Data collection over the air for machine learning model training

Abstract

Example embodiments of the present disclosure are directed to data collection over the air for machine learning (ML) model training A method includes receiving, from a second apparatus, a transmission allocation via downlink control information, DCI; determining, based on the DCI, a first modulation and coding scheme, MCS, designated for data transmission for data collection for ML model training; and performing, using the first MCS, the data transmission to the second apparatus.

Description

DATA COLLECTION OVER THE AIR FOR MACHINE LEARNING MODEL TRAINING CROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority from, and the benefit of US Provisional Application No.63 / 750839, filed January 29, 2025, which is hereby incorporated by reference in its entirety.FIELD

[0002] Various example embodiments of the present disclosure generally relate to the field of telecommunication and in particular, to methods, devices, apparatuses and computer readable storage medium for data collection over the air for machine learning (ML) model training.BACKGROUND

[0003] Data collection for ML-based physical layer algorithms will be a crucial part of 6thgeneration (6G), which is envisioned to adopt an Artificial Intelligence (Al)-native air interface. The collected data can be used for training the algorithms, validating their performance, or for debugging of malfunctioning ML-based algorithms.SUMMARY

[0004] In a first aspect of the present disclosure, there is provided a first apparatus. The first apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus at least to: receive, from a second apparatus, a transmission allocation via downlink control information, DCI; determine, based on the DCI, a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training; and perform, using the first MCS, the data transmission to the second apparatus.

[0005] In a second aspect of the present disclosure, there is provided a second apparatus. The second apparatus comprises at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus at least to: transmit, to a first apparatus, a transmission allocation via downlink control information, DCI; and receive, from the first apparatus, data transmission performed using a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training, the first MCS being determined based on the DCI.

[0006] In a third aspect of the present disclosure, there is provided a method. The method comprises: receiving, from a second apparatus, a transmission allocation via downlink control information, DCI; determining, based on the DCI, a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training; and performing, using the first MCS, the data transmission to the second apparatus.

[0007] In a fourth aspect of the present disclosure, there is provided a method. The method comprises: transmitting, to a first apparatus, a transmission allocation via downlink control information, DCI; and receiving, from the first apparatus, data transmission performed using a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training, the first MCS being determined based on the DCI.

[0008] In a fifth aspect of the present disclosure, there is provided a first apparatus. The first apparatus comprises means for receiving, from a second apparatus, a transmission allocation via downlink control information, DCI; means for determining, based on the DCI, a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training; and means for performing, using the first MCS, the data transmission to the second apparatus.

[0009] In a sixth aspect of the present disclosure, there is provided a second apparatus. The second apparatus comprises means for transmitting, to a first apparatus, a transmission allocation via downlink control information, DCI; and means for receiving, from the first apparatus, data transmission performed using a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training, the first MCS being determined based on the DCI.

[0010] In a seventh aspect of the present disclosure, there is provided a computer readable medium. The computer readable medium comprises instructions stored thereon for causing an apparatus to perform at least the method according to the third aspect.

[0011] In an eighth aspect of the present disclosure, there is provided a computer readable medium. The computer readable medium comprises instructions stored thereon for causing an apparatus to perform at least the method according to the fourth aspect.

[0012] It is to be understood that the Summary section is not intended to identify key or essential features of embodiments of the present disclosure, nor is it intended to be used to limit the scope of the present disclosure. Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Some example embodiments will now be described with reference to the accompanying drawings, where:

[0014] FIG. 1 illustrates an example communication environment in which example embodiments of the present disclosure can be implemented;

[0015] FIG. 2 illustrates a signaling flow for data collection for ML model training in accordance with some example embodiments of the present disclosure;

[0016] FIG. 3 illustrates a signaling flow for data collection for ML model training using an MCSindicated by an index in a first MCS table in accordance with some example embodiments of the present disclosure;

[0017] FIG. 4 illustrates a flowchart of an example procedure 400 for usage of collected data in accordance with some example embodiments of the present disclosure;

[0018] FIG. 5 illustrates a signaling flow for data collection for ML model training using an MCS indicated by an index in a second MCS table in accordance with some example embodiments of the present disclosure;

[0019] FIG. 6 illustrates a flowchart of a method implemented at a first apparatus in accordance with some example embodiments of the present disclosure;

[0020] FIG. 7 illustrates a flowchart of a method implemented at a second apparatus in accordance with some example embodiments of the present disclosure;

[0021] FIG. 8 illustrates a simplified block diagram of a device that is suitable for implementing example embodiments of the present disclosure; and

[0022] FIG. 9 illustrates a block diagram of an example computer readable medium in accordance with some example embodiments of the present disclosure.

[0023] Throughout the drawings, the same or similar reference numerals represent the same or similar element.DETAILED DESCRIPTION

[0024] Principle of the present disclosure will now be described with reference to some example embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure. Embodiments described herein can be implemented in various manners other than the ones described below.

[0025] In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.

[0026] References in the present disclosure to “one embodiment,” “an embodiment,” “an example embodiment,” and the like indicate that the embodiment described may include a particular feature, structure, or characteristic, but it is not necessary that every embodiment includes the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

[0027] It shall be understood that although the terms “first,” “second,”..., etc. in front of noun(s)and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another and they do not limit the order of the noun(s). For example, a first element could be termed a second element, and similarly, a second element could be termed a first element, without departing from the scope of example embodiments. As used herein, the term “and / or” includes any and all combinations of one or more of the listed terms.

[0028] As used herein, “at least one of the following: ” and “at least one of ” and similar wording, where the list of two or more elements are joined by “and” or “or”, mean at least any one of the elements, or at least any two or more of the elements, or at least all the elements.

[0029] As used herein, unless stated explicitly, performing a step “in response to A” does not indicate that the step is performed immediately after “A” occurs and one or more intervening steps may be included.

[0030] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms “a”, “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises”, “comprising”, “has”, “having”, “includes” and / or “including”, when used herein, specify the presence of stated features, elements, and / or components etc., but do not preclude the presence or addition of one or more other features, elements, components and / or combinations thereof.

[0031] As used in this application, the term “circuitry” may refer to one or more or all of the following:(a) hardware-only circuit implementations (such as implementations in only analog and / or digital circuitry) and(b) combinations of hardware circuits and software, such as (as applicable):(i) a combination of analog and / or digital hardware circuit(s) with software / firmware and(ii) any portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a mobile phone or server, to perform various functions) and (c) hardware circuit(s) and or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation.

[0032] This definition of circuitry applies to all uses of this term in this application, including in anyclaims. As a further example, as used in this application, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware. The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in server, a cellular network device, or other computing or network device.

[0033] As used herein, the term “communication network” refers to a network following any suitable communication standards, such as New Radio (NR), Long Term Evolution (LTE), LTE-Advanced (LTE-A), Wideband Code Division Multiple Access (WCDMA), High-Speed Packet Access (HSPA), Narrow Band Internet of Things (NB-loT) and so on. Furthermore, the communications between a terminal device and a network device in the communication network may be performed according to any suitable generation communication protocols, including, but not limited to, the first generation (1G), the second generation (2G), 2.5G, 2.75G, the third generation (3G), the fourth generation (4G), 4.5G, the fifth generation (5G), 5.5G, the sixth generation (6G) communication protocols, and / or any other protocols either currently known or to be developed in the future. Embodiments of the present disclosure may be applied in various communication systems. Given the rapid development in communications, there will of course also be future type communication technologies and systems with which the present disclosure may be embodied. It should not be seen as limiting the scope of the present disclosure to only the aforementioned system.

[0034] As used herein, the term “network device” refers to a node in a communication network via which a terminal device accesses the network and receives services therefrom. The network device may refer to a base station (BS) or an access point (AP), for example, a node B (NodeB or NB), an evolved NodeB (eNodeB or eNB), an NR NB (also referred to as a gNB), a Remote Radio Unit (RRU), a radio header (RH), a remote radio head (RRH), a relay, an Integrated Access and Backhaul (I AB) node, a low power node such as a femto, a pico, a non-terrestrial network (NTN) or non-ground network device such as a satellite network device, a low earth orbit (LEO) satellite and a geosynchronous earth orbit (GEO) satellite, an aircraft network device, and so forth, depending on the applied terminology and technology. In some example embodiments, radio access network (RAN) split architecture comprises a Centralized Unit (CU) and a Distributed Unit (DU) at an IAB donor node. An IAB node comprises a Mobile Terminal (IAB-MT) part that behaves like a UE toward the parent node, and a DU part of an IAB node behaves like a base station toward the next-hop IAB node.

[0035] The term “terminal device” refers to any end device that may be capable of wireless communication. By way of example rather than limitation, a terminal device may also be referred to as a communication device, user equipment (UE), a Subscriber Station (SS), a Portable Subscriber Station, a Mobile Station (MS), or an Access Terminal (AT). The terminal device may include, but notlimited to, a mobile phone, a cellular phone, a smart phone, voice over IP (VoIP) phones, wireless local loop phones, a tablet, a wearable terminal device, a personal digital assistant (PDA), portable computers, desktop computer, image capture terminal devices such as digital cameras, gaming terminal devices, music storage and playback appliances, vehicle-mounted wireless terminal devices, wireless endpoints, mobile stations, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), USB dongles, smart devices, wireless customer-premises equipment (CPE), an Internet of Things (loT) device, a watch or other wearable, a head-mounted display (HMD), a vehicle, a drone, a medical device and applications (e.g., remote surgery), an industrial device and applications (e.g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain contexts), a consumer electronics device, a device operating on commercial and / or industrial wireless networks, and the like. The terminal device may also correspond to a Mobile Termination (MT) part of an IAB node (e.g., a relay node). In the following description, the terms “terminal device”, “communication device”, “terminal”, “user equipment” and “UE” may be used interchangeably.

[0036] As used herein, the term “resource,” “transmission resource,” “resource block,” “physical resource block” (PRB), “uplink resource,” or “downlink resource” may refer to any resource for performing a communication, for example, a communication between a terminal device and a network device, such as a resource in time domain, a resource in frequency domain, a resource in space domain, a resource in code domain, or any other combination of the time, frequency, space and / or code domain resource enabling a communication, and the like. In the following, unless explicitly stated, a resource in both frequency domain and time domain will be used as an example of a transmission resource for describing some example embodiments of the present disclosure. It is noted that example embodiments of the present disclosure are equally applicable to other resources in other domains.

[0037] FIG. 1 illustrates an example communication environment 100 in which example embodiments of the present disclosure can be implemented. In the communication environment 100, a plurality of communication devices, including a first apparatus 110 and a second apparatus 120, can communicate with each other. In the example of FIG. 1, the first apparatus 110 may include or be implemented as a terminal device and the second apparatus 120 may include or be implemented as a network device such as a base station serving the terminal device.

[0038] It is to be understood that the number of devices and their connections shown in FIG. 1 are only for the purpose of illustration without suggesting any limitation. The communication environment 100 may include any suitable number of devices configured to implementing example embodiments of the present disclosure. Although not shown, it would be appreciated that one or more additional devices may be located in the cell, and one or more additional cells may be deployed in the communication environment 100. It is noted that although illustrated as a network device, the second apparatus 120 may be another device than a network device. Although illustrated as a terminal device,the first apparatus 110 may be another device than a terminal device.

[0039] In the following, for the purpose of illustration, some example embodiments are described with the first apparatus 110 operating as a terminal device and the second apparatus 120 operating as a network device. However, in some example embodiments, operations described in connection with a terminal device may be implemented at a network device or other device, and operations described in connection with a network device may be implemented at a terminal device or other device.

[0040] In some example embodiments, a transmission direction from the second apparatus 120 to the first apparatus 110 is referred to as a downlink (DL), while a transmission direction from the first apparatus 110 to the second apparatus 120 is referred to as an uplink (UL). In DL, the second apparatus 120 is a transmitting (TX) device (or a transmitter) and the first apparatus 110 is a receiving (RX) device (or a receiver). In UL, the first apparatus 110 is a TX device (or a transmitter) and the second apparatus 120 is a RX device (or a receiver).

[0041] Communications in the communication environment 100 may be implemented according to any proper communication protocol (s), including, but not limited to, cellular communication protocols, wireless local network communication protocols such as Institute for Electrical and Electronics Engineers (IEEE) 802.11 and the like, and / or any other protocols currently known or to be developed in the future. Moreover, the communication may utilize any proper 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 technologies currently known or to be developed in the future.

[0042] As mentioned above, data collection for ML-based physical layer algorithms may be envisioned to adopt an Al native air interface. The collected data can be used for training the algorithms, validating their performance, or for debugging of malfunctioning ML-based algorithms. For example, AI / ML-based receivers of the network device may require UE assistance in training data collected in the network, but the same principle is applicable for UE receiver solutions requiring network (NW) assistance. In most cases, the best loss function is to compute error based on known transmitted bits and compare it to predicted bits. In practice, this requires that the bits are known by both ends. Solutions so far only involve offline training and data-collection via simulations. This works well to some extent, but in situations when it doesn’t work well, it is expected to propose a solution for data collection for training. When problems occur, the models would benefit from additional training data from the environment where it operates, to finetune for specific context (e.g., location, NW configuration, etc.).

[0043] Some solutions have been proposed to obtain training data for such ML-based receivers in a real network, or under challenging conditions. However, none of the existing works have considered the use of channel code to provide the original bits in a robust manner.

[0044] The central challenge is that the receiver needs to have prior knowledge of the bit stream which has been used for modulating the transmitted in-phase quadrature (IQ) signal, so that the receiver can measure accurately the bit error rate for orthogonal frequency-division multiplexing (OFDM) symbols. The original bits are also needed for calculating the loss term in the case that the data is used for training. In current wireless systems, the target is to have a block error rate (BLER) of 10%, which means that a large portion of the packets will fail. For those packets, the original transmit bits are not known before the hybrid automatic repeat request (HARQ) process is completed. Moreover, typically the most useful collected data samples correspond to cases where the data bits cannot be decoded, which is troublesome for obtaining the original transmit bits.

[0045] Moreover, in regular MCS tables, code rates used by the larger modulation orders are quite high, which often prevents reconstructing the original transmit data when collecting samples outside the regular channel conditions. For efficient ML model training, it would be important to have a mechanism which can be expected to be able to reconstruct the transmitted bits with any modulation order under any channel conditions.

[0046] In accordance with some example embodiments of the present disclosure, a solution for data collection over the air for ML model training is proposed. In the solution, a first apparatus receives, from a second apparatus, a transmission allocation via DCI. The first apparatus determines a first MCS based on the DCI. The first MCS is designated for data transmission for data collection for ML model training. Then the first apparatus performs the data transmission to the second apparatus using the first MCS.

[0047] This solution is designed for model validation and training purposes and removes the need for delivering online knowledge of the bit stream or IQ signal. The present disclosure relies on existing channel coding mechanisms in order to add additional redundancy to the transmit signal, which ensures that it can be decoded with extremely high likelihood. This allows the receiver to extract the original transmit bits without any additional overhead.

[0048] The key aspect is that forward error coding is transparent for the neural receiver, because error correction happens after the neural receiver output. In data collection, the received IQ samples have to be associated with underlying data bits, and stronger error codes increase the likelihood of obtaining those bits. Thus, after having collected enough data and using that to retrain the ML model, the receiver can utilize the acquired knowledge to be more spectrally efficient.

[0049] Example embodiments of the present disclosure will be described in detail below with reference to FIGS. 2 to 7.

[0050] Reference is now made to FIG. 2, which illustrates a signaling flow 200 for data collection for ML model training in accordance with some example embodiments of the present disclosure. The signaling flow 200 involves the first apparatus 110 and the second apparatus 120. In some example embodiments, the first apparatus 110 may include a terminal device, e.g., a UE, and the second apparatus 120 may include a network device, e.g., a BS or gNB.

[0051] The second apparatus 120 transmits 202 a transmission allocation via DCI to the first apparatus 110. The transmission allocation is used to allocate resources for subsequent data transmission.

[0052] The first apparatus 110 receives 204 the transmission allocation and determines 206 a MCS (referred to as a first MCS for purpose of discussion) based on the DCI. The first MCS is designated for data transmission for data collection for ML model training. The data collected may be further used to finetune the ML model, validate the performance of the ML model, and the like, which is not limited in the present disclosure.

[0053] After determining the first MCS, the first apparatus 110 performs 208 the data transmission to the second apparatus 120. The second apparatus 120 receives the data transmission and then collects training data from the data transmission. The following will describe the details for the first MCS determination with reference to FIGS. 3 and 5 and describe an example procedure for the usage of collected data with reference to FIG. 4.

[0054] In some other example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may indicate a code rate. The data transmission may be performed using the code rate. Alternatively, or additionally, the first MCS may indicate a modulation order. If the first MCS indicates the code rate, then the code rate may be below a code rate threshold. In this way, the data transmission is performed using a relatively lower code rate to enhance the reconstruction of the original transmitted data bits.

[0055] The value of the code rate threshold may be various. For example, the code rate threshold may have a small value, such as 1%. This means that the first code rate may be a low code rate to perform the data transmission even with poor radio conditions.

[0056] In view of the above, the second apparatus 120 may request the first apparatus 110 to use the first MCS to perform the data transmission, which may be configured with a very small code rate, e.g., 1%.

[0057] In some alternative examples, there may be a MCS table named “first MCS table”. The first MCS table may include one or more MCSs designated for the data transmission for the data collection for the ML model training, without any MCS which is not designated for the data transmission for the data collection for the ML model training.

[0058] In some alternative embodiments, the first apparatus 110 may transmit capability informationto the second apparatus 120. The capability information may indicate the first apparatus 110 supports the first MCS designated for the data transmission for the data collection for the ML model training.

[0059] In an example, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in the first MCS table, and the first MCS table does not include an index for any MCS that is not designated for the data transmission for the data collection for the ML model training. For example, if the first apparatus 110 and the second apparatus 120 performs regular data transmission in a real network, which means carry normal payload data, the corresponding regular MCS table may be used. If the first apparatus 110 and the second apparatus 120 performs the data transmission for the data collection for the ML model training or receiver monitoring in the real network, it is required to use the first MCS table, which may be introduced to the 6G specifications.

[0060] The data transmission for the data collection for the ML model training performed with the proposed first MCS is more suitable for training data collection. For example, the data transmission for training purposes may be performed using a relatively lower code rate. However, if the regular data transmission is performed, using lower code rates would be rather inefficient due to bad spectral efficiency. Hence, lower code rate should be used only when the data transmission is performed for the data collection for the ML model training.

[0061] For the purposes of testing and data collection, it is enough to introduce a single robust code rate for each modulation order, e.g., the first code rate = 10 / 1024 « 0.01. In some example embodiments, assumed modulation orders range may include from QPSK to 1024-QAM. If higher modulation orders are applicable, such modulation orders could be introduced.

[0062] In some examples, the first MCS table may be also referred to as “MCS Table 5”or “5thMCS table”, which is different from the other four MCS tables which have already been defined. In some example embodiments, each index in the first MCS table (i.e., MCS Table 5) may indicate a modulation order and a code rate. An example of the proposed MCS Table 5 is shown as Table 1.Table 1: Example of the proposed MCS Table 5

[0063] Table 1 includes multiple MCS indices and the corresponding modulation orders and targetcode rate. For example, if the MCS index = 1 , the corresponding modulation order is quadrature phase shift keying (QPSK), and the corresponding code rate is 10 / 1024. If the MCS index = 5, the corresponding modulation order is 1024-state quadrature amplitude modulation (1024-QAM), and the corresponding code rate is 10 / 1024.

[0064] In some examples, the target code rate for all MCS indices may be different, e.g., 100 / 1024, if that is considered robust enough. In some examples, different MCS indices may correspond to different code rates. For example, for the MCS index = 1 , the corresponding modulation order is QPSK, and the target code rate is 100 / 1024; for the MCS index = 5, the corresponding modulation order is 1024-QAM, and the corresponding target code rate is 10 / 1024. The scope of the present disclosure is not limited in this respect.

[0065] Moreover, it should be noted that in the case of proposed MCS Table 5, the number of bits reserved for the MCS index can be reduced to 3.

[0066] When using such a first MCS, the transmitted data over the data transmission may be regular user data. Optionally, it may be specifically tailored dummy data consisting of some useful information, such as Global Positioning System (GPS) velocity, GPS coordinates, time of transmission, and the like. In some examples, if the transmitted data is dummy data, the second apparatus 120 may terminate the data processing pipeline after Layer 1, as the transmitted data is not associated with the regular user data.

[0067] In some examples, each index in the first MCS table may indicate a modulation order, and a code rate may be predefined at the first apparatus 110. In some examples, each index in the first MCS table may indicate a modulation order, and a code rate is determined separately, for example, based on configuration transmitted from the second apparatus 120.

[0068] In some examples, each index in the first MCS table may indicate a code rate, and a modulation order may be predefined at the first apparatus 110 or determined based on configuration transmitted from the second apparatus 120. For example, the modulation order may be determined by the second apparatus 120 based on the previous transmission, e.g., different from the modulation order that the previous transmission used. Then the modulation order determined would be indicated to the first apparatus 110 via a scheduling grant.

[0069] Alternatively, or additionally, each index in the first MCS table may indicate a time resource and / or frequency resource. For example, if not using regular user data, pre-defined time / frequency resources can be configured for the data transmission for data collection for ML model training in the 6G frame-structure similar manner to synchronization signals etc. in the 5G frame-structure. This may be defined in the standards and / or part configuration.

[0070] The following will describe the associated signaling if the first MCS designated for the data transmission is indicated by an index in the first MCS table (e.g., the MCS Table 5). More detailsrelated to the MCS Table 5 will be discussed with reference to FIG. 3.

[0071] In some example embodiments, the DCI transmitted by the second apparatus 120 may include an indication indicating the first apparatus 110 to perform the data transmission using the first MCS table. For example, the first MCS table includes one or two indices and the corresponding modulation order(s) and code rate(s). The transmission of the whole table would not increase additional overhead. Alternatively, or additionally, the first MCS indicated by an index in the first MCS table. At this point, the first apparatus 110 may be informed the first MCS table previously. Then, only an index is needed to be indicated, the first apparatus 110 may determine the first MCS.

[0072] As an example, the second apparatus 120 selects the first MCS based on its own logic or needs. In such an example, the first MCS is chosen so that the modulation order would be one step higher than the modulation order used for the particular first apparatus most recently. In this way, the collected data is likely to be most useful for enhancing or training an ML-based receiver.

[0073] In some example embodiments, the indication may be included in a scheduling grant in the DCI, e.g., the modified scheduling grant with the DCI message transmitted at Step 310.

[0074] In some example embodiments, the indication may further indicate that MCSs indicated by all indices in the first MCS table are to be used to perform the data transmission. The second apparatus 120 may select the first MCS based on its own logic or needs, and other decisions logic may be used as well. For example, for performance testing, the first MCS may be chosen to go through all options in consecutive transmissions with the first MCS table.

[0075] In some other example embodiments, there may be a MCS table named “second MCS table”. The second MCS table may include one or more MCSs designated for the data transmission for the data collection for the ML model training, and one or more MCSs which are not designated for the data transmission for the data collection for the ML model training.

[0076] The first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in the second MCS table, and the second MCS table may further include at least one index for at least one second MCS that is not designated for the data transmission for the data collection for the ML model training. The second MCS table may be a MCS table including one or more reserved MCS indexes.

[0077] With the second MCS table, the first MCS may be determined based on the DCI. Specifically, the DCI may include an index of the first MCS in the second MCS table. For example, the second MCS table is a MCS table reserved. The second apparatus 120 selects an index from the second MCS table and indicates the index to the first apparatus 110 via the DCI. At this point, the modulation order and the code rate corresponding to the index may be predetermined.

[0078] In some examples, at least one of a modulation order or a code rate may be indicated by an index in the second MCS table, and the index indicates the first MCS designated for the datatransmission for the data collection for the ML model training.

[0079] In some alternative embodiments, both a modulation order and a code rate may be indicated by an index in the second MCS table. Alternatively, one of the modulation order or the code rate may be indicated by an index in the second MCS table, and the other one may be indicated in another manner. Alternatively, the code rate may be predefined at the first apparatus 110. Alternatively, the code rate may be determined based on configuration information received from the second apparatus 120.

[0080] In some embodiments, the configuration information may include one or more MCS indices designated for the data transmission. For example, the second apparatus 120 may inform the first apparatus 110 the whole MCS indices designated for the data transmission for training data collection. When the second apparatus 120 wants to collect data for training, it may only transmit the DCI indicating a selected index. The first apparatus 110 may determine the corresponding modulation order and code rate.

[0081] In some examples, the configuration information may be received via a RRC signaling from the second apparatus 120.

[0082] More details related to the MCS Table 5 will be discussed with reference to FIG. 5.

[0083] In some example embodiments, the reserved values of the currently used MCS table may be reserved for the data transmission for the data collection for the ML training with very low code rate. The code rate (e.g., 1%) can be predefined in the specifications, and the system can be specified to assume that in the case of receiving the specified data transmission (non-retransmission), the reserved index may be used for such a purpose. There may be one or more first MCS corresponding to each modulation order (4 altogether, assuming QPSK, 16-QAM, 64-QAM and 256-QAM are used). For example, in 5G, these reserved MCS indices are 28, 29, 30, and 31. It may be assumed that reserved MCS indices will be specified again for 6G as well.

[0084] FIG. 3 illustrates a signaling flow 300 for data collection for ML model training using an MCS indicated by an index in the first MCS table in accordance with some example embodiments of the present disclosure. The signaling flow 300 includes implementations in FIG. 2, in particular, with regard to performing the data transmission using the first MCS indicated by an index in the first MCS table (e.g., MCS Table 5). The signaling flow 300 involves the first apparatus 110 and the second apparatus 120. The signaling flow 300 may be used for data collection in a commercial network, which means the signaling flow 300 may be compatible with the existing mechanism.

[0085] At Step 302, a communication connection is set up between the first apparatus 110 and the second apparatus 120. In some example embodiments, the first apparatus 110 may transmit capability information to the second apparatus 120. The capability information may indicate that the first apparatus 110 supports the first MCS table. For example, the first apparatus 110 may indicate itscapability to support the proposed MCS Table 5 when setting up the communication connection.

[0086] At Step 304, the first apparatus 110 transmits a scheduling request to the second apparatus 120. At Step 306, the second apparatus 120 transmits a scheduling grant to the first apparatus 110. At Step 308, in response to the scheduling grant, the first apparatus 110 transmits user data to the second apparatus 120.

[0087] The above steps are related to regular transmission of user data. Then, once the network considers that it is necessary to collect a slot of data for ML model training, at Step 310, the second apparatus 120 transmits a modified scheduling grant with a DCI message to the first apparatus 110 to perform the data transmission for the data collection for ML model training.

[0088] At Step 312, the second apparatus 120 receives the slot of physical uplink shared channel (PUSCH) data via the data transmission with the first MCS table (e.g., MCS Table 5). If the data transmission carries user data, the second apparatus 120 may perform regular processing with the MCS Table 5. In the case that the data transmission consists of other data, e.g., dummy data, the processing should terminate after Layer 1.

[0089] Once the second apparatus 120 has received the PUSCH data transmission with the first MCS table, it would request the first apparatus 110 to revert to the transmission(s) with a regular MCS table. For example, at Step 314, the second apparatus 120 transmits a scheduling grant to the first apparatus 110 to instruct the first apparatus 110 to use the regular MCS table on the consequent transmission. At Step 316, the first apparatus 110 transmits the PUSCH data with the regular MCS table.

[0090] Note that the names of the different control signals might be different in 6G, the intention of this example is to try to provide a generic description of the process.

[0091] In some examples, a request to use the MCS Table 5 may further be transmitted with a separate control message, e.g., with an RRC signaling.

[0092] By using the MCS Table 5, the data transmission for the data collection for the ML training may be performed even with poor radio conditions and without extra overhead. Data carried on the data transmission can be user data. Such a solution is compatible with the existing mechanisms for error correction and link adaptation.

[0093] FIG. 4 illustrates an example procedure 400 for usage of collected data in accordance with some example embodiments of the present disclosure. The example procedure 400 may be implemented at the second apparatus 120.

[0094] At Block 402, the second apparatus 120 may determine the first MCS by using the modulation order that is one step higher than the modulation order used in the previous regular PUSCH transmission.

[0095] At Block 404, the second apparatus 120 may generate a scheduling grant for requesting thefirst apparatus 110 to perform the subsequent data transmission with the MCS Table 5.

[0096] At Block 406, the second apparatus 120 may transmit the generated scheduling grant to the first apparatus 110. An index in the MCS Table 5 may be transmitted with the scheduling grant. Optionally, the whole MCS Table 5 may be transmitted with the scheduling grant. Upon receiving the scheduling grant, the first apparatus 110 may determine a modulation order and a code rate based on the MCS Table 5. Optionally, the first apparatus 110 may determine a modulation order based on the MCS Table 5 and determine a code rate based on a predefined code rate or based on a code rate indicated separately. Then the first apparatus 110 may perform the subsequent data transmission with the modulation order and the code rate.

[0097] At Block 408, the second apparatus 120 may receive the data transmission and collects data for training based on the data transmission. In some example embodiments, a Block 410, the second apparatus 120 may perform a cyclic redundancy check on the collected data. If it is determined that the cyclic redundancy check is passed, proceed to Block 412. That is, the second apparatus 120 may reconstruct data bits associated with the collected data and determine a training data sample based on the collected data and the data bits. At Block 416, the second apparatus 120 may store the training data sample in a database.

[0098] In some example embodiments, if it is determined that the cyclic redundancy check is failed, the second apparatus 120 may transmit, to the first apparatus 110, a request for a retransmission of the data transmission. In response to the request, the first apparatus 110 may perform the retransmission of the data transmission using the modulation order and the code rate used for the initial transmission. The retransmission may also be performed using the regular HARQ procedure.

[0099] If the second apparatus 120 receives, from the first apparatus 110, the retransmission of the data transmission successfully, it may perform operations at Blocks 410 to 416.

[0100] In general, the procedure 400 may be implemented from BS perspective when utilizing the MCS Table 5 for data collection. An indication to use MCS Table 5, together with the MCS index selected by the second apparatus 120, is signaled to the first apparatus 110, who would transmit the next PUSCH transmission with the requested MCS. Then, once the second apparatus 120 has received the PUSCH IQ samples and carried out the receiver processing, it may check whether the CRC is passed, after which it can reconstruct the originally transmitted bits by the first apparatus 110. These bits are then stored in the database, together with the received PUSCH IQ samples. In the highly unlikely case that the CRC fails, regular HARQ retransmission procedure may be performed. However, this will happen extremely rarely due to the high robustness of the MCS Table 5.

[0101] The above has described the embodiments that the first MCS designated for the data transmission for the data collection for the ML model training is indicated by an index in the first MCS table. In some other cases, the first MCS designated for the data transmission for the data collectionfor the ML model training may be indicated by an index in a second MCS table. The second MCS table may include one or more indexes for MCSs that are not designated for the data transmission for the data collection for the ML model training, and may also include one or more indexes for MCSs that are designated for the data transmission for the data collection for the ML model training. Details will be described below.

[0102] FIG. 5 illustrates a signaling flow 500 for data collection for ML model training using a MCS indicated by an index in a second MCS table in accordance with some example embodiments of the present disclosure. The signaling flow 300 described above includes implementations in FIG. 2, in particular, with regard to performing the data transmission using the first MCS indicated by an index in the first MCS table (e.g., MCS Table 5). The signaling flow 500 includes implementations in FIG. 2, in particular, with regard to performing the data transmission using the first MCS indicated by an index in the second MCS table (e.g., reserved MCS table). The signaling flow 500 involves the first apparatus 110 and the second apparatus 120. In some example embodiments, the first apparatus may include a terminal device, and the second apparatus may include a network device. The signaling flow 500 may be compatible with the existing mechanism.

[0103] Multiple steps in the signaling flow 500 are similar to the steps in the signaling flow 300, with the exception of not having to resort to the first MCS table. Furthermore, the implementations at the second apparatus 120 are essentially similar to the signaling flow 300, with the exception of not relying on the first MCS table.

[0104] Specifically, at Step 502, a communication connection is set up between the first apparatus 110 and the second apparatus 120.

[0105] At Step 504, the first apparatus 110 transmits a scheduling request to the second apparatus 120. At Step 506, the second apparatus 120 transmits a scheduling grant to the first apparatus 110. At Step 508, in response to the scheduling grant, the first apparatus 110 transmits user data to the second apparatus 120.

[0106] The above steps are related to regular transmission of user data. Then, once the network considers that it is necessary to collect a slot of data for ML model training, at Step 510, the second apparatus 120 transmits a scheduling grant with an index in the second MCS table (e.g., reserved MCS index) to the first apparatus 110 to perform the data transmission for the data collection for ML model training.

[0107] At Step 512, the second apparatus 120 receives the slot of PUSCH data via the data transmission with the modulation order and the code rate predefined at the first apparatus 110 or determined based on the index in the second MCS table. If the data transmission carries user data, the second apparatus 120 may perform regular processing. In the case that the data transmission consists of other data, e.g., dummy data, the processing should terminate after Layer 1.

[0108] In some example embodiments, the code rate (assigned by using reserved MCS indices for new data) may be provided to the first apparatus 110 via the RRC signaling, which means that different code rates may be used.

[0109] In some example embodiments, the reserved MCS indices may be indicated based on a new data indicator (NDI) given in the DCI. In particular, if the NDI indicates an initial transmission of the data transmission for training data collection, and one of the reserved MCS indices is in use, the code rate should be applied with the corresponding modulation order. On the other hand, in the case that the data transmission is being retransmitted and one of the reserved MCS indices is used, the code rate is obtained from the previous transmission. With such embodiments, it is possible to utilize the first MCS without having to specify a separate MCS table, which would reduce the associated signaling overhead.

[0110] This embodiment may be implemented to 5G specifications by adding one additional condition to 38.214 section 6.1.4.2 (and accordingly to expected 6G specification 39.214). If the code rate is configurable by the network entity, then 38.331 (and correspondingly 39.331 for 6G) needs one new parameter for configuring the desired code rate value.

[0111] Once the second apparatus 120 has received PUSCH data transmission with the second MCS table, it would request the first apparatus 110 to revert to the transmission(s) with a regular MCS table. For example, at Step 514, the second apparatus 120 transmits a scheduling grant to the first apparatus 110 to instruct the first apparatus 110 to use the regular MCS table on the consequent transmission. At Step 516, the first apparatus 110 transmits the PUSCH data with the regular MCS table.

[0112] In summary, the present disclosure provides solutions for data collection over the air for validation and testing for ML model training. The second apparatus 120 requests the first apparatus 110 to use the first MCS designed for the data transmission for the data collection for the ML model training, which may be configured with a very small code rate, e.g., 1%. The first apparatus 110 transmits the next data packet with this MCS. The second apparatus 120 decodes the collected data, verifies CRC, and reproduces the original encoded transmit bits. The second apparatus 120 associates the reproduced data bits with the received IQ samples, which constitutes one training data sample. After this, normal operation ensues with the regular MCSs, until a new data transmission for training data collection is requested.

[0113] The training data samples corresponding to such low-MCS transmissions do not differ from regular transmissions from ML receiver’s point of view, assuming that it does not perform channel decoding (and assuming that the information bit sequence is at least some tens of bits). A practical example of such an ML receiver is DeepRx. The main advantages are that data collection can be performed even with poor radio conditions and without significant extra overhead (the transmitted datacan be useful data); and that standardized data collection is introduced which is compatible with the existing mechanisms for error correction and link adaptation.

[0114] FIG. 6 shows a flowchart of an example method 600 implemented at a first apparatus in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the method 600 will be described from the perspective of the first apparatus 110 in FIG. 1.

[0115] At block 610, the first apparatus 110 receives, from a second apparatus, a transmission allocation via downlink control information, DOI.

[0116] At block 620, the first apparatus 110 determines, based on the DCI, a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training.

[0117] At block 630, the first apparatus 110 performs, using the first MCS, the data transmission to the second apparatus.

[0118] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in a first MCS table, and the first MCS table does not comprise an index for a second MCS that is not designated for the data transmission for the data collection for the ML model training.

[0119] In some example embodiments, the method 600 may further comprise: transmitting, to the second apparatus, capability information indicating that the first apparatus supports the first MCS table.

[0120] In some example embodiments, each index in the first MCS table may indicate at least one of: a modulation order, a code rate, a time resource, or a frequency resource.

[0121] In some example embodiments, the DCI may comprise an indication indicating the first apparatus to perform the data transmission using at least one of: the first MCS table, or the first MCS indicated by an index in the first MCS table.

[0122] In some example embodiments, the indication may be comprised in a scheduling grant in the DCI.

[0123] In some example embodiments, the indication may further indicate that MCSs indicated by all indices in the first MCS table are to be used to perform the data transmission.

[0124] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in a second MCS table, and the second MCS table may further comprise at least one index for at least one second MCS that is not designated for the data transmission for the data collection for the ML model training.

[0125] In some example embodiments, the DCI may comprise the index of the first MCS in the second MCS table.

[0126] In some example embodiments, a code rate for an initial transmission of the datatransmission for the data collection for the ML training may be predefined at the first apparatus, or determined based on configuration information received from the second apparatus.

[0127] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be determined based on configuration information received from the second apparatus, and the configuration information may comprise one or more MCS indices designated for the data transmission.

[0128] In some example embodiments, the configuration information may be received via a radio resource control, RRC, signaling from the second apparatus.

[0129] In some example embodiments, at least one of a modulation order or a code rate may be indicated by an index in a MCS table, and the index may indicate the first MCS designated for the data transmission for the data collection for the ML model training.

[0130] In some example embodiments, the method 600 may further comprise: transmitting, to the second apparatus, capability information indicating that the first apparatus supports the first MCS designated for the data transmission for the data collection for the ML model training.

[0131] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may indicate at least one of a code rate or a modulation order, and the code rate is below a code rate threshold.

[0132] In some example embodiments, the code rate threshold may be 1%.

[0133] In some example embodiments, the first apparatus may comprise a terminal device, and the second apparatus may comprise a network device.

[0134] FIG. 7 shows a flowchart of an example method 700 implemented at a second apparatus in accordance with some example embodiments of the present disclosure. For the purpose of discussion, the method 700 will be described from the perspective of the second apparatus 120 in FIG. 1.

[0135] At block 710, the second apparatus 120 transmits, to a first apparatus, a transmission allocation via downlink control information, DOI.

[0136] At block 720, the second apparatus 120 receives, from the first apparatus, data transmission performed using a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training, the first MCS being determined based on the DCI.

[0137] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in a first MCS table, and the first MCS table may not comprise an index for a second MCS that is not designated for the data transmission for the data collection for the ML model training.

[0138] In some example embodiments, the method 700 may further comprise: receiving, from the first apparatus, capability information indicating that the first apparatus supports the first MCS table.

[0139] In some example embodiments, each index in the first MCS table may indicate at least one of: a modulation order, a code rate, a time resource, or a frequency resource.

[0140] In some example embodiments, the DCI may comprise an indication indicating the first apparatus to perform the data transmission using at least one of: the first MCS table, or the first MCS indicated by an index in the first MCS table.

[0141] In some example embodiments, the indication may be comprised in a scheduling grant in the DCI.

[0142] In some example embodiments, the indication may further indicate that MCSs indicated by all indices in the first MCS table are to be used to perform the data transmission.

[0143] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in a second MCS table, and the second MCS table may further comprise at least one index for at least one second MCS that is not designated for the data transmission for the data collection for the ML model training.

[0144] In some example embodiments, the DCI may comprise the index of the first MCS in the second MCS table.

[0145] In some example embodiments, a code rate for an initial transmission of the data transmission for the data collection for the ML model training may be predefined at the first apparatus, or determined based on configuration information received from the second apparatus.

[0146] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be determined based on configuration information received from the second apparatus, and the configuration information may comprise one or more MCS indices designated for the data transmission.

[0147] In some example embodiments, the configuration information may be transmitted via a radio resource control, RRC, signaling to the first apparatus.

[0148] In some example embodiments, at least one of a modulation order or a code rate may be indicated by an index in a MCS table, and the index may indicate the first MCS designated for the data transmission for the data collection for the ML model training.

[0149] In some example embodiments, the method 700 may further comprise: receiving, from the first apparatus, capability information indicating that the first apparatus supports the first MCS designated for the data transmission for the data collection for the ML model training.

[0150] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may indicate at least one of a code rate or a modulation order, and the code rate is below a code rate threshold.

[0151] In some example embodiments, the code rate threshold may be 1%.

[0152] In some example embodiments, the method 700 may further comprise: performing a cyclicredundancy check on the collected data; in accordance with a determination that the cyclic redundancy check is passed, reconstructing data bits associated with the collected data; and determining a training data sample based on the collected data and the data bits.

[0153] In some example embodiments, the method 700 may further comprise: performing a cyclic redundancy check on the collected data; in accordance with a determination that the cyclic redundancy check is failed, transmitting to the first apparatus, a request for a retransmission of the data transmission; and receiving, from the first apparatus, the retransmission of the data transmission.

[0154] In some example embodiments, the first apparatus may comprise a terminal device, and the second apparatus may comprise a network device.

[0155] In some example embodiments, a first apparatus capable of performing any of the method 600 (for example, the first apparatus 110 in FIG. 1 may comprise means for performing the respective operations of the method 600. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module. The first apparatus may be implemented as or included in the first apparatus 110 in FIG. 1.

[0156] In some example embodiments, the first apparatus comprises means for receiving, from a second apparatus, a transmission allocation via downlink control information, DCI; means for determining, based on the DCI, a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training; and means for performing, using the first MCS, the data transmission to the second apparatus.

[0157] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in a first MCS table, and the first MCS table may not comprise an index for a second MCS that is not designated for the data transmission for the data collection for the ML model training.

[0158] In some example embodiments, the first apparatus may further comprise: means for transmitting, to the second apparatus, capability information indicating that the first apparatus supports the first MCS table.

[0159] In some example embodiments, each index in the first MCS table may indicate at least one of: a modulation order, a code rate, a time resource, or a frequency resource.

[0160] In some example embodiments, the DCI may comprise an indication indicating the first apparatus to perform the data transmission using at least one of: the first MCS table, or the first MCS indicated by an index in the first MCS table.

[0161] In some example embodiments, the indication may be comprised in a scheduling grant in the DCI.

[0162] In some example embodiments, the indication may further indicate that MCSs indicated by all indices in the first MCS table are to be used to perform the data transmission.

[0163] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in a second MCS table, and the second MCS table may further comprise at least one index for at least one second MCS that is not designated for the data transmission for the data collection for the ML model training.

[0164] In some example embodiments, the DCI may comprise the index of the first MCS in the second MCS table.

[0165] In some example embodiments, a code rate for an initial transmission of the data transmission for the data collection for the ML training may be predefined at the first apparatus, or determined based on configuration information received from the second apparatus.

[0166] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be determined based on configuration information received from the second apparatus, and the configuration information may comprise one or more MCS indices designated for the data transmission.

[0167] In some example embodiments, the configuration information may be received via a radio resource control, RRC, signaling from the second apparatus.

[0168] In some example embodiments, at least one of a modulation order or a code rate may be indicated by an index in a MCS table, and the index may indicate the first MCS designated for the data transmission for the data collection for the ML model training.

[0169] In some example embodiments, the first apparatus may further comprise: means for transmitting, to the second apparatus, capability information indicating that the first apparatus supports the first MCS designated for the data transmission for the data collection for the ML model training.

[0170] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may indicate at least one of a code rate or a modulation order, and the code rate may be below a code rate threshold.

[0171] In some example embodiments, the code rate threshold may be 1%.

[0172] In some example embodiments, the first apparatus may comprise a terminal device, and the second apparatus may comprise a network device.

[0173] In some example embodiments, a second apparatus capable of performing any of the method 700 (for example, the second apparatus 120 in FIG. 1 may comprise means for performing the respective operations of the method 700. The means may be implemented in any suitable form. For example, the means may be implemented in a circuitry or software module. The second apparatus may be implemented as or included in the second apparatus 120 in FIG. 1.

[0174] In some example embodiments, the second apparatus comprises means for transmitting, to a first apparatus, a transmission allocation via downlink control information, DCI; and means forreceiving, from the first apparatus, data transmission performed using a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training, the first MCS being determined based on the DCI.

[0175] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in a first MCS table, and the first MCS table may not comprise an index for a second MCS that is not designated for the data transmission for the data collection for the ML model training.

[0176] In some example embodiments, the second apparatus may further comprise: means for receiving, from the first apparatus, capability information indicating that the first apparatus supports the first MCS table.

[0177] In some example embodiments, each index in the first MCS table may indicate at least one of: a modulation order, a code rate, a time resource, or a frequency resource.

[0178] In some example embodiments, the DCI may comprise an indication indicating the first apparatus to perform the data transmission using at least one of: the first MCS table, or the first MCS indicated by an index in the first MCS table.

[0179] In some example embodiments, the indication may be comprised in a scheduling grant in the DCI.

[0180] In some example embodiments, the indication may further indicate that MCSs indicated by all indices in the first MCS table are to be used to perform the data transmission.

[0181] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be indicated by an index in a second MCS table, and the second MCS table may further comprise at least one index for at least one second MCS that is not designated for the data transmission for the data collection for the ML model training.

[0182] In some example embodiments, the DCI may comprise the index of the first MCS in the second MCS table.

[0183] In some example embodiments, a code rate for an initial transmission of the data transmission for the data collection for the ML model training may be predefined at the first apparatus, or determined based on configuration information received from the second apparatus.

[0184] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may be determined based on configuration information received from the second apparatus, and the configuration information may comprise one or more MCS indices designated for the data transmission.

[0185] In some example embodiments, the configuration information may be transmitted via a radio resource control, RRC, signaling to the first apparatus.

[0186] In some example embodiments, at least one of a modulation order or a code rate may beindicated by an index in a MCS table, and the index may indicate the first MCS designated for the data transmission for the data collection for the ML model training.

[0187] In some example embodiments, the second apparatus may further comprise: means for receiving, from the first apparatus, capability information indicating that the first apparatus supports the first MCS designated for the data transmission for the data collection for the ML model training.

[0188] In some example embodiments, the first MCS designated for the data transmission for the data collection for the ML model training may indicate at least one of a code rate or a modulation order, and the code rate is below a code rate threshold.

[0189] In some example embodiments, the code rate threshold may be 1%.

[0190] In some example embodiments, the second apparatus may further comprise: means for performing a cyclic redundancy check on the collected data; means for in accordance with a determination that the cyclic redundancy check is passed, reconstructing data bits associated with the collected data; and means for determining a training data sample based on the collected data and the data bits.

[0191] In some example embodiments, the second apparatus may further comprise: means for performing a cyclic redundancy check on the collected data; means for in accordance with a determination that the cyclic redundancy check is failed, transmitting to the first apparatus, a request for a retransmission of the data transmission; and means for receiving, from the first apparatus, the retransmission of the data transmission.

[0192] In some example embodiments, the first apparatus may comprise a terminal device, and the second apparatus may comprise a network device.

[0193] FIG. 8 is a simplified block diagram of a device 800 that is suitable for implementing example embodiments of the present disclosure. The device 800 may be provided to implement a communication device, for example, the first apparatus 110 or the second apparatus 120 as shown in FIG. 1. As shown, the device 800 includes one or more processors 810, one or more memories 820 coupled to the processor 810, and one or more communication modules 840 coupled to the processor 810.

[0194] The communication module 840 is for bidirectional communications. The communication module 840 has one or more communication interfaces to facilitate communication with one or more other modules or devices. The communication interfaces may represent any interface that is necessary for communication with other network elements. In some example embodiments, the communication module 840 may include at least one antenna.

[0195] The processor 810 may be of any type suitable to the local technical network and may include one or more of the following: general purpose computers, special purpose computers, microprocessors, digital signal processors (DSPs) and processors based on multicore processorarchitecture, as non-limiting examples. The device 800 may have multiple processors, such as an application specific integrated circuit chip that is slaved in time to a clock which synchronizes the main processor.

[0196] The memory 820 may include one or more non-volatile memories and one or more volatile memories. Examples of the non-volatile memories include, but are not limited to, a Read Only Memory (ROM) 824, an electrically programmable read only memory (EPROM), a flash memory, a hard disk, a compact disc (CD), a digital video disk (DVD), an optical disk, a laser disk, and other magnetic storage and / or optical storage. Examples of the volatile memories include, but are not limited to, a random-access memory (RAM) 822 and other volatile memories that will not last in the power-down duration.

[0197] A computer program 830 includes computer executable instructions that are executed by the associated processor 810. The instructions of the program 830 may include instructions for performing operations / acts of some example embodiments of the present disclosure. The program 830 may be stored in the memory, e.g., the ROM 824. The processor 810 may perform any suitable actions and processing by loading the program 830 into the RAM 822.

[0198] The example embodiments of the present disclosure may be implemented by means of the program 830 so that the device 800 may perform any process of the disclosure as discussed with reference to FIGS. 1 to 7. The example embodiments of the present disclosure may also be implemented by hardware or by a combination of software and hardware.

[0199] In some example embodiments, the program 830 may be tangibly contained in a computer readable medium which may be included in the device 800 (such as in the memory 820) or other storage devices that are accessible by the device 800. The device 800 may load the program 830 from the computer readable medium to the RAM 822 for execution. In some example embodiments, the computer readable medium may include any types of non-transitory storage medium, such as ROM, EPROM, a flash memory, a hard disk, CD, DVD, and the like. The term “non-transitory,” as used herein, is a limitation of the medium itself (i.e. , tangible, not a signal) as opposed to a limitation on data storage persistency (e.g., RAM vs. ROM).

[0200] FIG. 9 shows an example of the computer readable medium 900 which may be in form of CD, DVD or other optical storage disk. The computer readable medium 900 has the program 830 stored thereon.

[0201] Generally, various embodiments of the present disclosure may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. Some aspects may be implemented in hardware, and other aspects may be implemented in firmware or software which may be executed by a controller, microprocessor or other computing device. Although various aspects of embodiments of the present disclosure are illustrated and described as block diagrams, flowcharts, orusing some other pictorial representations, it is to be understood that the block, apparatus, system, technique or method described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.

[0202] Some example embodiments of the present disclosure also provide at least one computer program product tangibly stored on a computer readable medium, such as a non-transitory computer readable medium. The computer program product includes computer-executable instructions, such as those included in program modules, being executed in a device on a target physical or virtual processor, to carry out any of the methods as described above. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, or the like that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or split between program modules as desired in various embodiments. Machineexecutable instructions for program modules may be executed within a local or distributed device. In a distributed device, program modules may be located in both local and remote storage media.

[0203] Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. The program code may be provided to a processor or controller of a general-purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program code, when executed by the processor or controller, cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may execute entirely on a machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

[0204] In the context of the present disclosure, the computer program code or related data may be carried by any suitable carrier to enable the device, apparatus or processor to perform various processes and operations as described above. Examples of the carrier include a signal, computer readable medium, and the like.

[0205] The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable medium may include but not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the computer readable storage medium would include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random-access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0206] Further, although operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Likewise, although several specific implementation details are contained in the above discussions, these should not be construed as limitations on the scope of the present disclosure, but rather as descriptions of features that may be specific to particular embodiments. Unless explicitly stated, certain features that are described in the context of separate embodiments may also be implemented in combination in a single embodiment. Conversely, unless explicitly stated, various features that are described in the context of a single embodiment may also be implemented in a plurality of embodiments separately or in any suitable subcombination.

[0207] Although the present disclosure has been described in languages specific to structural features and / or methodological acts, it is to be understood that the present disclosure defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.

Claims

WHAT IS CLAIMED IS:

1. A first apparatus, comprising:at least one processor; andat least one memory storing instructions that, when executed by the at least one processor, cause the first apparatus at least to:receive, from a second apparatus, a transmission allocation via downlink control information, DCI;determine, based on the DCI, a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training; andperform, using the first MCS, the data transmission to the second apparatus.

2. The first apparatus of claim 1 , wherein the first MCS designated for the data transmission for the data collection for the ML model training is indicated by an index in a first MCS table, and the first MCS table does not comprise an index for a second MCS that is not designated for the data transmission for the data collection for the ML model training.

3. The first apparatus of claim 2, wherein the first apparatus is further caused to:transmit, to the second apparatus, capability information indicating that the first apparatus supports the first MCS table.

4. The first apparatus of claim 2 or 3, wherein each index in the first MCS table indicates at least one of:a modulation order,a code rate,a time resource, ora frequency resource.

5. The first apparatus of any of claims 2 to 4, wherein the DCI comprises an indication indicating the first apparatus to perform the data transmission using at least one of: the first MCS table, or the first MCS indicated by an index in the first MCS table.

6. The first apparatus of claim 5, wherein the indication is comprised in a scheduling grant in the DCI.

7. The first apparatus of claim 5 or 6, wherein the indication further indicates that MCSs indicated by all indices in the first MCS table are to be used to perform the data transmission.

8. The first apparatus of claim 1 , wherein the first MCS designated for the data transmission for the data collection for the ML model training is indicated by an index in a second MCS table, and the second MCS table further comprises at least one index for at least one second MCS that is not designated for the data transmission for the data collection for the ML model training.

9. The first apparatus of claim 8, wherein the DCI comprises the index of the first MCS in the second MCS table.

10. The first apparatus of claim 8 or 9, wherein a code rate for an initial transmission of the data transmission for the data collection for the ML training is predefined at the first apparatus, or determined based on configuration information received from the second apparatus.

11. The first apparatus of claim 8 or 9, wherein the first MCS designated for the data transmission for the data collection for the ML model training is determined based on configuration information received from the second apparatus, and the configuration information comprises one or more MCS indices designated for the data transmission.

12. The first apparatus of claim 10 or 11 , wherein the configuration information is received via a radio resource control, RRC, signaling from the second apparatus.

13. The first apparatus of any of claims 1 to 12, wherein at least one of a modulation order or a code rate is indicated by an index in a MCS table, and the index indicates the first MCS designated for the data transmission for the data collection for the ML model training.

14. The first apparatus of any of claims 1 to 12, wherein the first apparatus is further caused to: transmit, to the second apparatus, capability information indicating that the first apparatus supports the first MCS designated for the data transmission for the data collection for the ML model training.

15. The first apparatus of any of claims 1 to 12, wherein the first MCS designated for the data transmission for the data collection for the ML model training indicates at least one of a code rate or a modulation order, and the code rate is below a code rate threshold.

16. The first apparatus of claim 15, wherein the code rate threshold is 1%.

17. The first apparatus of any of claims 1 to 12, wherein the first apparatus comprises a terminal device, and the second apparatus comprises a network device.

18. A second apparatus, comprising:at least one processor; andat least one memory storing instructions that, when executed by the at least one processor, cause the second apparatus at least to:transmit, to a first apparatus, a transmission allocation via downlink control information, DCI; andreceive, from the first apparatus, data transmission performed using a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training, the first MCS being determined based on the DCI.

19. The second apparatus of claim 18, wherein the first MCS designated for the data transmission for the data collection for the ML model training is indicated by an index in a first MCS table, and the first MCS table does not comprise an index for a second MCS that is not designated for the data transmission for the data collection for the ML model training.

20. The second apparatus of claim 19, wherein the second apparatus is further caused to: receive, from the first apparatus, capability information indicating that the first apparatus supports the first MCS table.

21. The second apparatus of claim 19 or 20, wherein each index in the first MCS table indicates at least one of:a modulation order,a code rate,a time resource, ora frequency resource.

22. The second apparatus of any of claims 19 to 21 , wherein the DCI comprises an indication indicating the first apparatus to perform the data transmission using at least one of: the first MCS table, or the first MCS indicated by an index in the first MCS table.

23. The second apparatus of claim 22, wherein the indication is comprised in a scheduling grant in the DCI.

24. The second apparatus of claim 22 or 23, wherein the indication further indicates that MCSs indicated by all indices in the first MCS table are to be used to perform the data transmission.

25. The second apparatus of claim 18, wherein the first MCS designated for the data transmission for the data collection for the ML model training is indicated by an index in a second MCS table, and the second MCS table further comprises at least one index for at least one second MCS that is not designated for the data transmission for the data collection for the ML model training.

26. The second apparatus of claim 25, wherein the DCI comprises the index of the first MCS in the second MCS table.

27. The second apparatus of claim 25 or 26, wherein a code rate for an initial transmission of the data transmission for the data collection for the ML model training is predefined at the first apparatus, or determined based on configuration information received from the second apparatus.

28. The second apparatus of claim 25 or 26, wherein the first MCS designated for the data transmission for the data collection for the ML model training is determined based on configuration information received from the second apparatus, and the configuration information comprises one or more MCS indices designated for the data transmission.

29. The second apparatus of claim 27 or 28, wherein the configuration information is transmitted via a radio resource control, RRC, signaling to the first apparatus.

30. The second apparatus of any of claims 18 to 29, wherein at least one of a modulation order or a code rate is indicated by an index in a MCS table, and the index indicates the first MCS designated for the data transmission for the data collection for the ML model training.

31. The second apparatus of any of claim 18 to 29, wherein the second apparatus is further caused to:receive, from the first apparatus, capability information indicating that the first apparatus supports the first MCS designated for the data transmission for the data collection for the ML model training.

32. The second apparatus of any of claims 18 to 29, wherein the first MCS designated for the data transmission for the data collection for the ML model training indicates at least one of a code rate or a modulation order, and the code rate is below a code rate threshold.

33. The second apparatus of claim 32, wherein the code rate threshold is 1%.

34. The second apparatus of any of claims 18 to 29, wherein the second apparatus is further caused to:perform a cyclic redundancy check on the collected data;in accordance with a determination that the cyclic redundancy check is passed, reconstruct data bits associated with the collected data; anddetermine a training data sample based on the collected data and the data bits.

35. The second apparatus of any of claims 18 to 29, wherein the second apparatus is further caused to:perform a cyclic redundancy check on the collected data;in accordance with a determination that the cyclic redundancy check is failed, transmit, to the first apparatus, a request for a retransmission of the data transmission; andreceive, from the first apparatus, the retransmission of the data transmission.

36. The second apparatus of any of claims 18 to 29, wherein the first apparatus comprises a terminal device, and the second apparatus comprises a network device.

37. A method comprising:receiving, from a second apparatus, a transmission allocation via downlink control information, DCI;determining, based on the DCI, a first modulation and coding scheme, MCS, designated for data transmission for data collection for machine learning, ML, model training; andperforming, using the first MCS, the data transmission to the second apparatus.

38. A method comprising:transmitting, to a first apparatus, a transmission allocation via downlink control information, DCI; andreceiving, from the first apparatus, data transmission performed using a first modulation andcoding scheme, MCS, designated for data transmission for data collection for machine learning, ML,model training, the first MCS being determined based on the DCI.

Images