Training data collection

Abstract

An apparatus comprising: a coordinator function configured to send a first configuration message to a first collector function, the first configuration message comprising information, the information is used to configure the first collector function to collect the first training data, to transform the first training data to generate transformed first training data, and to report the transformed first training data.

Description

Technical Field

[0001] Various example embodiments are directed to collecting training data. Background Art

[0002] In wireless telecommunications networks, network nodes collect and process training data to train machine learning (ML) models to improve the operation of the network. Although techniques exist for collecting and processing training data, unexpected consequences may occur. Therefore, it is desirable to provide an improved technique for collecting and processing training data. Summary of the invention

[0003] The scope of protection sought by various exemplary embodiments of the invention is set out in the independent claims.Example embodiments and features described in this specification that do not fall within the scope of the independent claims, if any, should be interpreted as examples that aid in understanding the various embodiments of the invention.

[0004] According to various but not necessarily all example embodiments of the present invention, a device is provided, comprising: a coordinator function configured to send a first configuration message to a first collector function, the first configuration message comprising information for configuring the first collector function to collect first training data, to transform the first training data to generate transformed first training data, and to report the transformed first training data.

[0005] The first training data may include an N-dimensional matrix of values ​​collected by a first collector function.

[0006] The value may include channel information. The channel information may relate to a wireless link between an entity hosting the collector function and a transmitter. The channel information may include beamforming and / or channel values.

[0007] The first configuration message may include information to configure the first collector function to transform the first training data by removing specified vendor specific data and / or artifacts to generate transformed first training data. The specified vendor specific data and / or artifacts may include RF receiver delay, number of RF chains, etc.

[0008] The first configuration message may include a message to configure the first collector function to transform the first training data to generate transformed first training data having a target reconfiguration.

[0009] The first configuration message may include information to configure the first collector function to report the transformed first training data in a transformed training data format and / or at a specified reporting period.

[0010] The transformed first training data may include an N-dimensional matrix of values ​​collected by the first collector function.

[0011] The first configuration message may include information to configure the first collector function to report the transformed first training data to the coordinator function and / or the training function.

[0012] The first configuration message may include information to configure the first collector function to report the transformed first training data to both the first training function and the second training function.

[0013] The coordinator function may be configured to send a second configuration message to the second collector function, the second configuration message comprising information for configuring the second collector function to collect second training data, to reconfigure the second training data to generate transformed second training data, and to report the transformed second training data.

[0014] The coordinator function may be configured to combine the received transformed training data to form combined transformed training data.The received transformed training data may come from multiple collector functions and / or from the same collector function at different times.

[0015] The coordinator function may be configured to combine the received transformed training data by stacking, averaging, filtering, pruning, puncturing, scaling, and / or normalizing.

[0016] The coordinator function may be configured to send the combined transformed training data to the training function.

[0017] The coordinator function may be configured to send a configuration message to a collector function provided by a common vendor with the coordinator function and the collector function, the configuration message including instructions to configure the collector function to collect training data and to report the training data.

[0018] The coordinator function may be configured to send a conversion configuration message to the training function, the conversion configuration message comprising information to configure the training function to convert the received transformed training data and / or the received combined transformed training data into converted training data.

[0019] The conversion configuration message may include information to configure the training function to convert the received transformed training data and / or the received combined transformed training data into transformed training data by enhancing, averaging, filtering, pruning, puncturing, normalizing, scaling and / or translating. The conversion may be performed by suitable filtering or projection operations.

[0020] The coordinator function may be configured to send a first conversion configuration message to a first training function and a second conversion configuration message to a second training function, the first conversion configuration message comprising information for configuring the first training function to convert the received transformed training data and / or the received combined transformed training data into converted first training data, the second conversion configuration message comprising information for configuring the second training function to convert the received transformed training data and / or the received combined transformed training data into converted second training data.

[0021] Messages and / or data can be transmitted on a physical side link shared channel, a physical downlink shared channel, a physical uplink shared channel and / or any wireless medium channel. Messages and / or data can be transmitted on a physical side link control channel, a physical downlink common control channel and / or a physical uplink common channel. Messages and / or data can be transmitted on other interfaces other than the air interface, such as through an F1 interface, an Xn interface and / or an NG interface. Messages and / or data can be sent through an O-RAN A1 interface, an E1 interface, an E2 interface, and an F1 interface. It should be understood that these are applicable to 4G systems, 5G systems, and 6G systems.

[0022] The training data, transformed training data and / or transformed training data may include a data format specified by a combination of parameters associated with the training data, such as: sampling resolution; array shape: scalar, vector, matrix; and / or length, etc.

[0023] According to various but not necessarily all example embodiments of the present invention, a method is provided, the method comprising: sending a first configuration message to a first collector function, the first configuration message comprising information for configuring the first collector function to collect first training data, to transform the first training data to generate transformed first training data, and to report the transformed first training data.

[0024] The first training data may include an N-dimensional matrix of values ​​collected by a first collector function.

[0025] The value may include channel information. The channel information may relate to a wireless link between an entity hosting the collector function and a transmitter. The channel information may include beamforming and / or channel values.

[0026] The first configuration message may include information to configure the first collector function to transform the first training data by removing specified vendor specific data and / or artifacts to generate transformed first training data. The specified vendor specific data and / or artifacts may include RF receiver delay, number of RF chains.

[0027] The first configuration message may include a message to configure the first collector function to transform the first training data to generate transformed first training data having a target reconfiguration.

[0028] The first configuration message may include information to configure the first collector function to report the transformed first training data in a transformed training data format and / or at a specified reporting period.

[0029] The transformed first training data may include an N-dimensional matrix of values ​​collected by the first collector function.

[0030] The first configuration message may include information to configure the first collector function to report the transformed first training data to the coordinator function and / or the training function.

[0031] The first configuration message may include information to configure the first collector function to report the transformed first training data to both the first training function and the second training function.

[0032] The method may include sending a second configuration message to the second collector function, the second configuration message including information to configure the second collector function to collect second training data, to reconfigure the second training data to generate transformed second training data, and to report the transformed second training data.

[0033] The method may comprise combining the received transformed training data to form combined transformed training data.The received transformed training data may be from a plurality of collector functions and / or from the same collector function at different times.

[0034] The method may include combining the received transformed training data by stacking, averaging, filtering, clipping, puncturing, scaling and / or normalizing.

[0035] The method may include sending the combined transformed training data to a training function.

[0036] The sending may be performed by a coordinator function, and the method may include sending a configuration message to a collector function provided by a common vendor with the coordinator function and the collector function, the configuration message including instructions to configure the collector function to collect training data and to report the training data.

[0037] The method may include sending a conversion configuration message to the training function, the conversion configuration message including information to configure the training function to convert the received transformed training data and / or the received combined transformed training data into converted training data.

[0038] The conversion configuration message may include information to configure the training function to convert the received transformed training data and / or the received combined transformed training data into transformed training data by enhancing, averaging, filtering, pruning, puncturing, normalizing, scaling and / or translating. The conversion may be performed by suitable filtering or projection operations.

[0039] The method may include: sending a first conversion configuration message to a first training function and sending a second conversion configuration message to a second training function, the first conversion configuration message including information for configuring the first training function to convert received transformed training data and / or received combined transformed training data into converted first training data, and the second conversion configuration message including information for configuring the second training function to convert received transformed training data and / or received combined transformed training data into converted second training data.

[0040] Messages and / or data can be transmitted on a physical side link shared channel, a physical downlink shared channel, a physical uplink shared channel and / or any wireless medium channel. Messages and / or data can be transmitted on a physical side link control channel, a physical downlink common control channel and / or a physical uplink common channel. Messages and / or data can be transmitted on other interfaces other than the air interface, such as through an F1 interface, an Xn interface and / or an NG interface. Messages and / or data can be sent through an O-RAN A1 interface, an E1 interface, an E2 interface, and an F1 interface. It should be understood that these are applicable to 4G systems, 5G systems, and 6G systems.

[0041] The training data, transformed training data and / or transformed training data may include a data format specified by a combination of parameters associated with the training data, such as: sampling resolution; array shape: scalar, vector, matrix; and / or length, etc.

[0042] According to various but not necessarily all example embodiments of the present invention, there is provided an apparatus comprising: at least one processor; and at least one memory storing instructions, which when executed by the at least one processor causes the apparatus to at least perform the above method and its example embodiments.

[0043] According to various but not necessarily all exemplary embodiments of the present invention, there is provided a non-transitory computer-readable medium including program instructions stored thereon for executing the above-described method and exemplary embodiments thereof.

[0044] According to various but not necessarily all example embodiments of the present invention, a device is provided, comprising: a collector function, configured to receive a configuration message from a coordinator function, the configuration message comprising information for configuring the collector function to collect training data, to transform the training data to generate transformed training data, and to report the transformed training data.

[0045] The first training data may include an N-dimensional matrix of values ​​collected by a first collector function.

[0046] The value may include channel information. The channel information may relate to a wireless link between an entity hosting the collector function and a transmitter. The channel information may include beamforming and / or channel values.

[0047] The first configuration message may include information to configure the first collector function to transform the first training data by removing specified vendor specific data and / or artifacts to generate transformed first training data. The specified vendor specific data and / or artifacts may include RF receiver delay, number of RF chains, etc.

[0048] The configuration message may include a message to configure the first collector function to transform the training data to generate transformed training data having a target reconfiguration.

[0049] The configuration message may include information to configure the collector function to report the transformed training data in a transformed training data format and / or at a specified reporting period.

[0050] The transformed training data may include an N-dimensional matrix of values ​​collected by the collector function.

[0051] The configuration message may include information to configure the collector function to report the transformed training data to the coordinator function and / or the training function.

[0052] The configuration message may include information to configure the first collector function to report the transformed first training data to both the first training function and the second training function.

[0053] The collector function may be configured to receive a configuration message from a coordinator function provided by a common vendor with the collector function, the configuration message including instructions to configure the collector function to collect training data and to report the training data.

[0054] Messages and / or data can be transmitted on a physical side link shared channel, a physical downlink shared channel, a physical uplink shared channel and / or any wireless medium channel. Messages and / or data can be transmitted on a physical side link control channel, a physical downlink common control channel and / or a physical uplink common channel. Messages and / or data can be transmitted on other interfaces other than the air interface, such as through an F1 interface, an Xn interface and / or an NG interface. Messages and / or data can be sent through an O-RAN A1 interface, an E1 interface, an E2 interface, and an F1 interface. It should be understood that these are applicable to 4G systems, 5G systems, and 6G systems.

[0055] The training data, transformed training data and / or transformed training data may include a data format specified by a combination of parameters associated with the training data, such as: sampling resolution; array shape: scalar, vector, matrix; and / or length, etc.

[0056] According to various but not necessarily all example embodiments of the present invention, a method is provided that receives a configuration message from a coordinator function, the configuration message including information for configuring a collector function to collect training data, to transform the training data to generate transformed training data, and to report the transformed training data.

[0057] The training data may comprise an N-dimensional matrix of values ​​collected by a collector function.

[0058] The value may include channel information. The channel information may relate to a wireless link between an entity hosting the collector function and a transmitter. The channel information may include beamforming and / or channel values.

[0059] The configuration message may include information to configure the collector function to transform the training data by removing specified vendor-specific data and / or artifacts to generate transformed training data. The specified vendor-specific data and / or artifacts may include RF receiver delay, number of RF chains, etc.

[0060] The configuration message may include a message to configure the collector function to transform the training data to generate transformed training data with a target reconfiguration.

[0061] The configuration message may include information to configure the collector function to report the transformed training data in a transformed training data format and / or at a specified reporting period.

[0062] The transformed training data may include an N-dimensional matrix of values ​​collected by the collector function.

[0063] The configuration message may include information to configure the collector function to report the transformed training data to the coordinator function and / or the training function.

[0064] The configuration message may include information to configure the collector function to report the transformed training data to both the first training function and the second training function.

[0065] The method may include receiving a configuration message from a coordinator function provided by a common vendor with the collector function, the configuration message including instructions to configure the collector function to collect training data and to report the training data.

[0066] Messages and / or data can be transmitted on a physical side link shared channel, a physical downlink shared channel, a physical uplink shared channel and / or any wireless medium channel. Messages and / or data can be transmitted on a physical side link control channel, a physical downlink common control channel and / or a physical uplink common channel. Messages and / or data can be transmitted on other interfaces other than the air interface, such as through an F1 interface, an Xn interface and / or an NG interface. Messages and / or data can be sent through an O-RAN A1 interface, an E1 interface, an E2 interface, and an F1 interface. It should be understood that these are applicable to 4G systems, 5G systems, and 6G systems.

[0067] The training data, transformed training data and / or transformed training data may include a data format specified by a combination of parameters associated with the training data, such as: sampling resolution; array shape: scalar, vector, matrix; and / or length, etc.

[0068] According to various but not necessarily all example embodiments of the present invention, there is provided an apparatus comprising: at least one processor; and at least one memory storing instructions, which when executed by the at least one processor causes the apparatus to at least perform the above method and its example embodiments.

[0069] According to various but not necessarily all exemplary embodiments of the present invention, there is provided a non-transitory computer-readable medium including program instructions stored thereon for executing the above-described method and exemplary embodiments thereof.

[0070] According to various but not necessarily all example embodiments of the present invention, a device is provided, comprising: a training function, configured to receive a conversion configuration message from a coordinator function, the conversion configuration message comprising the following information, which is used to configure the training function to convert received transformed training data and / or received combined transformed training data into converted training data.

[0071] The conversion configuration message may include information to configure the training function to convert the received transformed training data and / or the received combined transformed training data into transformed training data by enhancing, averaging, filtering, pruning, puncturing, normalizing, scaling and / or translating. The conversion may be performed by suitable filtering or projection operations.

[0072] The training data may comprise an N-dimensional matrix of values ​​collected by a collector function.

[0073] The value may include channel information. The channel information may relate to a wireless link between an entity hosting the collector function and a transmitter. The channel information may include beamforming and / or channel values.

[0074] The transformed first training data may include an N-dimensional matrix of values ​​collected by the collector function.

[0075] The training function may be configured to combine the received transformed training data to form combined transformed training data.

[0076] The training function may be configured to combine the received transformed training data by stacking, averaging, filtering, clipping, puncturing, scaling, and / or normalizing.

[0077] Messages and / or data can be transmitted on a physical side link shared channel, a physical downlink shared channel, a physical uplink shared channel and / or any wireless medium channel. Messages and / or data can be transmitted on a physical side link control channel, a physical downlink common control channel and / or a physical uplink common channel. Messages and / or data can be transmitted on other interfaces other than the air interface, such as through an F1 interface, an Xn interface and / or an NG interface. Messages and / or data can be sent through an O-RAN A1 interface, an E1 interface, an E2 interface, and an F1 interface. It should be understood that these are applicable to 4G systems, 5G systems, and 6G systems.

[0078] The training data, transformed training data and / or transformed training data may include a data format specified by a combination of parameters associated with the training data, such as: sampling resolution; array shape: scalar, vector, matrix; and / or length, etc.

[0079] According to various but not necessarily all example embodiments of the present invention, a method is provided, the method comprising: receiving a conversion configuration message from a coordinator function, the conversion configuration message comprising the following information, which is used to configure the training function to convert received transformed training data and / or received combined transformed training data into converted training data.

[0080] The conversion configuration message may include information to configure the training function to convert the received transformed training data and / or the received combined transformed training data into transformed training data by enhancing, averaging, filtering, pruning, puncturing, normalizing, scaling and / or translating. The conversion may be performed by suitable filtering or projection operations.

[0081] The training data may comprise an N-dimensional matrix of values ​​collected by a collector function.

[0082] The value may include channel information. The channel information may relate to a wireless link between an entity hosting the collector function and a transmitter. The channel information may include beamforming and / or channel values.

[0083] The transformed first training data may include an N-dimensional matrix of values ​​collected by the collector function.

[0084] The method may include combining the received transformed training data to form combined transformed training data.

[0085] The method may include combining the received transformed training data by stacking, averaging, filtering, clipping, puncturing, scaling and / or normalizing.

[0086] Messages and / or data can be transmitted on a physical side link shared channel, a physical downlink shared channel, a physical uplink shared channel and / or any wireless medium channel. Messages and / or data can be transmitted on a physical side link control channel, a physical downlink common control channel and / or a physical uplink common channel. Messages and / or data can be transmitted on other interfaces other than the air interface, such as through an F1 interface, an Xn interface and / or an NG interface. Messages and / or data can be sent through an O-RAN A1 interface, an E1 interface, an E2 interface, and an F1 interface. It should be understood that these are applicable to 4G systems, 5G systems, and 6G systems.

[0087] The training data, transformed training data and / or transformed training data may include a data format specified by a combination of parameters associated with the training data, such as: sampling resolution; array shape: scalar, vector, matrix; and / or length, etc.

[0088] According to various but not necessarily all example embodiments of the present invention, there is provided an apparatus comprising: at least one processor; and at least one memory storing instructions, which when executed by the at least one processor causes the apparatus to at least perform the above method and its example embodiments.

[0089] According to various but not necessarily all exemplary embodiments of the present invention, there is provided a non-transitory computer-readable medium including program instructions stored thereon for executing the above-described method and exemplary embodiments thereof.

[0090] Further particular and preferred aspects are set out in the accompanying independent and dependent claims. Features of the dependent claims may be combined with features of the independent claims as appropriate, and in combinations other than those explicitly set out in the claims.

[0091] Where an apparatus feature is described as being operable to provide a function, it will be understood that this includes apparatus features that provide that function or that are adapted or configured to provide that function. BRIEF DESCRIPTION OF THE DRAWINGS

[0092] Some example embodiments will now be described with reference to the accompanying drawings, in which:

[0093] Figure 1 It shows that training data is scarce;

[0094] Figure 2 is a signalling diagram signalling diagram where data from supplier 2 is used to augment data from supplier 1;

[0095] Figure 3 is a signaling diagram in which data from vendor 2 is used to enhance data from vendor 1, and a controller function (NR-C) performs combining of vendor agnostic / domain invariant data (DID1 and DID2);

[0096] Figure 4 is a signaling diagram when multiple NR_Es report DIDs to multiple NR_Ts. This method can be used with Figure 2 and / or Figure 3 Combination of methods in

[0097] Figure 5 is a signaling diagram when multiple NR_Es report DIDs to multiple NR_Ts. This method can be used with Figure 2 , Figure 3 and / or Figure 4 Combination of methods in

[0098] Figure 6 is a signaling diagram of an SL positioning example; and

[0099] Figure 7 An enhanced positioning use case is shown. The entity DVD(a, b) in the DVD (or DID) matrix represents a UE-specific positioning measurement obtained at a frequency resource a*Fs and a time resource b*Ts, where Fs and Ts are the UE-specific sampling frequency and sampling time, respectively. DETAILED DESCRIPTION

[0100] Before discussing example embodiments in more detail, an overview will first be provided. Some example embodiments provide a technique whereby network nodes within a wireless telecommunications network are provided with the functionality of coordinating, collecting, and using training data to train ML models to perform various network and / or device specific tasks, commonly referred to as radio resource management (RRM). Typically, a collection function within a network node provided using training data by the same vendor as the network node with the training function can receive their training data, which has values ​​and is in a format known to the training function. In some embodiments, a uniform collection function within a network node provided using training data by the same vendor as the network node with the training function provides their training data in an agnostic or unchanging form. A collector function within a network node provided by a vendor (who uses training data differently from the network node with the training function) provides their training data in an agnostic or unchanging form that does not disclose vendor-specific information about the capabilities of the entity collecting the data. The training data may be provided to a coordinator function that combines the received data or a training function for combining the received data. The training functions are typically provided with information such as details of a transform that can then transform or convert the combined data into a format that can then be used by the training functions to train their (vendor-specific) ML models. This approach helps collect a diverse range of training data from network nodes provided by other vendors in a consistent manner.

[0101] Some example embodiments relate to the Rel-18 Study Item (SI) on Artificial Intelligence (AI) / Machine Learning (ML) for New Radio (NR) Air Interface [3GPP RP-213599]. The SI aims to explore the benefits of enhancing the air interface with features that enable support for AI / ML-based algorithms to enhance performance and / or reduce complexity / overhead. The goal of this SI is to lay down a foundation for future air interface use cases that leverage AI / ML techniques. The initial set of use cases to be covered include channel state information (CSI) feedback enhancement (e.g., overhead reduction, improved accuracy, prediction, etc.), beam management (e.g., beam prediction in time, and / or spatial domain for overhead and latency reduction, beam selection accuracy improvement, etc.), positioning accuracy enhancement, etc. For those use cases, the benefits should be evaluated (with developed methods and defined key performance indicators (KPIs)) and the potential impact on the specifications should be evaluated, including physical (PHY) layer aspects, protocol aspects, etc. One of the key expected results of the SI is that "the AI / ML approaches for selected sub-use cases need to be diverse enough to support the various requirements for gNode B-User Equipment (gNB-UE) collaboration levels."

[0102] It must be noted that additional other use cases may also be addressed in the work item (WI) phase of “AI / ML for air interface”. From Release 18 onwards, it is likely that multiple use cases and applications on ML in gNB and UE will be proposed. Release 17. Positioning Reference Unit (PRU) - A PRU is a 5G network entity that can be specified by the 5G NR network to assist in one or more positioning sessions. A PRU is a device or network node with a known location (e.g., a roadside unit, another UE, etc.) that can be activated as needed by the Location Management Function (LMF) to perform specific positioning operations, such as measurement and / or transmission of specific positioning signals. PRU [R2-2106920] - RAN1 Channel Status (LS) on RAN1 The use of a Positioning Reference Unit (PRU) with a known location for positioning has been evaluated and improvements in using PRUs to enhance positioning performance have been observed. Note: The term “Positioning Reference Unit (PRU)” is used as a term only in this discussion. PRU does not necessarily mean the introduction of a new network node. The PRU may support at least some of the Release 16 positioning functions of the UE, up to RAN2 if agreed. The positioning functions may include, but are not limited to, the following: providing positioning measurements (e.g., Reference Signal Time Difference (RSTD), Reference Signal Received Power (RSRP), Receive-Transmit (Rx-Tx) time difference); the LMF may request the sending of an Uplink (UL) Sounding Reference Signal (SRS) -PRU for positioning to provide the LMF with its own known location coordinate information. If the antenna orientation information of the PRU is known, then that information may also be requested by the LMF. Release 18. The ML models for Radio Resource Management (RRM) are expected to be vendor specific and therefore trained on vendor specific data. The foreseeable RAN outcome is that companies agree to train vendor specific ML models for the same RRM function using only vendor specific training data, so that training data is not exchanged between vendors. This is for several reasons why vendors (UE and / or gNB) do not want to share their data: it is UE specific and in many cases sensitive; and it provides them with a competitive edge by enabling them to generate and deploy their ML based solutions that outperform competitor solutions.

[0103] Therefore, if Figure 1 As shown, data collection for training vendor-specific ML models is expected to become a tedious process, which will most likely result in maintaining suboptimal training datasets that are: imbalanced, i.e., a large imbalance between minority and majority labels; and sparse, i.e., the collected data does not well characterize all scenarios of interest.

[0104] In order to mitigate at least some of the above limitations and ensure training of robust but vendor-specific ML models, vendor-specific data would benefit from being artificially diversified and amplified on a per-vendor basis before being used to train vendor-based ML models. The process of artificially enhancing training data is known as data augmentation, and the success of the process depends on two main factors: the amount and quality of the initial training data; and the augmentation algorithm and its design assumptions. However, at this stage there are no specific proposals on how to collect the required training data from different UEs in the network in order to enable a vendor-agnostic ML enablement solution.

[0105] Some example embodiments provide a technique by which vendor-specific training data (hereinafter referred to as domain variant data) is diversified by using other vendor data without exposing / sharing domain variant datasets between vendors. To this end, domain variant data is first predicted, i.e., stripped from vendor-specific attributes. Three types of NR elements are involved, and a combination of functions can be performed by one NR element: ML Coordinator Function (NR-C) -NR network element that plays the role of aggregating training data collected by different UEs and / or from multiple UE vendors and defining the vendor-agnostic or transformed training data format that each UE should transfer back to the NR-C. The NR-C can be a gNB-CU, NRT-RIC, NWDAF, etc. ML Data Collector Function (NR-E) -NR network element that collects / modifies the raw data indicated by NR-C in a first or vendor specific format and transmits the data to NR-T or NR-C using a vendor agnostic or converted format. NR-E can be UE, gNB-CU, RT-RIC, RSU, etc. By NR-Ek, we mean the NR-E that collects vendor-k specific training data. ML training function (NR-T) -NR network element that combines training data from different sources (e.g., multiple NR-Es) and trains vendor-specific ML functions. The NR-T can be a NWDAF, LMF, serving gNB, or UE. The NR-T can be in the same network element as the corresponding NR-E, such as a gNB or UE. By NR-Tm, we mean a NR-T that trains an ML model for vendor m. Provide training data from supplier 2 to supplier 1

[0106] Figure 2An example embodiment is shown in FIG, where a coordinator function configures a data collector function to provide training data to a training function. One of the data collector functions is from the same vendor as the training function and is therefore able to provide its training data in the form expected by the training function. Another data collector function is from a different vendor and is therefore instructed to collect specified training data, transform the training data into a specified format, and provide the transformed training data to the training function. The training function then converts the transformed training data to match the form of the training data provided by the data collector function from the same vendor as the training function, combines the training data, and uses the combined training data to train the ML model.

[0107] NR-C configures the elements NR-Ek, k=1...K of each supplier k to provide their training data in a given format.

[0108] Therefore, in step S10, when the training data and ML training are for the same vendor, NR_C instructs NR_E1 to provide the training data to NR_T1 in a vendor specific format, called Domain Variant Data (DVD). In step S20, NR_E1 collects the training data DVD1 in the vendor specific format, and in step S30, reports the training data DVD1 to NT_T1.

[0109] At step S40, if the vendor that needs training is different from the vendor that collected the data, NR_C instructs NR_E2 to provide the training data to NR_T2 in a vendor-agnostic format, called the Domain Invariant Data (DID) format. NR_C defines how to obtain the DID on each NR-Ek side by providing the details of a vendor-invariant conversion (VIC) or transformation, as described in more detail below.

[0110] In step S50, each NR-Ek, k=1...,K, transforms the domain-varying data k (DVD(k)) into a DID using the VIC provided by the NR_C. The DID has a unique format (format, size, type such as integer, real value, etc.) and can be scenario-specific, for example: the DID for beamforming can be a real-valued 2D matrix or 3D matrix, where each entity is the RSRP of the RS at a given (time, frequency) or (time, frequency, space) position; the DID for positioning enhancement can be a complex-valued 3D matrix, where each entity is the channel gain at a given (time, delay, space) position; generate a commonly understood / agreed DID format that can be used between different UE vendors / domains. The DVD to DID conversion (VIC) is configured by the NR-C and generally has the following goals: strip the DVD from sensitive information (UE-specific data payload, symbols, identifiers); strip the DVD from vendor-specific artifacts (e.g., UE-specific TX / RX delays, beam offsets, etc.). Note that the exact form of the VIC transformation to be applied to the DVD to obtain the DID is derived in the NR_E that performs the transformation. VIC parameterization may be fully / entirely configurable by NR-C and may constrain: DID format; DID reporting period; target VIC performance, where the performance metric depends on the use case.

[0111] At step S60, NR-Ek sends its DIDk to NR-Tm. Alternatively, the DID may be first sent to NR-C, which forwards it to each NR-Tm.

[0112] At step S70, DVDm is diversified using DIDs from domain k, k=1:K, NR-Tm, where m≠k: each DID(k), k=1…K is used to reconstruct DID(k) into DVD(m). The reconstruction or conversion process includes inputting DID(k) into a module that applies a domain-m-specific transformation to the DID and outputs an approximate DVD(m) (called reconstructed DVD: R-DVD(k→m)). The conversion from DID to R-DVD (Invariant to Variant Conversion (IVC)) is the opposite of VIC and includes converting the DID to DVD format and also includes available domain-specific information (use case related). This conversion ensures that the R-DVD has the same properties and format as DVDm for a specific UE vendor m. The exact form of the IVC transform only needs to be known at the vendor-specific functions (NR_T and / or NR_E).

[0113] In step S80, NR-Tm uses all R-DVD(k→m), k=1...K to combine with the original DVD(m) into a superset C-DVD(m)=combined {R-DVD(k→m), DVD(m)}. Functional combinations can include various operations such as superimposing data sets, averaging, filtering, etc.

[0114] In step S90, C-DVD(m) is enhanced to obtain a final training dataset for domain m, and in step S100, a domain m-specific ML model is trained. Such enhancement typically includes concatenation of datasets, random mixing of datasets, etc.

[0115] In other words, vendor 1 needs to train an ML module. NR_T1 is a function that trains a vendor 1 specific ML model. NR_T1 is configured by NR_C to collect: DVD1 from NR_E1, where NR_E1 belongs to vendor 1 - here, the data can be shared directly because both training and data belong to the same vendor; DID2 from NR_E2, where NR_E2 belongs to vendor 2, so the data needs to be predicted to the vendor before sharing. NR_E1 and NR_E2 are configured by NR_C to collect training data and share it with NR_T1. NR_E2 is configured by NR_C to apply a specific VIC2 to convert its own DVD2 to DID2. NR_T1 is configured by NR_C to apply IVC1 to DID2, so that DID2 is converted to R-DVD(2→1), a reconstructed DVD, i.e., data that vendor 1 can use for training, and originates from vendor 2 NR elements. NR_T1 then combines DVD1 and R-DVD(2→1), where this combination function is generally denoted as combine1. Function combination 1 can perform any of the following operations: superposition of DVD1 and R-DVD (2→1); averaging; filtering; cropping; puncturing; scaling; normalization; a combination of the above operations.

[0116] Figure 3 An example embodiment is shown in , where a coordinator function configures a data collector function to provide training data to the coordinator function. One of the data collector functions is from the same vendor as the training function and is therefore able to provide its training data in the form expected by the training function, but is instructed to collect specified training data, transform the training data into a specified format, and provide the transformed training data to the coordinator function. Another data collector function is from a different vendor and is instructed to collect specified training data, transform the training data into a specified format, and provide the transformed training data to the coordinator function. The coordinator function then combines the training data and provides it to the training function. The training function converts the transformed training data to match the form of the training data provided by the data collector function from the same vendor as the training function, and uses the training data to train the ML model.

[0117] Specifically, the NR-C function collects DID(k) and combines them into a combined DID (C-DID), which is then sent to the NR-T of a specific vendor. The NR_C configures the target DID format for all NR_Es that collect data. It is assumed that each NR_E is able to derive the corresponding VIC transform that knows the DID format and its own DVD format. The NR_T knows the inverse transform IVC corresponding to the vendor for which the training data is to be generated.

[0118] Therefore, when the training data and ML training are for the same vendor, NR_C instructs NR_E1 to provide the training data to NR_T1 in a vendor specific format called Domain Variant Data (DVD). In step S20, NR_E1 collects the training data DVD1 in the vendor specific format and in step S30, reports the training data DVD1 to NT_T1.

[0119] At steps S110 and S140, NR_C instructs NR_E1 and NR_E2 to provide training data to NR_C in a vendor-agnostic format, referred to as the Domain Invariant Data (DID) format. NR_C defines how to obtain the DID on each NR-Ek side by providing the details of a vendor-invariant conversion (VIC) or transformation, as described in more detail below.

[0120] At steps S120 and S150, each NR-Ek, k=1...,K, transforms the domain variation data k (DVD(k)) into a DID using the VIC provided by the NR_C. The DID has a unique format (format, size, type such as integer, real value, etc.) and can be scenario-specific, for example: the DID for beamforming can be a real-valued 2D matrix or 3D matrix, where each entity is the RSRP of the RS at a given (time, frequency) or (time, frequency, space) position; the DID for positioning enhancement can be a complex-valued 3D matrix, where each entity is the channel gain at a given (time, delay, space) position; generate a commonly understood / agreed DID format that can be used between different UE vendors / domains. The DVD to DID conversion (VIC) is configured by the NR-C and generally has the following goals: strip the DVD from sensitive information (UE-specific data payload, symbols, identifiers); strip the DVD from vendor-specific artifacts (e.g., UE-specific TX / RX delays, beam offsets, etc.). Note that the exact form of the VIC transform to be applied to the DVD to obtain the DID is derived in NR_E which performs the transform. The VIC parameterization can be fully / entirely configurable by NR-C and can constrain: DID format; DID reporting period; target VIC performance, where the performance metric depends on the use case.

[0121] At steps S130 and S160, each NR-Ek sends its DIDk to NR-C.

[0122] In step S170, NR-C combines all DIDk into a superset C-DID(m)=combination{DID(k)}. Functional combination may include various operations such as superimposing data sets, averaging, filtering, etc.

[0123] In step S180, the C-DID is reported to NR_T1.

[0124] At step S190, using the C-DID, NR-T1 diversifies the C-DID to reconstruct the C-DID into R-DVD1. The reconstruction or conversion process includes inputting the C-DID to a module that applies domain-specific transformations to the DID and outputs an approximate DVD (called a reconstructed DVD: R-DVD). The conversion from DID to R-DVD (Invariant to Variant Conversion (IVC)) is the opposite of VIC and includes converting the DID to a DVD format and also includes available domain-specific information (use case related). This conversion ensures that the R-DVD has the same properties and format as DVDm for a specific UE vendor m. The exact form of the IVC transform only needs to be known at the vendor-specific functions (NR_T and / or NR_E).

[0125] In step S200, R-DVD(m) is enhanced to obtain a final training dataset for domain m, and in step S210, a domain m-specific ML model is trained. Provide training data from supplier 2 to supplier 1 and from supplier 1 to supplier 2

[0126] Figure 4 An example embodiment is shown in , where a coordinator function configures a data collector function to provide training data to a training function of multiple vendors. This example embodiment can be combined with the above example embodiments. The data collector function is instructed to collect specified training data, transform the training data into a specified format, and provide the transformed training data to the training function. The training function then combines the training data to match the format of the vendor's training data and uses the training data to train its ML model.

[0127] Therefore, NR_C instructs NR_E1 and NR_E2 to provide training data in a vendor specific format, called domain variant data (DVD), to NR_T1 and NR_T2. At step S20, NR_E1 collects the training data DVD1 in the vendor specific format, and at step S30, reports the training data DVD1 to NT_T1. At step S220, NR_E2 collects the training data DVD2 in the vendor specific format, and at step S230, reports the training data DVD1 to NT_T2.

[0128] At steps S240 and S280, NR_C instructs NR_E1 and NR_E2 to provide training data to NR_T1 and NR_T2 in a vendor agnostic format, referred to as the Domain Invariant Data (DID) format. NR_C defines how to obtain the DID on each NR-Ek side by providing details of a vendor-invariant conversion (VIC) or transformation, as described in more detail below.

[0129] At steps S250 and S290, each NR-Ek, k=1...,K, transforms the domain variation data k (DVD(k)) into a DID using the VIC provided by the NR_C. The DID has a unique format (format, size, type such as integer, real value, etc.) and can be scenario-specific, for example: the DID for beamforming can be a real-valued 2D matrix or 3D matrix, where each entity is the RSRP of the RS at a given (time, frequency) or (time, frequency, space) position; the DID for positioning enhancement can be a complex-valued 3D matrix, where each entity is the channel gain at a given (time, delay, space) position; generate a commonly understood / agreed DID format that can be used between different UE vendors / domains. The DVD to DID conversion (VIC) is configured by the NR-C and generally has the following goals: strip the DVD from sensitive information (UE-specific data payload, symbols, identifiers); strip the DVD from vendor-specific artifacts (e.g., UE-specific TX / RX delays, beam offsets, etc.). Note that the exact form of the VIC transform to be applied to the DVD to obtain the DID is derived in NR_E which performs the transform. The VIC parameterization can be fully / entirely configurable by NR-C and can constrain: DID format; DID reporting period; target VIC performance, where the performance metric depends on the use case.

[0130] At steps S260, S270, S300 and S310, each NR-Ek sends its DIDk to both NR_T1 and NR_T2.

[0131] At steps S320 and S330, NR-T1 and NR_T2 combine all DIDk into a superset C-DID(m)=combination{DID(k)}. Functional combination may include various operations such as superimposing data sets, averaging, filtering, and the like.

[0132] At steps S340 and S350 , C-DID(m) is used to train a domain-m specific ML model.

[0133] Optionally, the C-DID is diversified using C-DID, NR-T1 and NR_T2 to reconstruct the C-DID into R-DVD1 and R-DVD1. The reconstruction or conversion process includes inputting the C-DID to a module that applies domain-specific transformations to the DID and outputs an approximate DVD (called a reconstructed DVD: R-DVD). The conversion from DID to R-DVD (Invariant to Variant Conversion (IVC)) is the opposite of VIC and includes converting the DID to DVD format and also includes available domain-specific information (use case related). This conversion ensures that the R-DVD has the same properties and format as DVDm for a specific UE vendor m. The exact form of the IVC transform only needs to be known at the vendor-specific functions (NR_T and / or NR_E). Enhance R-DVD(m) to obtain the final training dataset for domain m, and train a domain m-specific ML model. Providing training data from supplier Y to supplier A

[0134] Figure 5 An example embodiment is shown in , where a coordinator function from one vendor configures data collector functions from other vendors to provide training data to the training function of the other vendor. This example embodiment can be combined with the example embodiments described above. The data collector function is instructed to collect specified training data, transform the training data into a specified format, and provide the transformed training data to the training function. The training function then combines the training data to match the format of the vendor's training data and uses the training data to train its ML model.

[0135] In step S20, NR_EY collects training data DVDY in a vendor specific format.

[0136] At step S370, NR_C instructs NR_EY to provide training data to NR_TA in a vendor agnostic format, called Domain Invariant Data (DID) format. As described above, NR_C defines how to obtain DID on each NR-Ek side by providing details of a vendor-invariant conversion (VIC) or transformation.

[0137] At step S380, NR_C reports the conversion from DID to R-DVD (Invariant to Variant Conversion (IVC)), which includes available domain-specific information (use case related). This conversion ensures that the R-DVD has the same properties and format as DVDm of a specific UE vendor m. In this example, IVC_A provides the conversion from DID to R-DVD_A.

[0138] At step S390, each NR-Ek, k=1...,K, transforms the domain-varying data k (DVD(k)) into a DID using the VIC provided by the NR_C. The DID has a unique format (format, size, type such as integer, real value, etc.) and can be scenario-specific, for example: the DID for beamforming can be a real-valued 2D matrix or 3D matrix, where each entity is the RSRP of the RS at a given (time, frequency) or (time, frequency, space) position; the DID for positioning enhancement can be a complex-valued 3D matrix, where each entity is the channel gain at a given (time, delay, space) position; generate a commonly understood / agreed DID format that can be used between different UE vendors / domains. The DVD to DID conversion (VIC) is configured by the NR-C and generally has the following goals: strip the DVD from sensitive information (UE-specific data payload, symbols, identifiers); strip the DVD from vendor-specific artifacts (e.g., UE-specific TX / RX delays, beam offsets, etc.). Note that the exact form of the VIC transformation to be applied to the DVD to obtain the DID is derived in the NR_E that performs the transformation. VIC parameterization may be fully / entirely configurable by NR-C and may constrain: DID format; DID reporting period; target VIC performance, where the performance metric depends on the use case.

[0139] At step S400, each NR-Ek sends its DIDk to NR_TA.

[0140] At step S410, using the DID, NR-TA diversifies the DID to reconstruct the DID into R-DVD_A. The reconstruction or conversion process includes inputting the DID to a module that applies a domain-specific transformation to the DID and outputs an approximate DVD (called a reconstructed DVD: R-DVD). The transformation ensures that the R-DVD has the same properties and format as DVDm for a specific UE vendor m. R-DVD_A is enhanced to obtain a final training data set for domain m, and at step S400, a domain m-specific ML model is trained. Sidelink (SL) positioning

[0141] like Figure 6 As shown, in an example embodiment, for SL positioning, the vendor 1 UE uses its own data DVD1 and DID2 collected from the vendor 2 UE to train the ML-based position estimator. Although this example relates to positioning, it should be understood that the technology is applicable to other use cases.

[0142] At step S430, the provider 1 UE collects the time-frequency measurements of the DL PRS and stores them in the matrix DVD1.

[0143] At step S440, the provider 1 UE instructs the provider 2 UE to collect the time-frequency measurements of the DL PRS and store them in the matrix DVD2, where: DVD2(i,j) = channel frequency response at frequency i*Fs2 and time j*Ts2, where Fs2 and Ts2 are sampling frequency and time, both specific to Vendor2UE, and i≠a,j ≠b.

[0144] Using the new SL PSSCH IE, the Vendor 1 UE configures the Vendor 2 UE to report DID2, where DID2 is generated by the converter VIC2 for all Vendor 2 UEs.

[0145] The vendor UE configures VIC2 parameters. For example, VIC2 should output DID2, where: DID2(i,j) = CFR at i*Fs1 and time i*Ts1, in other words, DVD2 should be resampled at rate Fs1 and its resolution changed from Ts2 to Ts1. Remove the RX beam response of Vendor 2 UE, namely W2, from DVD2. For example, VIC2 should apply the transformation of DVD2 as: DID2 = inv(W2)*DVD2.

[0146] At step S450, provider 2 UE collects DVD2, applies VIC2 according to the instructions, and reports DID2 to the LMF at step S460.

[0147] At step S470, the Vendor 1 UE uses DID2 and converter IVC2 to reconstruct R-DVD (2→1). In other words, it applies its own response to DID2 to artificially generate how the Vendor 1 UE data looks like at resource index (i, j).

[0148] Next, it combines DVD1 with R-DVD(2→1) into C-DVD(1) at step 480. For example, it can superimpose the two matrices, or calculate the average response of the two.

[0149] In step S490, it generates enhanced DVD1 using C-DVD1 and preferred prior art enhancement methods (scaling, panning, etc.).

[0150] At step S500, the boosted set is then used to train a preferred prior art ML-based position estimator (eg, a deep neural network (DNN) with a rectified linear unit (ReLU) activation function). UE-assisted DL positioning

[0151] like Figure 7As shown, in an example embodiment, UE-assisted DL positioning is involved, where NR_T (or the associated network data analysis function (NWDAF) function) uses the DVD of the vendor 1 UE and the commonly used Figure 3 The technique described in

[0066] trains an ML-based location estimator for UE vendor 1 using DID2 collected from vendor 2.

[0152] NR-C instructs Vendor 1 UE to collect time-frequency measurements of DL PRS and store them in matrix DVD1, where: DVD1(a,b) = channel frequency response at frequency a*Fs1 and time b*Ts1, where Fs1 and Ts1 are the sampling frequency and time, both specific to Vendor 1 UE.

[0153] New LPP IE is used for Vendor 1 UE to report DVD to NR_T.

[0154] NR_C instructs the Vendor 2 UE to collect the time-frequency measurements of the DL PRS and store them in the matrix DVD2, where: DVD2(i,j) = channel frequency response at frequency i*Fs2 and time j*Ts2, where Fs2 and Ts2 are the sampling frequency and time, both specific to the vendor 2 UE, and i≠a, j≠b.

[0155] Using the new LPP IE, NR_C therefore configures the Vendor 2 UE to report DID2, where DID2 is generated by the translator VIC2 for all Vendor 2 UEs.

[0156] NR_C configures VIC2 parameters. For example, VIC2 should output DID2, where: DID2(i,j) = CFR at i*Fs1 and time i*Ts1, in other words, DVD2 should be resampled at rate Fs1 and its resolution changed from Ts2 to Ts1. Remove the RX beam response of Vendor 2 UE, namely W2, from DVD2. For example, VIC2 should apply the transformation of DVD2 as: DID2 = inv(W2)*DVD2.

[0157] The provider 2 UE collects DVD2, applies VIC2 according to the instructions and reports DID2 to NR_C.

[0158] NR_T uses DID2 and converter IVC2 to reconstruct R-DVD(2→1). In other words, NR_T applies the Vendor 1 UE specific response to DID2 to artificially generate how the Vendor 1 UE data looks like at resource index (i, j).

[0159] NR_T combines DVD1 with R-DVD(2→1) to form C-DVD(1). For example, NR_T can superimpose the two matrices, or calculate the average response of the two.

[0160] NR_T uses C-DVD1 and preferred prior art enhancement methods (scaling, panning, etc.) to produce an enhanced DVD1.

[0161] The augmented set is then used to train a preferred state-of-the-art ML-based location estimator (e.g., a DNN with ReLU activation function).

[0162] Those skilled in the art will readily appreciate that the steps of various above-mentioned methods can be performed by a programmed computer.Herein, some embodiments are also intended to encompass program storage devices, such as digital data storage media, which are machine or computer readable and encode machine executable or computer executable instruction programs, wherein the instructions perform some or all steps of the above-mentioned methods.Program storage devices can be, for example, digital memories, magnetic storage media such as disks and tapes, hard drives, or optically readable digital data storage media.Embodiments are also intended to encompass computers programmed to perform the steps of the above-mentioned methods.Contrary to the limitations on data storage persistence (e.g., RAM to ROM), the term "non-transient" used herein is a limitation of the medium itself (i.e., tangible, rather than signal).

[0163] As used in this application, the term "circuitry" may refer to one or more or all of the following: (a) hardware circuit implementation only (such as implementation only in analog and / or digital circuits) and (b) a combination of hardware circuitry and software such as (where applicable): (i) a combination of analog and / or digital hardware circuits and software / firmware; and (ii) any portion of a hardware processor with software (including a digital signal processor, software and memory that work together to enable a device such as a mobile phone or server to perform various functions); and (c) A hardware circuit and / or processor, such as a microprocessor or portion of a microprocessor, that requires software (e.g., firmware) for operation, but in which the software may not be present when not required for operation.

[0164] This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation of only a hardware circuit or processor (or multiple processors) or a portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. For example and if applicable to the particular claim element, the term circuitry also covers a baseband integrated circuit or processor integrated circuit for a mobile device, or a similar integrated circuit in a server, cellular network device, or other computing or network device.

[0165] Although example embodiments of the present invention have been described in the preceding paragraphs with reference to various examples, it should be appreciated that modifications to the examples given can be made without departing from the scope of the invention as claimed.

[0166] Features described in the preceding description may be used in combinations other than the combinations explicitly described.

[0167] Although functions have been described with reference to certain features, those functions may be performable by other features whether described or not.

[0168] Although features have been described with reference to certain embodiments, those features may also be present in other embodiments whether described or not.

[0169] While attention has been avoided in the foregoing description to those features of the present invention which are regarded as particularly important, it is to be understood that the applicant claims protection for any patentable feature or combination of features mentioned above and / or shown in the accompanying drawings, whether or not specific emphasis has been placed thereon.

Claims

1. A device comprising: A coordinator function is configured to send a first configuration message to a first collector function, wherein the first configuration message includes information for configuring the first collector function to collect first training data, to transform the first training data to generate transformed first training data, and to report the transformed first training data.

2. The apparatus according to claim 1, wherein the first configuration message comprises the following information, wherein the information is used to configure the first collector function as: transforming the first training data by removing specified vendor specific data and / or artifacts to generate the transformed first training data; and / or The transformed first training data is reported in a transformed training data format and / or at a specified reporting period.

3. The apparatus according to claim 1 or 2, wherein the first configuration message includes information for configuring the first collector function as follows: reporting the transformed first training data to the coordinator function and / or the training function; and / or The transformed first training data is reported to both the first training function and the second training function.

4. An apparatus according to any preceding claim, wherein the coordinator function is configured to send a second configuration message to a second collector function, the second configuration message comprising information for configuring the second collector function to collect second training data, for reconfiguring the second training data to generate transformed second training data, and for reporting the transformed second training data.

5. An apparatus according to any preceding claim, wherein the coordinator function is configured to combine the received transformed training data to form combined transformed training data, and preferably to send the combined transformed training data to the training function.

6. An apparatus according to any preceding claim, wherein the coordinator function is configured to send a configuration message to a collector function provided by a common vendor for the coordinator function and the collector function, the configuration message comprising instructions for configuring the collector function to collect training data and for reporting the training data.

7. An apparatus according to any preceding claim, wherein the coordinator function is configured to send a conversion configuration message to the training function, the conversion configuration message comprising information for configuring the training function to convert received transformed training data and / or received combined transformed training data into converted training data.

8. An apparatus according to claim 7, wherein the conversion configuration message includes information for configuring the training function to convert the received transformed training data and / or the received combined transformed training data into transformed training data by enhancing, averaging, filtering, pruning, puncturing, normalizing, scaling and / or translating.

9. An apparatus according to any preceding claim, wherein the coordinator function is configured to send a first conversion configuration message to the first training function and to send a second conversion configuration message to the second training function, the first conversion configuration message comprising information used to configure the first training function to convert the received transformed training data and / or the received combined transformed training data into converted first training data, and the second conversion configuration message comprising information used to configure the second training function to convert the received transformed training data and / or the received combined transformed training data into converted second training data.

10. The apparatus according to any preceding claim, wherein the message and / or data is transmitted on a physical sidelink shared channel, a physical downlink shared channel, a physical uplink shared channel and / or any wireless medium channel.

11. A method comprising: A first configuration message is sent to a first collector function, the first configuration message including information to configure the first collector function to collect first training data, to transform the first training data to generate transformed first training data, and to report the transformed first training data.

12. An apparatus comprising: A collector function is configured to receive a configuration message from a coordinator function, the configuration message comprising information for configuring the collector function to collect training data, to transform the training data to generate transformed training data, and to report the transformed training data.

13. A method comprising: A configuration message is received from a coordinator function, the configuration message comprising information to configure a collector function to collect training data, to transform the training data to generate transformed training data, and to report the transformed training data.

14. An apparatus comprising: A training function is configured to receive a conversion configuration message from a coordinator function, the conversion configuration message comprising information for configuring the training function to convert received transformed training data and / or received combined transformed training data into converted training data.

15. A method comprising: A conversion configuration message is received from the coordinator function, the conversion configuration message comprising information to configure the training function to convert the received transformed training data and / or the received combined transformed training data into converted training data.

Images