Data augmentation for machine learning training in positioning

Abstract

An example embodiment of the present disclosure relates to data augmentation for training a model. In this solution, a first device receives a configuration for data augmentation from a second device. The first device determines a data augmentation parameter set based on the configuration. The first device trains a model based on data acquired based on the data augmentation parameter set. In this way, the first device increases the amount of quality training data, enabling robust model training. Furthermore, the first device can evaluate the completeness of the augmented data set by using a probability value for each data point or data segment.

Description

Technical Field

[0001] Various embodiments of the present disclosure generally relate to the field of telecommunications, and more specifically, to methods, devices, apparatuses, and computer-readable storage media for data augmentation for machine learning training in positioning.

Background Art

[0002] In the telecommunications industry, techniques for improving the performance of telecommunications systems have been proposed. For example, in telecommunications systems, artificial intelligence / machine learning (AI / ML) models have been adopted to improve the performance of telecommunications systems. Therefore, it is valuable to consider the training of AI / ML models adopted in telecommunications systems.

Summary of the Invention

[0003] In a first aspect of the present disclosure, a first device is provided. The first device includes at least one processor and at least one memory storing instructions, which, when executed by the at least one processor, cause the first device to at least receive, from a second device, a configuration for data augmentation for training a positioning model in the first device; determine a data augmentation parameter set based on the configuration for data augmentation; obtain data including measurement data and augmented data of the first device based on the data augmentation procedure and the data augmentation parameter set; and train the positioning model based on a combination of the measurement data and the augmented data.

[0004] A second aspect of the present disclosure provides a second device, the second device comprising at least one processor and at least one memory for storing instructions, the instructions causing the second device to at least perform the following actions when executed by the at least one processor: receiving radio measurements from at least one third device; determining a configuration for a data augmentation to train a positioning model in a first device based on the measurements from the at least one third device; and transmitting the data augmentation configuration to the first device.

[0005] A third aspect of the present disclosure provides a first device, the first device comprising at least one processor and at least one memory for storing instructions, the instructions causing the first device to at least: send an assistance request for data augmentation to a second device, receive augmentation data from the second device for training a positioning model, and train a positioning model based on a combination of augmentation data and measurement data, when executed by the at least one processor.

[0006] A fourth aspect of the present disclosure provides a second device, the second device comprising at least one processor and at least one memory for storing instructions, the instructions causing the second device to at least perform the following actions when executed by the at least one processor: receiving a request for assistance for data augmentation from the first device; acquiring measurement data from the third device; generating augmentation data for training a positioning model in the first device based on the measurement data; and transmitting the augmentation data to the first device.

[0007] A fifth aspect of this disclosure provides a method, which includes receiving a data augmentation configuration from a second device for training a positioning model in a first device; determining a data augmentation parameter set based on the data augmentation configuration; acquiring data including measurement data and augmentation data from the first device based on the data augmentation procedure and data augmentation parameter set; and training a positioning model based on the combination of measurement data and augmentation data.

[0008] A sixth aspect of the present disclosure provides a method, which includes receiving radio measurements from at least one third device, determining a configuration for a data augmentation to train a positioning model in a first device based on the measurements from at least one third device, and transmitting the data augmentation configuration to the first device.

[0009] A seventh aspect of this disclosure provides a method, which includes sending a request for assistance for data augmentation to a second device, receiving augmentation data from the second device for training a positioning model, and training the positioning model based on a combination of the augmentation data and measurement data.

[0010] An eighth aspect of this disclosure provides a method, which includes receiving a request for assistance for data augmentation from a first device, acquiring measurement data from a third device, generating augmentation data for training a positioning model in the first device based on the measurement data, and transmitting the augmentation data to the first device.

[0011] A ninth aspect of the present disclosure provides an apparatus, which includes means for receiving a data augmentation configuration from a second device for training a positioning model in a first device; means for determining a data augmentation parameter set based on the data augmentation configuration; means for acquiring data including measurement data and augmentation data from the first device based on the data augmentation procedure and the data augmentation parameter set; and means for training a positioning model based on a combination of measurement data and augmentation data.

[0012] A tenth aspect of the present disclosure provides an apparatus, which includes means for receiving radio measurements from at least one third device, means for determining a configuration of a data augmentation for training a positioning model in a first device based on the measurements from at least one third device, and means for transmitting the data augmentation configuration to the first device.

[0013] An eleventh aspect of this disclosure provides an apparatus, which includes means for transmitting a request for assistance for data augmentation to a second device; means for receiving augmentation data from the second device for training a positioning model; and means for training a positioning model based on a combination of augmentation data and measurement data.

[0014] A twelfth aspect of this disclosure provides an apparatus, which includes means for receiving assistance requests for data augmentation from a first device; means for acquiring measurement data from a third device; means for determining augmentation data for training a positioning model in the first device based on the measurement data; and means for transmitting the augmentation data to the first device.

[0015] Item 13 of this disclosure provides a computer-readable medium that stores instructions causing a device to perform a method according to at least one of the fifth, sixth, seventh, or eighth aspects.

[0016] This summary should be understood as not intended to identify any key or fundamental features of the embodiments of the Disclosure, nor to be used to limit the scope of the Disclosure. Other features of the Disclosure will be readily apparent through the following description.

[0017] Next, several embodiments will be described with reference to the attached drawings. [Brief explanation of the drawing]

[0018] [Figure 1] This figure shows an example of a communication environment in which the embodiments of this disclosure can be implemented. [Figure 2] This is a schematic diagram of data collection according to some embodiments of the present disclosure. [Figure 3] This is a schematic diagram of a structure for model training with data augmentation, according to some embodiments of the present disclosure. [Figure 4] These are signaling diagrams for communication according to some embodiments of the present disclosure. [Figure 5] This is a schematic diagram of further measurements in the measurement requirements according to some embodiments of the present disclosure. [Figure 6A] These are signaling diagrams for communication according to some embodiments of the present disclosure. [Figure 6B] These are signaling diagrams for communication according to some embodiments of the present disclosure. [Figure 7] These are signaling diagrams for communication according to some embodiments of the present disclosure. [Figure 8] This figure shows a flowchart of a method according to some embodiments of the present disclosure. [Figure 9] This figure shows a flowchart of a method according to some embodiments of the present disclosure. [Figure 10] This figure shows a flowchart of a method according to some embodiments of the present disclosure. [Figure 11] This figure shows a flowchart of a method according to some embodiments of the present disclosure. [Figure 12] It is a simplified block diagram of a device suitable for implementing an example embodiment of the present disclosure. [Figure 13] It is a block diagram of an example of a computer-readable medium according to some example embodiments of the present disclosure.

Best Mode for Carrying Out the Invention

[0019] Throughout the drawings, the same or similar reference numerals represent the same or similar elements.

[0020] Hereinafter, the principles of the present disclosure will be described with reference to some example embodiments. It should be understood that these embodiments are merely illustrative and are useful for those skilled in the art to understand and implement the present disclosure, and do not imply any limitation on the scope of the present disclosure. The embodiments described in this specification can also be implemented in various ways other than those described below.

[0021] In the following description and claims, unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by those skilled in the art to which the present disclosure pertains.

[0022] References to "one embodiment", "an embodiment", "an example embodiment", etc. in the present disclosure indicate that the described embodiment can include specific features, structures, or characteristics, but not all embodiments need to include these specific features, structures, or characteristics. Further, such expressions do not necessarily refer to the same embodiment. Furthermore, when describing specific features, structures, or characteristics in relation to one embodiment, it is considered within the knowledge of those skilled in the art to affect such features, structures, or characteristics in relation to other embodiments, whether or not it is clearly described.

[0023] Furthermore, while this specification may use terms such as “first” and “second” to describe various elements, it should be understood that these elements should not be limited by these terms. These terms are merely used to distinguish one element from another. For example, without departing from the scope of the embodiments, the first element may be called the second element, and similarly, the second element may be called the first element. The term “and / or” as used herein includes any combination of one or more of the terms listed.

[0024] In this specification, "at least one of the following: <list of two or more elements>" is used. )" and "at least one of " The expressions ")" and similar expressions in which a list of two or more elements is connected by "and" or "or" mean at least one of these elements, at least two or three or more of these elements, or at least all of these elements.

[0025] Unless otherwise expressly stated, the “in response to A” step as used herein does not mean that the step is performed immediately after the occurrence of “A,” and may include one or more intervening steps.

[0026] The terms used herein are for the sole purpose of describing specific embodiments and are not intended to limit the examples of embodiments. The singular forms “one” and “it” used herein are intended to include the plural form unless otherwise explicitly indicated in the context. Furthermore, the terms “comprises, comprising, has, having, includes, and / or including” used herein specify the presence of the described features, elements, and / or components, but are not intended to exclude the presence or addition of one or more other features, elements, components, and / or combinations thereof.

[0027] As used in this application, the term "circuitry" may mean one, two or more, or all of the following: (a) Circuit implementation using only hardware (such as implementation using only analog and / or digital circuits), and, (b) (If applicable) The following combinations of hardware circuits and software, (i) combinations of analog and / or digital hardware circuits and software / firmware, (ii) Any part of a (one or multiple) hardware processor having software (including (one or multiple) digital signal processors, software, and (one or multiple) memory that work together to cause a device such as a mobile phone or server to perform various functions), (c) Hardware circuits (one or multiple) that require software (e.g., firmware) to operate, but which may not be present when the software is not required for operation, and / or processors (one or multiple) such as a microprocessor or a part of a microprocessor.

[0028] This definition of circuit applies to all uses of this term in this application, including in the claims. Further examples include, as used in this application, the term circuit encompasses only hardware circuits or processors (or more processors), or parts of hardware circuits or processors and their associated software and / or firmware implementations. The term circuit also encompasses, for example, baseband integrated circuits or processor integrated circuits for mobile devices, or similar integrated circuits in servers, cellular network devices, or other computer or network devices, where applicable to the elements of a particular claim.

[0029] In this specification, the term "communication network" means a network conforming to one of the following preferred communication standards: New Radio (NR), Long Term Evolution (LTE), LTE-Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High-Speed ​​Packet Access (HSPA), and Narrow Band Internet of Things (NB-IoT). Furthermore, communication between terminal devices and network devices in a communication network may be carried out in accordance with any preferred generation of communication protocol, including, but not limited to, first-generation (1G), second-generation (2G), 2.5G, 2.75G, third-generation (3G), fourth-generation (4G), 4.5G, fifth-generation (5G) communication protocols, and / or any other protocols currently known or to be developed in the future. Embodiments of this disclosure can be applied to a variety of communication systems. Given the rapid development of communications, it is natural that future communication technologies and systems capable of realizing this disclosure will also exist. The scope of this disclosure should not be considered to be limited solely to the systems described above.

[0030] As used herein, the term “network device” means a node in a communications network from which a terminal device accesses the network and receives services. Depending on the terminology and technology applied, a network device may also mean a base station (BS) or access point (AP), such as a node B (NodeB or NB), an evolved node B (eNodeB or eNB), an NR NB (also called a gNB), a remote radio unit (RRU), a radio header (RH), a remote radio head (RRH), a relay, an Integrated Access and Backhaul (IAB) node, a low-power node such as a femto or pico, a satellite network device, a non-terrestrial network (NTN) or non-terrestrial network device such as a low-earth orbit (LEO) satellite or a geosynchronous earth orbit (GEO) satellite, and an aircraft network device. In some embodiments, the radio access network (RAN) split architecture includes a centralized unit (CU) and a distributed unit (DU) in the IAB donor node. The IAB node includes a mobile terminal (IAB-MT) portion that behaves like a UE toward the parent node, while the DU portion of the IAB node behaves like a base station toward the next-hop IAB node.

[0031] The term "terminal device" refers to any end device capable of wireless communication. While not an exhaustive list, terminal devices may also be called communication devices, user equipment (UE), subscriber stations (SS), portable subscriber stations, mobile stations (MS), or access terminals (AT). Terminal devices include, but are not limited to, mobile phones, cellular phones, smartphones, voice over IP (VoIP) phones, wireless local loop phones, tablets, wearable devices, personal digital assistants (PDAs), portable computers, desktop computers, image capture devices such as digital cameras, game devices, music storage and playback devices, in-vehicle wireless terminal devices, wireless endpoints, mobile stations, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), USB dongles, smart devices, wireless customer-premises equipment (CPE), Internet of Things (IoT) devices, watches or other wearables, head-mounted displays (HMDs), vehicles, drones, medical devices and applications (e.g., remote surgery), industrial devices and applications (e.g., robots and / or other wireless devices operating in the context of industrial chains and / or automated processing chains), and consumer electronics. Examples include devices, as well as devices operating on commercial and / or industrial wireless networks. Terminal devices can also correspond to the mobile termination (MT) portion of IAB nodes (e.g., relay nodes). In the following description, the terms “terminal device,” “communication device,” “terminal,” “user equipment,” and “UE” may be used synonymously.

[0032] As used herein, the terms “resource,” “transmit resource,” “resource block,” “physical resource block (PRB),” “uplink resource,” or “downlink resource” may mean any resource for performing communication, such as communication between a terminal device and a network device, including resources in the time domain, resources in the frequency domain, resources in the spatial domain, resources in the code domain, or any other resource that enables communication. Hereinafter, unless otherwise specified, resources in both the frequency domain and the time domain are used as examples of transmit resources to describe some embodiments of this disclosure. The embodiments of this disclosure are similarly applicable to other resources in other domains.

[0033] As mentioned above, AI / ML can be employed in communication systems. For example, positioning accuracy can be improved by using AI / ML. Furthermore, explanations regarding data collection for training mainly focus on whether data can be collected from the entire simulation area or from a series of grids located within the simulation area. The basic premise seems to be that positioning data is readily available within the system. However, this premise is only valid for the simulation environment and cannot be applied to real-world environments where only a limited amount of data is available.

[0034] Assuming that supervised AI / ML models are run on the UE side, a certain amount (generally a large amount) of labeled data may be required for AI / ML training to enable AI / ML inference to be performed with a certain level of accuracy. The UE alone may take a long time to collect all the necessary data. In this case, the network can support the UE by sending additional labeled data, however, this comes at the cost of additional signaling.

[0035] Supervised learning is a crucial technique for extracting value from big data. However, the effectiveness of supervised learning requires a large amount of high-quality training data. Often, the size of the training data is not large enough to effectively train a supervised learning classifier. Data augmentation is a widely adopted approach to increase the amount of training data. However, the quality of augmented data is questionable and may depend on the use case.

[0036] With the large-scale deployment of 5G cellular infrastructure, traffic forecasting is becoming an essential part of cellular resource management systems to provide reliable and high-speed communication services that can meet increasingly demanding quality of service requirements. A promising approach to address this problem is to introduce intelligent methods that use AI / ML models to perform highly effective and efficient radio resource management procedures. On the other hand, integrating a multiaccess edge computing framework into 5G cellular networks facilitates the application of intelligent traffic forecasting models by enabling these implementations at the network edge. However, data scarcity and privacy issues can still be obstacles to training robust and accurate forecasting models at the edge. Therefore, it is worthwhile to consider training AI / ML models for use in telecommunications systems.

[0037] Environment example Figure 1 shows an example of a communication environment 100 that can implement an embodiment of the present disclosure. The communication environment 100 includes devices 110-1, 110-2, 110-3, ..., and 110-N, which can be collectively called "device 110". The communication environment also includes devices 120 and 130. Devices 110, 120, and 130 (one or more of them) can communicate with each other.

[0038] In the example in Figure 1, device 110 may include a terminal device, and device 130 may include a network device that provides services to the terminal device. Device 120 may include a core network device. For example, device 120 may include a device that can implement a location management function (LMF).

[0039] It should be understood that the number of devices and their connections shown in Figure 1 are for illustrative purposes only and not to imply limitation. The communication environment 100 may include any preferred number of devices configured to implement an embodiment of the present disclosure. Although not shown, one or more additional devices may exist within the cell of device 130, and one or more additional cells may be deployed within the communication environment 100. Although device 130 is shown as a network device, it may be a device other than a network device. Although device 110 is shown as a terminal device, it may be a device other than a terminal device.

[0040] In the following, for illustrative purposes, several examples of embodiments will be described in which device 110 operates as a terminal device and device 130 operates as a network device. However, in some examples of embodiments, the operations described in relation to a terminal device can be performed on a network device or other device, and the operations described in relation to a network device can be performed on a terminal device or other device.

[0041] In some embodiments, when device 110 is a terminal device and device 130 is a network device, the link from device 130 to device 110 is called a downlink (DL), and the link from device 110 to device 130 is called an uplink (UL). In a DL, device 130 is a transmitting (TX) device (or transmitter) and device 110 is a receiving (RX) device (or receiver). In a UL, device 110 is a TX device (or transmitter) and device 130 is an RX device (or receiver).

[0042] Communication in communication environment 100 may be carried out in accordance with any appropriate (one or more) communication protocol, including, but not limited to, cellular communication protocols such as first generation (1G), second generation (2G), third generation (3G), fourth generation (4G), fifth generation (5G), and sixth generation (6G), wireless local network communication protocols such as Institute for Electrical and Electronics Engineers (IEEE) 802.11, and / or other protocols currently known or to be developed in the future. Furthermore, communication may utilize any suitable wireless communication technology, including, but not limited to, Code Division Multiple Access (CDMA), Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Frequency Division Duplex (FDD), Time Division Duplex (TDD), Multiple-Input Multiple-Output (MIMO), Orthogonal Frequency Division Multiple (OFDM), Discrete Fourier Transform spread OFDM (DFT-s-OFDM), and / or any other technology currently known or to be developed in the future.

[0043] Operating principles and signaling examples for communication According to some embodiments of this disclosure, a solution for data augmentation for training AI / ML models is provided. In this solution, a first device receives a configuration for data augmentation from a second device. The first device determines a data augmentation parameter set based on the configuration. The first device trains the model based on a combination of measured data and augmented data acquired based on the data augmentation parameter set. In this way, the first device increases the amount of good quality training data and enables robust model training. Furthermore, the first device can assess the completeness of the augmented dataset by using probability values ​​for each data point or data segment.

[0044] Figure 2 is a schematic diagram of data acquisition examples according to some embodiments of the present disclosure. Device 110-1 can collect a series of measurements. For example, device 110-1 can acquire measurement data 210 as shown in Figure 2. The size of the collected data may not be sufficient to perform AI / ML model training. In this case, data augmentation can be applied. As used herein, the term “data augmentation” can mean a technique to increase the amount of data within a region of interest (ROI). Furthermore, spatial interpolation techniques can be applied to derive additional samples, which represent estimated radio measurements for missing locations. In this case, device 110-1 can use data augmentation to acquire augmented data 220 for missing locations. Furthermore, the efficiency of this data augmentation technique depends on tuning the spatial interpolation function and relates to selecting appropriate parameters that can be performed with network assistance. Thus, the amount of data, including the measurement data 210 and augmented data 220, becomes sufficient to train a model.

[0045] Figure 3 is a schematic diagram of an example structure 300 for model training with data augmentation, according to some embodiments of the present disclosure. Structure 300 can be implemented in the device 110 shown in Figure 1. As shown in Figure 3, in some embodiments, after data augmentation is performed using spatial interpolation to reach a target data size, a labeled dataset module 310 can add an indication that the data is interpolated or measured, so that AI / ML model training can further consider this information. The spatial interpolation method can also provide a level of data augmentation and associate a probability value for each predicted value or each augmented data segment. This value can indicate, for example, that the augmented data is likely to be equal to or less accurate than the measured data. In this way, the AI / ML model can prioritize measured / real data over interpolated or augmented data when minimizing training errors.

[0046] The embodiments of this disclosure will be described in detail below with reference to the attached drawings.

[0047] Next, we refer to signaling diagram 4 for interactions 400 according to some embodiments of the present disclosure. As shown in Figure 4, this signaling diagram shows interactions 400 between devices 410, 420 and 430. For illustrative purposes, we refer to Figure 1 to illustrate signaling diagram 200. For example, device 410 may represent or include device 110-1 shown in Figure 1, and device 430 may represent or include device 110-2 shown in Figure 1. Device 420 may represent or include device 120 shown in Figure 1. Note that devices 410, 420 and 430 may represent or include any suitable device.

[0048] Device 410 can report its capabilities to device 420. For example, the capabilities may include one or more supported models of device 410. Alternatively, or in addition to this, the capabilities may include the memory resources of device 410. In some other embodiments, the capabilities may include the computing capabilities of device 410. The capabilities may also include the computing resources in device 410. In some embodiments, the capabilities can be reported via long-term positioning protocol (LPP) messages.

[0049] In some embodiments, device 410 may proactively initiate data acquisition requests (e.g., Layer 1 (L1) measurements) to train a model in device 410. For example, device 410 may send an assistance request for data augmentation to device 420 (4005). In some embodiments, the assistance request may indicate the percentage of available data relative to the target data size used in training the positioning model. As used herein, the term “positioning model” may mean a processing model or AIML model used to position a device. In some embodiments, the assistance request may be sent along with the capabilities of device 410. For example, the assistance request and the capabilities to be reported may be included in an LPP message.

[0050] In some embodiments, the support request may include an instruction on the percentage of currently available data relative to the target data size for model training. For example, if the support request indicates 75%, device 420 can understand that 75% of the target data size has been collected and 25% still needs to be collected, i.e., 25% of the data needs to be interpolated to reach the target data size for training. In this way, measurement collection and model training can be accelerated.

[0051] Alternatively, the support request may not include an indication of the percentage of currently available data. In this case, device 420 can determine the percentage of data required that device 410 requests. In some embodiments, device 420 can determine the required data based on feedback regarding data augmentation performance received in device 420. Alternatively, it may determine the required data without feedback.

[0052] In some embodiments, device 410 can send a request for assistance to device 420 when it newly joins the network of device 420. Alternatively, device 410 can send a request for assistance to device 420 when it determines that it cannot train a model with the required accuracy using the current dataset. In this case, in some embodiments, the required accuracy can be based on a key performance indicator (KPI) of positioning accuracy. Alternatively, or in addition to this, the required accuracy can be based on one or more intermediate KPIs related to the intermediate model used for line-of-sight (LOS) or non-LOS (NLOS) classification. In some other embodiments, the required accuracy can be based on the positioning latency of device 410.

[0053] One or more devices can transmit their radio measurements to device 420. As shown in Figure 4, device 430 transmits its radio measurements to device 420 (4010). One or more devices can be located within the area designated by device 410. For example, device 430 can be located within the area surrounding device 410. In other words, device 430 can be a neighboring UE of device 410.

[0054] In some embodiments, the radio measurement may represent the reference signal received power (RSRP) measured by device 430. Alternatively, or in addition to this, the radio measurement may represent channel state information. In some other embodiments, the radio measurement may represent a channel response, such as the channel impulse response (CIR). Alternatively, or in addition to this, the radio measurement may represent one or more of the angle of arrival (AoA), angle of departure (AoD), time difference of arrival (TDoA), or round trip time (RTT).

[0055] Device 420 determines the configuration of data augmentation for training a positioning model based on radio measurements (4015). In some embodiments, device 420 can determine a data augmentation parameter set based on radio measurements. For example, an optimal list of data augmentation parameters can be estimated using radio measurements or feedback from other devices. As just one example, the data augmentation parameter set may include one or more parameters related to an interpolation function used to predict the data.

[0056] Device 420 transmits the data augmentation configuration to device 410 (4020). In some embodiments, the configuration may include a set of data augmentation parameters determined by device 420. Alternatively, the configuration may include radio measurements from device 430. For example, if device 420 determines that device 430 is located within the vicinity of device 410 based on its rough location, the data augmentation configuration may include radio measurements from device 430. As just one example, neighbor measurement data 530 from device 430 can be provided to device 410, as shown in Figure 5. In some other embodiments, the data augmentation configuration may include instructions on whether the data for training the positioning model is interpolated or measured. The data augmentation configuration may also indicate the (e.g., temporal or spatial) dimensions of the data for training the positioning model.

[0057] In some embodiments, device 430 can transmit wireless measurements directly to device 410 (4025). For example, wireless measurements can be transmitted via a sidelink between device 410 and device 430.

[0058] In some embodiments, device 420 can proactively transmit data augmentation configurations. In other words, the transmission of data augmentation configurations may not be based on assistance requests.

[0059] In some embodiments, the data augmentation configuration can be transmitted aperiodically. Figure 6A shows a signaling diagram for an interaction 600 according to some embodiments of the present disclosure. As shown in Figure 6A, this signaling diagram shows an interaction 600 between device 410 and device 420. Device 420 can monitor the performance of device 410 (6010). If device 420 detects or observes a performance degradation of device 410, it can refine or update the data augmentation parameter set (6020). For example, device 420 can refine the data augmentation parameter set if it observes that the positional uncertainty of device 410 is systematically high. Device 420 can transmit the data augmentation configuration to device 410 (6030). In this case, the data augmentation configuration may further indicate the validity period of the data augmentation configuration. For example, this new parameterization may have a validity period, i.e., the maximum duration for which device 420 considers the configuration valid. This validity period can be expressed as the number of subframes during which the configuration is valid, starting from the moment the message is received on device 410.

[0060] In some embodiments, the configuration of the data extension can be transmitted periodically. Figure 6B shows a signaling diagram for an interaction 601 according to some embodiments of the present disclosure. As shown in Figure 6B, this signaling diagram shows an interaction 601 between device 410 and device 420. Device 420 can refine or update the data extension parameter set (6120). Device 420 can transmit the configuration of the data extension to device 410 (6130). In this case, the configuration of the data extension may further indicate the validity period of the configuration of the data extension. For example, this new parameterization may be accompanied by a validity period, i.e., the maximum duration for which device 420 considers the configuration to be valid. This validity period can be expressed as the number of subframes for which the configuration is valid, starting from the time the message is received on the device 410 side.

[0061] Returning to Figure 4, device 410 determines the data augmentation parameter set based on the data augmentation configuration (4030). For example, in some embodiments, if the data augmentation configuration includes a data augmentation parameter set, device 410 can extract the data augmentation parameter set from the data augmentation configuration.

[0062] Alternatively, if the data augmentation configuration includes radio measurements from device 430, device 410 can determine the data augmentation parameter set based on the radio measurements from device 430. For example, device 410 can adjust the data augmentation parameter set based on the radio measurements. In some embodiments, the data augmentation parameter set (e.g., spatial interpolation parameters) can be adjusted by minimizing the error between the interpolated data and the data reported from device 430.

[0063] In some embodiments, as described above, wireless measurements can be transmitted via a sidelink between devices 410 and 430. In this case, device 410 can determine a data augmentation parameter set based on the wireless measurements from device 430.

[0064] Device 410 acquires data for training the positioning model based on a data augmentation procedure and a data augmentation parameter set (4035). In other words, device 410 can perform data augmentation using the configuration of data augmentation from device 420. For example, device 410 can select a kriging procedure as the data augmentation procedure. The data augmentation procedure can be any suitable procedure that can increase the amount of training data. The data for training the positioning model includes measurement data from device 410 and augmented data generated based on the data augmentation parameter set. Alternatively, the data for training the positioning model may also include radio measurement data from device 430. For example, as shown in Figure 5, the data for training the positioning model may include measurement data 510, augmented data 520, and radio measurement data 530. The augmented data 520 can estimate locations not measured by device 410. In this way, the amount of high-quality training data can be increased.

[0065] Device 410 trains a positioning model based on a combination of measured data and augmented data (4040). For example, the combination of measured data and augmented data may indicate that data at a particular location is measured, and data at another particular location is interpolated (i.e., augmented). In some embodiments, the combination of measured data and augmented data may be based on the weights of these data. For example, measured data may have a higher weight than augmented data. This enables robust model training. Furthermore, it is also possible to assess the completeness of the augmented dataset by using probability values ​​for each data point or data segment. For example, if a large data segment is only predicted, location estimation may not be used for critical applications based on the associated probability values.

[0066] As just one example of spatial interpolation, the kriging method can be used. Kriging assumes that missing values ​​can be estimated using weighted linear combinations of available neighboring values. The calculation of kriging weights can be based on the relationships between observations expressed through a spatial autocovariance function.

[0067] In some embodiments, device 410 can transmit feedback information to device 420 indicating the current data augmentation performance (4045). For example, this feedback information may indicate the interpolation accuracy achieved in device 410. In this case, device 420 can update the data augmentation configuration based on the feedback information (4050). For example, device 420 can expand the area from which it collects wireless measurements from neighboring devices. As just one example, if the achieved interpolation accuracy indicated in the feedback information falls below a threshold, device 420 can expand the area from which it collects wireless measurements.

[0068] Examples of embodiments of this disclosure propose methods and procedures targeting data augmentation for AI / ML-based positioning using AI / ML inference performed on the UE side. For example, this disclosure proposes a method that enables an increase in the amount of raw labeled data, such as radio measurements and corresponding geographic locations. Further signaling can ensure efficient data augmentation operation on the UE with network assistance. Regarding data augmentation methods, examples of supervised learning methods performed on the UE side can be considered as either estimating features useful for estimating locality, such as LOS / NLOS classification, or directly estimating locality.

[0069] Herein, we refer to Figure 7, which shows a signaling diagram for an interaction 700 according to some embodiments of the present disclosure. As shown in Figure 7, this signaling diagram shows the interaction 700 between devices 710, 720, and 730. For illustrative purposes, we refer to Figure 1 to illustrate the signaling diagram 700. For example, device 710 may represent or include device 110-1 shown in Figure 1, and device 730 may represent or include device 110-2 or device 130 shown in Figure 1. Device 720 may represent or include device 120 shown in Figure 1. Note that devices 710, 720, and 730 may represent or include any suitable device.

[0070] Device 710 can report its capabilities to device 720. For example, the capabilities may include one or more supported models of device 710. Alternatively, or in addition to this, the capabilities may also include the memory resources of device 710. In some other embodiments, the capabilities may include the computing capabilities of device 710. The capabilities may also include the computing resources of device 710. In some embodiments, the capabilities can be reported via Long-Term Positioning Protocol (LPP) messages.

[0071] In some embodiments, device 710 may proactively initiate data acquisition requests (e.g., Layer 1 (L1) measurements) to train a model in device 710. For example, device 710 may send an assistance request for data augmentation to device 720 (7005). In some embodiments, the assistance request may indicate the percentage of available data relative to the target data size used in training the positioning model. As used herein, the term “positioning model” may mean a processing model or AIML model used to position the device. In some embodiments, the assistance request may be sent along with the capabilities of device 710. For example, the assistance request and the capabilities to be reported may be included in an LPP message.

[0072] In some embodiments, the support request may include an instruction for the percentage of currently available data relative to the target data size for model training. For example, if the support request indicates 75%, device 420 can understand that 75% of the target data size has been collected and 25% still needs to be collected. In this way, measurement acquisition and model training can be accelerated.

[0073] Alternatively, the support request may not include an indication of the percentage of currently available data. In this case, device 720 can determine the percentage of data required from device 410 based on the data augmentation performance feedback received at device 720.

[0074] In some embodiments, device 710 can send a request for assistance to device 720 when it newly joins the network of device 720. Alternatively, device 710 can send a request for assistance to device 720 when it determines that it cannot train the model with the required accuracy using the current dataset. In this case, in some embodiments, the required accuracy can be based on key performance indicators (KPIs) of positioning accuracy. Alternatively, or in addition to this, the required accuracy can be based on one or more intermediate KPIs relating to the intermediate model used for line-of-sight (LOS) or non-line-of-sight (NLOS) classification. In some other embodiments, the required accuracy can be based on the positioning latency of device 410.

[0075] One or more devices can transmit their radio measurements to device 720. As shown in Figure 7, device 730 transmits its radio measurements to device 720 (7010). In some embodiments, device 430 may be located within the vicinity of device 410. In other words, device 430 may be a neighboring UE of device 410. In some embodiments, the radio measurement may represent the reference signal received power (RSRP) measured by device 730. In addition to or instead of this, the radio measurement may also represent channel state information. In some other embodiments, the radio measurement may represent a channel response, such as channel impulse response (CIR). In addition to or instead of this, the radio measurement may also represent one or more of the angle of arrival (AoA), angle of departure (AoD), time difference of arrival (TDoA), or round-trip time (RTT).

[0076] Alternatively, device 730 may be a network device associated with device 710. In this case, data acquisition is performed by network devices participating in the positioning process of device 710 (i.e., device 730), namely the serving gNB and adjacent gNBs. The gNBs can specifically collect measurements of the UL sounding reference signal (SRS) (or UL SRS for positioning: SRS-P) emitted by device 710 while device 710 is present at a designated location. The gNBs can then provide such measurements to device 720, and other devices (e.g., device 110-2) can provide device 730 with the location from which the SRS was transmitted. In this way, the SRS can be applied to DL positioning, UL positioning, and UL+DL positioning.

[0077] Device 720 determines data for training the positioning model based on radio measurements (7015). In other words, device 720 can perform data augmentation. For example, device 720 may select a kriging procedure as the data augmentation procedure. The data augmentation procedure can be any appropriate procedure that can increase the amount of training data. The data for training the positioning model includes measurement data and augmented data from device 730. Alternatively, the data for training the positioning model may also include radio measurement data from device 730.

[0078] Device 720 sends data to device 710 (7020) to train the positioning model. For example, the data can be sent in LPP messages.

[0079] Device 710 trains a positioning model based on a combination of measured data and augmented data (7025). For example, the combination of measured data and augmented data may indicate that data at a particular location is measured and data at another particular location is interpolated (i.e., augmented). In some embodiments, the combination of measured data and augmented data may be based on the weights of these data. For example, measured data may have a higher weight than augmented data. This enables robust model training. Furthermore, it is possible to assess the completeness of the augmented dataset by using probability values ​​for each data point or data segment. For example, if a large data segment is only predicted, location estimation may not be used for critical applications based on the associated probability values.

[0080] Example of method Figure 8 shows a flowchart of an example method 800 implemented in or performed by a first device, according to several embodiments of the present disclosure. For illustrative purposes, method 800 will be described in terms of device 110 in Figure 1.

[0081] In block 810, the first device receives a configuration for data augmentation for training a positioning model in the first device from the second device. In some embodiments, the first device may send an assistance request for data augmentation to the second device. For example, the assistance request may further indicate the ratio of available data to the target data size used in training the positioning model. In some embodiments, the data augmentation configuration may include instructions on whether the data for training the positioning model is interpolated or measured.

[0082] In block 820, the first device determines the data augmentation parameter set based on the data augmentation configuration. In some embodiments, the data augmentation configuration may include a data augmentation parameter set based on the wireless measurement values ​​of the third device. In this case, the first device can extract the data augmentation parameter set from the data augmentation configuration.

[0083] Alternatively, the data augmentation configuration may include radio measurements from a third device. In this case, the first device can determine the data augmentation parameter set based on the radio measurements from the third device.

[0084] In other embodiments, the first device can receive wireless measurements from the third device via a sidelink between the first and third devices. In this case, the first device can determine a data augmentation parameter set based on the wireless measurements from the third device.

[0085] In block 830, the first device acquires data based on a data augmentation procedure and a data augmentation parameter set. The data includes measurement data and augmented data from the first device.

[0086] In block 840, the first device trains a positioning model based on a combination of measurement data and augmented data. For example, the combination of measurement data and augmented data may indicate that data at a particular location is measured and data at another particular location is interpolated (i.e., augmented). In some embodiments, the combination of measurement data and augmented data may be based on the weights of these data. For example, the measurement data may have a higher weight than the augmented data. This enables robust model training. In some embodiments, the first device may transmit feedback information to the second device indicating the current data augmentation performance in the first device.

[0087] Figure 9 shows a flowchart of a method example 900 implemented in or performed by a second device, according to some embodiments of the present disclosure. For illustrative purposes, method 900 will be described in terms of device 120 in Figure 1.

[0088] In block 910, the second device receives wireless measurements from the third device.

[0089] In block 920, the second device determines the configuration of the data augmentation for training the positioning model in the first device based on measurements from the third device. In some embodiments, the data augmentation configuration may further indicate the lifespan of the data augmentation configuration.

[0090] In block 930, the second device transmits the data augmentation configuration to the first device. In some embodiments, the second device may determine the data augmentation parameter set based on radio measurements. In this case, the data augmentation configuration may include the data augmentation parameter set. Alternatively, the data augmentation configuration may include radio measurements from the third device.

[0091] In some embodiments, a second device may receive a request for assistance from the first device for data augmentation. For example, the request may further indicate the ratio of available data to the target data size used in training the positioning model.

[0092] In some embodiments, a second device can monitor the performance of a first device. In this case, if a performance degradation of the first device is detected, the second device can send a data enhancement configuration to the first device. Alternatively, the second device can send a data enhancement configuration periodically.

[0093] In some embodiments, a second device can receive feedback information from the first device indicating the current data augmentation performance of the first device. In this case, the second device can update the data augmentation configuration based on the feedback information.

[0094] Figure 10 shows a flowchart of Method Example 1000, which is implemented in or performed by the First Device, according to some embodiments of the present disclosure. For illustrative purposes, Method 1000 will be described from the perspective of Device 110 in Figure 1.

[0095] In block 1010, the first device sends a request for assistance for data augmentation to the second device. In some embodiments, the request for assistance may further indicate the ratio of available data to the target data size used in training the positioning model.

[0096] In block 1020, the first device receives augmented data from the second device for training a positioning model.

[0097] In block 1030, the first device trains a positioning model based on a combination of augmented data and measured data. For example, the combination of measured data and augmented data may indicate that data at a particular location is measured, and data at another particular location is interpolated (i.e., augmented). In some embodiments, the combination of measured data and augmented data may be based on the weights of these data. For example, the measured data may have a higher weight than the augmented data. In this way, robust model training becomes possible.

[0098] Figure 11 shows a flowchart of an example method 1100 implemented in or performed by a second device, according to some embodiments of the present disclosure. For illustrative purposes, method 1100 will be described from the perspective of device 120 in Figure 1.

[0099] In block 1110, the second device receives a request for assistance for data augmentation from the first device. In some embodiments, the request for assistance may further indicate the ratio of available data to the target data size used in training the positioning model.

[0100] In block 1120, the second device acquires measurement data from the third device. In some embodiments, the measurement data may include at least one of uplink measurement data or downlink measurement data.

[0101] In block 1130, the second device generates augmented data based on the measurement data to train the positioning model in the first device.

[0102] In block 1140, the second device transmits extended data to the first device.

[0103] Examples of devices, apparatus, and media In some embodiments, a first apparatus (e.g., device 110 in Figure 1) capable of performing any of the methods 800 may include means for performing each operation of the methods 800. The means can be implemented in any preferred form. For example, the means can be implemented in a circuit or a software module.

[0104] In one embodiment, the first device includes means for receiving a data augmentation configuration from a second device for training a positioning model in a first device; means for determining a data augmentation parameter set based on the data augmentation configuration; means for acquiring data including measurement data and augmentation data from the first device based on the data augmentation procedure and data augmentation parameter set; and means for training a positioning model based on a combination of measurement data and augmentation data.

[0105] In some embodiments, the first device includes means for transmitting a request for assistance for data augmentation to a second device.

[0106] In some embodiments, the support request further indicates the ratio of available data to the target data size used in training the positioning model.

[0107] In some embodiments, the data augmentation configuration includes a data augmentation parameter set based on wireless measurements of a third device. In some embodiments, the means for determining the data augmentation parameter set includes means for extracting the data augmentation parameter set from the data augmentation configuration.

[0108] In some embodiments, the data augmentation configuration includes wireless measurements of a third device.

[0109] In some embodiments, the means for determining the data augmentation parameter set includes means for determining the data augmentation parameter set based on wireless measurements of a third device.

[0110] In some embodiments, the first device includes means for receiving radio measurements of the third device from the third device via a side link between the first device and the third device, and means for determining a data augmentation parameter set based on the radio measurements of the third device.

[0111] In some embodiments, the configuration of the data augmentation includes instructions on whether the data for training the positioning model is interpolated or measured.

[0112] In some embodiments, the first device includes means for transmitting feedback information to a second device indicating the current data augmentation performance of the first device.

[0113] In some embodiments, the first device includes a terminal device, and the second device includes a network device.

[0114] In some embodiments, a second device (e.g., device 120 in Figure 1) capable of performing any of the methods 900 may include means for performing each operation of the methods 900. The means can be implemented in any preferred form. For example, the means can be implemented in a circuit or a software module.

[0115] In some embodiments, the second device includes means for receiving wireless measurements from a third device, means for determining a data augmentation configuration for training a positioning model in the first device based on the measurements from the third device, and means for transmitting the data augmentation configuration to the first device.

[0116] In some embodiments, the second device includes means for determining a data augmentation parameter set based on wireless measurements, and means for transmitting the data augmentation configuration, including the data augmentation parameter set, to the first device.

[0117] In some embodiments, the data augmentation configuration includes wireless measurements of a third device.

[0118] In some embodiments, the second device includes means for receiving assistance requests for data augmentation from the first device.

[0119] In some embodiments, the support request further indicates the ratio of available data to the target data size used in training the positioning model.

[0120] In some embodiments, the means for transmitting a data extension configuration includes means for monitoring the performance of a first device and means for transmitting a data extension configuration to the first device in accordance with a determination that a performance degradation of the first device has been detected.

[0121] In some embodiments, the means for transmitting the data extension configuration includes means for periodically transmitting the data extension configuration.

[0122] In some embodiments, the data augmentation configuration further indicates the lifespan of the data augmentation configuration.

[0123] In some embodiments, the second device includes means for receiving feedback information from the first device indicating the current data augmentation performance of the first device, and means for updating the data augmentation configuration based on the feedback information.

[0124] In some embodiments, the first device includes a terminal device, the second device includes a network device, and the third device includes another terminal device.

[0125] In some embodiments, a first apparatus (e.g., device 110 in Figure 1) capable of performing any of the methods 1000 may include means for performing each operation of the methods 1000. The means can be implemented in any preferred form. For example, the means can be implemented in a circuit or a software module.

[0126] In some embodiments, the first device includes means for transmitting a request for assistance for data augmentation to a second device, means for receiving augmentation data from the second device for training a positioning model, and means for training the positioning model based on a combination of augmentation data and measurement data.

[0127] In some embodiments, the support request further indicates the ratio of available data to the target data size used in training the positioning model.

[0128] In some embodiments, the first device includes a terminal device, and the second device includes a network device.

[0129] In some embodiments, a second device (e.g., device 120 in Figure 1) capable of performing any of the methods 1100 may include means for performing each operation of the methods 1100. The means can be implemented in any preferred form. For example, the means can be implemented in a circuit or a software module.

[0130] In some embodiments, the second device includes means for receiving assistance requests for data augmentation from the first device, means for acquiring measurement data from the third device, means for generating augmentation data for training a positioning model in the first device based on the measurement data, and means for transmitting the augmentation data to the first device.

[0131] In some embodiments, the support request further indicates the ratio of available data to the target data size used in training the positioning model.

[0132] In some embodiments, the measurement data includes at least one of uplink measurement data or downlink measurement data.

[0133] In some embodiments, the first device includes a terminal device, the second device includes a network device, and the third device includes another terminal device or another network device.

[0134] Figure 12 is a simplified block diagram of a device 1200 suitable for carrying out an embodiment of the present disclosure. The device 1200 can be provided to implement a communication device such as the device 110 or device 120 shown in Figure 1, for example. As shown, the device 1200 includes one or more processors 1210, one or more memories 1220 coupled to the processor 1210, and one or more communication modules 1240 coupled to the processor 1210.

[0135] The communication module 1240 is for bidirectional communication. The communication module 1240 has one or more communication interfaces to facilitate communication with one or more other modules or devices. The communication interfaces can represent any interface necessary for communication with other network elements. In some embodiments, the communication module 1240 may include at least one antenna.

[0136] The processor 1210 can be any type suitable for a local technology network and, in non-limiting examples, may include one or more of the following: a general-purpose computer, a dedicated computer, a microprocessor, a digital signal processor (DSP), and a processor based on a multi-core processor architecture. The device 1200 may include multiple processors, such as application-specific integrated circuit chips that are time-following a clock that synchronizes the main processor.

[0137] Memory 1220 may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memories include, but are not limited to, read-only memory (ROM) 1224, electrically programmable read-only memory (EPROM), flash memory, hard disks, compact discs (CDs), digital video discs (DVDs), optical discs, laser discs, and other magnetic and / or optical storage. Examples of volatile memories include, but are not limited to, random-access memory (RAM) 1222 and other volatile memories that do not persist during power-down periods.

[0138] The computer program 1230 includes computer-executable instructions that are executed by the associated processor 1210. The instructions in program 1230 may include instructions for performing actions / operations as described in some embodiments of the present disclosure. Program 1230 can be stored in memory, such as ROM 1224. The processor 1210 can perform any preferred actions and operations by loading program 1230 into RAM 1222.

[0139] An example embodiment of the present disclosure can be implemented by program 1230 so that device 1200 can perform any of the processes of the present disclosure as described with reference to Figures 4 to 11. An example embodiment of the present disclosure can also be implemented by hardware, or by a combination of software and hardware.

[0140] In some embodiments, program 1230 can be tangibly contained in a computer-readable medium that can be contained in device 1200 (such as in memory 1220) or in another storage device accessible by device 1200. Device 1200 can load program 1230 from the computer-readable medium into RAM 1222 for execution. In some embodiments, the computer-readable medium can include any type of non-transitory storage medium, such as ROM, EPROM, flash memory, hard disk, CD, or DVD. As used herein, the term “non-transitory” refers to a limitation of the medium itself (i.e., tangible and not signaling) as opposed to a limitation of data storage persistence (e.g., RAM vs. ROM).

[0141] Figure 13 shows an example of a computer-readable medium 1300, which may be in the form of a CD, DVD, or other optical storage disc. The computer-readable medium 1300 stores a program 1230.

[0142] In general, various embodiments of this disclosure can be implemented in hardware, dedicated circuitry, software, logic, or any combination thereof. Some embodiments can be implemented in hardware, while others can be implemented in firmware or software that can be executed by a controller, microprocessor, or other computer device. Various embodiments of some of the embodiments of this disclosure are illustrated and described using block diagrams, flowcharts, or other graphical representations, but it should be understood that the blocks, apparatus, systems, techniques, or methods described herein can be implemented in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers, or other computer devices, or any combination thereof, as non-limiting examples.

[0143] Some embodiments of this disclosure also provide at least one computer program product tangibly stored in a computer-readable medium, such as a non-temporary computer-readable medium. The computer program product includes computer-executable instructions that perform one of the methods described above, such as being contained in a program module and executed on a target physical or virtual processor in a device. Generally, a program module includes routines, programs, libraries, objects, classes, components, or data structures that perform a specific task or implement a specific abstract data type. In various embodiments, the functionality of program modules can be combined or divided among program modules as desired. The machine-executable instructions for a program module can be executed in a local or distributed device. In a distributed device, program modules can be located on both local and remote storage media.

[0144] Program code for performing the method of this disclosure can be written in any combination of one or more programming languages. By providing this program code to the processor or controller of a general-purpose computer, a dedicated computer, or other programmable data processing device, the functions / operations specified in the flowchart and / or block diagram can be performed when the program code is executed by the processor or controller. The program code can be executed entirely on a machine, partially on a machine, as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0145] In the context of this disclosure, computer program code or associated data can be carried by any suitable carrier in order to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals and computer-readable media.

[0146] A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or a suitable combination thereof. Specific examples of computer-readable storage media include electrical connections having one or more wires, portable computer diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or a suitable combination thereof.

[0147] Furthermore, while the operations are shown in a specific order, it should be understood that such operations do not necessarily need to be performed in a specific order or sequentially as shown, or that all illustrated operations must be performed, in order to achieve the desired result. In some situations, multitasking and parallel processing may be advantageous. Similarly, although the above description includes several specific implementation details, these should not be interpreted as limiting the scope of this disclosure, but rather as descriptions of features that may be specific to a particular embodiment. Unless otherwise specified, several features described in the context of a different embodiment may be implemented in combination in a single embodiment. Conversely, unless otherwise specified, various features described in the context of a single embodiment may be implemented individually or in any preferred partial combination in multiple embodiments.

[0148] While this disclosure has been described using language specific to structural features and / or methodological actions, it should be understood that the disclosure set forth in the attached claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are disclosed as exemplary forms for carrying out the claims. [Explanation of Symbols]

[0149] 400 interactions 410-430 devices 4005 Submitted a request for assistance 4010 Transmit wireless measurement values 4015 Configuration decided Send 4020 configuration 4025 Transmit wireless measurement values Determine the 4030 data augmentation parameter set. 4035 Data acquisition based on data extension parameter set Train the 4040 positioning model. 4045 Send feedback information 4050 configuration updated

Claims

1. The first device, At least one processor, At least one memory for storing instructions, The instruction is provided that, when executed by the at least one processor, The configuration for training the positioning model in the first device is received from the second device, The data extension parameter set is determined based on the data extension configuration, Based on the data augmentation procedure and the data augmentation parameter set, data including the measurement data and augmented data of the first device is acquired. Training the positioning model based on the combination of the measurement data and the extended data, To cause the first device to perform at least the above, A first device characterized by the following:

2. The first device is Sending a request for assistance for data augmentation to the second device, The first device according to claim 1, which performs the following:

3. The aforementioned support request further indicates the ratio of available data to the target data size used in training the positioning model, The first device according to claim 2.

4. The data augmentation configuration includes the data augmentation parameter set based on the wireless measurement values ​​of the third device, Determining the aforementioned data expansion parameter set means that Extracting the data extension parameter set from the data extension configuration, The first device according to claim 1, including

5. The data augmentation configuration includes wireless measurement values ​​of the third device, The first device according to claim 1.

6. Determining the aforementioned data expansion parameter set means that Determining the data augmentation parameter set based on the wireless measurement values ​​of the third device, The first device according to claim 5, including

7. The first device is Receiving wireless measurement values ​​from the third device via a side link between the first device and the third device, Determining the data augmentation parameter set based on the wireless measurement values ​​of the third device, The first device according to claim 1, further performing the above.

8. The configuration of the aforementioned data expansion is, Instructions regarding whether the data for training the positioning model is interpolated data or measured data, The first device according to claim 1, including

9. The first device is To transmit feedback information indicating the current data expansion performance of the first device to the second device, The first device according to claim 1, further performing the above.

10. The first device includes a terminal device, and the second device includes a network device. The first device according to claim 1.

11. The second device is At least one processor, At least one memory for storing instructions, The instruction is provided that, when executed by the at least one processor, Receiving wireless measurement values ​​from a third device, Based on the measurements from the third device, determine the configuration of the data augmentation for training the positioning model in the first device, The configuration of the data extension is transmitted to the first device, To cause the second device to perform at least the following: A second device characterized by the following.

12. The second device described above is Determining the data augmentation parameter set based on the aforementioned wireless measurement values, Transmitting the data extension configuration, including the data extension parameter set, to the first device, A second device according to claim 11, further performing the above.

13. Sending the aforementioned data extension configuration means Monitoring the performance of the first device, In accordance with the determination that a performance degradation of the first device has been detected, the configuration of the data extension is transmitted to the first device, The second device according to claim 11, including the following:

14. Sending the aforementioned data extension configuration means The configuration of the aforementioned data extension will be sent periodically. The second device according to claim 11, including the following:

15. The configuration of the data extension further indicates the validity period of the configuration of the data extension. The second device according to claim 13.

16. The second device described above is Receiving feedback information from the first device indicating the current data augmentation performance of the first device, The configuration of the data extension is updated based on the aforementioned feedback information, A second device according to claim 11, further performing the above.

17. The first device includes a terminal device, the second device includes a network device, and the third device includes another terminal device. The second device according to claim 11.

18. The first device, At least one processor, At least one memory for storing instructions, The instruction is provided that, when executed by the at least one processor, Sending a request for assistance for data augmentation to a second device, Receiving augmented data for training the positioning model from the second device, Training the positioning model based on the combination of the aforementioned extended data and measurement data, To cause the first device to perform at least the above, A first device characterized by the following:

19. The aforementioned support request further indicates the ratio of available data to the target data size used in training the positioning model, The first device according to claim 18.

20. The first device includes a terminal device, and the second device includes a network device. The first device according to claim 18.

Images