Methods for enabling training of an artificial intelligence, ai, model for positioning of a wireless device, related nodes

Abstract

A method is disclosed, performed by a location network node, for enabling training of an Artificial Intelligence, AI, model for positioning of wireless devices. The method comprises communicating, with the positioning measurement node, a message indicative of a channel response report format for a channel between a WD and a radio network node. The method comprises receiving, from a positioning measurement node, a channel response report for the channel between the WD and the radio network node using the communicated channel response format. The method comprises receiving, from a WD to be positioned, a label report indicative of the position of the WD wherein the label report is associated with the channel response report.

Description

[0001] METHODS FOR ENABLING TRAINING OF AN ARTIFICIAL INTELLIGENCE, Al, MODEL FOR POSITIONING OF A WIRELESS DEVICE, RELATED NODES

[0002] The present disclosure pertains to the field of wireless communications. The present disclosure relates to methods for enabling training of an Artificial Intelligence, Al, model for positioning of a wireless device, related location network nodes, related positioning measurement nodes and related label generating nodes.

[0003] BACKGROUND

[0004] Positioning in dense multipath environments or NLOS channel where a Non-Line-of-Sight (NLOS) paths can be rich and dominant than a Line-of-Sight (LOS) path, such as in an indoor factory scenario, has been a challenging problem. Traditional positioning algorithm sometimes fail in determining the accurate position or location of a wireless device in NLOS channel due to the timing measurement from NLOS propagation channel deviating from the ground truth first path Time of Arrival (TOA). Consequently, traditional localization algorithm sometimes provides an inaccurate positioning estimation (e.g., wrong coordinate), resulting in non-negligible positioning errors.

[0005] There is a need for improving positioning in NLOS channels.

[0006] SUMMARY

[0007] This disclosure finds that Artificial Intelligence (Al) I Machine Learning (ML)-based positioning can be an effective method to improve positioning estimation accuracy in a scenario where NLOS is dominant.

[0008] Al-based positioning mitigates errors in positioning in NLOS channels by learning the relationship between LOS and NLOS TOA measurements and can compensate, such as precompensate, the differences.

[0009] Accordingly, there is a need for location network nodes, positioning measurement nodes and methods for enabling training of an Artificial Intelligence, Al, model for positioning of a wireless device, which may mitigate, alleviate, or address the shortcomings existing and may improve the accuracy of positioning especially in NLOS situations.

[0010] A method is disclosed, performed by a location network node, for enabling training of an Artificial Intelligence, Al, model for positioning of wireless devices. The method comprises communicating, with the positioning measurement node, a message indicative of a channel response report format for a channel between a WD and a radio network node. The method comprises receiving, from a positioning measurement node, a channel response report for the channel between the WD and the radio network node using the communicated channel response format. The method comprises receiving, from a WD to be positioned, a label report indicative of the state of the WD wherein the label report is associated with the channel response report and / or one or more positioning resources.

[0011] Further, a location network node is provided. The location network node comprises memory circuitry, processor circuitry, and a wireless interface, wherein the location network node is configured to perform any of the methods according to the current disclosure relating to the location network node.

[0012] It is an advantage of the present disclosure that channel response information received from a first node, such as a positioning measurement node, and label data from a second node, such as a label generating node, can be received by the location network node, and respectively provided as input for training of an Al model for positioning of wireless devices. The channel response report may be provided as input data to the Al model and the label report as the desired output from the Al model based on the channel response report input. The first node and the second node can be the same or different nodes. By associating the label report with the channel response report and / or the positioning resources, the label report can be associated with a corresponding measurement report. Thereby, pairs of input data and corresponding output data can be generated to train the Al model to be used in determining and / or estimating an actual state of a WD based on a channel response report for a channel between the WD and a radio network node. Associating the label report with the channel response report further enables a training of the model on assisted AI / ML-based positioning where the channel response report is provided by a radio network node acting as a positioning measurement node which is not aware of the WDs actual state. Thereby, training data can be provided which can improve an accuracy of the Al model for positioning of wireless devices, especially in NLOS situations.

[0013] A method is disclosed, performed by a positioning measurement node, for enabling training of an Al model for positioning of wireless devices. The method comprises communicating, with the location network node, a message indicative of a channel response report format for the channel between a WD and a radio network node. The method comprises transmitting, to the location network node, a channel response report for the channel between the WD and the radio network node using the communicated channel response report format. The channel response report may be associated with a label report indicative of an actual state of the WD. Further, a positioning measurement node is disclosed, the positioning measurement node comprising memory circuitry, processor circuitry, and a wireless interface. The positioning measurement node is configured to perform any of the methods disclosed herein relating to the positioning measurement node.

[0014] It is an advantage of the present disclosure that the positioning measurement node can provide channel response information for a channel between a WD and a radio network node, to a location network node. This enables the location network node to provide the channel response report as input for training of an Al model for positioning of wireless devices. By associating the channel response report with a label report, pairs of input data and corresponding output data can be generated to train the Al model to determine an actual state of a WD based on a channel response report for a channel between the WD and a radio network node. Associating the label report with the channel response report further enables training of the model on assisted Al-based positioning where the channel response report is provided by a radio network node. Thereby, the accuracy of the Al model for positioning of wireless devices can be improved, especially in NLOS situations. Furthermore, the channel response size can also be properly configured based on the positioning measurement node capability and the need to minimize the signaling overhead.

[0015] BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The above and other features and advantages of the present disclosure will become readily apparent to those skilled in the art by the following detailed description of examples thereof with reference to the attached drawings, in which:

[0017] Fig. 1 is a diagram illustrating an example wireless communication system comprising an example network node and an example wireless device according to this disclosure,

[0018] Figs. 2A-B are diagrams illustrating an example wireless communication system for Downlink and Uplink positioning respectively according to this disclosure,

[0019] Fig. 3 is a diagram illustrating an example solution for reducing a channel response report size using a moving search time-window according to the current disclosure,

[0020] Fig. 4 is a signaling diagram illustrating an example signaling between a location network node, a radio network node and a wireless device for downlink data collection according to the current disclosure, Fig. 5 is a signaling diagram illustrating an example signaling between a location network node, a radio network node, and a wireless device for uplink data collection according to the current disclosure,

[0021] Fig. 6 is a flow-chart illustrating an example method, performed in a location network node of a wireless communication system, for enabling training of an Al model for positioning of wireless devices according to this disclosure,

[0022] Fig. 7 is a flow-chart illustrating an example method, performed in a positioning measurement node of a wireless communication system, for enabling training of an Al model for positioning of wireless devices according to this disclosure,

[0023] Fig. 8 is a flow-chart illustrating an example method, performed in a label generating node of a wireless communication system, for enabling training of an Al model for positioning of wireless devices according to this disclosure,

[0024] Fig. 9 is a block diagram illustrating an example location network node according to this disclosure,

[0025] Fig. 10 is a block diagram illustrating an example positioning measurement node according to this disclosure, and

[0026] Fig. 11 is a block diagram illustrating an example label generating node according to this disclosure.

[0027] DETAILED DESCRIPTION

[0028] Various examples and details are described hereinafter, with reference to the figures when relevant. It should be noted that the figures may or may not be drawn to scale and that elements of similar structures or functions are represented by like reference numerals throughout the figures. It should also be noted that the figures are only intended to facilitate the description of the examples. They are not intended as an exhaustive description of the disclosure or as a limitation on the scope of the disclosure. In addition, an illustrated example needs not have all the aspects or advantages shown. An aspect or an advantage described in conjunction with a particular example is not necessarily limited to that example and can be practiced in any other examples even if not so illustrated, or if not so explicitly described.

[0029] The figures are schematic and simplified for clarity, and they merely show details which aid understanding the disclosure, while other details have been left out. Throughout, the same reference numerals are used for identical or corresponding parts. Fig. 1 is a diagram illustrating an example wireless communication system 1 comprising an example radio network node 400 and an example wireless device 300 according to this disclosure.

[0030] As discussed in detail herein, the present disclosure relates to a wireless communication system 1 comprising a cellular system, for example, a 3GPP wireless communication system. The wireless communication system 1 comprises one or more wireless devices 300, 300A, such as a mobile wireless device 300 and / or a Positioning Reference Unit (PRU) 300A, and / or a radio network node 400, and / or a location network node 600, such as a Location Management Function (LMF).

[0031] A radio network node disclosed herein refers to a radio access network node operating in the radio access network, such as a base station, a transmission and reception point (TRP), an evolved Node B, eNB, Next-Generation Node B (gNB) in NR. In one or more examples, the RAN node is a functional unit which may be distributed in several physical units. In uplink (UL) based positioning, the radio node is responsible for receiving UL Sounding Reference Signal (UL-SRS) transmitted from the WD and performing positioning measurement. The radio network node is thus acting as positioning measurement node in UL.

[0032] A wireless device may refer to a mobile device and / or a user equipment, UE. The wireless device can be seen as an end-user uses for mobile communication, such as a phone, that connects to the radio network node (e.g., gNBs) and communicate over the wireless network. In downlink (DL) based positioning, the wireless device is responsible for receiving Positioning Reference Signal (PRS) transmitted from the radio network node and performing positioning measurement. The WD is thus acting as positioning measurement node in DL.

[0033] The wireless device 300 is configured to communicate with the radio network node 400 via a wireless link (or radio access link) 10. The wireless device 300 is configured to communicate with other wireless devices via sidelink (SL) communication. In SL positioning, the positioning procedure between the wireless devices is performed over PC5-link.

[0034] A core network, CN, node disclosed herein refers to a network node operating in the core network, such as in the Evolved Packet Core Network, EPC, and / or a 5G Core Network, 5GC. Examples of CN nodes in EPC include a Location Management Function.

[0035] A Positioning Reference Unit 300A can be seen as a positioning measurement unit for which the location is known (such as fixed). For example, the PRU 300A is a functional node for Uplink (UL) -Sounding Reference Signal (SRS) transmission, DL-PRS reception and measurement and is able to provide a precise location information by a manual configuration, Radio Access Technology (RAT) dependent or RAT independent methods, such as Global Positioning System (GPS). The PRU 300A is configured to communicate with the radio network node 400 via a wireless link (or radio access link) 10A.

[0036] A location network node 600 can be seen as a functional node of the 5G positioning architecture. In one or more examples, the location network node 600 is a Location Management Function (LMF). For example, the location network node 600 receives measurements and assistance information from the radio network nodes and the wireless device to compute the position of the wireless device. In another example, the location network node 600 provides the configuration to the radio network nodes and the wireless device. In some examples, an AI / ML model is trained in location network node 600 and subsequently the location network node 600 collects the information data and construct the training data set for training the AI / ML model. In this disclosure, AI / ML-model, Al-model, and ML-model are used interchangeably for a machine learning model trained using artificial intelligence. In some examples, the ML model is trained in a separate node (such as a dedicated ML computation node) and connected to the location network node 600, such as the LMF, within a core network.

[0037] Figs. 2A-B are diagrams illustrating an example wireless communication system for Downlink and Uplink positioning respectively according to this disclosure.

[0038] Fig. 2A shows a DL-based positioning involving an example wireless device 300, an example radio network node 400, an example PRU 300A, and an example location network node 600.

[0039] In Fig. 2A, the radio network node 400 sends Downlink Positioning Reference Signal (DL-PRS) to the wireless device 300. The wireless device 300 receives the DL-PRS signals and performs positioning measurements, such as DL-Time Different of Arrival (DL-TDOA) and Reference Signal Timing difference (RSTD), DL-Angle of Departure (AoD) of the received signal. The positioning measurement results, such as timing information is reported back to location network node 600, via LTE Positioning Protocol (LPP), for location determination. In other words, in DL, the wireless device 300 acts as a positioning measurement node. The timing information may be reported back to the location network node in a channel response report, such as a CIR report. The channel response report, such as the CIR report, differs from legacy positioning reports in that timing information is reported for all arrival paths of the channel, rather than for only the first arrival path of the channel in legacy positioning report.

[0040] Fig. 2B shows an UL positioning involving an example wireless device 300, an example radio network node 400, an example PRU 300A, and an example location network node 600. In Fig. 2B, the wireless device 300 transmits Uplink Sounding Reference Signal (UL-SRS) to the radio network node 400. The radio network node 400 performs positioning measurement, such as measures the timing of the received UL-SRS, e.g., Relative Time of Arrival (RTOA), Angle of Arrival (AoA) of the received UL-SRS and reports the measurement results to the location network node 600. The location network node 600 will further calculate the estimated WD position (such as coordinate). In other words, in UL, the radio network node 400 acts as a positioning measurement node.

[0041] UL-based positioning uses Time Difference of Arrival (TDOA) measurements, which relies on measuring the time difference it takes for a signal to travel from the wireless device to multiple radio network nodes. For the DL-based positioning, the TDOA measurements rely on measuring the time difference it takes for a signal to travel from multiple radio network nodes to the wireless device. By comparing these time differences, the location network node can multi- laterate the wireless device location. For example, to measure TDOA, the wireless device first calculates a channel response, such as a Channel Impulse Response (CIR). For example, the channel response includes information about the multipath characteristic, signal reflections, and attenuation due to obstacles. By analyzing the channel response, the wireless device (WD) can estimate a time delay, such as a propagation delay, such as a propagation time, and further calculates the TDOA. The channel response measurements can be normalized to provide power and time, such as delay, information, which is also be referred to as a Power Delay Profile (PDP).

[0042] In some examples, raw channel response measurement (such as unprocessed channel response measurement) can be expressed in a N-dimensional matrix with size

[0043] NTRPx Nsampx N, where NTRPindicates the number of the radio network nodes used for one positioning occasion, Nsampdenotes the number of samples in the channel response and N denotes the radio characteristic parameter of each channel response sample. N can be assumed equal to 3 to represent the delay, power and phase information of each channel response sample. In some examples, NTRPcan be up to 18, for example in an indoor scenario, and the determination of Nsampis up to implementation, such as the positioning measurement node capability. For example, Nsampcan be 256. In legacy positioning, determining the wireless device’s location involves identifying the first path by analyzing channel response, extracting corresponding timing information and subsequently calculate WD’s positioning estimation based on multiple timing information or other positioning measurement information, such as angle information. However, in Al-based positioning, the Al model directly learns the relationship between channel response(s) and the wireless device coordinate. The disclosed approach simplifies the localization process and improves the positioning accuracy.

[0044] For example, Al-based positioning operates in DL and UL directions. Al-based positioning comprises for example two sub-categories, differentiated by the specific usage of the Al model: Direct Al-based positioning and Assisted Al-based positioning.

[0045] In direct Al-based positioning, for example, the positioning based on an Al model directly outputs the coordinate of the WD and the model is trained based on at least the ground truth label of the WD / PRU’s coordinate.

[0046] In assisted Al-based positioning, for example, the positioning based on Al model does not directly output the coordinate of the WD, but provides an intermediate estimation, such as positioning measurement results, including TOA or information about the LOS or NLOS states of the wireless channel. These intermediate estimations can help in the calculation of the WD’s coordinate by the location network node. In assisted Al-based positioning, for example, the model output varies depending on the specific function, which also leads to various labeling types in data collection accordingly. For example, when the model aims for NLOS mitigation, label generation entities provide ground truth LOS probability or LOS indicator. For example, in the other scenarios, when the model focuses on TOA estimation, the label generation entities provide ground truth range differences from the WD to radio network node.

[0047] Al-based positioning comprises 3 steps: Data collection, Model training and Model deployment. The present disclosure may be seen as addressing data collection aspects.

[0048] Data collection supports for example the development and training of Al model. Data collection is an operation that gathers information and observations that are used to train, validate, and test the Al model(s). In the context of Al-based positioning, for example, the location network node is responsible for data collection and collects information data from a positioning measurement node (which can be any of WD / PRU / radio network node) to develop an Al model dedicated for positioning. The collected data can include raw measurement(s), and / or some features extracted from the radio reference signal.

[0049] The data collection may include a labelling of the collected data. For example, the labeling ensures that each information data is associated with its corresponding ground truth location of the WD. This linkage between information data and accurate ground truth location supports advantageously and effectively the training of Al model(s) for achieving a more precise positioning calculation. In other words, for example, the data collection is based on cooperation between a positioning measurement node and location network node.

[0050] The positioning measurement node can be seen as an information generation entity (WD / PRU / radio network node). The positioning measurement node for example provides positioning measurements and generates information data needed in the Al model training. In some examples, the positioning measurement node reports the information data over the air to location network node. For example, location network node collects the information data and proceed with data analysis, post-processing and model training.

[0051] It may be appreciated that in NR positioning protocol, Al-based positioning is based on a significant data collection in term of the size and the number of parameters, which requires the position measurement node to report channel response measurements, such as CIR measurements, in a channel response report to the location network node. For example, the location network node holds information the ground truth label associated with each channel response report (e.g., CIR measurement report) to execute, train, and update the Al model. In some examples, the channel response reports (e.g., CIR measurements) and the corresponding labels are generated by the position measurement node (e.g., WD, PRU or radio network node depending on the signaling direction, DL or UL). The present disclosure provides signaling enabling reporting of channel response reports and associated label reports for training of an Al model over the air to the location network node.

[0052] However, reporting raw channel response, such as CIR, measurement can be seen as causing a heavy signaling overhead. For example, channel response measurement is implemented by the WD in DL positioning and by the radio network node in UL. For example, one raw CIR measurement covers the wireless channel of interest, which duration depends on the deployment scenario (outdoor or indoor) and the environment’s delay spread. For example, in an indoor scenario, one raw CIR measurement may comprise 256x3 samples per WD-radio network node link, recording the delay, power and phase information of the channel over the range span between 0 meters to 624 meters in FR1 with SCS = 30kHz. This range is the equivalent distance / range of the delay information. For example, in FR2 with SCS = 120kHz, 256 samples in raw CIR measurement only cover range up to 156m. For example, when the size of the CIR is converted into information bits and each information element is represented with 8 bits as an example, the CIR measurements are of a size of 6144 bits (0.77 KB) and become even of a larger size (3.85 KB) when considering 5 WD-radio network node links used for one positioning occasion. 8 bits here is just an example, other sizes can also be applied with better granularity, such as 16, and / or 32 bits. It can be seen that it is not efficient to report such a large data packet at every positioning occasion.

[0053] The present disclosure addresses the signaling overhead situation by reducing the report size without compromising the positioning accuracy. The disclosure provides a radio channel characteristic reporting that is more compact in size and / or can be adjusted, such as based on the device capability.

[0054] Fig. 3 illustrates an example solution for reducing a channel response report size according to the current disclosure. According to the current disclosure, an enhanced channel response report format, such as CIR measurement report format, with a flexible report size may be used to reduce the signaling overhead. According to the example solution a moving search window 20 is used to identify a channel of interest. Reporting a raw channel response measurement, such as CIR measurement, may require a heavy signaling overhead. A raw channel response measurement can herein be seen as the entire measurement performed by the positioning measurement node, such as all Nsampmeasured samples. A raw channel response measurement may comprise channel information within a range spanning from 0 to 624 m in 3GPP Frequency Range 1 (FR1). However, not all this channel information from 0 m to 624 m may be relevant for the location network node, such as for the training of the Al model. For instance, in an example scenario where a WD is located approximately 500 m from a radio network node, the channel information from 0 m to 400 m does not provide any relevant information of environment surrounding the WD. Including this portion of the channel response in the channel response report is therefore unnecessary and creates unnecessary signaling overhead. In such case, it is more efficient to report a subset of the raw channel response data which captures channel information from a narrower range, for example from 450 m to 550 m. This subset of channel response is sufficient for the location network node to analyze the relevant channel characteristics and determine the position of the WD. To determine the narrower range in the channel response, the positioning measurement node (such as the WD in DL or the radio network node in UL) can deploy a moving search window 20 in a time domain, such as a moving time window. The search window 20 may have a size of Nwinsamples where Nwin<Nsamp- Fig- 3 is a schematic diagram illustrating the process of searching for a signal arrival region. The search window 20 moves, such as traverses, from a first measured delay sample to a last measured delay sample (such as window time stamp from 1 to 256). During the movement, the WD continuously calculates a summation of the measured power level within the search window. By identifying a timestamp Twinat which the search window 20 provides a maximum summation of power, the WD can identify the channel of interest. Using this approach, the WD then reports the search window time stamp Twinand a subset of channel response measurements with sample number Nwinfrom the time stamp Twin.

[0055] Fig. 4 shows a signaling diagram illustrating an example signaling between a wireless device 300, 300A, a radio network node 400, and a location network node 600 for DL-based data collection for enabling training of an Al model for positioning of WDs according to the current disclosure. In the DL-based data collection the radio network node 400 is a transmission point transmitting positioning reference signals, such as DL Positioning Reference Signals (PRS). The WD 300, 300A measures on the reference signals and therefore acts as a positioning measurement node. The example signaling shown in Fig. 4 may be used to train the Al model on direct Al based positioning.

[0056] In one or more examples, the location network node 600 transmits a capability request 1001 to the positioning measurement node, such as WD 300, 300A, requesting the WD 300, 300A to provide its capability for collecting data for training an Al model for positioning of WDs, such as its capability to provide a channel response report and / or a label report.

[0057] In one or more examples, the WD 300, 300A provides information 1002 indicative of its capability of collecting data for training the Al model, such as its capability to provide a channel response report and / or a label report, to the location network node 600.

[0058] The location network node 600 may send a request for a channel response report 1003 for a channel between the radio network node 400 and the WD 300, 300A. The request may be sent to the radio network node 400 and the WD 300, 300A. In one or more example methods, the request 1003 may be forwarded to the WD 300, 300A by the radio network node 400. In another example, the request 1003 is directly sent from the location network node 600 to the WD 300, 300A (i.e. , transparent to the radio network node 400).

[0059] The radio network node 400 transmits reference signals for positioning either periodically, aperiodically, or semi-static, such as DL PRS 1004, to the WD 300, 300A.

[0060] Upon receiving the request, the WD 300, 300A measures on the reference signals and performs a positioning measurement 1005 and / or a channel response measurement 1006.

[0061] In one or more examples, the location network node 600 may send a request for label collection 1007 to the WD 300, 300A. The label is indicative of a state, such as an actual state, of the WD, such as a ground truth coordinate of the WD 300, 300A. The label may be associated with the channel response measurement, such as a channel response resource used for measuring the channel response. The state of the WD may comprise information indicative of a current state, such as situation, of the WD. The information indicative of the state of the WD may comprise one or more of an actual position of the WD, such as its ground truth coordinate, an orientation of the WD, LOS / NLOS information, and an actual delay, such as a timing delay of the WD.

[0062] The WD 300, 300A may send positioning and / or channel response reports comprising a label report, such as label information indicative of the WDs actual state, associate with the channel response report and / or the channel response resource(s). By associating the label report with the channel response report, the channel response report may be provided as input for training an Al model for positioning while the associated label report may be provided as the output of the Al model. The Al model can thus be trained to determine a position of wireless devices based on their measured channel responses.

[0063] Fig. 5 shows a signaling diagram illustrating an example signaling between the wireless device 300, 300A, the radio network node 400, and the location network node 600 for UL-based data collection for enabling training of an Al model for positioning of WDs according to the current disclosure. In the UL-based data collection the WD 300, 300A is a transmission point transmitting positioning reference signals, such as UL Sounding Reference Signals (SRS). The radio network node 400 measures on the reference signals transmitted from the WD 300, 300A and therefore acts as a positioning measurement node. The example signaling shown in Fig. 5 may be used to train the Al model on assisted Al based positioning.

[0064] In one or more examples, the location network node 600 transmits a capability request 2001 A to the WD 300, 300A, requesting the WD 300, 300A to provide its capability for collecting data for training an Al model for positioning of WDs, such as collecting label information and / or providing a label report.

[0065] In one or more examples, the location network node 600 transmits a capability request 2001 B to the radio network node 400, requesting the radio network node 400 to provide its capability for collecting data for training the Al model for positioning of WDs, such as for collecting channel response information and / or providing a channel response report. In one or more example methods, the capability request 2001 B may also request the radio network node to provide its capability for collecting label information and / or providing a label report, particularly label information related to the positioning measurement (such as LOS / NLOS information, and an actual delay, such as a timing delay of the WD).

[0066] In one or more examples, the WD 300, 300A provides information 2002A indicative of its capability of collecting data for training the Al model to the location network node 600, such as for collecting label information and / or providing the label report. In one or more examples, the radio network node 400 provides information 2002B indicative of its capability of collecting data for training the Al model to the location network node 600, such as for collecting channel response information and / or providing a channel response report. In one or more example methods, the information 2002B may also be indicative of the radio network node’s capability for collecting label information and / or providing a label report, particularly label information related to the positioning measurement (such as LOS / NLOS information, and an actual delay, such as a timing delay of the WD).

[0067] The location network node 600 may send a request 2003 for UL reference signal resource allocation and label information to the radio network node 400.

[0068] The radio network node 400 sends a request 2004 for UL reference signal transmission and label information to the WD 300, 300A. The request 2004 may comprise an UL reference signal transmission configuration.

[0069] Upon receiving the request 2004, the WD 300, 300A transmits reference signals 2005 for positioning, such as UL SRS. The reference signals may be transmitted in dedicated resources, such as time and / or frequency resources. The dedicated resources may herein be referred to as positioning resources.

[0070] The WD 300, 300A further sends a label report 2006 comprising information indicative of the actual state of the WD 300, 300A, such as a ground truth coordinate of the WD 300, 300A. The label report 2006 may be associated with a positioning resource, such as the resources for transmission of the UL SRS 2005. By associating the label report 2006 with the positioning resources, the label report can be associated with a channel measurement report provided to the location network node by a different node, such as the radio network node 400.

[0071] In one or more example methods, the radio network node 400 performs a positioning measurement 2007 on the reference signals 2005, such as the UL SRS, transmitted by the WD in the positioning resources.

[0072] The location network node 600 sends a channel response report request 2008 to the radio network node 400.

[0073] Upon receiving the channel response report request 2008, the radio network node 400 performs a channel response measurement 2009 on the reference signals 2005 transmitted by the WD 300, 300A.

[0074] The radio network node 400 sends a channel response report 2010 comprising channel response measurement information to the location network node 600. The channel response report 2010 may be associated with the label report 2006 transmitted by the WD 300, 300A based on the resources, such as positioning resources, used for transmission of the reference signals 2005. The channel response report may, in one or more example methods, comprise information indicative of the positioning resources used for transmission of the reference signals 2005 By associating the label report 2006 with the channel response report 2010, the channel response report 2010 may be provided as input for training an Al model for positioning while the associated label report 2006 may be provided as the output of the Al model. The Al model can thus be trained to determine a position of wireless devices based on their measured channel responses.

[0075] Fig. 6 shows a flow diagram of an example method 100, performed by a location network node according to the disclosure, for enabling training of an Al model for positioning of a wireless device. The location network node is the location network node, such as an LMF disclosed herein, such as location network node 600 of Fig. 1 , Fig. 2A-2B, Fig. 4, Fig. 5, and Fig. 9.

[0076] In one or more example methods, the method comprises receiving S101 , from the positioning measurement node, such as the WD, information indicative of the positioning measurement node’s capability for collecting data, such as Al training data. In one or more example methods, receiving S101 comprises receiving, from a label generating node, such as the WD and / or the radio network node, information indicative of the label generating node’s capability of providing label information for training of the Al model.

[0077] As part of the data collection procedure, the positioning measurement node and / or the label generating node provide their capability of collecting data, such as input data for the Al model, to the location network node. In one or more example methods, the information indicative of the positioning measurement node’s capability of collecting data comprises information indicative of a capability of performing a channel response measurement. In one or more example methods, the information indicative of the label generating node’s capability of collecting data comprises information indicative of a capability of providing label information associated with a positioning resource and / or a channel response measurement. The information indicative of the capability of channel response measurement may indicate whether the positioning measurement node is capable to perform channel response measurements over a reference signal for positioning, such as a DL-PRS or UL-SRS signal. In one or more example methods, the information indicative of the positioning measurement node’s capability for collecting data comprises information indicative of a capability of channel response reporting, such as a reporting of the channel response measurement. In one or more example methods, the information indicative of the positioning measurement node’s capability may specify a preferred reporting index, such as a CIRReportFormatlndex. The reporting index may represent a reporting size, such as a dimension of the reporting, and one or more measurement types, such as one or more of power, time (such as a delay), and phase.

[0078] The method 100 comprises communicating S102, with a positioning measurement node, a message indicative of a channel response report format for a channel between a WD and a radio network node. In one or more example methods, the positioning measurement node may select a channel response report format from a predetermined look-up table comprising a plurality of different channel response report formats and may indicate the selected channel response report format to the location network node. The predetermined look-up table may be transmitted to the positioning measurement node by the location network node or may be hard- coded, such as installed upon manufacturing or setup, in the positioning measurement node.

[0079] In one or more example methods, communicating S102 comprises receiving S102A, from the positioning measurement node, a message indicative of the channel response report format used by the positioning measurement node. In one or more example methods, the channel response report format used by the positioning node may be indicated by the message comprising an index corresponding to a specific channel response format indicated in the lookup table. In one or more example methods, the message indicative of the channel response report format used by the positioning measurement node may comprise a confirm of a request from the location network node to use a specific channel response report format. In one or more example methods, such as when the positioning measurement node cannot provide the format requested from the location network node, the message indicative of the channel response report format used by the positioning measurement node may comprise information indicative of alternative channel response report format, such as a less complex, and / or simplified, and / or lower size channel response report format.

[0080] In one or more example methods, communicating S102 comprises transmitting S102B, to the positioning measurement node, a message indicative of a channel response report format, such as a CIR format, to be used by the positioning measurement node. The location network node, such as the LMF, and the positioning measurement node, such as the WD for DL positioning or the radio network node for UL positioning, may agree on a channel response format to use. For example, the positioning measurement node can inform the location network node about the report format via Information Elements (lEs), such as CIRReportFormat-NumberOfTRP, CIRReportFormat-NumberOfCIRSample and / or CIRReportFormat-MeasurementType. In one or more example methods, the positioning measurement node may report this information individually to the location network node. In one or more example methods, the channel response report format is indicative of an index associated with the channel response report format. In one or more example methods, the location network node predefines a lookup table which may specify combinations of parameters to report. In such case, the positioning measurement node may report only an Index of the combination, for example via the IE CIRReportFormatlndex. An example look-up table is described below. According to one or more example look-up tables, the look-up table comprises entries for the index (CIRReportFormatlndex), a number of transmission points to report channel responses for (CIRReportFormat-NumberOfTRP), a number of channel response samples to report (CIRReportFormat-NumberOfCIRSample), and the measurement types to report (CIRReportFormat-MeasurementType). In one or more examples, the maximum number of TRPs to report may be 18 and the maximum number of samples to report may be 256.

[0081] In one or more example methods, the look-up table, comprising for example parameters to be reported and ranges and / or options of each parameter, is defined in a specification, such as a 3GPP specification. In one or more example methods, when the location network node requests a positioning measurement node, the location network node specifically asks the positioning measurement node to provide a certain parameter, such as for example number of transmission points, or number of samples to report, with a certain value X. The location network node may for example request a measurement report with the number of transmission points TRP = 6, and the number of channel response samples = 100, etc. In one or more example methods, the channel response measurement report may report all channel response measurements available, which may herein be referred to as a reporting raw CIR measurements. In other words, the positioning measurement node does not reduce the measurement report and reports data in the entire IE reserved for channel response reporting.

[0082] In one or more examples, the IE for channel response reporting, such as the CIR Report IE, has a reserved data size of 18 x 256 x 3, assuming the maximum number of TRPs, NTRP= 18, and the maximum number of samples, Nsamp= 256

[0083] In one or more example methods, the channel response report format is indicative of a number of transmission points to report a channel response from. The transmission points may be radio network nodes in DL positioning when the positioning measurement node is a WD. The transmission point may be a WD in UL positioning when the positioning measurement nodes are the radio network nodes. In one or more example methods, the positioning measurement node, such as the WD in DL or the radio network node in UL, can choose to report the channel response, such as the CIR, from a subset of the available transmission points. For example, the positioning measurement node may report a subset NTRP' out of NTRPavailable transmission points, where NPRP< NTRP.

[0084] In one or more example methods, the channel response report format is indicative of a number of samples to be reported for each transmission point. In one or more example methods, the positioning measurement node may report a reduced sample number. In one or more example methods, the subset of samples reported may be determined by the size of a time window, such as a moving time window. In one or more example methods, the positioning measurement node reports a window time stamp Twinand a reduced size channel response report having a size of for example 18 x Nwinx 3. In one or more example methods, the positioning measurement node may choose the size reduction percentage by selecting a corresponding Nwin. For instance, the channel response report, such as the CIR report, package size can be reduced by 90%, if reducing the number of samples from the reserved data size of 256 to the time window of 26, e.g., Nwin= 26. In other words, by reducing the number of samples from the reserved data size of 256 to the time window of Nwin, the package size of the channel response report can be reduced by (1 - Nwin / 256). The raw CIR measurement report can hereby be considered as a special case where Twin= 0 and Nwin= 256.

[0085] In one or more example methods, the channel response report format is indicative of a channel information type to report. The channel information type may comprise one or more of: time information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response. In one or more example methods, the positioning measurement node may report all or a subset of the channel information, such as time (such as delay), power, and / or phase information. In one or more example methods, the positioning measurement node may report time, power, and phase. In one or more example methods, the positioning measurement node may report time (such as delay) and power. In one or more example methods, the positioning measurement node may report time only.

[0086] In one or more example methods, the channel response report format is indicative of a granularity of the channel information. In one or more example methods, the positioning measurement node may report time information (such as delay information), power information and / or phase information with their distinct level of granularities. In one or more example methods, the positioning measurement node may select from three different granularities, such as a first granularity for time information, a second granularity for power information, and / or a third granularity for phase information. These granularities may be indicated by IE ‘CIRtimingReportingGranularityFactor’, IE ‘CIRpowerReportingGranularityFactor’, and IE ‘CIRphaseReportingGranularityFactor’ correspondingly.

[0087] In one or more example methods, the method comprises sending S103, to the positioning measurement node, a request for a channel response report for the channel between a WD and a radio network node. In one or more example methods, such as when the positioning is performed in UL, the positioning measurement node is the radio network node. In one or more example methods, such as when the positioning is performed in DL, the positioning measurement node is the WD.

[0088] In one or more example methods, the method comprises sending S104, to a label generating node, such as the WD, a request for a label report indicative of a state of the WD. In one or more example methods, such as when the positioning is performed in DL, the positioning measurement node and the label generating node is the same node, such as the WD. In one or more example methods, such as when the positioning is performed in UL, the positioning measurement node and the label generating node are different nodes, where the positioning measurement node is the radio network node and the label generating node is the WD. In one or more example methods, the request for a label report is provided together with the request for a channel response measurement, such as a request for an SRS transmission sent to the WD from the radio network node.

[0089] The method 100 comprises receiving S106, from the positioning measurement node, a channel response report, such as a CIR report, for the channel between the WD and the radio network node, using the communicated channel response report format, such as the channel response format communicated in S102, and / or S102, and / or S102B. In other words, the channel response report may be reported according to the channel response format communicated in S102, and / or S102, and / or S102B. For DL-based positioning, such as DL based data collection, the channel response report can be received from the WD via an LTE Positioning Protocol (LPP). For UL-based positioning, such as UL based data collection, the channel response report can be received from the radio network node via a NR Positioning Protocol A (NRPPa).

[0090] In one or more example methods, the channel response report comprises a subset of channel response measurement samples, such as a number Nwinof the available number of samples NSamp- The subset may be a subset of an entire set of samples measured by the positioning measurement node. In one or more example methods, the subset comprises one or more of time information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response.

[0091] In one or more example methods, the channel response report comprises the time stamp, such as time stamp Twin, indicative of a timing of the first time parameter, and / or power parameter, and / or phase parameter in the subset of channel measurement samples. The timing of the first time parameter may be indicated as an absolute timing or relative to a reference timing in units of second / millisecond or number of time sample.

[0092] In one or more example methods, the channel response report comprises a size, such as information indicative of the size, of a time window of the reported subset of channel measurement samples.

[0093] In one or more example methods, the channel response report comprises information indicative of one or more resource(s) in which the reference signals for positioning was / were transmitted, such as one or more positioning resources. In one or more example methods, the resources may be explicitly and / or implicitly provided in the channel response report. This may allow the location network node to associate the channel response report with a label report based on the one or more resources, such as positioning resources.

[0094] In one or more example methods, the channel response report is associated to one or more of the positioning measurement(s). In one or more example methods, such as for DL based data collection, the channel response report is associated with measurements from one or more transmission points, such as transmission points transmitting reference signals from positioning measurements. In other words, when the positioning measurement node is a WD, the channel response report may be associated with, such as may comprise information associated with a plurality of radio network nodes transmitting reference signals. In one or more example methods, such as for UL-based data collection when the positioning measurement node is the radio network node, the channel response report may comprise channel response measurements specific for one WD-radio network node link. In this case, the IE CIRReportFormat-NumberOfTRP is equal to 1.

[0095] The method 100 comprises receiving S108, from a label generating node, such as the WD or the radio network node, a label report indicative of the state of the WD. The label report may be associated with one or more channel response report(s) and / or one or more positioning resource(s). The positioning resources may be one or more time and / or frequency resources for transmission of reference signals for positioning, such as positioning reference signals in DL (DL-PRS) or Sounding reference signals in UL (UL-SRS). The association between the label report and the channel response report and / or the one or more positioning resources, can be conveyed implicitly, or by explicitly sending one or more resource IDs together with the label report. In one or more example methods, the label report is received from the label generating node, such as the WD, via LPP. The label report may, in one or more example methods, be signaled using a ‘LPP Location Information Delivery Procedure’ as defined in 3GPP TS 37.355 V17.3.0.

[0096] The state of the WD may comprise information indicative of a current state, such as situation, of the WD. The information indicative of the state of the WD may comprise one or more of an actual position of the WD, such as its ground truth coordinate, an orientation of the WD, LOS / NLOS information, and an actual delay, such as a timing delay of the WD.

[0097] The label types, such as the information comprised in the label report, may differ based on the Al model’s functionality. In one or more example methods, the label report comprises information indicative of ground truth (GT) coordinates of the WD, such as of a WD for which positioning training data is to be collected. The GT coordinates may be coordinates in a Global Cartesian System (GCS). In one or more example methods, such as for direct AI / ML-based positioning, the label generating node, such as the WD, provides its GT coordinate to the location network node. In one or more example methods, only one GT coordinate for the WD is provided in the label report for each channel response report.

[0098] In one or more example methods, the label report comprises a Line-of Sight (LOS), or non-Line- of-Sight (NLOS) channel indication, such as information indicative of whether a channel between the WD and a radio network node is LOS or NLOS. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning, where the label generation node provides information specific to each link between the WD and a radio network node. In one or more example methods, the label report comprises information indicative of a time, such as a delay, such as a propagation delay, on the channel between the WD and a radio network node. The time delay may be a time delay of the LOS path on the channel between the WD and the radio network node. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning.

[0099] In one or more example methods, the label report comprises information indicative of a range, such as a LOS distance, of the channel between the WD and a radio network node. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning. In one or more example methods, the information indicative of the time delay may be indicative of the range.

[0100] In one or more example methods, the label report comprises information indicative of a timing error. The timing error may be a timing error at the positioning measurement node and / or at the transmission point transmitting reference signals for positioning. In other words, the timing error may be a WD timing error and / or a radio network node timing error. In one or more example methods, this information may be provided from a specific WD, such as a PRU. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning.

[0101] In one or more example methods, the label report comprises information indicative of an association between the label report and the channel response report. The information indicative of the association between the label report and the channel response report can be used to map the label report to the channel response report, such as to map an expected output from the Al model to an input to the Al model. For UL data collection, the label report and the channel response report may be received from different entities, such as from different nodes. The label report may for example be received from the WD and the channel response report from the radio network node. In this case the location network node may receive the label report from the label generating node, such as the WD, with information associating the label report to the channel response report received from the positioning measurement node, such as the radio network node.

[0102] In one or more example methods, the label report is transmitted from the WD and received by the location network node as soon as the WD transmits reference signals for positioning, such as SRS. In this case, the association between the label report and the channel response report is implicitly provided by the WD by informing the location network node about the association between the label report and the SRS transmission, such as resources of the SRS transmission. Thereby, the location network node is made aware of the association between the label report and the corresponding channel response report from that SRS transmission.

[0103] In one or more example methods, the information indicative of the association between the label report and the channel response report may be explicitly provided in the label report. In one or more example methods, the label report may comprise information indicative of the positioning resources used for the channel response measurement, such as an SRS resource ID or an PRS resource ID, associated with the label report.

[0104] In one or more example methods, the label report comprises information indicative of a label generation source, such as information indicative of a method and / or technology used for label information generation. The label generation source and / or the label generation node, can be seen as the source and / or node that is generating the label report. Label information may be generated from different sources, such as from different nodes. In the context of direct AI / ML positioning, different types of WDs, such as mobile UEs and Positioning Reference Units (PRUs), may use different methods for obtaining their location information, such as information indicative of the actual state of the WD. For PRUs, which have fixed and known locations, the GT PRU coordinate should remain constant over time and the uncertainty of this location is therefore negligible. However, for regular mobile UEs for which the position varies over time, GT coordinates are typically determined using Radio Access Technology (RAT) dependent methods (such as 3GPP-based DL Time Difference of Arrival (TDOA), sidelink (SL) ranging, and / or a Carrier Phase Positioning (CPP)) or RAT-independent methods (such as Global Navigation Satellite System (GNSS) Real-Time Kinematic (RTK)). A quality of the label report, such as an uncertainty in obtaining the actual position of the WD indicated in the label report, such as the GT coordinate of the WD, may be highly dependent on the chosen method for obtaining the location information.

[0105] In one or more example methods, the label generating node may thus provide information indicative of the source of label generation in the label report. This may enable the location network node to assign different confidence levels and / or judgements to the label report, such as the information provided in the label report, thereby improving the model training performance. In one or more example methods, the label report may comprise a dedicated IE, such as an IE referred to as ‘ Labe / Source Index’ where an index value may correspond to a specific label generation source, as illustrated in the following table:

[0106] In one or more example methods, the label report comprises information indicative of a label quality, such as a quality of the indicated position of the WD. In one or more example methods, the information indicative of the label quality may be explicitly indicated in the label report, for example as an uncertainty of the WD coordinate in an actual distance, such as in meters.

[0107] In one or more example methods, the label report comprises information indicative of a time stamp, such as a label time stamp. The time stamp, such as the label time stamp, may be a time stamp when a WD receives DL-PRS for performing channels response measurements, or a time stamp when the WD transmits UL SRS for positioning used by one or more radio network nodes to measure a channel response.

[0108] In one or more example methods, the label report comprises information indicative of a time window, such as a label time window. The time window, such as the label time window, may be indicative of a duration of label validity, such as a maximum time that the label report is valid. In case the label generation node detects that there are changes related to reported label information, then the label generating node can transmit a new label report before the time window has expired.

[0109] In one or more example methods, such as when the data collection, such as the positioning measurement, is performed in DL, the positioning measurement node is a WD.

[0110] In one or more example methods, such as when the data collection, such as the positioning measurement, is performed in UL, the positioning measurement node is a radio network node. In one or more example methods, the method comprises providing 110 the channel response information and / or the label report for training of an Al model for positioning of WDs. Al model training can be seen as a process of supervised learning, which learns the relationship between a model input and an expected model output using for example a neural network. To train the Al model for positioning, the channel response report, such as CIR information data, may be provided as an input to the Al model, and the label report, such as the corresponding labels, can be provided as an expected Al model output.

[0111] Fig. 7 shows a flow diagram of an example method 200, performed by a positioning measurement node according to the disclosure, for enabling training of an Al model for positioning of wireless devices. The positioning measurement node is the positioning measurement node disclosed herein, such as the positioning measurement node 700 of Fig. 10, such as the wireless device 300, 300A of Fig. 1 , Figs. 2A-2B, Fig. 4, Fig. 5 when the data is collected in DL, or the radio network node 400 of Fig. 1 , Fig. 2A-2B, Fig. 4 and Fig. 5 when the data is collected in UL. The positioning measurement node may herein also be referred to as a data collecting node, such as a data collecting node for collecting input data to an Al model for positioning of WDs.

[0112] In one or more example methods, the method comprises transmitting S201 , to the location network node, information indicative of the positioning measurement node’s capability for collecting data, such as Al training data. In one or more example methods, transmitting S201 is performed in response to receiving a request for capability of performing data collection, such as capability of providing label information and / or a channel response report, from the location network node. The information indicative of the positioning measurement node’s capability for collecting data may be indicative of the positioning measurement node’s capability of providing a channel response report and / or a label report.

[0113] Transmitting S201 corresponds to receiving S101 performed by the location network node in method 100 of Fig. 6, to 1002 of Fig. 4 and to 2002B of Fig. 5.

[0114] The method 200 comprises communicating S202, with the location network node, a message indicative of a channel response report format for the channel between the WD and a radio network node.

[0115] In one or more example methods, the channel response report format is indicative of, such as comprises information indicative of, a number of transmission points to report a channel response from. In one or more example methods, the channel response report format is indicative of, such as comprises information indicative of, a number of samples to be reported for each transmission point.

[0116] In one or more example methods, the channel response report format is indicative of, such as comprises information indicative of, a channel information type to report. In one or more example methods, the channel information type comprises one or more of time information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response.

[0117] In one or more example methods, the channel response report format is indicative of, such as comprises information indicative of, a granularity of the channel information.

[0118] In one or more example methods, the channel response report format is indicative of, such as comprises information indicative of, an index associated with the channel response report format.

[0119] In one or more example methods, communicating S202 comprises transmitting S202A, to the location network node, a message indicative of the channel response report format used by the positioning measurement node. In one or more example methods, the positioning measurement node may select a channel response report format from a predetermined look-up table comprising a plurality of different channel response report formats and may indicate the selected channel response report format to the location network node. The predetermined lookup table may be transmitted to the positioning measurement node by the location network node or may be hard-coded, such as installed upon manufacturing or setup, in the positioning measurement node. In one or more example methods, the channel response report format used by the positioning node may be indicated by the message comprising an index corresponding to a specific channel response format indicated in the look-up table. Transmitting S202A corresponds to the receiving S102A performed by the location network node in method 100 of Fig. 6.

[0120] In one or more example methods, communicating S202 comprises receiving S202B, from the location network node, a message indicative of a plurality of channel response report formats available to be used by the positioning measurement node. In other words, the message may comprise information indicative of the plurality of channel response report formats available to be used by the positioning measurement node. Receiving S202B corresponds to the transmitting S102B performed by the location network node in method 100 of Fig. 6. In other words, the message received by in S202B corresponds to the message transmitted in S102B. In one or more example methods, the method comprises receiving S203, from the location network node, a request for a channel response report for a channel between a WD and a radio network node. Receiving S203 corresponds to the sending S103 performed by the location network node in method 100 of Fig. 6.

[0121] In one or more example methods, the method comprises receiving S204, from the location network node, a request for a label report indicative of a state of the WD. Receiving S204 corresponds to the sending S104 performed by the location network node in method 100 of Fig. 6.

[0122] In one or more example methods, the method comprises measuring S205 a set of channel response measurement samples. In one or more example methods, the set of channel response measurement samples are measured on a reference signal received from a transmission point, such as DL-PRS when the positioning measurement node is a WD, or UL- SRS when the positioning measurement node is a radio network node.

[0123] In one or more example methods, the method comprises determining S206 a subset of the set of channel response measurement samples to report to the location network node. The subset of the channel response measurement samples to report may be determined based on the message indicative of a plurality of channel response report formats available to be used by the positioning measurement node, such as based on the information indicative of a subset of channel response measurement samples. In one or more example methods, the subset of samples reported may be determined by the size of a time window, such as a moving time window. In one or more example methods, the positioning measurement node reports a window time stamp Twinand a reduced size channel response report having a size of for example 18 x Nwinx 3. In one or more example methods, the positioning measurement node may choose the size reduction percentage by selecting a corresponding Nwin. For instance, the channel response report, such as the CIR report, package size can be reduced by 90%, if Nwin= 26. In other words, by reducing the number of samples from the reserved data size of 256 to the time window of 26, the package size of the channel response report can be reduced by 256 / Nwbl%. The raw CIR measurement report can hereby be considered as a special case where Tw,„ = 0 and Nwir)= 256.

[0124] In one or more example methods, determining S206 comprises determining S206A the subset by detecting a maximum summation of power within a predetermined time window of samples, such as based on a size of the predetermined time window. To reduce the signaling overhead, the positioning measurement node may report a subset of samples corresponding to a range of interest for the location network node. The range of interest may be a narrower range than the range measured by the positioning measurement node. To find this narrower range in the channel response, the positioning measurement node may deploy a moving search window in a time domain with the size of Nwinwhere Nwin< Nsamp, as illustrated in Fig. 3. The search window may traverse from a first delay sample to the last delay sample, such as window time stamp from 1 to 256. In the meantime, the positioning measurement node may continuously calculate a summation of a power level of the measured reference signal within the time window. By identifying a timestamp at which the time window provides a maximum summation of power, the positioning measurement node can identify a channel of interest. By using this approach, the positioning measurement node only has to report the window time stamp Twinand a subset of the channel response with sample numberNwm fromthe window time stamp, which reduces the signaling overhead of the channel response report.

[0125] The method 200 comprises transmitting S207, to the location network node, a channel response report, such as a CIR report, for the channel between the WD and the radio network node using the communicated channel response report format, such as the channel response format communicated in S202, and / or S202, and / or S202B. The channel response report transmitted in S207 corresponds to the channel response report received by the location network node in S106 of method 100 of Fig. 6, 2010 of Fig. 5 and is similar to 1007 of Fig. 4.

[0126] In one or more example methods, the channel response report comprises information indicative of a subset of channel response measurement samples, such as a number Nwinof the available number of samples Nsamp. The subset may be a subset of an entire set of samples measured by the positioning measurement node. In one or more example methods, the subset comprises one or more of time information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response.

[0127] In one or more example methods, the channel response report comprises a time stamp, such as time stamp Twin, indicative of a timing of the first time parameter, and / or power parameter, and / or phase parameter in the subset of channel measurement samples. The timing of the first time parameter may be indicated as an absolute timing or relative to a reference timing.

[0128] In one or more example methods, the channel response report comprises a size, such as information indicative of the size, of a time window of the reported subset of channel measurement samples. In one or more example methods, the method comprises transmitting S208, to the location network node, a label report indicative of the state of the WD, wherein the label report is associated with a channel response report of a channel between the WD and the radio network node. In other words, in one or more example methods, the positioning measurement node may correspond to the label generating node. The state of the WD may comprise information indicative of a current state, such as situation, of the WD. The information indicative of the state of the WD may comprise one or more of an actual position of the WD, such as its ground truth coordinate, an orientation of the WD, LOS / NLOS information, and an actual delay, such as a timing delay of the WD. Transmitting S208 corresponds to receiving S108 performed by the location network node in method 100 of Fig. 6, 2006 of Fig. 5 and is similar to 1007 of Fig. 4.

[0129] The label types, such as the information comprised in the label report, may differ based on the Al model’s functionality. In one or more example methods, the label report comprises information indicative of ground truth (GT) coordinates of the WD, such as of a WD for which positioning training data is to be collected. The GT coordinates may be coordinates in a Global Cartesian System (GCS). In one or more example methods, such as for direct AI / ML-based positioning, the label generating node, such as the WD, provides its ground truth coordinate to the location network node. In one or more example methods, only one GT coordinate for the WD is provided in the label report for each channel response report.

[0130] In one or more example methods, the label report comprises a Line-of Sight (LOS), or non-Line- of-Sight (NLOS) channel indication, such as information indicative of whether a channel between the WD and a radio network node is LOS or NLOS. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning, where the label generation node provides information specific to each link between the WD and a radio network node.

[0131] In one or more example methods, the label report comprises information indicative of a time delay on the channel between the WD and a radio network node. The time delay may be a time delay of the LOS path on the channel between the WD and the radio network node. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning.

[0132] In one or more example methods, the label report comprises information indicative of a range, such as a LOS distance, of the channel between the WD and a radio network node. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning. In one or more example methods, the label report comprises information indicative of a timing error. The timing error may be a timing error at the positioning measurement node and / or at the transmission point transmitting reference signals for positioning. In other words, the timing error may be a WD timing error and / or a radio network node timing error. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning.

[0133] In one or more example methods, the label report comprises information indicative of an association between the label report and the channel response report. In one or more example methods, the label report is transmitted from the WD and received by the location network node as soon as the WD transmits reference signals for positioning, such as SRS. In this case, the association between the label report and the channel response report is implicitly provided by the WD by informing the location network node about the association between the label report and the SRS transmission, such as resources of the SRS transmission.

[0134] In one or more example methods, the information indicative of the association between the label report and the channel response report may be explicitly provided in the label report. In one or more example methods, the label report may comprise information indicative of the positioning resources used for the channel response measurement, such as an SRS resource ID or an PRS resource ID, associated with the label report.

[0135] In one or more example methods, the label report comprises information indicative of a label generation source, such as the label generation node.

[0136] In one or more example methods, the label generating node may provide information indicative of the source of label generation in the label report. In one or more example methods, the label report may comprise a dedicated IE, such as an IE referred to as ‘ LabelSourcelndex’ where an index value may correspond to a specific label generation source.

[0137] In one or more example methods, the label report comprises information indicative of a label quality, such as a quality of the indicated position of the WD. In one or more example methods, the information indicative of the label quality may be explicitly indicated in the label report, for example as an uncertainty of the WD coordinate in an actual distance, such as in meters.

[0138] In one or more example methods, the label report comprises information indicative of a time stamp, such as a label time stamp. The time stamp, such as the label time stamp, may be a time stamp when a WD receives DL-PRS for performing channels response measurements, or a time stamp when the WD transmits UL SRS for positioning used by one or more radio network nodes to measure a channel response. In one or more example methods, the label report comprises information indicative of a time window, such as a label time window. The time window, such as the label time window, may be indicative of a duration of label validity, such as a maximum time that the label report is valid. In case the label generation node detects that there are changes related to reported label information, then the label generating node can transmit a new label report before the time window has expired.

[0139] The information comprised in the label report may be dependent on the type of label generating mode that is generating the label report. The label report may comprise one or more of the types of information disclosed herein. In other words, the label report may be reported, such as may comprise information, based on the information indicative of the label generating source, such as the label generating node.

[0140] In one or more example methods, such as when data is collected for training of an Al model for assisted AI / ML positioning, the label generating node is the radio network node.

[0141] In one or more example methods, such as when data is collected for training of an Al model for direct AI / ML-based positioning, the label generating node is a WD.

[0142] Fig. 8 shows a flow diagram of an example method 900, performed by a label generating node according to the disclosure, for enabling training of an Al model for positioning of wireless devices. The label generating node is the label generating node disclosed herein, such as the label generating node 800 of Fig. 11 , such as the wireless device 300, 300A of Fig. 1 , Figs. 2A- 2B, Fig. 4, Fig. 5, or the radio network node 400 of Fig. 1 , Fig. 2A-2B, Fig. 4 and Fig. 5.

[0143] In one or more example methods, the method comprises transmitting S901 , to the location network node, information indicative of the label generating node’s capability for collecting data, such as Al training data, such as a capability of providing label information for training of the Al model. In one or more example methods, the information indicative of the label generating node’s capability of collecting data comprises information indicative of a capability of providing label information associated with a positioning resource and / or a channel response measurement. In other words, the label generating node may indicate its capability to associate label information, such as information indictive of a WDs actual position to a channel response report. Transmitting S201 is similar to receiving S101 performed by the location network node in method 100 of Fig. 6. In one or more example methods, transmitting S901 is performed in response to receiving a request for capability of performing data collection, such as capability of providing label information, from the location network node. In one or more example methods, the method comprises receiving S903, from the location network node, a request for a label report indicative of a state of the WD. Receiving S903 corresponds to the sending S104 performed by the location network node in method 100 of Fig. 6. In one or more example methods, receiving S903 comprises receiving S903A the request for the label report from, such as via, a radio network node. Receiving S903 corresponds to sending S104 performed by the location network node in method 100.

[0144] In one or more example methods, the method comprises transmitting S906, to the location network node, a label report indicative of the position of the WD, wherein the label report is associated with a channel response report of a channel between the WD and the radio network node, and / or associated with a positioning resource, such as a resource used for transmitting reference signals on which the channel response is measured. In other words, in one or more example methods, the positioning measurement node may correspond to the label generating node. Transmitting S906 corresponds to receiving S108 performed by the location network node in method 100 of Fig. 6 and 2006 of Fig. 5, and is similar to transmitting S108 performed by the positioning measurement node in method 200.

[0145] The label types, such as the information comprised in the label report, may differ based on the Al model’s functionality. In one or more example methods, the label report comprises information indicative of GT coordinates of the WD, such as of a WD for which positioning training data is to be collected. The GT coordinates may be coordinates in the GCS. In one or more example methods, such as for direct AI / ML-based positioning, the label generating node, such as the WD, provides its ground truth coordinate to the location network node. In one or more example methods, only one GT coordinate for the WD is provided in the label report for each channel response report.

[0146] In one or more example methods, the label report comprises a LOS, or NLOS channel indication, such as information indicative of whether a channel between the WD and a radio network node is LOS or NLOS. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning, where the label generation node provides information specific to each link between the WD and a radio network node.

[0147] In one or more example methods, the label report comprises information indicative of a time delay on the channel between the WD and a radio network node. The time delay may be a time delay of the LOS path on the channel between the WD and the radio network node. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning. In one or more example methods, the label report comprises information indicative of a range, such as a LOS distance, of the channel between the WD and a radio network node. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning.

[0148] In one or more example methods, the label report comprises information indicative of a timing error. The timing error may be a timing error at the positioning measurement node and / or at the transmission point transmitting reference signals for positioning. In other words, the timing error may be a WD timing error and / or a radio network node timing error. In one or more example methods, this information may be provided for training of an Al model for assisted AI / ML positioning.

[0149] In one or more example methods, the label report comprises information indicative of an association between the label report and the channel response report. In one or more example methods, the label report is transmitted from the WD and received by the location network node as soon as the WD transmits reference signals for positioning, such as SRS. In this case, the association between the label report and the channel response report is implicitly provided by the WD by informing the location network node about the association between the label report and the SRS transmission, such as resources of the SRS transmission.

[0150] In one or more example methods, the information indicative of the association between the label report and the channel response report may be explicitly provided in the label report. In one or more example methods, the label report may comprise information indicative of the positioning resources used for the channel response measurement, such as an SRS resource ID or an PRS resource ID, associated with the label report.

[0151] In one or more example methods, the label report comprises information indicative of a label generation source, such as the label generation node. In one or more example methods, the label generating node may provide information indicative of the source of label generation in the label report. In one or more example methods, the label report may comprise a dedicated IE, such as an IE referred to as ‘ Labe / Source Index’ where an index value may correspond to a specific label generation source.

[0152] In one or more example methods, the label report comprises information indicative of a label quality, such as a quality of the indicated position of the WD. In one or more example methods, the information indicative of the label quality may be explicitly indicated in the label report, for example as an uncertainty of the WD coordinate in an actual distance, such as in meters. In one or more examples, the quality represents a quality of the estimated measurement results (such as of a timing error estimation, a time delay estimation, a range estimation, etc.) associated with the label report.

[0153] In one or more example methods, the label report comprises information indicative of a time stamp, such as a label time stamp. The time stamp, such as the label time stamp, may be a time stamp when a WD receives DL-PRS for performing channels response measurements, or a time stamp when the WD transmits UL SRS for positioning used by one or more radio network nodes to measure a channel response.

[0154] In one or more example methods, the label report comprises information indicative of a time window, such as a label time window. The time window, such as the label time window, may be indicative of a duration of label validity, such as a maximum time that the label report is valid. In case the label generation node detects that there are changes related to reported label information, then the label generating node can transmit a new label report before the time window has expired.

[0155] The information comprised in the label report may be dependent on the type of label generating mode that is generating the label report. The label report may comprise one or more of the information disclosed herein. In other words, the label report may be reported, such as may comprise information, based on the information indicative of the label generating source, such as the label generating node.

[0156] In one or more example methods, such as when data is collected for training of an Al model for assisted AI / ML positioning, the label generating node is a radio network node.

[0157] In one or more example methods, such as when data is collected for training of an Al model for direct AI / ML-based positioning, the label generating node is a WD.

[0158] Fig. 9 shows a block diagram of an example location network node 600 according to the disclosure. The location network node 600 comprises memory circuitry 601 , processor circuitry 602, and an interface 603, such as a wired and / or wireless interface. The location network node 600 may be configured to perform any of the methods disclosed in Fig. 6. In other words, the location network node 600 may be configured for enabling training of an Al model for positioning of wireless devices.

[0159] The location network node 600 is configured to communicate with a positioning measurement node, such as the positioning measurement node 700, and / or with a label generating node, such as label generating node 800, disclosed herein, using a wireless communication system. In one or more examples, the location network node may be configured to communicate with the positioning measurement node and / or the label generating node.

[0160] The interface 603 is configured for communication via a wireless communication system, such as a 3GPP system, such as a 3GPP system supporting one or more of: Beyond 5G system, New Radio, NR, Long Term Evolution, LTE, Narrow-band loT, NB-loT, and Long Term Evolution - enhanced Machine Type Communication, LTE-M, and 3GPP system operated in licensed bands or unlicensed bands.

[0161] The location network node 600 is configured to communicate, for example via the interface 603, with a positioning measurement node, a message indicative of a channel response report format for a channel between the WD and a radio network node.

[0162] The location network node 600 is configured to receive, for example via the interface 603, from the positioning measurement node, a channel response report for the channel between the WD and the radio network node using the communicated channel response report format.

[0163] The location network node 600 is configured to receive, for example via the interface 603, from a label generating node, a label report indicative of the position of the WD wherein the label report is associated with the channel response report.

[0164] Processor circuitry 602 is optionally configured to perform any of the operations disclosed in Fig. 6 (such as any one or more of S101 , S102, S102A, S102B, S103, S104, S105, S106, S108, S110). The operations of the location network node 600 may be embodied in the form of executable logic routines (for example, lines of code, software programs, etc.) that are stored on a non-transitory computer readable medium (for example, memory circuitry 601 ) and are executed by processor circuitry 602).

[0165] Furthermore, the operations of the location network node 600 may be considered a method that the location network node 600 is configured to carry out. Also, while the described functions and operations may be implemented in software, such functionality may also be carried out via dedicated hardware or firmware, or some combination of hardware, firmware and / or software.

[0166] Memory circuitry 601 may be one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, a random access memory (RAM), or other suitable device. In a typical arrangement, memory circuitry 601 may include a nonvolatile memory for long term data storage and a volatile memory that functions as system memory for processor circuitry 602. Memory circuitry 601 may exchange data with processor circuitry 602 over a data bus. Control lines and an address bus between memory circuitry 601 and processor circuitry 602 also may be present (not shown in Fig. 9). Memory circuitry 601 is considered a non-transitory computer readable medium.

[0167] Memory circuitry 601 may be configured to store information, such as information indicative of a channel response, information indicative of a channel response report format, information indicative of a label report, and / or information indicative of the positioning measurement node’s capability for collecting Al training data, in a part of the memory.

[0168] Fig. 10 shows a block diagram of an example positioning measurement node 700 according to the disclosure. For DL-based data collection, such as DL-based positioning, the positioning measurement node 700 may be a WD, such as the WD 300, 300A of Fig. 1 , Figs. 2A-2B and Fig. 4. For UL-based data collection, such as UL-based positioning, the positioning measurement node 700 may be a radio network node, such as the radio network node 400 of Fig. 1 , Figs. 2A-2B and Fig. 5. The positioning measurement node may herein also be referred to as a data collecting node, such as a data collecting node for collecting input data to an Al model for positioning of WDs. The positioning measurement node 700 comprises memory circuitry 701 , processor circuitry 702, and an interface 703, such as a wired interface and / or wireless interface. The positioning measurement node 700 may be configured to perform any of the methods disclosed in Fig. 7. In other words, the positioning measurement node 700 may be configured for enabling training of an Al model for positioning of wireless devices.

[0169] The positioning measurement node 700 is configured to communicate with a location network node, such as the location network node 600 disclosed herein, and / or a transmission point transmitting reference signals for positioning, using a wireless communication system.

[0170] The positioning measurement node 700 is configured to communicate (such as via the interface 703), with the location network node, a message indicative of a channel response report format for the channel between a WD and a radio network node.

[0171] The positioning measurement node 700 is configured to transmit (such as via the interface 703), to the location network node, a channel response report for the channel between the WD and the radio network node using the communicated channel response report format.

[0172] The interface 703 is configured for communications via a wireless communication system, such as a 3GPP system, such as a 3GPP system supporting one or more of: Beyond 5G system, New Radio, NR, Long Term Evolution, LTE, Narrow-band loT, NB-loT, and Long Term Evolution - enhanced Machine Type Communication, LTE-M, and 3GPP system operated in licensed bands or unlicensed bands. The positioning measurement node 700 is optionally configured to perform any of the operations disclosed in Fig. 7 (such as any one or more of S201 , S202, S202A, S202B, S203, S204, S205, S206, S206A, S207, S208). The operations of the positioning measurement node 700 may be embodied in the form of executable logic routines (for example, lines of code, software programs, etc.) that are stored on a non-transitory computer readable medium (for example, memory circuitry 701) and are executed by processor circuitry 702).

[0173] Furthermore, the operations of the positioning measurement node 700 may be considered a method that the positioning measurement node 700 is configured to carry out. Also, while the described functions and operations may be implemented in software, such functionality may also be carried out via dedicated hardware or firmware, or some combination of hardware, firmware and / or software.

[0174] Memory circuitry 701 may be one or more of a buffer, a flash memory, a hard drive, a removable media, a volatile memory, a non-volatile memory, a random access memory (RAM), or other suitable device. In a typical arrangement, memory circuitry 701 may include a nonvolatile memory for long term data storage and a volatile memory that functions as system memory for processor circuitry 702. Memory circuitry 701 may exchange data with processor circuitry 702 over a data bus. Control lines and an address bus between memory circuitry 701 and processor circuitry 702 also may be present (not shown in Fig. 10). Memory circuitry 701 is considered a non-transitory computer readable medium.

[0175] Memory circuitry 701 may be configured to store information (such as information indicative of a channel response, information indicative of a channel response report format, information indicative of a label report, and / or information indicative of the positioning measurement node’s capability for collecting Al training data) in a part of the memory.

[0176] In this disclosure, the positioning measurement node’s capability may be seen as a capability of the positioning measurement node, such as a capability for collecting Al training data, and / or e.g. for providing label information for training of the Al model.

[0177] Examples of methods and products (location network node, positioning measurement node, and label generating node) according to the disclosure are set out in the following items:

[0178] Item 1 . A method performed by a location network node, for enabling training of an Artificial Intelligence, Al, model for positioning of wireless devices, the method comprising: receiving (S108), from a label generating node, a label report indicative of the position of the WD wherein the label report is associated with the channel response report and / or one or more positioning resource.

[0179] Item 2. The method according to Item 1 , wherein the method comprises: communicating (S102), with a positioning measurement node, a message indicative of a channel response report format for a channel between a WD and a radio network node, and receiving (S106), from the positioning measurement node, a channel response report for the channel between the WD and the radio network node using the communicated channel response report format.

[0180] Item 3. The method according to Item 2, wherein communicating (S102) comprises: receiving (S102A), from the positioning measurement node, a message indicative of the channel response report format used by the positioning measurement node.

[0181] Item 4. The method according to Item 2 or 3, wherein communicating (S102) comprises: transmitting (S102B), to the positioning measurement node, a message indicative of a channel response report format to be used by the positioning measurement node.

[0182] Item 5. The method according to any one of the Items 2 to 4, wherein the channel response report format is indicative of one or more of: a number of radio network nodes to report a channel response from, a number of samples to be reported for each transmission point, a channel information type to report, a granularity of the channel information, and an index associated with the channel response report format.

[0183] Item 6. The method according to Item 5, wherein the channel information type comprises one or more of: time information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response. Item 7. The method according to any one of the Items 2 to 6, wherein the channel response report comprises one or more of: a subset of channel response measurement samples, wherein the subset is a subset of the entire set of samples measured by the positioning measurement node, the subset comprising one or more of: information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response, a time stamp indicative of a timing of the first time, power and / or phase parameter in the subset of channel measurement samples, and a size of a time window of the reported subset of channel measurement samples.

[0184] Item 8. The method according to any one of the previous Items, wherein the method comprises: sending (S103), to the positioning measurement node, a request for a channel response report for a channel between a WD and a radio network node, and sending (S104), to the label generating node, a request for a label report indicative of a state of the WD.

[0185] Item 9. The method according to any one of the Items 2 to 8, wherein the channel response report is associated to one or more of the positioning measurement.

[0186] Item 10. The method according to any one of the previous Items, wherein the label report comprises one or more of: information indicative of ground truth coordinates of the WD, information indicative of whether a channel between the WD and a radio network node is Line-of Sight, LOS, or non-Line-of-Sight, NLOS, information indicative of a time delay on the channel between the WD and a radio network node, information indicative of a range of the channel between the WD and a radio network node, information indicative of a timing error. Item 11. The method according to any one of the previous Items, wherein the label report comprises one or more of: information indicative of an association between the label report and the channel response report, information indicative of a label generation source, information indicative of a label quality, information indicative of a label time stamp, and information indicative of a label time window.

[0187] Item 12. The method according to Item 11 , wherein the label quality is associated to an uncertainty in obtaining the position of the WD indicated in the label report.

[0188] Item 13. The method according to Items 11 or 12, wherein the label report is reported based on the information indicative of the label generation node.

[0189] Item 14. The method according to any one of the previous Items, wherein the label report is associated to one or more channel response reports transmitted by respective positioning measurement nodes.

[0190] Item 15. The method according to any one of the previous Items, wherein the method comprises: receiving (S101 ), from the positioning measurement node, information indicative of the positioning measurement node’s capability for collecting Al training data, and / or from the label generating node, information indicative of the label generating node’s capability of providing label information for training of the Al model.

[0191] Item 16. The method according to any one of the Items 2-15, wherein the positioning measurement node is a WD.

[0192] Item 17. The method according to any one of the Items 2-15, wherein the positioning measurement node is a radio network node.

[0193] Item 18. The method according to any one of the previous Items, wherein the label generating node is the same node as the positioning measurement node. Item 19. The method according to any one of the Items 1-17, wherein the label generating node is a different node than the positioning measurement node.

[0194] Item 20. A method performed by a positioning measurement node, for enabling training of an Artificial Intelligence, Al, model for positioning of wireless devices, the method comprising: communicating (S202), with the location network node, a message indicative of a channel response report format for the channel between a WD and a radio network node, and transmitting (S207), to the location network node, a channel response report for the channel between the WD and the radio network node using the communicated channel response report format.

[0195] Item 21. The method according to Item 20, wherein the channel response report comprises one or more of: a subset of channel response measurement samples, wherein the subset is a subset of the entire set of samples measured by the positioning measurement node, the subset comprising one or more of information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response, a time stamp indicative of a timing of the first time, power and / or phase parameter in the subset of channel measurement samples, and a size of a time window of the reported subset of channel measurement samples.

[0196] Item 22. The method according to Item 20 or 21 , wherein the method comprises: measuring (S205) a set of channel response measurement samples, and determining (S206) a subset of the set of channel response measurement samples to report to the location network node.

[0197] Item 23. The method according to Item 22, wherein determining (S206) comprises determining (S206A) the subset by detecting a maximum summation of power within a predetermined time window of samples. Item 24. The method according to any one of the Items 16-19, wherein communicating (S202) comprises: transmitting (S202A), to the location network node, a message indicative of the channel response report format used by the positioning measurement node.

[0198] Item 25. The method according to any one of the Items 20 to 24, wherein communicating (S202) comprises: receiving (S202B), from the location network node, a message indicative of a plurality of channel response report formats available to be used by the positioning measurement node.

[0199] Item 26. The method according to any one of the Items 20 to 25, wherein the channel response report format is indicative of one or more of: a number of transmission points to report a channel response from, a number of samples to be reported for each transmission point, a channel information type to report, a granularity of the channel information, and an index associated with the channel response report format.

[0200] Item 27. The method according to Item 26, wherein the channel information type comprises one or more of: time information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response.

[0201] Item 28. The method according to any one of the Items 20 to 27, wherein the method comprises: receiving (S203), from the location network node, a request for a channel response report for a channel between a WD and a radio network node.

[0202] Item 29. The method according to any one of the Items 20 to 28, wherein the method comprises: transmitting (S208), to the location network node, a label report indicative of the position of the WD, wherein the label report is associated with a channel response report of a channel between the WD and the radio network node. Item 30. The method according to Item 29, wherein the method comprises: receiving (S204), from the location network node, a request for a label report indicative of a state of the WD.

[0203] Item 31. The method according to any one of the Items 29 or 30, wherein the label report comprises one or more of: information indicative of ground truth coordinates of the WD to be positioned, information indicative of whether a channel between the WD and a radio network node is Line-of Sight, LOS, or non-Line-of-Sight, NLOS, information indicative of a delay on the channel between the WD and a radio network node, information indicative of a range of the channel between the WD and a radio network node, information indicative of a timing error.

[0204] Item 32. The method according to any one of the Items 30 to 31 , wherein the label report comprises one or more of: information indicative of an association between the label report and the channel response report, information indicative of a label generation source, information indicative of a label quality, information indicative of a label time stamp, and information indicative of a label time window.

[0205] Item 33. The method according to any one of the Items 20 to 32, wherein the method comprises: transmitting (S201), to the location network node, information indicative of the positioning measurement node’s capability for collecting Al training data. Item 34. The method according to any one of the Items 20 to 33, wherein the positioning measurement node is the radio network node.

[0206] Item 35. The method according to any one of the Items 20 to 33, wherein the positioning measurement node is the WD.

[0207] Item 36. A method performed by a label generating node, for enabling training of an Artificial Intelligence, Al, model for positioning of wireless devices, the method comprising: transmitting (S906), to the location network node, a label report indicative of the position of the WD wherein the label report is associated with one or more positioning resources.

[0208] Item 37. The method according to Item 36, wherein the label report is associated with a channel response report of a channel between the WD and the radio network node.

[0209] Item 38. The method according to Item 36 or 37, wherein the method comprises: receiving (S903), from the location network node, a request for a label report indicative of a state of the WD.

[0210] Item 39. The method according to any one of the Items 36 to 38, wherein the method comprises: receiving (S903A), from a radio network node, a request to provide a label report to location network node.

[0211] Item 40. The method according to any one of the Items 36 to 39, wherein the label report comprises one or more of: information indicative of ground truth coordinates of the WD to be positioned, information indicative of whether a channel between the WD and a radio network node is Line-of Sight, LOS, or non-Line-of-Sight, NLOS, information indicative of a time delay on the channel between the WD and a radio network node, information indicative of a range of the channel between the WD and a radio network node, information indicative of a timing error.

[0212] Item 41 . The method according to any one of the Items 36 to 40, wherein the label report comprises one or more of: information indicative of an association between the label report and one or more positioning measurement nodes, information indicative of the identification of the associated positioning measurement nodes, information indicative of a label generation source, and information indicative of a label quality information indicative of a label time stamp, and information indicative of a label time window.

[0213] Item 42. The method according to Item 41 , wherein the label quality is associated to an uncertainty in obtaining the position of the WD indicated in the label report.

[0214] Item 43. The method according to Items 41 or 42, wherein the label report is reported based on the information indicative of the label generation node.

[0215] Item 44. The method according to any one of the Items 36 to 43, wherein the label report is associated to one or more channel response reports transmitted by respective positioning measurement nodes.

[0216] Item 45. The method according to any one of the Items 36 to 44, wherein the method comprises: transmitting (S901 ), to the location network node, information indicative of the positioning measurement node’s capability for providing label information for training of the Al model.

[0217] Item 46. A location network node comprising memory circuitry, processor circuitry, and a wireless interface, wherein the location network node is configured to perform any of the methods according to any of Items 1-19.

[0218] Item 47. A positioning measurement node comprising memory circuitry, processor circuitry, and a wireless interface, wherein the positioning measurement node is configured to perform any of the methods according to any of Items 20-35. Item 48. A label generating node comprising memory circuitry, processor circuitry, and a wireless interface, wherein the wireless device is configured to perform any of the methods according to any of Items 36-45.

[0219] The use of the terms “first”, “second”, “third” and “fourth”, “primary”, “secondary”, “tertiary” etc. does not imply any particular order, but are included to identify individual elements. Moreover, the use of the terms “first”, “second”, “third” and “fourth”, “primary”, “secondary”, “tertiary” etc. does not denote any order or importance, but rather the terms “first”, “second”, “third” and “fourth”, “primary”, “secondary”, “tertiary” etc. are used to distinguish one element from another. Note that the words “first”, “second”, “third” and “fourth”, “primary”, “secondary”, “tertiary” etc. are used here and elsewhere for labelling purposes only and are not intended to denote any specific spatial or temporal ordering. Furthermore, the labelling of a first element does not imply the presence of a second element and vice versa.

[0220] It may be appreciated that Figures 1-9 comprise some circuitries or operations which are illustrated with a solid line and some circuitries, components, features, or operations which are illustrated with a dashed line. Circuitries or operations which are comprised in a solid line are circuitries, components, features or operations which are comprised in the broadest example. Circuitries, components, features, or operations which are comprised in a dashed line are examples which may be comprised in, or a part of, or are further circuitries, components, features, or operations which may be taken in addition to circuitries, components, features, or operations of the solid line examples. It should be appreciated that these operations need not be performed in order presented. Furthermore, it should be appreciated that not all of the operations need to be performed. The example operations may be performed in any order and in any combination. It should be appreciated that these operations need not be performed in order presented. Circuitries, components, features, or operations which are comprised in a dashed line may be considered optional.

[0221] Other operations that are not described herein can be incorporated in the example operations. For example, one or more additional operations can be performed before, after, simultaneously, or between any of the described operations.

[0222] Certain features discussed above as separate implementations can also be implemented in combination as a single implementation. Conversely, features described as a single implementation can also be implemented in multiple implementations separately or in any suitable sub-combination. Moreover, although features may be described above as acting in certain combinations, one or more features from a claimed combination can, in some cases, be excised from the combination, and the combination may be claimed as any sub-combination or variation of any sub-combination

[0223] It is to be noted that the word "comprising" does not necessarily exclude the presence of other elements or steps than those listed.

[0224] It is to be noted that the words "a" or "an" preceding an element do not exclude the presence of a plurality of such elements.

[0225] It is to be noted that the term "indicative of may be seen as “associated with”, “related to”, “descriptive of’, “characterizing”, and / or “defining”. The terms “indicative of”, “associated with”, “related to”, “descriptive of’, “characterizing”, and “defining” can be used interchangeably. The term “indicative of” can be seen as indicating a relation. For example, weight data indicative of weight may comprise one or more weight parameters.

[0226] It is to be noted that the word "based on" may be seen as “as a function of” and / or “derived from”. The terms “based on” and “as a function of’ can be used interchangeably. For example, a parameter determined “based on” a data set can be seen as a parameter determined “as a function of” the data set. In other words, the parameter may be an output of one or more functions with the data set as an input.

[0227] A function may be characterizing a relation between an input and an output, such as mathematical relation, a database relation, a hardware relation, logical relation, and / or other suitable relations.

[0228] It should further be noted that any reference signs do not limit the scope of the claims, that the examples may be implemented at least in part by means of both hardware and software, and that several "means", "units" or "devices" may be represented by the same item of hardware.

[0229] Language of degree used herein, such as the terms “approximately,” “about,” “generally,” and “substantially” as used herein represent a value, amount, or characteristic close to the stated value, amount, or characteristic that still performs a desired function or achieves a desired result. For example, the terms “approximately”, “about”, “generally,” and “substantially” may refer to an amount that is within less than or equal to 10% of, within less than or equal to 5% of, within less than or equal to 1% of, within less than or equal to 0.1% of, and within less than or equal to 0.01% of the stated amount. If the stated amount is 0 (e.g., none, having no), the above recited ranges can be specific ranges, and not within a particular % of the value. For example, within less than or equal to 10 wt. / vol. % of, within less than or equal to 5 wt. / vol. % of, within less than or equal to 1 wt. / vol. % of, within less than or equal to 0.1 wt. / vol. % of, and within less than or equal to 0.01 wt. / vol. % of the stated amount.

[0230] The various example methods, devices, nodes and systems described herein are described in the general context of method steps or processes, which may be implemented in one aspect by a computer program product, embodied in a computer-readable medium, including computerexecutable instructions, such as program code, executed by computers in networked environments. A computer-readable medium may include removable and non-removable storage devices including, but not limited to, Read Only Memory (ROM), Random Access Memory (RAM), compact discs (CDs), digital versatile discs (DVD), etc. Generally, program circuitries may include routines, programs, objects, components, data structures, etc. that perform specified tasks or implement specific abstract data types. Computer-executable instructions, associated data structures, and program circuitries represent examples of program code for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represents examples of corresponding acts for implementing the functions described in such steps or processes.

[0231] Although features have been shown and described, it will be understood that they are not intended to limit the claimed disclosure, and it will be made obvious to those skilled in the art that various changes and modifications may be made without departing from the scope of the claimed disclosure. The specification and drawings are, accordingly, to be regarded in an illustrative rather than restrictive sense. The claimed disclosure is intended to cover all alternatives, modifications, and equivalents.

Claims

CLAIMS1 . A method performed by a location network node, for enabling training of an Artificial Intelligence, Al, model for positioning of wireless devices, the method comprising: communicating (S102), with a positioning measurement node, a message indicative of a channel response report format for a channel between a WD and a radio network node, receiving (S106), from the positioning measurement node, a channel response report for the channel between the WD and the radio network node using the communicated channel response report format, and receiving (S108), from a label generating node, a label report indicative of the position of the WD wherein the label report is associated with the channel response report.

2. The method according to claim 1 , wherein communicating (S102) comprises: receiving (S102A), from the positioning measurement node, a message indicative of the channel response report format used by the positioning measurement node, or transmitting (S102B), to the positioning measurement node, a message indicative of a channel response report format to be used by the positioning measurement node.

3. The method according to claim 1 or 2, wherein the channel response report format is indicative of one or more of: a number of radio network nodes to report a channel response from, a number of samples to be reported for each transmission point, a channel information type to report, a granularity of the channel information, and an index associated with the channel response report format.

4. The method according to claim 3, wherein the channel information type comprises one or more of: time information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response.

5. The method according to any one of the previous claims, wherein the channel response report comprises one or more of: a subset of channel response measurement samples, wherein the subset is a subset of the entire set of samples measured by the positioning measurement node, the subset comprising one or more of: information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response, a time stamp indicative of a timing of the first time, power and / or phase parameter in the subset of channel measurement samples, and a size of a time window of the reported subset of channel measurement samples.

6. The method according to any one of the previous claims, wherein the method comprises: sending (S103), to the positioning measurement node, a request for a channel response report for a channel between a WD and a radio network node, and sending (S104), to the label generating node, a request for a label report indicative of a state of the WD.

7. The method according to any one of the previous claims, wherein the channel response report is associated to one or more of the positioning measurements.

8. The method according to any one of the previous claims, wherein the label report comprises one or more of: information indicative of ground truth coordinates of the WD, information indicative of whether a channel between the WD and a radio network node is Line-of Sight, LOS, or non-Line-of-Sight, NLOS, information indicative of a time delay on the channel between the WD and a radio network node, information indicative of a range of the channel between the WD and a radio network node, information indicative of a timing error,information indicative of an association between the label report and the channel response report, information indicative of a label generation source, information indicative of a label quality, information indicative of a label time stamp, and information indicative of a label time window.

9. The method according to claim 8, wherein the label quality is associated to an uncertainty in obtaining the position of the WD indicated in the label report.

10. The method according to any of claims 8 -9, wherein the label report is reported based on the information indicative of the label generation node.11 . The method according to any one of the previous claims, wherein the label report is associated to one or more channel response reports transmitted by respective positioning measurement nodes.

12. The method according to any one of the previous claims, wherein the method comprises: receiving (S101 ), from the positioning measurement node, information indicative of the positioning measurement node’s capability for collecting Al training data, and / or from the label generating node, information indicative of the label generating node’s capability of providing label information for training of the Al model.

13. The method according to any one of the previous claims, wherein the positioning measurement node is a WD.

14. The method according to any one of claims 1-12, wherein the positioning measurement node is a radio network node.

15. The method according to any one of the previous claims, wherein the label generating node is the same node as the positioning measurement node.

16. The method according to any one of the claims 1-14, wherein the label generating node is a different node than the positioning measurement node.

17. A method performed by a positioning measurement node, for enabling training of an Artificial Intelligence, Al, model for positioning of wireless devices, the method comprising: communicating (S202), with the location network node, a message indicative of a channel response report format for the channel between a WD and a radio network node, and transmitting (S207), to the location network node, a channel response report for the channel between the WD and the radio network node using the communicated channel response report format.

18. The method according to claim 17, wherein the channel response report comprises one or more of: a subset of channel response measurement samples, wherein the subset is a subset of the entire set of samples measured by the positioning measurement node, the subset comprising one or more of information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response, a time stamp indicative of a timing of the first time, power and / or phase parameter in the subset of channel measurement samples, and a size of a time window of the reported subset of channel measurement samples.

19. The method according to claim 17 or 18, wherein the method comprises: measuring (S205) a set of channel response measurement samples, and determining (S206) a subset of the set of channel response measurement samples to report to the location network node.

20. The method according to claim 19, wherein determining (S206) comprises determining (S206A) the subset by detecting a maximum summation of power within a predetermined time window of samples.

21. The method according to any one of the claims 17-20, wherein communicating (S202) comprises:transmitting (S202A), to the location network node, a message indicative of the channel response report format used by the positioning measurement node.

22. The method according to any one of the claims 17 to 21 , wherein communicating (S202) comprises: receiving (S202B), from the location network node, a message indicative of a plurality of channel response report formats available to be used by the positioning measurement node.

23. The method according to any one of the claims 17 to 22, wherein the channel response report format is indicative of one or more of: a number of transmission points to report a channel response from, a number of samples to be reported for each transmission point, a channel information type to report, a granularity of the channel information, and an index associated with the channel response report format.

24. The method according to claim 23, wherein the channel information type comprises one or more of: time information associated with the channel response, power information associated with the channel response, and phase information associated with the channel response.

25. A method performed by a label generating node, for enabling training of an Artificial Intelligence, Al, model for positioning of wireless devices, the method comprising: transmitting (S906), to the location network node, a label report indicative of the position of the WD wherein the label report is associated with one or more positioning resources.

Images