground truth data. In some examples, the apparatus further comprises means for transmitting, to the first apparatus, the dataset comprising the ground truth data.LIST OF THE DRAWINGS

[0020] In the following, the disclosure will be described in greater detail with reference to the embodiments and the accompanying drawings, in which:

[0021] Fig. 1 shows an example of a communication network to which examples disclosed herein may be applied;

[0022] Fig. 2 shows an example of a method;

[0023] Fig. 3 shows an example of an equation;

[0024] Fig. 4 shows an example of an equation;

[0025] Fig. 5 shows an example of a flow chart;

[0026] Fig. 6 shows an example of an equation;

[0027] Fig. 7 shows an example of a flow chart;

[0028] Fig. 8 shows an example of a signaling flow diagram;

[0029] Fig. 9 shows an example of a signaling flow diagram;

[0030] Fig. 10 shows an example of a signaling flow diagram;

[0031] Fig. 11 shows an example of a method;

[0032] Fig. 12 shows an example of a method;

[0033] Fig. 13 shows an example of a method;

[0034] Fig. 14 shows an example of a method; and

[0035] Fig. 15 shows an example of an apparatus.DESCRIPTION OF EMBODIMENTS

[0036] The following embodiments are exemplary. Although the specification may refer to "an”, "one”, or "some” embodiment(s) in several locations of the text, this does not necessarily mean that each reference is made to the same embodiment(s), or that a particular feature only applies to a single embodiment. Single features of different embodiments may also be combined to provide other embodiments. Further, when a particular feature, structure, or characteristic is described in connection of an embodiment, it is within the knowledge of one skilled in the art to apply such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described. It shall be understood that although the terms "first”, "second”, and / or the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another.

[0037] For the purposes of the present disclosure, the phrases "at least one of A or B”, "at least one of A and B”, and "A and / or B” mean (A), (B), or (A and B). For the purposes of the present disclosure, the phrase "A, B, and / or C” means (A), (B), (C), (A and B), (A and C), (B and C), or (A, B, and C).

[0038] Embodiments described may be implemented in a communication network, such as any of the following radio access technologies (RATs): Worldwide Interoperability for Micro-wave Access (WiMAX), Global System for Mobile communications (GSM, 2G), GSM EDGE radio access Network (GERAN), General Packet Radio Service (GRPS), Universal Mobile Telecommunication System (UMTS, 3G) based on basic wideband-code division multiple access (W-CDMA), high-speed packet access (HSPA), Long Term Evolution (LTE), LTE-Advanced, and enhanced LTE (eLTE), 5G (also called NR), or any future RAT such as 6G. Moreover, communications within the communication network may utilize any proper wireless communication technology, comprising but not limited to: Code Division Multiple Access (CDMA), Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Frequency Division Duplex (FDD), Time Division Duplex (TDD), Multiple-Input Multiple-Output (MIMO), Orthogonal Frequency Division Multiple (OFDM), and / or Discrete Fourier Transform spread OFDM (DFT-s-OFDM).

[0039] As used herein, the term "network device”, "network node”, or "network entity” refers to a node in a communication network via which user equipment may access the network and / or which is capable of controlling radio communication and managing radio resources within a cell. The network node or networkdevice may be referred to as a base station (BS), an access point (AP), or an access node. The network device may be, depending on the applied technology, for example, a node B (NodeB or NB), an evolved NodeB (eNodeB or eNB), an NR NB (also referred to as a gNB), a Remote Radio Unit (RRU), a radio head (RH), a remote radio head (RRH), a relay, an Integrated Access and Backhaul (IAB) node, a low power node, a non-terrestrial network (NTN) or non-ground network device such as a satellite network device, a low earth orbit (LEO) satellite and a geosynchronous earth orbit (GEO) satellite, or an aircraft network device.

[0040] Moreover, in connection of split radio access network (RAN), the network device may refer to a centralized unit (OU) or a base station and / or a distributed unit (DU) of a base station. An interface between OU and DU may be referred to as an F1 interface in NR. In the split RAN architecture, node operations may be carried out, at least partly, in the central / centralized unit, OU (e.g., server, host, or node) operationally coupled to the DU, (e.g., a radio head / node). One CU may control one or more DUs, acting at least as transmit / receive (Tx / Rx) nodes. In some embodiments, the DUs may comprise, for example, a radio link control (RLC), medium access control (MAC) layer and a physical (PHY) layer, whereas the CU may comprise the layers above the RLC layer, such as a packet data convergence protocol (PDCP) layer, a radio resource control (RRC) layer and an internet protocol (IP) layer. Other functional splits are also possible. In practice, any processing task may be performed in either the CU or the DU and the boundary where the responsibility is shifted between the CU and the DU may depend on the applied implementation.

[0041] The term "terminal device” refers to any end device that may be capable of wireless communication. By way of example, a terminal device may be referred to as a communication device, user equipment (UE), a Subscriber Station (SS), or a Mobile Station (MS). The terminal device may include a mobile phone, a cellular phone, a smart phone, voice over IP (VoIP) phones, wireless local loop phones, a tablet, a wearable terminal device, a personal digital assistant (PDA), portable computers, desktop computers, image capture terminal devices such as digital cameras, gaming terminal devices, music storage and playback appliances, vehicle-mounted wireless terminal devices, USB dongles, an Internet of Things (loT) device, a watch or other wearable, a head-mounted display (HMD), a vehicle, a drone, a medical device and applications (e.g., remote surgery), an industrial device and applications (e.g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain context), a consumer electronics device, a device operating on commercial and / or industrial networks, and / or the like.

[0042] The term "resource”, as used herein, may refer to radio resources in time domain, in frequency domain, in space domain, and / or in code domain. Some examples of resources include, for example, physical resource block (PRB), a radio frame, a subframe, a time slot, a sub-band, a frequency region, a sub-carrier, a beam, and / or the like. The term "transmission” and / or "reception” may refer to wirelessly transmitting and / or receiving via a wireless propagation channel on radio resources.

[0043] Fig. 1 illustrates an example of a communication network to which examples disclosed herein may be applied. The communication network or a cellular communication network may comprise a network node 110 providing one or more cells, such as cell 100, and a network node 112 providing one or more other cells, such as cell 102. A respective cell may be, for example, a macro cell, a micro cell, a femto cell, or a pico cell, for example. The respective cell may define a coverage area or a service area of the corresponding access node.

[0044] The network node 110 may provide a user equipment (UE) 120 (one or more UEs) with wireless access to the communication network. The wireless access may comprise downlink (DL) communication from the network node to the UE 120 and uplink (UL) communication from the UE 120 to the network node. Examples of uplink channels comprise physical uplink control channel (PUCCH) for transmitting control information and physical uplink shared channel (PUSCH) for transmitting data towards the network. Examples of downlink channels comprise physical downlink control channel (PDCCH) for transmitting control information and physical downlink shared channel (PDSCH) for transmitting data towards the user equipment.

[0045] There may be a plurality of UEs 120, 122, in the system. Each of them may be served by the same or by different network nodes 110, 112. UE may be configured with dual connectivity (DC), wherein the UE, for example, UE 120, may be connected to multiple network nodes 110, 112. The UEs 120, 122, may communicate with each other, in case device-to-device (D2D) communication interface is established between them via a so-called sidelink (SL). Such D2D communications may be referred to as machine-to- machine, peer-to-peer (P2P) communications, or vehicle-to-vehicle (V2V), for example.

[0046] In the case of multiple network nodes in the communication network, the network nodes may be connected to each other via an interface. LTE specifications call such an interface an X2 interface. An interface between an LTE node and a 5G node, or between two 5G nodes, may be called an Xn interface.

[0047] The network nodes 110 and 112 may be further connected via another interface to a core network 116 of the communication network. The core network 116 may be configured to comprise a location management function (LMF). The LTE specifications specify the core network as an evolved packet core (EPC), and the core network may comprise, for example, a mobility management entity (MME) and a gateway node. The MME may handle mobility of terminal devices in a tracking area encompassing a plurality of cells and handle signaling connections between the terminal devices and the core network. The gateway node may handle data routing in the core network and to / from the terminal devices. The 5G specifications specify the core network as a 5G core (5GC). The 5G core may comprise, for example, an access and mobility management function (AMF) and a use plane function / gateway (UPF) and other functions. The AMF may handle termination of non-access stratum (NAS) signaling, NAS ciphering and integrity protection, registration management, connection management, mobility management, accessauthentication and authorization, security context management, and / or the like. The UPF node may support packet routing and forwarding, packet inspection, and quality of service (QoS) handling, for example.

[0048] An LMF, for example, comprised within the core network 116, may be further configured to identify one or more device (e.g., terminal device, UE, etc.) positioning methods, implement one or more tasks related to terminal device positioning, receive and / or obtain one or more terminal device positioning measurements, determine one or more positions of a respective terminal device, and / or perform other tasks.

[0049] Artificial intelligence (Al) can be broadly defined as getting computers to perform tasks mimicking human brain. Machine learning (ML) is one category of Al techniques: computer algorithms able to automatically improve their performance without explicit programming. Al algorithms were first conceived in the 1950's but only in recent years AI / ML has become useful in vast area of real-world applications, partly due to advancements in computational power and in providing storage capacity for data.

[0050] AI / ML can help adjust and optimize radio access network (RAN) parameters and settings using (real-time) monitoring and prediction of network performance, quality, and demand. Additionally, AI / ML can identify and diagnose degradation in network performance, as well as provide protection from cyberattacks. AI / ML is usable in energy saving, load balancing, mobility optimization, link adaptation and security just to mention but a few.

[0051] It is envisioned that AI / ML will enable real-time analysis as well as automated operation and control in 5G and beyond RAN. This requires the availability of data streamed from wireless devices in a timely manner, especially in extremely time-critical applications such as real-time video monitoring and extended reality (XR). This may be reflected in network architecture, such as by placing and moving ML agents to the required locations in the network, for example for data collection. User devices (e.g., mobile devices) may assist network in decision-making in resource management, thus a user device may act as an infrastructure resource.

[0052] As the network evolves to programmable and flexible cloud native implementation, AI / ML- based network automation will be used to simplify network management and optimization. It is expected that parts of the air interface, in particular signal processing algorithms, are supported and eventually even replaced with machine learning models. Thus, a 6G wireless communication standard will natively support an Al-based air interface.

[0053] Machine learning algorithms are usually classified into four different types: supervised learning, unsupervised learning, semi-supervised learning and reinforcement learning.

[0054] In supervised learning the algorithm learns from labelled data. For training, the algorithm receives input data and corresponding correct output labels. The algorithm is trained to predict accurate labels for new data.

[0055] In unsupervised learning the algorithm analyses unlabeled data. The aim is to discover patterns, relationships, or structures within the data, for example, unsupervised learning algorithms make groups of similar data points.

[0056] Semi-supervised learning is a hybrid machine learning approach that combines labelled and unlabeled data for training. A limited amount of labelled data and a larger set of unlabeled data is used to improve training. This approach is useful when acquiring labelled data is expensive or time-consuming as is the case in many real-world applications. Semi-supervised learning techniques can be applied to various tasks, such as classification, regression, and anomaly detection, allowing models to make more accurate predictions and generalize better in real-world scenarios.

[0057] Reinforcement learning is a machine learning algorithm which learns from trial and error. An ML agent interacts with environment and learns from experience aiming to maximize cumulative rewards. The ML agent receives feedback through rewards or penalties based on its actions. The agent learns to take actions that lead to the most favorable outcomes over time. The algorithm adapts to changing environments and achieve long-term goals through a sequence of actions.

[0058] An example of an ML algorithm found applicable to adjust and optimize the radio access network (RAN) parameters and settings is deep learning. Deep learning is a subset of machine learning algorithms using a neural network. Neural networks are also known as artificial neural networks (ANNs) or simulated neural networks (SNNs). Deep learning can be based on supervised, semi-supervised or unsupervised learning.

[0059] Artificial neural networks (ANNs) are comprised of an input layer, one or more hidden layers, and an output layer. Each node of a layer, or an artificial neuron, connects to another one and has an associated weight as well as a threshold value. If the output of an individual node is above the threshold value specified to this node, the node is activated and sending or passing data to the next layer of the neural network.

[0060] In the case the supervised learning is applied in the training of a neural network, the training is carried out by using examples, each of which contains a known "input" and "result", forming probability- weighted associations between them. The training comprises determining the difference between the output of the neural network (a prediction) for an input and a target output for the same input. The difference is called an error value. The neural network then adjusts its weighted associations according to a learning rule and using this error value. Successive adjustments make the neural network produce output that is approaching the target output. After a sufficient number of these adjustments, the training can be terminated based on a certain criterion.

[0061] AI / ML models described herein may be implemented by various devices, for example, such as terminal devices (e.g., UEs, mobile devices, etc.), network entities (e.g., gNBs, LMFs, etc.), and / or other devices.

[0062] Other solutions, in some examples, may provide for device positioning. Some device positioning solutions, in some examples, may rely on assessment of the quality of timing information absent artificial intelligence and / or machine learning (AI / ML) models and / or techniques. Such device positioning solutions may not sufficiently capture timing measurement quality, for example, if timing information is estimated by AI / ML model(s). At least one problem with the above-mentioned other solutions is that AI / ML techniques, for example, for various aspects of device positioning (and / or improvements thereto), are not considered or included. For example, evaluation and / or assessment of timing information estimation via AI / ML techniques in AI / ML-assisted positioning schemes are absent in the other solutions.

[0063] To at least partially tackle this problem, there is proposed a solution for Al- and / or ML-based device positioning, particularly for assessment of timing information quality in AI / ML-based device positioning.

[0064] Referring now to Fig. 2, a method is shown. In the example of Fig. 2, the method is an AI / ML- based assisted positioning scheme. A first AI / ML model (labeled "AI / ML Model-1” in Fig. 2) may receive input comprising at least possible measurements and / or channel observations from a terminal device (and / or UE). The AI / ML Model-1 may be trained on labeled data, for example, including: (i) time of arrival (ToA) values measured by a terminal device (e.g., a UE) and / or by a network entity (e.g., a gNB); and / or (ii) corresponding measurements and / or channel observations performed by the terminal device and / or by the network entity.

[0065] Assessment of quality of timing information (e.g., the ToA values) may be based on various factors, including: (i) quality indicators and / or quality indicators of ground truth data used for training the AI / ML Model-1; (ii) quality of the AI / ML Model-1 inference and / or confidence in the AI / ML Model-1 output; and / or (iii) line of sight (LoS) and / or non-line of sight (NLoS) indicator(s), which may be output by the AI / ML Model-1 as an intermediate feature (e.g., as shown in Fig. 2). The assessment of the quality of the timing information may be indicated by a quality indicator parameter (e.g., timing information quality indicator value(s)). The quality indicator parameter may be configured to correlate the training of the first model and the assessment of the quality of the timing information.

[0066] Based on the intermediate feature output by the AI / ML Model-1, various techniques may be used to determine terminal device positioning, several of which are described in detail herein. In a first example, a second AI / ML model (labeled "AI / ML Model-2” in Fig. 2) may take the intermediate feature as input and determine, using AI / ML techniques, at least one position of the terminal device. In a secondexample, non-AI / ML methods and / or techniques may take the intermediate feature as input and determine at least one position of the terminal device.

[0067] Various examples provide methods for evaluating timing information quality indicator(s). In some examples, training of an AI / ML model configured to output one or more ToA values relies on ground truth data including one or more ground truth ToA values and / or one or more corresponding input radio measurements (e.g., as measured by a respective terminal device at respective positions within a respective region).

[0068] In at least some technologies, a respective ToA value may be associated with a quality metric nr-TimingQuality (e.g., as defined in the 3GPP TS 37.355). The quality metric nr-TimingQuality may be evaluated by a terminal device for a given measured ToA value. Moreover, the quality metric nr- TimingQuality may be based on at least one of the following: a. clock synchronization (e.g., poor synchronization may result in timing errors); b. propagation delays (e.g., introduced by environmental factors such as obstacles), which may impact the measured ToA value and / or timing quality; c. LoS and / or NLoS conditions; d. measurement noise (e.g., present at receivers), which may introduce errors in time measurements; e. signal-to-noise ratio (SNR) of received reference signals, which may affect accuracy of ToA estimation; and / or f. multipath propagation, which may affect accuracy of ToA quality estimation.

[0069] In some examples, AI / ML models may be configured to calculate a confidence level parameter which indicates the degree of truth of predicted values, for example, such as predicted ToA values.

[0070] As described herein, some examples provide for a quality indicator metric (hereafter "Uncertainty-Weighted Quality” or “UWQ”), wherein a UWQ may be determined based on at least two types of information: (1) a quality metric (e.g., nr-TimingQuality),' (ii) a confidence level parameter; and / or other types of information. Based on ToA values of at least some (e.g., all) of a plurality of respective transmission and reception points (TRPs), a UWQ may be evaluated for respective TRPs. TRP -specific UWQ values may be indicated via fields and / or information elements (IBs).

[0071] Referring now to Fig. 3, an example of an equation for a relative quality metric (hereafter "Relative Quality” or “RQ”) is provided, where: NRtimingquality is a quality metric (e.g., nr-TimingQuality) measured in meters, and MaxAllowederror is an error threshold (e.g., an NR timing quality threshold) measured in meters. The MaxAllowederror may be calculated by an LMF based at least in part on data collection specifications and / or targeted positioning accuracy. In some examples, if an estimated error is greater than the error threshold, a respective device (e.g., terminal device, network entity) may beconfigured to not report a respective measurement to the LMF (e.g., instead reporting a measurement error message, setting the respective measurement to zero, and / or the like). In some examples, if the estimated error is less than or equal to the error threshold, the RQ, which can range in value from 0 to 1, may be computed. The RQ may indicate a relative quality of the quality metric. In some examples, the RQ may be taken as input to an AI / ML model (e.g., a second AI / ML model "ML model (2)” in Fig. 5) configured to estimate quality of predicted timing information (e.g., ToA values).

[0072] Referring now to Fig. 4, an example of an equation for a UWQ value (e.g., UWQtraining) is provided, where: RQ is a Relative Quality, TP is a predicted ToA value, and T1is a true (e.g., ground truth) ToA value. As described herein, the UWQ may be evaluated based on at least two types of information: a quality metric and a confidence level parameter. In some examples, an AI / ML model (e.g., the second AI / ML model "ML model (2) in Fig. 5) may take predicted ToA values as input and output UWQtraining.

[0073] Referring now to Fig. 5, a flow chart for AI / ML-assisted device positioning is provided. In some examples, one or more AI / ML models may be relied upon to evaluate a quality indicator defined as the UWQ. A first AI / ML model "ML model (1)” may take as input one or more ground truth ToA values and / or one or more corresponding input radio measurements (e.g., channel impulse response (CIR), reference signal received path power (RSRPP), and / or the like). At step 1, the ML model (1) may be trained, for example, on at least one dataset, to predict one or more predicted ToA values based on the input data. The dataset may include at least one of the following: one or more ground truth ToA values, one or more input radio measurements, or one or more corresponding quality of a timing value (e.g., such as nr- TimingQuality), The ML model (1) may be a neural network based supervised learning model, for example, such as a regression model. At step 2, the ML model (1) may obtain the one or more predicted ToA values, for example, via running the ML model (1) for inference. At step 3, the second AI / ML model ML model (2) may take as input at least one of the following: the one or more predicted ToA values, the one or more ground truth ToA values, and / or one or more RQ values. The one or more RQ values may be evaluated based on the equation of Fig. 3. The ML model (2) may be trained based on the input data. The ML model (2) may be configured to calculate and / or output one or more UWQ values. The one or more UWQ values may be evaluated based on the equation of Fig. 4. In some examples, the ML model (2) may be trained to output values ranging from 0 to 1 based on input data comprising nr-TimingQuality.

[0074] Referring now to Fig. 6, an example of an equation for a UWQ value is provided, where NRtimingquality is a quality metric (e.g., nr-TimingQuality) measured in meters, confidence is a metric indicating confidence in a respective model's estimation, LoSJndicator is a LoS indicator value (e.g., which may be replaced by NLoSJndicator, an NLoS indicator value), and oi, 02, and 03 are weighting factors which may be variously tuned as suitable for various applications.

[0075] Referring now to Fig. 7, a flow chart for AI / ML-assisted device positioning is provided, particularly for timing information quality indicator evaluation. In the example of Fig. 7, a first AI / ML model ML model (1) may be configured to compute and / or output one or more predicted ToA values based at least in part on one or more ground truth ToA values and / or one or more corresponding input radio measurements. A second AI / ML model ML model (2) may be configured to compute and / or output a probability of the one or more predicted ToA values matching the one or more ground truth ToA values for a respective RQ.

[0076] In some examples, the ML model (2) may be configured to compute and / or output a confidence level based on the one or more predicted ToA values and the one or more ground truth ToA values. In some examples, the confidence level is only based on the one or more predicted ToA values and the one or more ground truth ToA values. In such examples, a UWQ may be evaluated based on the equation of Fig. 6.

[0077] Various examples provide for various types of signaling impacts arising from the various AI / ML-based device positioning schemes described herein. Figs. 8-10 show signaling flow diagrams corresponding to various signaling schemes for AI / ML-based device positioning.

[0078] Referring now to Fig. 8, a signaling flow diagram is provided. At step 1, a network entity (e.g., an LMF) may request ground truth data from a positioning reference unit (PRU), wherein the ground truth data includes at least one or more ground truth ToA values and / or one or more corresponding input radio measurements. The LMF may further request to evaluate and / or report a related quality metric (e.g., nr- TimingQuality) for the one or more ground truth ToA values.

[0079] At step 2, the PRU may collect the requested ground truth data and related quality metric data.

[0080] At step 3, the PRU may transmit the collected data to the LMF.

[0081] At step 4, the LMF may transmit, to a terminal device, a dataset of labeled data (e.g., wherein the labeled data is a labeled version of the collected data received from the PRU). In some examples, theLMF may further transmit, to the terminal device, information indicating one or more maximum allowed error threshold values (e.g., MaxAllowederror values).

[0082] At step 5, the terminal device may perform model training (e.g., following at least some steps of the flow chart of Fig. 7) of a first AI / ML model using the labeled data received from the LMF. In examples where the terminal device further received the information indicating the one or more maximum allowed threshold values, the terminal device may use the information for calculating one or more RQ values.

[0083] At step 6, the terminal device may use the trained first AI / ML model to run inference and / or determine intermediate features (e.g., one or more predicted ToA values) and / or corresponding timing information quality indicator(s) for one or more TRPs.

[0084] At step 7, the terminal device may transmit, to the LMF, the one or more predicted ToA values and / or the one or more corresponding quality indicators.

[0085] At step 8, the LMF may run a second AI / ML model based at least on the one or more predicted ToA values to evaluate at least one position of the terminal device.

[0086] Referring now to Fig. 9, a signaling flow diagram is provided. At step 1, a network entity (e.g., an LMF) may request ground truth data from a positioning reference unit (PRU), wherein the ground truth data includes at least one or more ground truth ToA values and / or one or more corresponding input radio measurements. The LMF may further request to evaluate and / or report a related quality metric (e.g., nr- TimingQuality) for the one or more ground truth ToA values.

[0087] At step 2, the PRU may collect the requested ground truth data and related quality metric data.

[0088] At step 3, the PRU may transmit the collected data to the LMF.

[0089] At step 4, the LMF may transmit, to a network entity (e.g., a gNB), a dataset of labeled data(e.g., wherein the labeled data is a labeled version of the collected data received from the PRU). In some examples, the LMF may further transmit, to the gNB, information indicating one or more maximum allowed error threshold values (e.g., MaxAllowederror values).

[0090] At step 5, the gNB may perform model training (e.g., following at least some steps of the flow chart of Fig. 7) of a first AI / ML model using the labeled data received from the LMF. In examples where the gNB further received the information indicating the one or more maximum allowed threshold values, the terminal device may use the information for calculating one or more RQ values.

[0091] At step 6, the LMF may request, from a terminal device, information indicating terminal device positioning measurement.At step 7, the terminal device may transmit, to the gNB, one or more positioning reference signal (PRS)- based radio measurements.

[0092] At step 8, the gNB may use the trained first AI / ML model to run inference and / or determine intermediate features (e.g., one or more predicted ToA values) and / or corresponding timing information quality indicator(s) for one or more TRPs.

[0093] At step 9, the gNB may transmit, to the LMF, the one or more predicted ToA values and / or the one or more corresponding quality indicators.

[0094] At step 10, the LMF may run a second AI / ML model based at least on the one or more predicted ToA values to evaluate at least one position of the terminal device.

[0095] Referring now to Fig. 10, a signaling flow diagram is provided. At step 1, a network entity (e.g., an LMF) may request ground truth data from a positioning reference unit (PRU), wherein the ground truth data includes at least one or more ground truth ToA values and / or one or more corresponding input radiomeasurements. The LMF may further request to evaluate and / or report a related quality metric (e.g., nr- TimingQuality) for the one or more ground truth ToA values.

[0096] At step 2, the PRU may collect the requested ground truth data and related quality metric data.

[0097] At step 3, the PRU may transmit the collected data to the LMF.

[0098] At step 4, the LMF may transmit, to a terminal device, a dataset of labeled data (e.g., wherein the labeled data is a labeled version of the collected data received from the PRU). In some examples, the LMF may further transmit, to the terminal device, information indicating one or more maximum allowed error threshold values (e.g., MaxAllowederror values). The LMF may further transmit, to the terminal device, one or more rules indicating that only ToA measurements, of respective TRPs, satisfying a predefined quality condition may be reported by the terminal device.

[0099] At step 5, the terminal device may perform model training (e.g., following at least some steps of the flow chart of Fig. 7) of a first AI / ML model using the labeled data received from the LMF. In examples where the terminal device further received the information indicating the one or more maximum allowed threshold values, the terminal device may use the information for calculating one or more RQ values.

[0100] At step 6, the terminal device may use the trained first AI / ML model to run inference and / or determine intermediate features (e.g., one or more predicted ToA values) and / or corresponding timing information quality indicator(s) for one or more TRPs.

[0101] At step 7, the terminal device may evaluate the one or more ToA values based on the predefined quality condition and / or determine which should be reported.

[0102] At step 8, the terminal device may transmit, to the LMF and for a subset of the one or more TRPs, the one or more predicted ToA values satisfying the predefined quality condition and / or the one or more corresponding quality indicators.

[0103] At step 9, the LMF may run a second AI / ML model based at least on the one or more predicted ToA values, for the subset of the one or more TRPs, satisfying the predefined quality condition to evaluate at least one position of the terminal device.

[0104] Figs. 11-14 show example methods.

[0105] Referring now to Fig. 11, an example of a method 1100 is provided. As shown in block 1110, a first apparatus (e.g., a terminal device and / or a network device) may receive, from a network entity, a dataset.

[0106] As shown in block 1120, the first apparatus may receive, from the network entity, a maximum allowed error value for quality indicator computation for training a first artificial intelligence (Al) and / or machine learning (ML) model.

[0107] As shown in block 1130, the first apparatus may predict, via running the first AI / ML model, and based on one or more input radio measurements, one or more predicted ToA values.

[0108] As shown in block 1140, the first apparatus may train the first AI / ML model, based on the dataset, to obtain one or more predicted time of arrival (ToA) values and one or more quality indicators.

[0109] As shown in block 1150, the first apparatus may obtain, via running the first AI / ML model, based on the input radio measurements, the one or more predicted ToA values for one or more transmission and reception points (TRPs).

[0110] As shown in block 1160, the first apparatus may (e.g., based on blocks 1140 and 1150) obtain, via running the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values.

[0111] As shown in block 1170, the first apparatus may transmit, to the network entity, the one or more predicted ToA values and the one or more quality indicators for the one or more TRPs.

[0112] Referring now to Fig. 12, an example of a method 1200 is provided. As shown in block 1210, a second apparatus (e.g., an LMF) may request, from a positioning reference unit (PRU), ground truth data including one or more ground truth time of arrival (ToA) values, one or more input radio measurements, and one or more corresponding quality of a timing value.

[0113] As shown in block 1220, the second apparatus may receive, from the PRU, the ground truth data.

[0114] As shown in block 1230, the second apparatus may transmit, to a first apparatus (e.g., the first apparatus of the method 1100 of Fig. 11), a dataset comprising the ground truth data.

[0115] As shown in block 1240, the second apparatus may transmit, to the first apparatus, a maximum allowed error value for quality indicator computation for training a first AI / ML model.

[0116] As shown in block 1250, the second apparatus may receive, from the first apparatus, one or more predicted ToA values for one or more transmission and reception points (TRPs) and one or more quality indicators for the one or more TRPs, wherein the one or more predicted ToA values and the one or more quality indicators are obtained by the first apparatus via the first artificial intelligence (Al) and / or machine learning (ML) model.

[0117] As shown in block 1260, the second apparatus may obtain, via a second AI / ML model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus.

[0118] Referring now to Fig. 13, an example of a method 1300 is provided. As shown in block 1310, a first apparatus (e.g., a terminal device) may receive, from a network entity, a reporting criterion.

[0119] As shown in block 1320, the first apparatus may receive, from the network entity, a dataset.

[0120] As shown in block 1330, the first apparatus may receive, from the network entity, a maximum allowed error value for quality indicator computation for training a first artificial intelligence (Al) and / or machine learning (ML) model.

[0121] As shown in block 1340, the first apparatus may train the first AI / ML model, based on the dataset, to obtain one or more predicted time of arrival (ToA) values and one or more quality indicators.

[0122] As shown in block 1350, the first apparatus may obtain, via running the first AI / ML model and based on the input radio measurements, one or more predicted ToA values for one or more transmission and reception points (TRPs).

[0123] As shown in block 1360, the first apparatus may (e.g., based on blocks 1340 and 1350) obtain, via the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or mor predicted ToA values.

[0124] As shown in block 1370, the first apparatus may determine which of the one or more quality indicators meet the reporting criterion.

[0125] As shown in block 1380, the first apparatus may transmit, to the network entity, one or more predicted ToA values and one or more quality indicators which meet the reporting criterion for a respective subset of the one or more TRPs.

[0126] Referring now to Fig. 14, an example of a method 1400 is provided. As shown in block 1410, a second apparatus (e.g., an LMF) may request, from a positioning reference unit (PRU), ground truth data including one or more ground truth time of arrival (ToA) values, one or more input radio measurements, and one or more corresponding quality of a timing value.

[0127] As shown in block 1420, the second apparatus may receive, from the PRU, the ground truth data.

[0128] As shown in block 1430, the second apparatus may transmit, to a first apparatus, a reporting criterion.

[0129] As shown in block 1440, the second apparatus may transmit, to the first apparatus, a dataset comprising the ground truth data.

[0130] As shown in block 1450, the second apparatus may transmit, to the first apparatus, a maximum allowed error value for quality indicator computation for training a first artificial intelligence (Al) and / or machine learning (ML) model.

[0131] As shown in block 1460, the second apparatus may receive, from the first apparatus, one or more predicted ToA values and one or more quality indicators which meet the reporting criterion for a respective subset of one or more transmission and reception points (TRPs), wherein the one or more predicted ToA values for the respective subset of TRPs and the one or more quality indicators for the respective subset of TRPs are obtained by the first apparatus via the first AI / ML model.

[0132] As shown in block 1470, the second apparatus may obtain, via a second AI / ML model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus.

[0133] Fig. 15 shows, by way of example, a block diagram of an apparatus 10. The apparatus 10 comprises, for example, at least one processor 12 and at least one memory 14 storing instructions 15 that, when executed by the at least one processor 12, cause the apparatus 10 at least to perform the method or methods as disclosed herein, and any of the embodiments thereof. In an example, the at least one memory 14 and instructions 15 (e.g., a computer program, code, software, and / or the like) are configured, with the at least one processor 12, to cause the apparatus 10 to perform the method or methods as disclosed herein, and any of the embodiments thereof.

[0134] A processor 12 may comprise circuitry, or be constituted as circuitry or circuitries, the circuitry or circuitries being configured to perform phases of methods in accordance with example embodiments described herein. As used in this application, the term "circuitry” may refer to one or more or all of the following: (a) hardware-only circuit implementations, such as implementations in only analog and / or digital circuitry, and (b) combinations of hardware circuits and software, such as, as applicable: (I) a combination of analog and / or digital hardware circuit(s) with software / f I rmware and (ii) any portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a user equipment, to perform various functions), and (c) hardware circuit(s) and / or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation. This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in server, a cellular network device, or other computing or network device.

[0135] The memory 14 may be implemented using any suitable data storage technology. The memory may comprise a database for storing data. The memory 14 may be at least in part external to apparatus 10 but accessible to apparatus 10.

[0136] The instructions 15 may be comprised in a computer-readable medium or a non-transitory computer readable medium. A term "non-transitory”, as used herein, is a limitation of the medium itself (e.g., tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., random access memory (RAM) vs. read only memory (ROM)).

[0137] For example, the apparatus 10 is a terminal device, such as the UE of Fig. 15. As another example, the apparatus is comprised in such a terminal device, for example, as a chipset configured tocontrol the terminal device. The apparatus 10 may be caused or configured to perform at least the methods of any of Figs. 1-14 and / or any one or more of the embodiments described.

[0138] As another example, the apparatus 10 is a network node, for example, the network node of any of Figs. 1-14. In another embodiment, the apparatus is comprised in such a network node, for example, as a chipset configured to control the network node. The apparatus 10 may be caused or configured to perform the methods of any of Figs. 1-14 and / or any one or more of the embodiments described herein.

[0139] The apparatus may comprise one or more entities of any protocol layers, such as a MAC entity, an RRC entity, an RLC entity, a PDCP entity, a PHY entity, and / or the like. In some embodiments, the entity is configured to perform the methods of any of Figs. 1-14, and / or any one or more of the embodiments described.

[0140] The apparatus 10 comprises a radio interface 16. The radio interface 16 may provide the apparatus 10 with communication capabilities. The radio interface 16 may comprise a receiver configured to receive information in accordance with at least one cellular or non-cellular standard. The radio interface 16 may comprise a transmitter configured to transmit information in accordance with at least one cellular or non-cellular standard. The receiver may comprise more than one receiver. The transmitter may comprise more than one transmitter. The radio interface 16 may comprise a transceiver configured to receive and transmit information in accordance with at least one cellular or non-cellular standard. The transceiver may comprise more than one transceiver.

[0141] The apparatus 10 may comprise a user interface 18 comprising, for example, at least one of a keypad, a microphone, a touch display, a display, a speaker, and / or the like. The user interface 18 may be used to control the apparatus by a user. The user interface 18 may be external to the apparatus 10. For example, the apparatus 10 may be connected to another device, such as a computer, either via wireless or wired connection, and the apparatus 10 is controlled by the user via the computer.

[0142] In an embodiment, at least some of the processes described herein may be carried out by an apparatus comprising means for carrying out at least some of the described processes. Means for performing method steps as disclosed herein may include software and / or hardware components of the apparatus 10. For example, the at least one processor 12, the memory 14, and the instructions 15 (e.g., computer program code) form means for carrying out the method or methods as disclosed herein, and any of the embodiments thereof. As used herein, the term "means” is to be construed in singular form, i.e., referring to a single element, or in plural form, i.e., referring to a combination of single elements. Therefore, terminology "means for [performing A, B, C]” is to be interpreted to cover an apparatus in which there is only one means for performing A, B, and C, or where there are separate means for performing A, B, and C, or partially or fully overlapping means for performing A, B, C. Further, terminology "means for performing A, means for performing B, means for performing C” is to be interpreted to cover an apparatus in which thereis only one means for performing A, B, and C, or where there are separate means for performing A, B, and C, or partially or fully overlapping means for performing A, B, C.

[0143] Even though the disclosure has been described above with reference to an example according to the accompanying drawings, it is clear that the disclosure is not restricted thereto but can be modified in several ways within the scope of the appended claims. Therefore, all words and expressions should be interpreted broadly, and they are intended to illustrate, not to restrict, the embodiment. It will be obvious to a person skilled in the art that, as technology advances, the inventive concept can be implemented in various ways. Further, it is clear to a person skilled in the art that the described embodiments may, but are not required to, be combined with other embodiments in various ways.

Claims

CLAIMS:What is claimed is:

1. A method, comprising: obtaining, via running a first artificial intelligence (Al) / machine learning (ML) model, based on a dataset, one or more predicted ToA values for one or more transmission and reception points (TRPs); obtaining, via running the first AI / ML model, one or more quality indicators for the one or more TRPs based on the one or more predicted ToA values; and transmitting, to a network entity, the one or more predicted ToA values and the one or more quality indicators for the one or more TRPs.

2. The method of claim 1, further comprising: predicting, via running the first AI / ML model, and based on one or more input radio measurements, the one or more predicted ToA values.

3. The method of claim 1 or claim 2, further comprising: receiving, from the network entity, a maximum allowed error value for quality indicator computation for training the first AI / ML model.

4. The method of any of claims 1-3, wherein the network entity comprises a location management function (LMF).

5. The method of any of claims 1-4, wherein the obtaining the one or more quality indicators is further based on one or more ground truth time of arrival (ToA) values and one or more corresponding quality measurements measured by the first apparatus.

6. The method of any of claims 1-5, further comprising: receiving, from the network entity, the dataset.

7. The method of any of claims 1-6, wherein the dataset comprises ground truth data, including at least one of the following: one or more ground truth ToA values; one or more input radio measurements; orone or more corresponding quality indicators of respective timing values.

8. The method of any of claims 1-7, further comprising: training the first AI / ML model, based on one or more ground truth ToA values, to obtain the one or more predicted ToA values and the one or more quality indicators.

9. The method of any of claims 1-8, wherein the first AI / ML model is used for at least one of the following: training, or inference.

10. The method of any of claims 1-9, wherein the first AI / ML model comprises a neural network based supervised learning model.11 . The method of any of claims 1 -10, wherein an apparatus performing the method comprises a terminal device or a network device.

12. A method, comprising: receiving, from a first apparatus, one or more predicted time of arrival (ToA) values for one or more transmission and reception points (TRPs) and one or more quality indicators for the one or more TRPs, wherein the one or more predicted ToA values and the one or more quality indicators are obtained by the first apparatus via a first artificial intelligence (Al) / machine learning (ML) model.

13. The method of claim 12, further comprising: transmitting, to the first apparatus, a maximum allowed error value for quality indicator computation for training the first AI / ML model.

14. The method of claim 12 or claim 13, further comprising: obtaining, via a second artificial intelligence (Al) / machine learning (ML) model and based on the received one or more predicted ToA values and the one or more quality indicators, at least one position of the first apparatus.

15. The method of any of claims 12-14, further comprising: requesting, from a positioning reference unit (PRU), ground truth data including one or more ground truth time ToA values, one or more input radio measurements, and one or more corresponding quality indicators of respective timing values; andreceiving, from the PRU, the ground truth data.

16. The method of any of claims 12-15, further comprising: transmitting, to the first apparatus, a dataset comprising the ground truth data.

17. An apparatus comprising: at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to perform a method according to any of claims 1-16.

18. An apparatus comprising: means for performing a method according to any of claims 1-16.

19. A non-transitory computer readable medium comprising program instructions that, when executed by an apparatus, cause the apparatus to perform at least a method according to any of claims 1-16.

20. A computer program comprising instructions, which, when executed by an apparatus, cause the apparatus to perform the method of any of claims 1-16.

Images