WO2025060348A1

WO2025060348A1 - Methods, devices, and computer readable medium for artificial intelligence (ai) service

Info

Publication number: WO2025060348A1
Application number: PCT/CN2024/078996
Authority: WO
Inventors: Hao Tang; Jianglei Ma; Peiying Zhu; Wen Tong
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2023-09-20
Filing date: 2024-02-28
Publication date: 2025-03-27
Anticipated expiration: 2026-03-20

Abstract

Example embodiments relate to artificial intelligence (AI) service. In a method, a terminal device receives a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature. The terminal device trains or updates the sub-model and reports information associated with the sub-model. In this way, a scheme of the life cycle management for AI as a service is designed. In particular, distributed learning by multiple devices is enabled, and each device may train or update part of the AI model, thereby improving the efficiency of the AI service management.

Description

METHODS, DEVICES, AND COMPUTER READABLE MEDIUM FOR ARTIFICIAL INTELLIGENCE (AI) SERVICE

CORSS-REFERENCES TO RELATED APPLICATIONS

This application claims the benefit and priority to U.S. Provisional Patent Application No. 63/584,066 filed September 20, 2023, the content of which is incorporated herein by reference in its entirety.

FIELD

Example embodiments of the present disclosure generally relate to the field of communications, and in particular, to methods, devices, and a non-transitory computer readable medium for artificial intelligence (AI) service.

BACKGROUND

Artificial intelligence (AI) , and in particular deep machine learning, is a wide-ranging branch of computer science concerned with building smart machines capable of performing tasks that typically require human intelligence. It is expected that the introduction of AI will create a paradigm shift in virtually every sector of the tech industry and AI is expected to play a role in advancement of network technologies. For example, existing communication techniques, which rely on classical analytical modeling of channels, have enabled wireless communications to take place at close to the theoretical Shannon limit. To further maximize efficient use of the signal space, existing techniques may be unsatisfactory. AI is expected to help address this challenge. Other aspects of wireless communication may benefit from the use of AI, particularly in future generations of wireless technologies, such as technologies in advanced fifth generation (5G) , future sixth generation (6G) systems, and beyond.

To support the use of AI in a wireless network, an appropriate AI framework is needed. However, the wireless technology mainly considers the AI use cases to improve network performance. Studies about supporting the network to provide AI services to the devices are needed.

SUMMARY

In general, example embodiments of the present disclosure provide a solution for AI service, especially for life cycle management of an AI service.

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.

In a first aspect, there is provided a method implemented at a terminal device. In the method, the terminal device receives a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature. The terminal device trains or updates the sub-model and reports information associated with the sub-model. In this way, a scheme of the life cycle management for AI as a service is designed. In particular, distributed learning by multiple devices is enabled, and each device may train or update part of the AI model, thereby improving the efficiency of the AI service management.

In some implementations, the sub-model is associated with one of the following: at least one parameter among a plurality of parameters associated with the AI model; a functionality among a plurality of functionalities associated with the AI model based on a one-to-one correspondence; at least one functionality among a plurality of functionalities associated with the AI model; an AI task among a plurality of AI tasks associated with the AI model based on a one-to-one correspondence; at least one AI task among a plurality of AI tasks associated with the AI model; a predefined size requirement; or a predefined neuron network algorithm. In this way, on the basis that the AI model is consisted of a plurality of distributed sub-models, the sub-models of the AI model can be divided in a flexible manner.

In some implementations, the sub-model is associated with a first subset of parameters among a plurality of parameters associated with the AI model. The sub-model is a first sub-model. The method further comprises: receiving a configuration of a second sub-model of the AI model, wherein the second sub-model is associated with a second subset of parameters among the plurality of parameters. The first subset of parameters and the second subset of parameters are non-overlapped; or the first subset of parameters and the second subset of parameters are overlapped. In this way, the terminal device is able to train or update different parts of the AI model either overlapped or non-overlapped, thus improving the flexibility of the AI service management.

In some implementations, receiving the configuration of the sub-model comprises: receiving at least one configuration of at least one sub-model, wherein the at least one sub-model comprises the sub-model. In this way, the terminal device may be configured with one or more sub-models to be trained or updated.

In some implementations, receiving the configuration of the sub-model comprises: receiving a plurality of configurations of a plurality of candidate sub-models; and receiving at least one configuration of at least one sub-model among the plurality of candidate sub-models, wherein the at least one sub-model comprises the sub-model. For example, the terminal device may be configured with candidate sub-models in a broadcast or a groupcast manner and may then be indicated about the sub-model (s) to be trained or updated among the configured candidate sub-models in a unicast or groupcast manner. In this way, the sub-model (s) to be trained or updated by the terminal device may be configured in a flexible manner.

In some implementations, the sub-model is identified based on an index. The index is unique in at least one set of sub-models of at least one AI models associated with one or more AI features. In this way, the sub-model may be identified by the corresponding index. The terminal device may identify the sub-model based on the sub-model index which is globally unique for multiple AI features. Thus, the resource overhead for configuring the sub-model (s) to be trained or updated may be reduced.

In some implementations, the sub-model is identified based on an index of the sub-model and an index of the AI feature. In this way, the sub-model may be identified by the corresponding sub-model index and AI feature index. The terminal device may identify the sub-model to be trained or updated based on the AI feature index and the sub-model index which is unique within the AI feature. Thus, the resource overhead for configuring the sub-model (s) to be trained or updated may be reduced.

In some implementations, training or updating the sub-model comprises: receiving an indication of activating the sub-model. In this way, the configured sub-model may be activated for the training or the updating based on the activation indication.

In some implementations, the sub-model is a first sub-model, and the method further comprises: upon determining that a second sub-model of the AI model is being activated when the indication is received, deactivating the second sub-model. In this way, a sub-model currently being activated may be deactivated if another sub-model is to be activated. Thus, only one sub-model may be activated for the training or the updating for a given time. The active sub-model switching may be triggered by the activation indication of the sub-model to be trained or updated.

In some implementations, training or updating the sub-model comprises: receiving an indication of activating the AI feature, wherein the sub-model is a default sub-model of the AI feature. In this way, an AI feature may be activated or deactivated. If an AI feature is activated and no activation indication of a sub-model for the AI feature is received, a default sub-model of the AI feature may be activated and then be trained or updated accordingly.

In some implementations, training or updating the sub-model comprises: receiving an indication of activating at least one sub-model of the AI model, wherein the at least one sub-model comprises the sub-model; and training or updating the at least one sub-model. In this way, multiple sub-models may be simultaneously trained or updated by the terminal device.

In some implementations, a number of the at least one sub-model is smaller than a pre-defined number. In this way, efficiency of the AI service management may be guaranteed.

In some implementations, the method further comprises: transmitting assistance information. The assistance information comprises at least one of the following: a computing capability of the terminal device; or a size of a dataset of the terminal device for model training or model updating. In this way, the network device may configure the sub-model (s) to be trained or updated by the terminal device based on the assistance information reported by the terminal device, thus guaranteeing the efficiency of the AI service management.

In some implementations, the sub-model is a first sub-model. Training or updating the sub-model comprises: training or updating the first sub-model with input data associated with the AI model. The method further comprises: training or updating a second sub-model of the AI model with the input data associated with the AI model. In this way, the terminal device may train or update multiple sub-models using the same input data.

In some implementations, the method further comprises: receiving an indication of at least one of a format of the input data for the AI model or a format of an output data for the AI model. In this way, different sub-models of the AI model may have the same input data format or the same output data format.

In some implementations, the sub-model is a first sub-model. Training or updating the sub-model comprises: training or updating the first sub-model with first input data. The method further comprises: training or updating a second sub-model of the AI model with second input data different from the first input data. In this way, the terminal device may train or update sub-models using different input data for the sub-models.

In some implementations, the method further comprises: receiving an indication of at least one of a format of the first input data or a format of an output data for the sub-model. In this way, different sub-models of the AI model may have different input data formats or different output data format.

In some implementations, reporting the information associated with the sub-model comprises: compressing at least one parameter of the sub-model in a first compression; and transmitting the at least one compressed parameter. The sub-model is a first sub-model. The method further comprises: receiving a configuration of a second sub-model of the AI model; compressing at least one parameter of the second sub-model in a second compression scheme; and transmitting the at least one compressed parameter of the second sub-model. In this way, information of the sub-model may be reported in a compression manner, thus reducing the resource overhead.

In some implementations, the first compression scheme is the same as the second compression scheme. Alternatively, the first compression scheme is different from the second compression scheme. In this way, information of different sub-models may be reported in the same compression scheme or different compression schemes.

In some implementations, the method further comprises: determining a performance of the sub-model based on a test dataset; transmitting an indication of the performance of the sub-model. Reporting the information associated with the sub-model comprise: transmitting at least one parameter of the sub-model upon determining that an indication to report the information associated with the sub-model is received. The method further comprises: continuing training or updating the sub-model upon determining that an indication to not report the information associated with the sub-model is received. In this way, the terminal device may validate the trained or updated sub-model and determine whether the trained or updated sub-model needs to be reported based on the performance of the sub-model. The information of the sub-model would not be reported until the performance of the sub-model is validated to be good. The resource overhead for the life cycle management of the AI service may thus be reduced.

In some implementations, determining a performance of the sub-model comprises: receiving the test dataset for the sub-model; receiving an indication to perform a validation operation for the sub-model; and determining the performance of the sub-model based on determining that the indication to perform the validation operation is received. In this way, the sub-model may be validated by the terminal device before its information is reported to the network.

In some implementations, the sub-model is associated with a dedicated radio network temporary identifier (RNTI) . In this way, a dedicated identifier for the sub-model may be provided.

In some implementations, the method further comprises: receiving downlink scheduling information (DCI) to schedule a resource for the sub-model, wherein a cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI; and receiving at least one parameter of the sub-model using the resource scheduled by the DCI. In this way, the terminal device configured with the RNTI for the sub-model may be able to decode the DCI and receive the sub-model. A scheme for delivering the sub-model to the terminal device (s) selected to train or update the sub-model may be provided.

In some implementations, the method further comprises: receiving downlink scheduling information (DCI) to schedule a resource for reporting the information associated with the sub-model, wherein a cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI. Reporting the information associated with the sub-model comprises: transmitting at least one parameter of the sub-model using the resource scheduled by the DCI. In this way, the terminal device configured with the RNTI for the sub-model may be able to decode the DCI and report the trained or updated sub-model. A scheme for reporting the sub-model by the terminal device (s) selected to train or update the sub-model may be provided.

In some implementations, the method further comprises: transmitting a scheduling request for reporting the information associated with the sub-model, wherein the scheduling request is associated with the sub-model. In this way, the terminal device may request to report the trained or updated sub-model and the network device may allocate resources for the reporting accordingly, thus reducing the resource overhead for the reporting the sub-model.

In some implementations, the method further comprises: performing a federated learning of the sub-model by iteratively receiving at least one parameter of the sub-model, training or updating the sub-model and transmitting the at least one trained or updated parameter. In this way, the AI model may be trained or updated based on the federated learning of the sub-models.

In a second aspect, there is provided a method implemented at a network device. In the method, the network device transmits a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature. The network device receives a report of information associated with the sub-model, wherein the sub-model is trained or updated by a terminal device. In this way, a scheme of the life cycle management for AI as a service is designed. In particular, distributed learning by multiple devices is enabled, and each device may train or update part of the AI model, thereby improving the efficiency of the AI service management.

In some implementations, the sub-model is associated with a first subset of parameters among a plurality of parameters associated with the AI model. The sub-model is a first sub-model. The method further comprises: transmitting a configuration of a second sub-model of the AI model, wherein the second sub-model is associated with a second subset of parameters among the plurality of parameters. The first subset of parameters and the second subset of parameters are non-overlapped; or the first subset of parameters and the second subset of parameters are overlapped. In this way, the terminal device is able to train or update different parts of the AI model either overlapped or non-overlapped, thus improving the flexibility of the AI service management.

In some implementations, transmitting the configuration of the sub-model comprises: transmitting at least one configuration of at least one sub-model, wherein the at least one sub-model comprises the sub-model. In this way, the selected device may be configured with one or more sub-models to be trained or updated.

In some implementations, transmitting the configuration of the sub-model comprises: transmitting a plurality of configurations of a plurality of candidate sub-models; and transmitting at least one configuration of at least one sub-model among the plurality of candidate sub-models, wherein the at least one sub-model comprises the sub-model. For example, the network device may be broadcast or a groupcast candidate sub-models and may then indicate to the selected device the sub-model (s) to be trained or updated among the configured candidate sub-models in a unicast or groupcast manner. In this way, the sub-model (s) to be trained or updated by the selected device may be configured in a flexible manner.

In some implementations, the sub-model is identified based on an index. The index is unique in at least one set of sub-models of at least one AI models associated with one or more AI features. In this way, the sub-model may be identified by the corresponding index which is globally unique for multiple AI features. Thus, the resource overhead for configuring the sub-model (s) to be trained or updated may be reduced.

In some implementations, the sub-model is identified based on an index of the sub-model and an index of the AI feature. In this way, the sub-model may be identified by the corresponding AI feature index and the sub-model index which is unique within the AI feature. Thus, the resource overhead for configuring the sub-model (s) to be trained or updated may be reduced.

In some implementations, the method further comprises: transmitting an indication of activating the sub-model. In this way, the configured sub-model may be activated for the training or the updating based on the activation indication.

In some implementations, the method further comprises: transmitting an indication of activating the AI feature, wherein the sub-model is a default sub-model of the AI feature. In this way, an AI feature may be activated or deactivated. If an AI feature is activated and no activation indication of a sub-model for the AI feature is transmitted, a default sub-model of the AI feature may be activated and then be trained or updated accordingly.

In some implementations, the method further comprises: transmitting an indication of activating at least one sub-model of the AI model, wherein the at least one sub-model comprises the sub-model. In this way, multiple sub-models may be simultaneously trained or updated by the selected device.

In some implementations, the number of the at least one sub-model is smaller than a pre-defined number. In this way, efficiency of the AI service management may be guaranteed.

In some implementations, the method further comprises: receiving assistance information, wherein the assistance information comprises at least one of the following: a computing capability of the network device; or a size of a dataset of the network device for model training or model updating. In this way, the network device may configure the sub-model (s) to be trained or updated by the terminal device based on the assistance information reported by the terminal device, thus guaranteeing the efficiency of the AI service management.

In some implementations, the method further comprises: transmitting an indication of at least one of a format of input data for the AI model or a format of an output data for the AI model. In this way, different sub-models of the AI model may have the same input data format or the same output data format.

In some implementations, the method further comprises: transmitting an indication of at least one of a format of input data for the sub-model or a format of an output data for the sub-model. In this way, different sub-models of the AI model may have different input data formats or different output data format.

In some implementations, receiving the report of the information associated with the sub-model comprises: receiving at least one parameter of the sub-model. The at least one parameter of the sub-model is compressed in a first compression scheme, the sub-model is a first sub-model. The method further comprises: receiving at least one parameter of a second sub-model of the AI model, wherein the at least one parameter of the second sub-model is compressed in a second compression scheme. In this way, information of the sub-model may be reported in a compression manner, thus reducing the resource overhead.

In some implementations, the method further comprises: receiving an indication of a performance of the sub-model; determining whether to report the information associated with the sub-model based on the performance of the sub-model; and transmitting an indication to report the information associated with the sub-model or an indication not to report the information associated with the sub-model based on the determination. In this way, the network device may determine whether the trained or updated sub-model needs to be reported based on the performance of the sub-model. The resource overhead for the life cycle management of the AI service may thus be reduced.

In some implementations, the method further comprises: transmitting a test dataset for determining the performance of the sub-model; and transmitting an indication to perform a validation operation for the sub-model. In this way, the sub-model may be validated by the selected device before its information is reported by the selected device. The resource overhead for the life cycle management of the AI service may thus be reduced.

In some implementations, the method further comprises: transmitting downlink scheduling information (DCI) to schedule a resource for the sub-model, wherein a cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI; and transmitting at least one parameter of the sub-model using the resource scheduled by the DCI. In this way, the device (s) configured with the RNTI for the sub-model may be able to decode the DCI and receive the sub-model. A scheme for delivering the sub-model to the device (s) selected to train or update the sub-model may be provided.

In some implementations, the method further comprises: transmitting downlink scheduling information (DCI) to schedule a resource for the report of the information associated with the sub-model, wherein a cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI. In this way, the device (s) configured with the RNTI for the sub-model may be able to decode the DCI and report the trained or updated sub-model. A scheme for reporting the sub-model by the terminal device (s) selected to train or update the sub-model may be provided.

In some implementations, the method further comprises: receiving a scheduling request for the report of the information associated with the sub-model, wherein the scheduling request is associated with the sub-model. In this way, the selected device (s) training or updating the sub-model may request to report the trained or updated sub-model and the network device may allocate resources for the reporting accordingly, thus reducing the resource overhead for the reporting the sub-model.

In a third aspect, there is provided a terminal device. The terminal device comprises a transceiver and a processor communicatively coupled with the transceiver. The processor is configured to receive, via the transceiver, a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature; train or update the sub-model; and report information associated with the sub-model. In this way, a scheme of the life cycle management for AI as a service is designed. In particular, distributed learning by multiple devices is enabled, and each device may train or update part of the AI model, thereby improving the efficiency of the AI service management.

In a fourth aspect, there is provided a network device. The network device comprises a transceiver and a processor communicatively coupled with the transceiver. The processor is configured to transmit, via the transceiver, a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature; and receive, via the transceiver, a report of information associated with the sub-model, wherein the sub-model is trained or updated by a terminal device. In this way, a scheme of the life cycle management for AI as a service is designed. In particular, distributed learning by multiple devices is enabled, and each device may train or update part of the AI model, thereby improving the efficiency of the AI service management.

In a fifth aspect, there is provided a non-transitory computer readable medium. The non-transitory computer readable medium comprises computer program stored thereon, the computer program, when executed on at least one processor, causing the at least one processor to perform the method of the first aspect, the second aspect, or any possible implementation of the first aspect or the second aspect.

In a sixth aspect, there is provided a chip. The chip comprising at least one processing circuit configured to perform the method of the first aspect, the second aspect, or any possible implementation of the first aspect or the second aspect.

In a seventh aspect, there is provided a system. The system comprising at least one terminal device of the third aspect and the at least one network device of the fourth aspect.

In an eighth aspect, there is provided a computer program product tangibly stored on a computer-readable medium and comprising computer-executable instructions which, when executed, cause an apparatus to perform the method of the first aspect, the second aspect, or any possible implementation of the first aspect or the second aspect.

BRIEF DESCRIPTION OF THE DRAWINGS

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

FIG. 1A illustrates an example environment in which some embodiments of the present disclosure can be implemented;

FIG. 1B illustrates an example communication system in which some embodiments of the present disclosure can be implemented;

FIG. 1C illustrates example devices in the example environments of FIG. 1A and FIG. 1B;

FIG. 1D illustrates example modules in the devices of the present disclosure;

FIG. 1E illustrates another example communication system in which some embodiments of the present disclosure can be implemented;

FIG. 1F illustrates an example sensing management function (SMF) of the present disclosure;

FIGS. 2A to 2C illustrate example distributed sub-models of an AI model according to some embodiments of the present disclosure;

FIG. 3 illustrates a signaling process for training or updating sub-models of an AI model according to some embodiments of the present disclosure;

FIG. 4 illustrates example distributed sub-models of an AI model associated with corresponding RNTIs according to some embodiments of the present disclosure;

FIG. 5 illustrates a flowchart of an example method implemented at a terminal device according to some embodiments of the present disclosure;

FIG. 6 illustrates a flowchart of an example method implemented at a network device according to some embodiments of the present disclosure;

FIG. 7 is a block diagram of a device that may be used for implementing some embodiments of the present disclosure;

FIG. 8 is a schematic diagram of a structure of an apparatus in accordance with some embodiments of the present disclosure; and

FIG. 9 is a schematic diagram of a structure of another apparatus in accordance with some embodiments of the present disclosure.

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

DETAILED DESCRIPTION

Principles 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 of the present disclosure described herein can be implemented in various manners other than the ones specifically described below.

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 skill in the art to which this disclosure belongs.

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. The term “another embodiment” is to be read as “at least one other embodiment. ” 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 within the knowledge of one skilled in the art to adapt or modify such feature, structure, or characteristic in connection with other embodiments, whether or not such adaptations are explicitly described.

It shall be understood that although the terms “first” and “second” etc. 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. 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. Other definitions, explicit and implicit, may be included below.

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.

As used herein, the term “communication network” refers to a network following any suitable communication standards, such as 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-IoT) , Wireless Fidelity (WiFi) 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 fourth generation (4G) , 4.5G, the future fifth generation (5G) , IEEE 802.11 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.

As used herein, the term “network device” refers to a device which is capable of providing or hosting a cell or coverage area where terminal devices can communicate. Examples of a network device include, but are not limited to, a Node B (NodeB or NB) , an evolved NodeB (eNodeB or eNB) , a next generation NodeB (gNB) , a transmission reception point (TRP) , a remote radio unit (RRU) , a radio head (RH) , a remote radio head (RRH) , an IAB node, a low power node such as a femto node or a pico node, a reconfigurable intelligent surface (RIS) , network-controlled repeaters, and the like.

As used herein, the term “terminal device” refers to any device having wireless or wired communication capabilities. Examples of the terminal device include, but not limited to, user equipment (UE) , personal computers, desktops, mobile phones, cellular phones, smart phones, personal digital assistants (PDAs) , portable computers, tablets, wearable devices, internet of things (IoT) devices, Ultra-reliable and Low Latency Communications (URLLC) devices, Internet of Everything (IoE) devices, machine type communication (MTC) devices, devices for vehicle to everything (V2X) communication, devices for Integrated Access and Backhaul (IAB) , devices for Small Data Transmission (SDT) , devices for mobility, devices for Multicast and Broadcast Services (MBS) , devices for positioning, devices for dynamic/flexible duplexing in commercial networks, reduced capability (RedCap) devices, space-borne vehicles or air-borne vehicles in non-terrestrial networks (NTN) including satellites and High Altitude Platforms (HAPs) encompassed in Unmanned Aircraft Systems (UAS) , eXtended Reality (XR) devices including different types of realities such as Augmented Reality (AR) , Mixed Reality (MR) and Virtual Reality (VR) , an unmanned aerial vehicle (UAV) , a drone, devices on high speed train (HST) , image capture devices such as digital cameras, sensors, gaming devices, music storage and playback devices, Internet-connected appliances, and the like. The terminal device may further include a “multicast/broadcast” feature to support public safety and/or mission critical applications. The terminal device may further include transparent IPv4/IPv6 multicast delivery such as for IPTV, smart TV, radio services, software delivery over wireless, group communications, and IoT applications. The terminal may be incorporate a Subscriber Identity Module (SIM) or multiple SIMs, also known as Multi-SIM. The term “terminal device” can also be used interchangeably with variations of some of all of the preceding terms, such as a UE, a mobile station, a subscriber station, a mobile terminal, a user terminal, a wireless device, or a reduced capability terminal device.

The terminal device or the network device may have artificial intelligence (AI) or machine learning (ML) capability. AI/ML generally refers to a model which has been trained from numerous collected data for a specific function, and can be used to predict some information. The terminal or the network device may function in several frequency ranges, e.g. FR1 (410 MHz –7125 MHz) , FR2 (24.25 GHz to 71 GHz) , 71 GHz to 114 GHz, and ranges of frequencies greater than 100 GHz, including Tera Hertz (THz) frequencies. The terminal or the network device can further function in licensed, unlicensed, or shared spectra. The terminal device may have multiple connections with multiple network devices, such as under a Multi-Radio Dual Connectivity (MR-DC) application scenario. The terminal device or the network device may be capable of advanced duplexing functions, such as full duplex, flexible duplex, and cross-division duplex (XDD) modes.

The network device may have functions or capabilities for network energy saving, self-organizing network (SON) automation, or minimization of drive tests (MDT) mechanisms. The terminal device may have functions or capabilities for power saving.

The embodiments of the present disclosure may be performed in test equipment, e.g. a signal generator, a signal analyzer, a spectrum analyzer, a network analyzer, a test terminal device, a test network device, and a channel emulator.

The embodiments of the present disclosure may be performed according to communication protocols of any generation either currently known or to be developed in the future. Examples of these communication protocols include, but are not limited to, cellular protocols including the first generation (1G) , the second generation (2G, 2.5G, 2.75G) , the third generation (3G) , the fourth generation (4G, sometimes known as “LTE” , 4.5G, sometimes known as “LTE Advanced” and “LTE Advanced Pro” ) , the fifth generation (5G, sometimes known as “NR” , 5.5G, 5G-Advanced) , and the sixth generation (6G) , as well as various generations of Wireless Fidelity (WiFi) , and Ultra Wideband (UWB) .

In one embodiment, the terminal device may be connected to a first network device and a second network device. One of the first network device and the second network device may be a master node and the other one may be a secondary node. The first network device and the second network device may use different radio access technologies (RATs) . In one embodiment, the first network device may be a first RAT device and the second network device may be a second RAT device. In one embodiment, the first RAT device is eNB and the second RAT device is gNB. In another embodiment, the first RAT device is 5G network device and the second RAT device is a 6G network device. Information related to different RATs may be transmitted to the terminal device from at least one of the first network device and the second network device. In one embodiment, first information may be transmitted to the terminal device from the first network device and second information may be transmitted to the terminal device from the second network device directly or via the first network device. In one embodiment, information related to configuration for the terminal device, and configured by the second network device, may be transmitted from the second network device via the first network device. Information related to reconfiguration for the terminal device, and configured by the second network device, may be transmitted to the terminal device from the second network device directly or via the first network device.

In some examples, values, procedures, or apparatus may be referred to as “best, ” “lowest, ” “highest, ” “minimum, ” “maximum, ” or the like. It will be appreciated that such descriptions are intended to indicate that a selection among many functional alternatives may be made; however, and such selections may be superlatives in some respects but need not be better, smaller, higher, or otherwise preferable to other selections in other respects.

In some embodiments of this disclosure, the AI service includes the service provided to the devices (for example, a terminal devices) , and the service is supported by AI capability. In some examples, the AI service may include, but not limited to, image recognition, voice recognition, intelligent question and answer and so on. The AI service may be also referred to the “AI feature” in some embodiments of this disclosure. Moreover, without any limitation, a device can be provided with one or more AI services (AI features) .

In some embodiments of this disclosure, one AI service/feature may be implemented by means of one or more AI models. As mentioned above, in the case that the AI model is trained in the distribution manner, an AI model of the one or more AI models may consist of a plurality of sub-models. In some embodiments of this disclosure, the sub-models of the AI model may be also referred to as “Model Part (MoP) ” of the AI model. Furthermore, the (whole) AI model may be also referred to as “big model” or “parent model” . In addition, the terms “AI model” and “AI/Machine Learning (ML) model” may be used interchangeably.

As mentioned above, in communication systems (for example, the 5G wireless technology) , the AI capabilities are utilized to improve network performance in most cases. However, the devices (for example, the terminal device, UE, customer premise equipment CPE, computing node) in the network cannot adequately benefit from the development of AI technology in these cases. Therefore, to support the use of AI in a wireless network, an appropriate AI framework is needed. Considering that there are massive devices in the communication network which have data and computing capability, by distributed training/inference, AI service may be provided without consuming considerable compute and storage resources. However, once the AI model for the AI service is distributed over the massive devices in the “MoP” manner, managing of the MoPs should be considered.

In view of the above, example embodiments of the present disclosure propose a solution for the AI service. In this solution, a terminal device receives a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature. The terminal device trains or updates the sub-model and reports information associated with the sub-model. Specifically, a (parent/big) AI model may consist of a plurality of “sub-models” of the AI model which are distributed over the massive devices. The AI model can be distributed and trained over massive devices in a network which have data and computing capability. A device in the network may be configured with one or more “sub-models” of the AI model. As such, in addition to the network performance enhancement, the devices in the network may also participate in the life cycle management of the AI service by training or updating the AI model distributed over the devices. In this way, a scheme of the life cycle management for AI as a service is designed. In particular, distributed learning by multiple devices is enabled, and each device may train or update part of the AI model, thereby improving the efficiency of the AI service management.

For illustrative purposes, principles and example embodiments of the present disclosure will be described below with reference to FIGS. 1A-9. However, it is to be noted that these embodiments are given to enable the person skilled in the art to understand inventive concepts of the present disclosure and implement the solution as proposed herein, and are not intended to limit the scope of the present disclosure in any way to explicitly illustrated structures and combinations of features.

FIG. 1A illustrates an example environment 100A in which some embodiments of the present disclosure can be implemented. Referring to FIG. 1A, as an illustrative example without limitation, a simplified schematic illustration of a communication system is provided. The communication system 100 comprises a radio access network 120. The radio access network 120 may be a next generation (e.g. sixth generation (6G) or later) radio access network, or another generation (e.g. 5G, 4G, 3G or 2G) of radio access network. One or more communication electric device (ED) 110a, 110b, 110c, 110d, 110e, 110f, 110g, 110h, 110i, 110j (generically referred to as 110) may be interconnected to one another or connected to one or more network nodes (170a, 170b, generically referred to as 170) in the radio access network 120. A core network 130 may be a part of the communication system and may be dependent or independent of the radio access technology used in the communication system 100. Also the communication system 100 comprises a public switched telephone network (PSTN) 140, the internet 150, and other networks 160.

FIG. 1B illustrates an example system 100B in which some embodiments of the present disclosure can be implemented. In general, the communication system 100B enables multiple wireless or wired elements to communicate data and other content. The purpose of the communication system 100B may be to provide content, such as voice, data, video, signaling and/or text, via broadcast, multicast and unicast, etc. The communication system 100B may operate by sharing resources, such as carrier spectrum bandwidth, between its constituent elements. The communication system 100B may include a terrestrial communication system and/or a non-terrestrial communication system. The communication system 100 may provide a wide range of communication services and applications (such as earth monitoring, remote sensing, passive sensing and positioning, navigation and tracking, autonomous delivery and mobility, etc. ) . The communication system 100 may provide a high degree of availability and robustness through a joint operation of a terrestrial communication system and a non-terrestrial communication system. For example, integrating a non-terrestrial communication system (or components thereof) into a terrestrial communication system can result in what may be considered a heterogeneous network comprising multiple layers. Compared to conventional communication networks, the heterogeneous network may achieve better overall performance through efficient multi-link joint operation, more flexible functionality sharing, and faster physical layer link switching between terrestrial networks and non-terrestrial networks.

The terrestrial communication system and the non-terrestrial communication system could be considered sub-systems of the communication system. In the example shown in FIG. 1B, the communication system 100 includes electronic devices (ED) 110a, 110b, 110c, 110d (generically referred to as ED 110) , radio access networks (RANs) 120a-120b, a non-terrestrial communication network 120c, a core network 130, a public switched telephone network (PSTN) 140, the Internet 150, and other networks 160. The RANs 120a-120b include respective base stations (BSs) 170a-170b, which may be generically referred to as terrestrial transmit and receive points (T-TRPs) 170a-170b. The non-terrestrial communication network 120c includes an access node 172, which may be generically referred to as a non-terrestrial transmit and receive point (NT-TRP) 172. Without any limitation, the above EDs 110, TRPs 170, RANs 120, core network 130, PSTN 140, Internet 150 and other networks 160 in FIG. 1B may be the corresponding devices, stations, RAN, networks in FIG. 1A. Alternatively, the above EDs 110, TRPs 170, RANs 120, core network 130, PSTN 140, Internet 150 and other networks 160 in FIG. 1B may be the devices, stations, RAN, networks other than FIG. 1A.

Any ED 110 may be alternatively or additionally configured to interface, access, or communicate with any T-TRP 170a-170b and NT-TRP 172, the Internet 150, the core network 130, the PSTN 140, the other networks 160, or any combination of the preceding. In some examples, ED 110a may communicate an uplink and/or downlink transmission over a terrestrial air interface 190a with T-TRP 170a. In some examples, the EDs 110a, 110b, 110c and 110d may also communicate directly with one another via one or more sidelink air interfaces 190b. In some examples, ED 110d may communicate an uplink and/or downlink transmission over a non-terrestrial air interface 190c with NT-TRP 172.

The air interfaces 190a and 190b may use similar communication technology, such as any suitable radio access technology. For example, the communication system 100 may implement one or more channel access methods, such as code division multiple access (CDMA) , space division multiple access (SDMA) , time division multiple access (TDMA) , frequency division multiple access (FDMA) , orthogonal FDMA (OFDMA) , Direct Fourier Transform spread OFDMA (DFT-OFDMA) or single-carrier FDMA (SC-FDMA) in the air interfaces 190a and 190b. The air interfaces 190a and 190b may utilize other higher dimension signal spaces, which may involve a combination of orthogonal and/or non-orthogonal dimensions.

The non-terrestrial air interface 190c can enable communication between the ED 110d and one or multiple NT-TRPs 172 via a wireless link or simply a link. For some examples, the link is a dedicated connection for unicast transmission, a connection for broadcast transmission, or a connection between a group of EDs 110 and one or multiple NT-TRPs 172for multicast transmission.

The RANs 120a and 120b are in communication with the core network 130 to provide the EDs 110a 110b, and 110c with various services such as voice, data, and other services. The RANs 120a and 120b and/or the core network 130 may be in direct or indirect communication with one or more other RANs (not shown) , which may or may not be directly served by core network 130, and may or may not employ the same radio access technology as RAN 120a, RAN 120b or both. The core network 130 may also serve as a gateway access between (i) the RANs 120a and 120b or EDs 110a 110b, and 110c or both, and (ii) other networks (such as the PSTN 140, the Internet 150, and the other networks 160) . In addition, some or all of the EDs 110a 110b, and 110c may include functionality for communicating with different wireless networks over different wireless links using different wireless technologies and/or protocols. Instead of wireless communication (or in addition thereto) , the EDs 110a 110b, and 110c may communicate via wired communication channels to a service provider or switch (not shown) , and to the Internet 150. PSTN 140 may include circuit switched telephone networks for providing plain old telephone service (POTS) . Internet 150 may include a network of computers and subnets (intranets) or both, and incorporate protocols, such as Internet Protocol (IP) , Transmission Control Protocol (TCP) , User Datagram Protocol (UDP) . EDs 110a 110b, and 110c may be multimode devices capable of operation according to multiple radio access technologies, and incorporate multiple transceivers necessary to support such.

FIG. 1C illustrates example devices in the example environments of FIG. 1A and FIG. 1B. Specifically, FIG. 1C illustrates another example of the ED 110 and a base station 170a, 170b and/or 170c according to some embodiments of this disclosure. The ED 110 is used to connect persons, objects, machines, etc. The ED 110 may be widely used in various scenarios, for example, cellular communications, device-to-device (D2D) , vehicle to everything (V2X) , peer-to-peer (P2P) , machine-to-machine (M2M) , machine-type communications (MTC) , Internet of things (IOT) , virtual reality (VR) , augmented reality (AR) , mixed reality (MR) , metaverse, digital twin, industrial control, self-driving, remote medical, smart grid, smart furniture, smart office, smart wearable, smart transportation, smart city, drones, robots, remote sensing, passive sensing, positioning, navigation and tracking, autonomous delivery and mobility, etc.

Each ED 110 represents any suitable end user device for wireless operation and may include such devices (or may be referred to) as a user equipment/device (UE) , a wireless transmit/receive unit (WTRU) , a mobile station, a fixed or mobile subscriber unit, a cellular telephone, a station (STA) , a machine type communication (MTC) device, a personal digital assistant (PDA) , a smartphone, a laptop, a computer, a tablet, a wireless sensor, a consumer electronics device, a smart book, a vehicle, a car, a truck, a bus, a train, or an IoT device, wearable devices such as a watch, head mounted equipment, a pair of glasses, an industrial device, or apparatus (e.g. communication module, modem, or chip) in the forgoing devices, among other possibilities. Future generation EDs 110 may be referred to using other terms. Each base station 170a and 170b is a T-TRP and will hereafter be referred to as T-TRP 170. Also shown in FIG. 1C, a NT-TRP will hereafter be referred to as NT-TRP 172. Each ED 110 connected to T-TRP 170 and/or NT-TRP 172 can be dynamically or semi-statically turned-on (i.e., established, activated, or enabled) , turned-off (i.e., released, deactivated, or disabled) and/or configured in response to one of more of: connection availability and connection necessity.

The ED 110 includes one or more antennas 104, a transmitter 111 and a receiver 113 coupled to the one or more antennas 104. Only one antenna 104 is illustrated. One, some, or all of the antennas 104 may alternatively be panels. The transmitter 111 and the receiver 113 may be integrated, e.g. as a transceiver. The transceiver is configured to modulate data or other content for transmission by at least one antenna 104 or network interface controller (NIC) . The transceiver is also configured to demodulate data or other content received by the at least one antenna 104. Each transceiver includes any suitable structure for generating signals for wireless or wired transmission and/or processing signals received wirelessly or by wire. Each antenna 104 includes any suitable structure for transmitting and/or receiving wireless or wired signals.

The ED 110 includes at least one memory 115. The memory 115 stores instructions and data used, generated, or collected by the ED 110. For example, the memory 115 could store software instructions or modules configured to implement some or all of the functionality and/or embodiments described herein and that are executed by one or more processing unit (s) (e.g., a processor 117) . Each memory 115 includes any suitable volatile and/or non-volatile storage and retrieval device (s) . Any suitable type of memory may be used, such as random access memory (RAM) , read only memory (ROM) , hard disk, optical disc, subscriber identity module (SIM) card, memory stick, secure digital (SD) memory card, on-processor cache, and the like.

The ED 110 may further include one or more input/output devices (not shown) or interfaces (such as a wired interface to the Internet 150 in FIG. 1A or FIG. 1B) . The input/output devices permit interaction with a user or other devices in the network. Each input/output device includes any suitable structure for providing information to or receiving information from a user, such as through operation as a speaker, a microphone, a keypad, a keyboard, a display, or a touch screen, including network interface communications.

The ED 110 includes the processor 117 for performing operations including those operations related to preparing a transmission for uplink transmission to the NT-TRP 172 and/or the T-TRP 170, those operations related to processing downlink transmissions received from the NT-TRP 172 and/or the T-TRP 170, and those operations related to processing sidelink transmission to and from another ED 110. Processing operations related to preparing a transmission for uplink transmission may include operations such as encoding, modulating, transmit beamforming, and generating symbols for transmission. Processing operations related to processing downlink transmissions may include operations such as receive beamforming, demodulating and decoding received symbols. Depending upon the embodiment, a downlink transmission may be received by the receiver 113, possibly using receive beamforming, and the processor 117 may extract signaling from the downlink transmission (e.g. by detecting and/or decoding the signaling) . An example of signaling may be a reference signal transmitted by the NT-TRP 172 and/or by the T-TRP 170. In some embodiments, the processor 117 implements the transmit beamforming and/or the receive beamforming based on the indication of beam direction, e.g. beam angle information (BAI) , received from the T-TRP 170. In some embodiments, the processor 117 may perform operations relating to network access (e.g. initial access) and/or downlink synchronization, such as operations relating to detecting a synchronization sequence, decoding and obtaining the system information, etc. In some embodiments, the processor 117 may perform channel estimation, e.g. using a reference signal received from the NT-TRP 172 and/or from the T-TRP 170.

Although not illustrated, the processor 117 may form part of the transmitter 111 and/or part of the receiver 113. Although not illustrated, the memory 115 may form part of the processor 117.

The processor 117, the processing components of the transmitter 111 and the processing components of the receiver 113 may each be implemented by the same or different one or more processors that are configured to execute instructions stored in a memory (e.g. in the memory 115) . Alternatively, some or all of the processor 117, the processing components of the transmitter 111 and the processing components of the receiver 113 may each be implemented using dedicated circuitry, such as a programmed field-programmable gate array (FPGA) , a graphical processing unit (GPU) , a Central Processing Unit (CPU) or an application-specific integrated circuit (ASIC) .

The T-TRP 170 may be known by other names in some implementations, such as a base station, a base transceiver station (BTS) , a radio base station, a network node, a network device, a device on the network side, a transmit/receive node, a Node B, an evolved NodeB (eNodeB or eNB) , a Home eNodeB, a next Generation NodeB (gNB) , a transmission point (TP) , a site controller, an access point (AP) , a wireless router, a relay station, a remote radio head, a terrestrial node, a terrestrial network device, a terrestrial base station, a base band unit (BBU) , a remote radio unit (RRU) , an active antenna unit (AAU) , a remote radio head (RRH) , a central unit (CU) , a distributed unit (DU) , a positioning node, among other possibilities. The T-TRP 170 may be a macro BS, a pico BS, a relay node, a donor node, or the like, or combinations thereof. The T-TRP 170 may refer to the forgoing devices or refer to apparatus (e.g. a communication module, a modem, or a chip) in the forgoing devices.

In some embodiments, the parts of the T-TRP 170 may be distributed. For example, some of the modules of the T-TRP 170 may be located remote from the equipment that houses the antennas 106 for the T-TRP 170, and may be coupled to the equipment that houses the antennas 106 over a communication link (not shown) sometimes known as front haul, such as common public radio interface (CPRI) . Therefore, in some embodiments, the term T-TRP 170 may also refer to modules on the network side that perform processing operations, such as determining the location of the ED 110, resource allocation (scheduling) , message generation, and encoding/decoding, and that are not necessarily part of the equipment that houses the antennas 106 of the T-TRP 170. The modules may also be coupled to other T-TRPs. In some embodiments, the T-TRP 170 may actually be a plurality of T-TRPs that are operating together to serve the ED 110, e.g. through the use of coordinated multipoint transmissions.

The T-TRP 170 includes at least one transmitter 181 and at least one receiver 183 coupled to one or more antennas 106. Only one antenna 106 is illustrated. One, some, or all of the antennas 106 may alternatively be panels. The transmitter 181 and the receiver 183 may be integrated as a transceiver. The T-TRP 170 further includes a processor 182 for performing operations including those related to: preparing a transmission for downlink transmission to the ED 110, processing an uplink transmission received from the ED 110, preparing a transmission for backhaul transmission to the NT-TRP 172, and processing a transmission received over backhaul from the NT-TRP 172. Processing operations related to preparing a transmission for downlink or backhaul transmission may include operations such as encoding, modulating, precoding (e.g. multiple input multiple output (MIMO) precoding) , transmit beamforming, and generating symbols for transmission. Processing operations related to processing received transmissions in the uplink or over backhaul may include operations such as receive beamforming, demodulating received symbols and decoding received symbols. The processor 182 may also perform operations relating to network access (e.g. initial access) and/or downlink synchronization, such as generating the content of synchronization signal blocks (SSBs) , generating the system information, etc. In some embodiments, the processor 182 also generates an indication of beam direction, e.g. BAI, which may be scheduled for transmission by a scheduler 184. The processor 182 performs other network-side processing operations described herein, such as determining the location of the ED 110, determining where to deploy the NT-TRP 172, etc. In some embodiments, the processor 182 may generate signaling, e.g. to configure one or more parameters of the ED 110 and/or one or more parameters of the NT-TRP 172. Any signaling generated by the processor 182 is sent by the transmitter 181. Note that “signaling” , as used herein, may alternatively be called control signaling. Dynamic signaling may be transmitted in a control channel, e.g. a physical downlink control channel (PDCCH) , and static or semi-static higher layer signaling may be included in a packet transmitted in a data channel, e.g. in a physical downlink shared channel (PDSCH) .

The scheduler 184 may be coupled to the processor 182. The scheduler 184 may be included within or operated separately from the T-TRP 170. The scheduler 184 may schedule uplink, downlink, and/or backhaul transmissions, including issuing scheduling grants and/or configuring scheduling-free ( “configured grant” ) resources. The T-TRP 170 further includes a memory 185 for storing information and data. The memory 185 stores instructions and data used, generated, or collected by the T-TRP 170. For example, the memory 185 could store software instructions or modules configured to implement some or all of the functionality and/or embodiments described herein and that are executed by the processor 182.

Although not illustrated, the processor 182 may form part of the transmitter 181 and/or part of the receiver 183. Also, although not illustrated, the processor 182 may implement the scheduler 184. Although not illustrated, the memory 185 may form part of the processor 182.

The processor 182, the scheduler 184, the processing components of the transmitter 181 and the processing components of the receiver 183 may each be implemented by the same or different one or more processors that are configured to execute instructions stored in a memory, e.g. in the memory 185. Alternatively, some or all of the processor 182, the scheduler 184, the processing components of the transmitter 181 and the processing components of the receiver 183 may be implemented using dedicated circuitry, such as a FPGA, a GPU, a CPU, or an ASIC.

Although the NT-TRP 172 is illustrated as a drone only as an example, the NT-TRP 172 may be implemented in any suitable non-terrestrial form, such as high altitude platforms, satellite, high altitude platform as international mobile telecommunication base stations and unmanned aerial vehicles, which forms will be discussed hereinafter. Also, the NT-TRP 172 may be known by other names in some implementations, such as a non-terrestrial node, a non-terrestrial network device, or a non-terrestrial base station. The NT-TRP 172 includes a transmitter 186 and a receiver 187 coupled to one or more antennas 108. Only one antenna 108 is illustrated. One, some, or all of the antennas may alternatively be panels. The transmitter 186 and the receiver 187 may be integrated as a transceiver. The NT-TRP 172 further includes a processor 188 for performing operations including those related to: preparing a transmission for downlink transmission to the ED 110, processing an uplink transmission received from the ED 110, preparing a transmission for backhaul transmission to T-TRP 170, and processing a transmission received over backhaul from the T-TRP 170. Processing operations related to preparing a transmission for downlink or backhaul transmission may include operations such as encoding, modulating, precoding (e.g. MIMO precoding) , transmit beamforming, and generating symbols for transmission. Processing operations related to processing received transmissions in the uplink or over backhaul may include operations such as receive beamforming, demodulating received symbols and decoding received symbols. In some embodiments, the processor 188 implements the transmit beamforming and/or receive beamforming based on beam direction information (e.g. BAI) received from the T-TRP 170. In some embodiments, the processor 188 may generate signaling, e.g. to configure one or more parameters of the ED 110. In some embodiments, the NT-TRP 172 implements physical layer processing, but does not implement higher layer functions such as functions at the medium access control (MAC) or radio link control (RLC) layer. As this is only an example, more generally, the NT-TRP 172 may implement higher layer functions in addition to physical layer processing.

The NT-TRP 172 further includes a memory 189 for storing information and data. Although not illustrated, the processor 188 may form part of the transmitter 186 and/or part of the receiver 187. Although not illustrated, the memory 189 may form part of the processor 188.

The processor 188, the processing components of the transmitter 186 and the processing components of the receiver 187 may each be implemented by the same or different one or more processors that are configured to execute instructions stored in a memory, e.g. in the memory 189. Alternatively, some or all of the processor 188, the processing components of the transmitter 186 and the processing components of the receiver 187 may be implemented using dedicated circuitry, such as a programmed FPGA, a GPU, a CPU, or an ASIC. In some embodiments, the NT-TRP 172 may actually be a plurality of NT-TRPs that are operating together to serve the ED 110, e.g. through coordinated multipoint transmissions. The T-TRP 170, the NT-TRP 172, and/or the ED 110 may include other components, but these have been omitted for the sake of clarity.

FIG. 1D illustrates example modules in the devices of the present disclosure. One or more steps of the embodiment methods provided herein may be performed by corresponding units or modules, according to FIG. 1D. FIG. 1D illustrates units or modules in a device, such as in the ED 110, in the T-TRP 170, or in the NT-TRP 172. For example, a signal may be transmitted by a transmitting unit or by a transmitting module. A signal may be received by a receiving unit or by a receiving module. A signal may be processed by a processing unit or a processing module. Other steps may be performed by an AI or ML module. The respective units or modules may be implemented using hardware, one or more components or devices that execute software, or a combination thereof. For instance, one or more of the units or modules may be an integrated circuit, such as a programmed FPGA, a GPU, a CPU, or an ASIC. It will be appreciated that where the modules are implemented using software for execution by a processor for example, the modules may be retrieved by a processor, in whole or part as needed, individually or together for processing, in single or multiple instances, and that the modules themselves may include instructions for further deployment and instantiation.

Additional details regarding the EDs 110, the T-TRP 170, and the NT-TRP 172 are known to those of skill in the art. As such, these details are omitted here.

An air interface generally includes a number of components and associated parameters that collectively specify how a transmission is to be sent and/or received over a wireless communications link between two or more communicating devices. For example, an air interface may include one or more components defining the waveform (s) , frame structure (s) , multiple access scheme (s) , protocol (s) , coding scheme (s) and/or modulation scheme (s) for conveying information (e.g. data) over a wireless communications link. The wireless communications link may support a link between a radio access network and user equipment (e.g. a “Uu” link) , and/or the wireless communications link may support a link between device and device, such as between two user equipments (e.g. a “sidelink” ) , and/or the wireless communications link may support a link between a non-terrestrial (NT) -communication network and user equipment (UE) . The followings are some examples for the above components:

A waveform component may specify a shape and form of a signal being transmitted. Waveform options may include orthogonal multiple access waveforms and non-orthogonal multiple access waveforms. Non-limiting examples of such waveform options include Orthogonal Frequency Division Multiplexing (OFDM) , Filtered OFDM (f-OFDM) , Time windowing OFDM, Filter Bank Multicarrier (FBMC) , Universal Filtered Multicarrier (UFMC) , Generalized Frequency Division Multiplexing (GFDM) , Wavelet Packet Modulation (WPM) , Faster Than Nyquist (FTN) Waveform, and low Peak to Average Power Ratio Waveform (low PAPR WF) .

A frame structure component may specify a configuration of a frame or group of frames. The frame structure component may indicate one or more of a time, frequency, pilot signature, code, or other parameter of the frame or group of frames. More details of frame structure will be discussed below.

A multiple access scheme component may specify multiple access technique options, including technologies defining how communicating devices share a common physical channel, such as: Time Division Multiple Access (TDMA) , Frequency Division Multiple Access (FDMA) , Code Division Multiple Access (CDMA) , Single Carrier Frequency Division Multiple Access (SC-FDMA) , Low Density Signature Multicarrier Code Division Multiple Access (LDS-MC-CDMA) , Non-Orthogonal Multiple Access (NOMA) , Pattern Division Multiple Access (PDMA) , Lattice Partition Multiple Access (LPMA) , Resource Spread Multiple Access (RSMA) , and Sparse Code Multiple Access (SCMA) . Furthermore, multiple access technique options may include: scheduled access vs. non-scheduled access, also known as grant-free access; non-orthogonal multiple access vs. orthogonal multiple access, e.g., via a dedicated channel resource (e.g., no sharing between multiple communicating devices) ; contention-based shared channel resources vs. non-contention-based shared channel resources, and cognitive radio-based access.

A hybrid automatic repeat request (HARQ) protocol component may specify how a transmission and/or a re-transmission is to be made. Non-limiting examples of transmission and/or re-transmission mechanism options include those that specify a scheduled data pipe size, a signaling mechanism for transmission and/or re-transmission, and a re-transmission mechanism.

A coding and modulation component may specify how information being transmitted may be encoded/decoded and modulated/demodulated for transmission/reception purposes. Coding may refer to methods of error detection and forward error correction. Non-limiting examples of coding options include turbo trellis codes, turbo product codes, fountain codes, low-density parity check codes, and polar codes. Modulation may refer, simply, to the constellation (including, for example, the modulation technique and order) , or more specifically to various types of advanced modulation methods such as hierarchical modulation and low PAPR modulation.

In some embodiments, the air interface may be a “one-size-fits-all concept” . For example, the components within the air interface cannot be changed or adapted once the air interface is defined. In some implementations, only limited parameters or modes of an air interface, such as a cyclic prefix (CP) length or a multiple input multiple output (MIMO) mode, can be configured. In some embodiments, an air interface design may provide a unified or flexible framework to support below 6 GHz and beyond 6 GHz frequency (e.g., mmWave) bands for both licensed and unlicensed access. As an example, flexibility of a configurable air interface provided by a scalable numerology and symbol duration may allow for transmission parameter optimization for different spectrum bands and for different services/devices. As another example, a unified air interface may be self-contained in a frequency domain, and a frequency domain self-contained design may support more flexible radio access network (RAN) slicing through channel resource sharing between different services in both frequency and time.

A frame structure is a feature of the wireless communication physical layer that defines a time domain signal transmission structure, e.g. to allow for timing reference and timing alignment of basic time domain transmission units. Wireless communication between communicating devices may occur on time-frequency resources governed by a frame structure. The frame structure may sometimes instead be called a radio frame structure.

Depending upon the frame structure and/or configuration of frames in the frame structure, frequency division duplex (FDD) and/or time-division duplex (TDD) and/or full duplex (FD) communication may be possible. FDD communication is when transmissions in different directions (e.g. uplink vs. downlink) occur in different frequency bands. TDD communication is when transmissions in different directions (e.g. uplink vs. downlink) occur over different time durations. FD communication is when transmission and reception occurs on the same time-frequency resource, i.e. a device can both transmit and receive on the same frequency resource concurrently in time.

One example of a frame structure is a frame structure in long-term evolution (LTE) having the following specifications: each frame is 10 ms in duration; each frame has 10 subframes, which are each 1 ms in duration; each subframe includes two slots, each of which is 0.5 ms in duration; each slot is for transmission of 7 OFDM symbols (assuming normal CP) ; each OFDM symbol has a symbol duration and a particular bandwidth (or partial bandwidth or bandwidth partition) related to the number of subcarriers and subcarrier spacing; the frame structure is based on OFDM waveform parameters such as subcarrier spacing and CP length (where the CP has a fixed length or limited length options) ; and the switching gap between uplink and downlink in TDD has to be the integer time of OFDM symbol duration.

Another example of a frame structure is a frame structure in new radio (NR) having the following specifications: multiple subcarrier spacings are supported, each subcarrier spacing corresponding to a respective numerology; the frame structure depends on the numerology, but in any case the frame length is set at 10 ms, and consists of ten subframes of 1 ms each; a slot is defined as 14 OFDM symbols, and slot length depends upon the numerology. For example, the NR frame structure for normal CP 15 kHz subcarrier spacing ( “numerology 1” ) and the NR frame structure for normal CP 30 kHz subcarrier spacing ( “numerology 2” ) are different. For 15 kHz subcarrier spacing a slot length is 1 ms, and for 30 kHz subcarrier spacing a slot length is 0.5 ms. The NR frame structure may have more flexibility than the LTE frame structure.

Another example of a frame structure is an example flexible frame structure, e.g. for use in a 6G network or later. In a flexible frame structure, a symbol block may be defined as the minimum duration of time that may be scheduled in the flexible frame structure. A symbol block may be a unit of transmission having an optional redundancy portion (e.g. CP portion) and an information (e.g. data) portion. An OFDM symbol is an example of a symbol block. A symbol block may alternatively be called a symbol. Embodiments of flexible frame structures include different parameters that may be configurable, e.g. frame length, subframe length, symbol block length, etc. A non-exhaustive list of possible configurable parameters in some embodiments of a flexible frame structure include:

(1) Frame: The frame length need not be limited to 10 ms, and the frame length may be configurable and change over time. In some embodiments, each frame includes one or multiple downlink synchronization channels and/or one or multiple downlink broadcast channels, and each synchronization channel and/or broadcast channel may be transmitted in a different direction by different beamforming. The frame length may be more than one possible value and configured based on the application scenario. For example, autonomous vehicles may require relatively fast initial access, in which case the frame length may be set as 5 ms for autonomous vehicle applications. As another example, smart meters on houses may not require fast initial access, in which case the frame length may be set as 20 ms for smart meter applications.

(2) Subframe duration: A subframe might or might not be defined in the flexible frame structure, depending upon the implementation. For example, a frame may be defined to include slots, but no subframes. In frames in which a subframe is defined, e.g. for time domain alignment, then the duration of the subframe may be configurable. For example, a subframe may be configured to have a length of 0.1 ms or 0.2 ms or 0.5 ms or 1 ms or 2 ms or 5 ms, etc. In some embodiments, if a subframe is not needed in a particular scenario, then the subframe length may be defined to be the same as the frame length or not defined.

(3) Slot configuration: A slot might or might not be defined in the flexible frame structure, depending upon the implementation. In frames in which a slot is defined, then the definition of a slot (e.g. in time duration and/or in number of symbol blocks) may be configurable. In one embodiment, the slot configuration is common to all UEs or a group of UEs. For this case, the slot configuration information may be transmitted to UEs in a broadcast channel or common control channel (s) . In other embodiments, the slot configuration may be UE specific, in which case the slot configuration information may be transmitted in a UE-specific control channel. In some embodiments, the slot configuration signaling can be transmitted together with frame configuration signaling and/or subframe configuration signaling. In other embodiments, the slot configuration can be transmitted independently from the frame configuration signaling and/or subframe configuration signaling. In general, the slot configuration may be system common, base station common, UE group common, or UE specific.

(4) Subcarrier spacing (SCS) : SCS is one parameter of scalable numerology which may allow the SCS to possibly range from 15 KHz to 480 KHz. The SCS may vary with the frequency of the spectrum and/or maximum UE speed to minimize the impact of the Doppler shift and phase noise. In some examples, there may be separate transmission and reception frames, and the SCS of symbols in the reception frame structure may be configured independently from the SCS of symbols in the transmission frame structure. The SCS in a reception frame may be different from the SCS in a transmission frame. In some examples, the SCS of each transmission frame may be half the SCS of each reception frame. If the SCS between a reception frame and a transmission frame is different, the difference does not necessarily have to scale by a factor of two, e.g. if more flexible symbol durations are implemented using inverse discrete Fourier transform (IDFT) instead of fast Fourier transform (FFT) . Additional examples of frame structures can be used with different SCSs.

(5) Flexible transmission duration of basic transmission unit: The basic transmission unit may be a symbol block (alternatively called a symbol) , which in general includes a redundancy portion (referred to as the CP) and an information (e.g. data) portion, although in some embodiments the CP may be omitted from the symbol block. The CP length may be flexible and configurable. The CP length may be fixed within a frame or flexible within a frame, and the CP length may possibly change from one frame to another, or from one group of frames to another group of frames, or from one subframe to another subframe, or from one slot to another slot, or dynamically from one scheduling to another scheduling. The information (e.g. data) portion may be flexible and configurable. Another possible parameter relating to a symbol block that may be defined is ratio of CP duration to information (e.g. data) duration. In some embodiments, the symbol block length may be adjusted according to: channel condition (e.g. mulit-path delay, Doppler) ; and/or latency requirement; and/or available time duration. As another example, a symbol block length may be adjusted to fit an available time duration in the frame.

(6) Flexible switch gap: A frame may include both a downlink portion for downlink transmissions from a base station, and an uplink portion for uplink transmissions from UEs. A gap may be present between each uplink and downlink portion, which is referred to as a switching gap. The switching gap length (duration) may be configurable. A switching gap duration may be fixed within a frame or flexible within a frame, and a switching gap duration may possibly change from one frame to another, or from one group of frames to another group of frames, or from one subframe to another subframe, or from one slot to another slot, or dynamically from one scheduling to another scheduling.

The concept of cell, carrier, bandwidth parts (BWPs) and occupied bandwidth will be described below.

A device, such as a base station, may provide coverage over a cell. Wireless communication with the device may occur over one or more carrier frequencies. A carrier frequency will be referred to as a carrier. A carrier may alternatively be called a component carrier (CC) . A carrier may be characterized by its bandwidth and a reference frequency, e.g. the center or lowest or highest frequency of the carrier. A carrier may be on licensed or unlicensed spectrum. Wireless communication with the device may also or instead occur over one or more bandwidth parts (BWPs) . For example, a carrier may have one or more BWPs. More generally, wireless communication with the device may occur over spectrum. The spectrum may comprise one or more carriers and/or one or more BWPs.

A cell may include one or multiple downlink resources and optionally one or multiple uplink resources, or a cell may include one or multiple uplink resources and optionally one or multiple downlink resources, or a cell may include both one or multiple downlink resources and one or multiple uplink resources. As an example, a cell might only include one downlink carrier/BWP, or only include one uplink carrier/BWP, or include multiple downlink carriers/BWPs, or include multiple uplink carriers/BWPs, or include one downlink carrier/BWP and one uplink carrier/BWP, or include one downlink carrier/BWP and multiple uplink carriers/BWPs, or include multiple downlink carriers/BWPs and one uplink carrier/BWP, or include multiple downlink carriers/BWPs and multiple uplink carriers/BWPs. In some embodiments, a cell may instead or additionally include one or multiple sidelink resources, including sidelink transmitting and receiving resources.

A BWP is a set of contiguous or non-contiguous frequency subcarriers on a carrier, or a set of contiguous or non-contiguous frequency subcarriers on multiple carriers, or a set of non-contiguous or contiguous frequency subcarriers, which may have one or more carriers.

In some embodiments, a carrier may have one or more BWPs, e.g. a carrier may have a bandwidth of 20 MHz and consist of one BWP, or a carrier may have a bandwidth of 80 MHz and consist of two adjacent contiguous BWPs, etc. In other embodiments, a BWP may have one or more carriers, e.g. a BWP may have a bandwidth of 40 MHz and consists of two adjacent contiguous carriers, where each carrier has a bandwidth of 20 MHz. In some embodiments, a BWP may comprise non-contiguous spectrum resources which consists of non-contiguous multiple carriers, where the first carrier of the non-contiguous multiple carriers may be in mmW band, the second carrier may be in a low band (such as 2 GHz band) , the third carrier (if it exists) may be in THz band, and the fourth carrier (if it exists) may be in visible light band. Resources in one carrier which belong to the BWP may be contiguous or non-contiguous. In some embodiments, a BWP has non-contiguous spectrum resources on one carrier.

Wireless communication may occur over an occupied bandwidth. The occupied bandwidth may be defined as the width of a frequency band such that, below the lower and above the upper frequency limits, the mean powers emitted are each equal to a specified percentage□/2 of the total mean transmitted power, for example, the value of□/2is taken as 0.5%.

The carrier, the BWP, or the occupied bandwidth may be signaled by a network device (e.g. base station) dynamically, e.g. in physical layer control signaling such as DCI, or semi-statically, e.g. in radio resource control (RRC) signaling or in the medium access control (MAC) layer, or be predefined based on the application scenario; or be determined by the UE as a function of other parameters that are known by the UE, or may be fixed, e.g. by a standard.

In current networks, frame timing and synchronization is established based on synchronization signals, such as a primary synchronization signal (PSS) and a secondary synchronization signal (SSS) . Notably, known frame timing and synchronization strategies involve adding a timestamp, e.g., (xx0: yy0: zz) , to a frame boundary, where xx0, yy0, zz in the timestamp may represent a time format such as hour, minute, and second, respectively.

It is anticipated that diverse applications and use cases in future networks may involve usage of different periods of frames, slots and symbols to satisfy the different requirements, functionalities and Quality of Service (QoS) types. It follows that usage of different periods of frames to satisfy these applications may present challenges for frame timing alignment among diverse frame structures. Consider, for example, frame timing alignment for a TDD configuration in neighboring carrier frequency bands or among sub-bands (or bandwidth parts) of one channel/carrier bandwidth.

The present disclosure relates, generally, to mobile, wireless communication and, in particular embodiments, to a frame timing alignment/realignment, where the frame timing alignment/realignment may comprise a timing alignment/realignment in terms of a boundary of a symbol, a slot or a sub-frame within a frame; or a frame (thus the frame timing alignment/realignment here is more general, not limiting to the cases where a timing alignment/realignment is from a frame boundary only) . Also, in the present disclosure, relative timing to a frame or frame boundary should be interpreted in a more general sense, i.e., the frame boundary means a timing point of a frame element with the frame such as (starting or ending of) a symbol, a slot or subframe within a frame, or a frame. In the following, the phrases “ (frame) timing alignment or timing realignment” and “relative timing to a frame boundary” are used in more general sense described in above.

In overview, aspects of the present disclosure relate to a network device, such as a base station 170, referenced hereinafter as a TRP 170, transmitting signaling that carries a timing realignment indication message. The timing realignment indication message includes information allowing a receiving UE 110 to determine a timing reference point. On the basis of the timing reference point, transmission of frames, by the UE 110, may be aligned. In some aspects of the present disclosure, the frames that become aligned are in different sub-bands of one carrier frequency band. In other aspects of the present disclosure, the frames that become aligned are found in neighboring carrier frequency bands.

On the TRP 170 side, aspects of the present disclosure relate to use of one or more types of signaling to indicate the timing realignment (or/and timing correction) message. Two example types of signaling are provided here to show the schemes. The first example type of signaling may be referenced as cell-specific signaling, examples of which include group common signaling and broadcast signaling. The second example type of signaling may be referenced as UE-specific signaling. One of these two types of signaling or a combination of the two types of signaling may be used to transmit a timing realignment indication message. The timing realignment indication message may be shown to notify one or more UEs 110 of a configuration of a timing reference point. References, hereinafter, to the term “UE 110” may be understood to represent reference to a broad class of generic wireless communication devices within a cell (i.e., a network receiving node, such as a wireless device, a sensor, a gateway, a router, etc. ) , that is, being served by the TRP 170. A timing reference point is a timing reference instant and may be expressed in terms of a relative timing, in view of a timing point in a frame, such as (starting or ending boundary of) a symbol, a slot or a sub-frame within a frame; or a frame. For a simple description in the following, the term “aframe boundary” is used to represent a boundary of possibly a symbol, a slot or a sub-frame within a frame; or a frame. Thus, the timing reference point may be expressed in terms of a relative timing, in view of a current frame boundary, e.g., the start of the current frame. Alternatively, the timing reference point may be expressed in terms of an absolute timing based on certain standards timing reference such as a GNSS (e.g., GPS) , Coordinated Universal Time ( “UTC” ) , etc. In the absolute timing version of the timing reference point, a timing reference point may be explicitly stated.

The timing reference point may be shown to allow for timing adjustments to be implemented at the UEs 110. The timing adjustments may be implemented for improvement of accuracy for a clock at the UE 110. Alternatively, or additionally, the timing reference point may be shown to allow for adjustments to be implemented in future transmissions made from the UEs 110. The adjustments may be shown to cause realignment of transmitted frames at the timing reference point. Note that the realignment of transmitted frames at the timing reference point may comprise the timing realignment from (the starting boundary of) a symbol, a slot or a sub-frame within a frame; or a frame at the timing reference point for one or more UEs and one or more BSs (in a cell or a group of cells) , which applies across the application below.

At UE 110 side, the UE 110 may monitor for the timing realignment indication message. Responsive to receiving the timing realignment indication message, the UE 110 may obtain the timing reference point and take steps to cause frame realignment at the timing reference point. Those steps may, for example, include commencing transmission of a subsequent frame at the timing reference point.

Furthermore, or alternatively, before monitoring for the timing realignment indication message, the UE 110 may cause the TRP 170 to transmit the timing realignment indication message by transmitting, to the TRP 170, a request for a timing realignment, that is, a timing realignment request message. Responsive to receiving the timing realignment request message, the TRP 170 may transmit, to the UE 110, a timing realignment indication message including information on a timing reference point, thereby allowing the UE 110 to implement a timing realignment (or/and a timing adjustment including clock timing error correction) , wherein the timing realignment is in terms of (e.g., a starting boundary of) a symbol, a slot or a sub-frame within a frame; or a frame for UEs and base station (s) in a cell (or a group of cells) .

According to aspects of the present disclosure, a TRP 170 associated with a given cell may transmit a timing realignment indication message. The timing realignment indication message may include enough information to allow a receiver of the message to obtain a timing reference point. The timing reference point may be used, by one or more UEs 110 in the given cell, when performing a timing realignment (or/and a timing adjustment including clock timing error correction) .

According to aspects of the present disclosure, the timing reference point may be expressed, within the timing realignment indication message, relative to a frame boundary (where, as previously described and to be applicable below across the application, a frame boundary can be a boundary of a symbol, a slot or a sub-frame with a frame; or a frame) . The timing realignment indication message may include a relative timing indication, Δt. It may be shown that the relative timing indication, Δt, expresses the timing reference point as occurring a particular duration, i.e., Δt, subsequent to a frame boundary for a given frame. Since the frame boundary is important to allowing the UE 110 to determine the timing reference point, it is important that the UE 110 be aware of the given frame that has the frame boundary of interest. Accordingly, the timing realignment indication message may also include a system frame number (SFN) for the given frame.

It is known, in 5G NR, that the SFN is a value in range from 0 to 1023, inclusive. Accordingly, 10 bits may be used to represent a SFN. When a SFN is carried by an SSB, six of the 10 bits for the SFN may be carried in a Master Information Block (MIB) and the remaining four bits of the 10 bits for the SFN may be carried in a Physical Broadcast Channel (PBCH) payload.

Optionally, the timing realignment indication message may include other parameters. The other parameters may, for example, include a minimum time offset. The minimum time offset may establish a duration of time preceding the timing reference point. The UE 110 may rely upon the minimum time offset as an indication that DL signaling, including the timing realignment indication message, will allow the UE 110 enough time to detect the timing realignment indication message to obtain information on the timing reference point.

A generic background for 6G integrated sensing and communication will now be described. User Equipment (UE) position information is often used in cellular communication networks to improve various performance metrics for the network. Such performance metrics may, for example, include capacity, agility, and efficiency. The improvement may be achieved when elements of the network exploit the position, the behavior, the mobility pattern, etc., of the UE in the context of a priori information describing a wireless environment in which the UE is operating.

A sensing system may be used to help gather UE pose information, including its location in a global coordinate system, its velocity and direction of movement in the global coordinate system, orientation information, and the information about the wireless environment. “Location” is also known as “position” and these two terms may be used interchangeably herein. Examples of well-known sensing systems include RADAR (Radio Detection and Ranging) and LIDAR (Light Detection and Ranging) . While the sensing system can be separate from the communication system, it could be advantageous to gather the information using an integrated system, which reduces the hardware (and cost) in the system as well as the time, frequency, or spatial resources needed to perform both functionalities. However, using the communication system hardware to perform sensing of UE pose and environment information is a highly challenging and open problem. The difficulty of the problem relates to factors such as the limited resolution of the communication system, the dynamicity of the environment, and the huge number of objects whose electromagnetic properties and position are to be estimated.

Accordingly, integrated sensing and communication (also known as integrated communication and sensing) is a desirable feature in existing and future communication systems

Any or all of the EDs 110 and BS 170 may be sensing nodes in the communication system 100E as illustrated in FIG. 1E, which is an example sensing system in accordance with some example embodiments of the present disclosure. Sensing nodes are network entities that perform sensing by transmitting and receiving sensing signals. Some sensing nodes are communication equipment that perform both communications and sensing. However, it is possible that some sensing nodes do not perform communications, and are instead dedicated to sensing. FIG. 1E illustrates another example communication system 100E in which some embodiments of the present disclosure can be implemented. FIG. 1E differs from FIG. 1B in that there is a sensing agent 174 in the communication system 100E, which is absent in FIG. 1B. The sensing agent 174 is an example of a sensing node that is dedicated to sensing. Unlike the EDs 110 and BS 170, the sensing agent 174 does not transmit or receive communication signals. However, the sensing agent 174 may communicate configuration information, sensing information, signaling information, or other information within the communication system 100E. The sensing agent 174 may be in communication with the core network 130 to communicate information with the rest of the communication system 100E. By way of example, the sensing agent 174 may determine the location of the ED 110a, and transmit this information to the base station 170a via the core network 130. Although only one sensing agent 174 is shown in FIG. 1E, any number of sensing agents may be implemented in the communication system 100E. In some embodiments, one or more sensing agents may be implemented at one or more of the RANs 120.

A sensing node may combine sensing-based techniques with reference signal-based techniques to enhance UE pose determination. This type of sensing node may also be known as a sensing management function (SMF) . In some networks, the SMF may also be known as a location management function (LMF) . The SMF may be implemented as a physically independent entity located at the core network 130 with connection to the multiple BSs 170. In other aspects of the present disclosure, the SMF may be implemented as a logical entity co-located inside a BS 170 through logic carried out by the processor 182.

FIG. 1F illustrates an example sensing management function (SMF) 176 of the present disclosure. As shown in FIG. 1F, the SMF 176, when implemented as a physically independent entity, includes at least one transmitter 192, at least one processor 194, one or more antennas 195, at least one receiver 196, a scheduler 198, and at least one memory 199. A transceiver, not shown, may be used instead of the transmitter 192 and receiver 196. The scheduler 198 may be coupled to the processor 194. The scheduler 198 may be included within or operated separately from the SMF 176. The processor 194 implements various processing operations of the SMF 176, such as signal coding, data processing, power control, input/output processing, or any other functionality. The processor 194 can also be configured to implement some or all of the functionality and/or embodiments described in more detail above. Each processor 194 includes any suitable processing or computing device configured to perform one or more operations. Each processor 194 could, for example, include a microprocessor, microcontroller, digital signal processor, field programmable gate array, or application specific integrated circuit.

A reference signal-based pose determination technique belongs to an “active” pose estimation paradigm. In an active pose estimation paradigm, the enquirer of pose information (i.e., the UE) takes part in process of determining the pose of the enquirer. The enquirer may transmit or receive (or both) a signal specific to pose determination process. Positioning techniques based on a global navigation satellite system (GNSS) such as Global Positioning System (GPS) are other examples of the active pose estimation paradigm.

In contrast, a sensing technique, based on radar for example, may be considered as belonging to a “passive” pose determination paradigm. In a passive pose determination paradigm, the target is oblivious to the pose determination process.

By integrating sensing and communications in one system, the system need not operate according to only a single paradigm. Thus, the combination of sensing-based techniques and reference signal-based techniques can yield enhanced pose determination.

The enhanced pose determination may, for example, include obtaining UE channel sub-space information, which is particularly useful for UE channel reconstruction at the sensing node, especially for a beam-based operation and communication. The UE channel sub-space is a subset of the entire algebraic space, defined over the spatial domain, in which the entire channel from the TP to the UE lies. Accordingly, the UE channel sub-space defines the TP-to-UE channel with very high accuracy. The signals transmitted over other sub-spaces result in a negligible contribution to the UE channel. Knowledge of the UE channel sub-space helps to reduce the effort needed for channel measurement at the UE and channel reconstruction at the network-side. Therefore, the combination of sensing-based techniques and reference signal-based techniques may enable the UE channel reconstruction with much less overhead as compared to traditional methods. Sub-space information can also facilitate sub-space based sensing to reduce sensing complexity and improve sensing accuracy.

In some embodiments of integrated sensing and communication, a same radio access technology (RAT) is used for sensing and communication. This avoids the need to multiplex two different RATs under one carrier spectrum, or necessitating two different carrier spectrums for the two different RATs.

In embodiments that integrate sensing and communication under one RAT, a first set of channels may be used to transmit a sensing signal, and a second set of channels may be used to transmit a communications signal. In some embodiments, each channel in the first set of channels and each channel in the second set of channels is a logical channel, a transport channel, or a physical channel.

At the physical layer, communication and sensing may be performed via separate physical channels. For example, a first physical downlink shared channel PDSCH-C is defined for data communication, while a second physical downlink shared channel PDSCH-Sis defined for sensing. Similarly, separate physical uplink shared channels (PUSCH) , PUSCH-C and PUSCH-S, could be defined for uplink communication and sensing.

In another example, the same PDSCH and PUSCH could be also used for both communication and sensing, with separate logical layer channels and/or transport layer channels defined for communication and sensing. Note also that control channel (s) and data channel (s) for sensing can have the same or different channel structure (format) , occupy same or different frequency bands or bandwidth parts.

In a further example, a common physical downlink control channel (PDCCH) and a common physical uplink control channel (PUCCH) is used to carry control information for both sensing and communication. Alternatively, separate physical layer control channels may be used to carry separate control information for communication and sensing. For example, PUCCH-Sand PUCCH-C could be used for uplink control for sensing and communication respectively, and PDCCH-Sand PDCCH-C for downlink control for sensing and communication respectively.

Different combinations of shared and dedicated channels for sensing and communication, at each of the physical, transport, and logical layers, are possible.

The term RADAR originates from the phrase Radio Detection and Ranging; however, expressions with different forms of capitalization (i.e., Radar and radar) are equally valid and now more common. Radar is typically used for detecting a presence and a location of an object. A radar system radiates radio frequency energy and receives echoes of the energy reflected from one or more targets. The system determines the pose of a given target based on the echoes returned from the given target. The radiated energy can be in the form of an energy pulse or a continuous wave, which can be expressed or defined by a particular waveform. Examples of waveforms used in radar include frequency modulated continuous wave (FMCW) and ultra-wideband (UWB) waveforms.

Radar systems can be monostatic, bi-static, or multi-static. In a monostatic radar system, the radar signal transmitter and receiver are co-located, such as being integrated in a transceiver. In a bi-static radar system, the transmitter and receiver are spatially separated, and the distance of separation is comparable to, or larger than, the expected target distance (often referred to as the range) . In a multi-static radar system, two or more radar components are spatially diverse but with a shared area of coverage. A multi-static radar is also referred to as a multisite or netted radar.

Terrestrial radar applications encounter challenges such as multipath propagation and shadowing impairments. Another challenge is the problem of identifiability because terrestrial targets have similar physical attributes. Integrating sensing into a communication system is likely to suffer from these same challenges, and more.

Communication nodes can be either half-duplex or full-duplex. A half-duplex node cannot both transmit and receive using the same physical resources (time, frequency, etc. ) ; conversely, a full-duplex node can transmit and receive using the same physical resources. Existing commercial wireless communications networks are all half-duplex. Even if full-duplex communications networks become practical in the future, it is expected that at least some of the nodes in the network will still be half-duplex nodes because half-duplex devices are less complex, and have lower cost and lower power consumption. In particular, full-duplex implementation is more challenging at higher frequencies (e.g. in the millimeter wave bands) , and very challenging for small and low-cost devices, such as femtocell base stations and UEs.

The limitation of half-duplex nodes in the communications network presents further challenges toward integrating sensing and communications into the devices and systems of the communications network. For example, both half-duplex and full-duplex nodes can perform bi-static or multi-static sensing, but monostatic sensing typically requires the sensing node have full-duplex capability. A half-duplex node may perform monostatic sensing with certain limitations, such as in a pulsed radar with a specific duty cycle and ranging capability.

Sensing signal waveform and frame structure will now be described. Properties of a sensing signal, or a signal used for both sensing and communication, include the waveform of the signal and the frame structure of the signal. The frame structure defines the time-domain boundaries of the signal. The waveform describes the shape of the signal as a function of time and frequency. Examples of waveforms that can be used for a sensing signal include ultra-wide band (UWB) pulse, Frequency-Modulated Continuous Wave (FMCW) or “chirp” , orthogonal frequency-division multiplexing (OFDM) , cyclic prefix (CP) -OFDM, and Discrete Fourier Transform spread (DFT-s) -OFDM.

In an embodiment, the sensing signal is a linear chirp signal with bandwidth B and time duration T. Such a linear chirp signal is generally known from its use in FMCW radar systems. A linear chirp signal is defined by an increase in frequency from an initial frequency, f_chirp0, at an initial time, t_chirp0, to a final frequency, f_chirp1, at a final time, t_chirp1 where the relation between the frequency (f) and time (t) can be expressed as a linear relation of f-f_chirp0=α (t-t_chirp0) , whereis defined as the chirp slope. The bandwidth of the linear chirp signal may be defined as B=f_chirp1-f_chirp0 and the time duration of the linear chirp signal may be defined as T=t_chirp1-t_chirp0. Such linear chirp signal can be presented asin the baseband representation.

Precoding as used herein may refer to any coding operation (s) or modulation (s) that transform an input signal into an output signal. Precoding may be performed in different domains, and typically transform the input signal in a first domain to an output signal in a second domain. Precoding may include linear operations.

A terrestrial communication system may also be referred to as a land-based or ground-based communication system, although a terrestrial communication system can also, or instead, be implemented on or in water. The non-terrestrial communication system may bridge the coverage gaps for underserved areas by extending the coverage of cellular networks through non-terrestrial nodes, which will be key to ensuring global seamless coverage and providing mobile broadband services to unserved/underserved regions, in this case, it is hardly possible to implement terrestrial access-points/base-stations infrastructure in the areas like oceans, mountains, forests, or other remote areas.

The terrestrial communication system may be a wireless communications using 5G technology and/or later generation wireless technology (e.g., 6G or later) . In some examples, the terrestrial communication system may also accommodate some legacy wireless technology (e.g., 3G or 4G wireless technology) . The non-terrestrial communication system may be a communications using the satellite constellations like Geo-Stationary Orbit (GEO) satellites which utilizing broadcast public/popular contents to a local server, Low earth orbit (LEO) satellites establishing a better balance between large coverage area and propagation path-loss/delay, stabilize satellites in very low earth orbits (VLEO) enabling technologies substantially reducing the costs for launching satellites to lower orbits, high altitude platforms (HAPs) providing a low path-loss air interface for the users with limited power budget, or Unmanned Aerial Vehicles (UAVs) (or unmanned aerial system (UAS) ) achieving a dense deployment since their coverage can be limited to a local area, such as airborne, balloon, quadcopter, drones, etc. In some examples, GEO satellites, LEO satellites, UAVs, HAPs and VLEOs may be horizontal and two-dimensional. In some examples, UAVs, HAPs and VLEOs coupled to integrate satellite communications to cellular networks emerging 3D vertical networks consist of many moving (other than geostationary satellites) and high altitude access points such as UAVs, HAPs and VLEOs.

Multiple input multiple-output (MIMO) technology allows an antenna array of multiple antennas to perform signal transmissions and receptions to meet high transmission rate requirement. The above ED110 and T-TRP 170, and/or NT-TRP use MIMO to communicate over the wireless resource blocks. MIMO utilizes multiple antennas at the transmitter and/or receiver to transmit wireless resource blocks over parallel wireless signals. MIMO may beamform parallel wireless signals for reliable multipath transmission of a wireless resource block. MIMO may bond parallel wireless signals that transport different data to increase the data rate of the wireless resource block.

In recent years, a MIMO (large-scale MIMO) wireless communication system with the above T-TRP 170, and/or NT-TRP 172 configured with a large number of antennas has gained wide attentions from the academia and the industry. In the large-scale MIMO system, the T-TRP 170, and/or NT-TRP 172 is generally configured with more than ten antenna units (such as 128 or 256) , and serves for dozens of the ED 110 (such as 40) in the meanwhile. A large number of antenna units of the T-TRP 170, and NT-TRP 172 can greatly increase the degree of spatial freedom of wireless communication, greatly improve the transmission rate, spectrum efficiency and power efficiency, and eliminate the interference between cells to a large extent. The increase of the number of antennas makes each antenna unit be made in a smaller size with a lower cost. Using the degree of spatial freedom provided by the large-scale antenna units, the T-TRP 170, and NT-TRP 172 of each cell can communicate with many ED 110 in the cell on the same time-frequency resource at the same time, thus greatly increasing the spectrum efficiency. A large number of antenna units of the T-TRP 170, and/or NT-TRP 172 also enable each user to have better spatial directivity for uplink and downlink transmission, so that the transmitting power of the T-TRP 170, and/or NT-TRP 172 and an ED 110 is obviously reduced, and the power efficiency is greatly increased. When the antenna number of the T-TRP 170, and/or NT-TRP 172 is sufficiently large, random channels between each ED 110 and the T-TRP 170, and/or NT-TRP 172 can approach to be orthogonal, and the interference between the cell and the users and the effect of noises can be eliminated. The plurality of advantages described above enable the large-scale MIMO to have a magnificent application prospect.

A MIMO system may include a receiver connected to a receive (Rx) antenna, a transmitter connected to transmit (Tx) antenna, and a signal processor connected to the transmitter and the receiver. Each of the Rx antenna and the Tx antenna may include a plurality of antennas. For instance, the Rx antenna may have an ULA antenna array in which the plurality of antennas are arranged in line at even intervals. When a radio frequency (RF) signal is transmitted through the Tx antenna, the Rx antenna may receive a signal reflected and returned from a forward target.

A non-exhaustive list of possible unit or possible configurable parameters or in some embodiments of a MIMO system include:

Panel: unit of antenna group, or antenna array, or antenna sub-array which can control its Tx or Rx beam independently.

Beam: A beam is formed by performing amplitude and/or phase weighting on data transmitted or received by at least one antenna port, or may be formed by using another method, for example, adjusting a related parameter of an antenna unit. The beam may include a Tx beam and/or a Rx beam. The transmit beam indicates distribution of signal strength formed in different directions in space after a signal is transmitted through an antenna. The receive beam indicates distribution of signal strength that is of a wireless signal received from an antenna and that is in different directions in space. The beam information may be a beam identifier, or antenna port (s) identifier, or CSI-RS resource identifier, or SSB resource identifier, or SRS resource identifier, or other reference signal resource identifier.

Artificial Intelligence technologies can be applied in communication, including artificial intelligence or machine learning (AI/ML) based communication in the physical layer and/or AI/ML based communication in the higher layer, e.g., medium access control (MAC) layer. For example, in the physical layer, the AI/ML based communication may aim to optimize component design and/or improve the algorithm performance. For the MAC layer, the AI/ML based communication may aim to utilize the AI/ML capability for learning, prediction, and/or making a decision to solve a complicated optimization problem with possible better strategy and/or optimal solution, e.g. to optimize the functionality in the MAC layer, e.g. intelligent TRP management, intelligent beam management, intelligent channel resource allocation, intelligent power control, intelligent spectrum utilization, intelligent modulation and coding scheme (MCS) , intelligent hybrid automatic repeat request (HARQ) strategy, intelligent transmit/receive (Tx/Rx) mode adaption, etc.

The following are some terminologies which are used in AI/ML field:

Data collection: Data is the very important component for AI/ML techniques. Data collection is a process of collecting data by the network nodes, management entity, or UE for the purpose of AI/ML model training, data analytics and inference.

AI/ML model training: AI/ML model training is a process to train an AI/ML Model by learning the input/output relationship in a data driven manner and obtain the trained AI/ML Model for inference.

AI/ML model inference: A process of using a trained AI/ML model to produce a set of outputs based on a set of inputs.

AI/ML model validation: As a sub-process of training, validation is used to evaluate the quality of an AI/ML model using a dataset different from the one used for model training. Validation can help selecting model parameters that generalize beyond the dataset used for model training. The model parameter after training can be adjusted further by the validation process.

AI/ML model testing: Similar with validation, testing is also a sub-process of training, and it is used to evaluate the performance of a final AI/ML model using a dataset different from the one used for model training and validation. Differently from AI/ML model validation, testing do not assume subsequent tuning of the model.

Online training: Online training means an AI/ML training process where the model being used for inference is typically continuously trained in (near) real-time with the arrival of new training samples.

Offline training: An AI/ML training process where the model is trained based on collected dataset, and where the trained model is later used or delivered for inference.

AI/ML model delivery/transfer: A generic term referring to delivery of an AI/ML model from one entity to another entity in any manner. Delivery of an AI/ML model over the air interface includes either parameters of a model structure known at the receiving end or a new model with parameters. Delivery may contain a full model or a partial model.

Life cycle management (LCM) : When the AI/ML model is trained and/or inferred at one device, it is necessary to monitor and manage the whole AI/ML process to guarantee the performance gain obtained by AI/ML technologies. For example, due to the randomness of wireless channels and the mobility of UEs, the propagation environment of wireless signals changes frequently. Nevertheless, it is difficult for an AI/ML model to maintain optimal performance in all scenarios for all the time, and the performance may even deteriorate sharply in some scenarios. Therefore, the lifecycle management (LCM) of AI/ML models needs to be studied for sustainable operation of AI/ML in NR air-interface.

Life cycle management covers the whole procedure of AI/ML technologies which applied on one or more nodes. In specific, it includes at least one of the following sub-process: data collection, model training, model identification, model registration, model deployment, model configuration, model inference, model selection, model activation, deactivation, model switching, model fallback, model monitoring, model update, model transfer/delivery and UE capability report. Model monitoring can be based on inference accuracy, including metrics related to intermediate key performance indicator (KPI) s, and it can also be based on system performance, including metrics related to system performance KPIs, e.g., accuracy and relevance, overhead, complexity (computation and memory cost) , latency (timeliness of monitoring result, from model failure to action) and power consumption. Moreover, data distribution may shift after deployment due to the environment changes, thus the model based on input or output data distribution should also be considered.

Supervised learning: The goal of supervised learning algorithms is to train a model that maps feature vectors (inputs) to labels (output) , based on the training data which includes the example feature-label pairs. The supervised learning can analyze the training data and produce an inferred function, which can be used for mapping the inference data. Supervised learning can be further divided into two types: Classification and Regression. Classification is used when the output of the AI/ML model is categorical i.e. with two or more classes. Regression is used when the output of the AI/ML model is a real or continuous value.

Unsupervised learning: In contrast to supervised learning where the AI/ML models learn to map the input to the target output, the unsupervised methods learn concise representations of the input data without the labelled data, which can be used for data exploration or to analyze or generate new data. One typical unsupervised learning is clustering which explores the hidden structure of input data and provide the classification results for the data.

Reinforce learning: Reinforce learning is used to solve sequential decision-making problems. Reinforce learning is a process of training the action of intelligent agent from input (state) and a feedback signal (reward) in an environment. In reinforce learning, an intelligent agent interacts with an environment by taking an action to maximize the cumulative reward. Whenever the intelligent agent takes one action, the current state in the environment may transfer to the new state, and the new state resulted by the action will bring to the associated reward. Then the intelligent agent can take the next action based on the received reward and new state in the environment. During the training phase, the agent interacts with the environment to collect experience. The environments often mimicked by the simulator since it is expensive to directly interact with the real system. In the inference phase, the agent can use the optimal decision-making rule learned from the training phase to achieve the maximal accumulated reward.

Federated learning: Federated learning (FL) is a machine learning technique that is used to train an AI/ML model by a central node (e.g., server) and a plurality of decentralized edge nodes (e.g., UEs, next Generation NodeBs, “gNBs” ) .

According to the wireless FL technique, a server may provide, to an edge node, a set of model parameters (e.g., weights, biases, gradients) that describe a global AI/ML model. The edge node may initialize a local AI/ML model with the received global AI/ML model parameters. The edge node may then train the local AI/ML model using local data samples to, thereby, produce a trained local AI/ML model. The edge node may then provide, to the serve, a set of AI/ML model parameters that describe the local AI/ML model.

Upon receiving, from a plurality of edge nodes, a plurality of sets of AI/ML model parameters that describe respective local AI/ML models at the plurality of edge nodes, the server may aggregate the local AI/ML model parameters reported from the plurality of UEs and, based on such aggregation, update the global AI/ML model. A subsequent iteration progresses much like the first iteration. The server may transmit the aggregated global model to a plurality of edge nodes. The above procedure is performed multiple iterations until the global AI/ML model is considered to be finalized, e.g., the AI/ML model is converged or the training stopping conditions are satisfied.

Notably, the wireless FL technique does not involve exchange of local data samples. Indeed, the local data samples remain at respective edge nodes.

AI technologies (which encompass ML technologies) may be applied in communication, including AI-based communication in the physical layer and/or AI-based communication in the MAC layer. For the physical layer, the AI communication may aim to optimize component design and/or improve the algorithm performance. For example, AI may be applied in relation to the implementation of: channel coding, channel modelling, channel estimation, channel decoding, modulation, demodulation, MIMO, waveform, multiple access, physical layer element parameter optimization and update, beam forming, tracking, sensing, and/or positioning, etc. For the MAC layer, the AI communication may aim to utilize the AI capability for learning, prediction, and/or making a decision to solve a complicated optimization problem with possible better strategy and/or optimal solution, e.g. to optimize the functionality in the MAC layer. For example, AI may be applied to implement: intelligent TRP management, intelligent beam management, intelligent channel resource allocation, intelligent power control, intelligent spectrum utilization, intelligent MCS, intelligent HARQ strategy, and/or intelligent transmission/reception mode adaption, etc.

An AI architecture may involve multiple nodes, where the multiple nodes may possibly be organized in one of two modes, i.e., centralized and distributed, both of which may be deployed in an access network, a core network, or an edge computing system or third party network. A centralized training and computing architecture is restricted by possibly large communication overhead and strict user data privacy. A distributed training and computing architecture may comprise several frameworks, e.g., distributed machine learning and federated learning. In some embodiments, an AI architecture may comprise an intelligent controller which can perform as a single agent or a multi-agent, based on joint optimization or individual optimization. New protocols and signaling mechanisms are desired so that the corresponding interface link can be personalized with customized parameters to meet particular requirements while minimizing signaling overhead and maximizing the whole system spectrum efficiency by personalized AI technologies.

New protocols and signaling mechanisms are provided for operating within and switching between different modes of operation, including between AI and non-AI modes, and for measurement and feedback to accommodate the different possible measurements and information that may need to be fed back, depending upon the implementation.

An air interface that uses AI as part of the implementation, e.g. to optimize one or more components of the air interface, will be referred to herein as an “AI enabled air interface” . In some embodiments, there may be two types of AI operation in an AI enabled air interface: both the network and the UE implement learning; or learning is only applied by the network.

The example communication environment, communication system, terminal device, network device, electronic device, UE, BS, sensing node, etc. of this disclosure have heretofore been discussed with reference to FIGS. 1A to 1F. Methods and procedures in accordance with embodiments of this disclosure are further discussed with reference to FIGS. 2A to 6.

In some embodiments of this disclosure, to provide the AI service for the wireless devices efficiently, the AI model may be distributed over the massive devices in the network which have data and computing capability. As mentioned above, the sub-model (s) distributed on a wireless device may also be called the “MoP” . In this case, the training of the big AI model which is usually implemented at the network side (within RAN, or outside RAN, e.g., core network, third party) can be implemented based on distributed learning by multiple devices, and each device trains one or more MoPs of the whole AI model. Similarly, the AI models may also be updated in the distribution manner based on the MoPs. As such, the life cycle management for AI service can be based on the MoPs, not the whole big model.

Only for discussion clarity, an example scenario of this disclosure is described as below. In the example scenario, there may be multiple learning devices in the network, including UE and BS, edge devices. As an example, the UEs can be smartphone, robot, Customer Premise Equipment (CPE) , computing node with capability of communication, etc. The network may manage the training of a big AI/ML model supporting multiple tasks, and provide AI service for UE or the third party. Specifically, to train the big AI/ML model, devices with learning capability may be indicated or configured to train some functionality of the big AI/ML model, e.g. train partial parameters of the big model or train one task of the big AI/ML model. After the local training at the device is completed, the device may report its trained model (e.g. small sub-model) to an aggregation node to generate the big AI/ML model, the aggregation node can be within RAN (e.g. BS, central processing unit within RAN) or outside RAN (e.g. core network or the third party) . Moreover, even if the big AI/ML model is initially trained already, the sub-models of the AI model may need to be updated by multiple devices over time during the AI service. In some implementations, the sub-models to be trained or updated by different learning devices may be the same or different. In some implementations, the AI model structures for different learning devices may be the same or different. In some implementations, the AI algorithms for different learning devices may be the same or different.

Before discussing life cycle management of AI service based on the training and/or updating of the distributed sub-models of an AI model, for discussion clarity, some example relationships between sub-models and the (whole) AI model are discussed with reference to FIGS. 2A to 2C.

In some embodiments of this disclosure, a sub-model is a part of a specific AI model. The specific AI model can be regarded as parent AI model, and the sub-model can be regarded as sub-model. Generally, the parent model is a bigger model (or called larger model) and the sub-model is a smaller model. The parent model is characterized by its large size and/or large number of functionalities/tasks, where the large model has a large number of model parameters. Furthermore, the sub-model can have the same or different neuron network (NN) structure as the parent model. In an example, the parent AI model is composed by multiple components, and the sub-model (i.e., MoP) may consist a partial of the multiple components. For an AI feature or an AI service or an AI model (i.e., a big AI model) , one or multiple AI/ML sub-models are associated to it.

In some embodiments, a sub-model of an AI model is associated with at least one parameter among a plurality of parameters associated with the AI model. For example, the AI model may be associated with a plurality of parameters, and a sub-model of this AI model may be associated with a portion of the plurality of parameters. In some embodiments, a first sub-model of the AI model is associated with a first subset of parameters among a plurality of parameters associated with the AI model, and a second sub-model of the AI model is associated with a second subset of parameters among the plurality of parameters. In some implementations, the first subset of parameters and the second subset of parameters are non-overlapped. In other words, different sub-models of the AI model may be associated with non-overlapped parameters. Alternatively, the first subset of parameters and the second subset of parameters are overlapped. In other words, different sub-models of the AI model may be associated with overlapped parameters. The expression “overlapped” herein means that a parameter of the AI model can be in both of two sub-models, i.e. the two sub-models can train the same parameter. In this way, the AI model can be divided into multiple sub-models based on the parameters.

FIG. 2A illustrates example distributed sub-models of an AI model according to some embodiments of the present disclosure. As shown in FIG. 2A, the AI model in block 210 is the “big AI model” , and Model Part 0 (in the block 211) , Model Part 1, …, and Model Part k are sub-models of the AI model in block 210 and may be distributed over multiple devices. In this example, the total number of parameters of the AI model is N. The parameters of the Model Part 0 are M0 parameters out of N parameters. Similarly, the parameters of the Model Part 1 are M1 parameters of the N parameters. In some implementations, the M0 parameters of the Model Part 0 may be completely different from the M1 parameters of the Model Part 1. That is, each of the M0 parameters is different from each of the M1 parameters. Alternatively, the M0 parameters of the Model Part 0 may be partially the same as the M1 parameters used for the Model Part 1. That is, the M0 parameters and the M1 parameters may have L identical parameters with L ≤ M0 and L ≤ M1. In this case, the Model Part 0 and the Model Part 1 may train the L identical parameters.

In some embodiments, a sub-model of an AI model may be defined based on the supported functionality (ies) . In some implementations, the sub-model is associated with one functionality among a plurality of functionalities associated with the AI model based on a one-to-one correspondence. In an example, a sub-model is associated with one functionality of the associated AI model or the associated AI feature.

FIG. 2B illustrates example distributed sub-models of an AI model according to some embodiments of the present disclosure. As shown in FIG. 2B, the AI model in block 220 is the “big AI model” , and Model Part 0 (in the block 221) , Model Part 1, …, and Model Part k are sub-models of the AI model in block 220. In this example, the AI model supports the functionalities 0, 1, …, N, the Model Part 0 may support the functionality 0, the Model Part 1 may support the functionality 1, and the Model Part k may support the functionality k.

Table 1 illustrates more specific examples of sub-models (i.e., MoPs) associated with a functionality of an AI model in a one-to-one correspondence. As shown in Table 1, for AI positioning service, the AI/ML model supports to provide 3D coordinate (X, Y, Z) of a location (e.g. UE location) . In the AI positioning service, there may be three functionalities, i.e., providing (X, Y) coordinates, providing (X, Z) coordinates, and providing (Y, Z) coordinates, where (X, Y, Z) is the coordinate in a three-dimensional coordinate system, e.g. there is one-to-one correspondence between a 3D location and a triple (x, y, z) . The values of X, Y and Z are the X dimension-, Y dimension-and Z dimension-coordinates of a location. In this case, the MoP with index 0 of the AI/ML model for positioning service may support (X, Y) coordinate provision. The MoP with index 1 of the AI/ML model for positioning service may support (X, Z) coordinate provision. Moreover, the MoP with index 2 of the AI/ML model for positioning service may support (Y, Z) coordinate provision. Similarly, for AI object detection service, the AI/ML model supports to detect an object. In the AI object detection service, there may be four functionalities, namely, existence judgement, range detection, Azimuth detection, and Elevation detection. The MoP with index 0 of the AI/ML model for object detection service may support existence judgement. The MoP with index 1 of the AI/ML model for object detection service may support range detection. The MoP with index 2 of the AI/ML model for object detection service may support Azimuth detection. Moreover, the MoP with index 3 of the AI/ML model for object detection service may support Elevation detection. That is, a MoP may support one functionality of an associated AI model. As such, by combining multiple MoPs of the AI model at a central processing unit, the whole big AI model may be obtained to support the complete functionalities for the AI service.

Table 1 Examples of MoPs associated with a functionality of an AI model

In some implementations, the sub-model is associated with at least one functionality among a plurality of functionalities associated with the AI model. In an example, a sub-model is associated with one or more functionalities of the associated AI model or the associated AI feature. In other words, multiple AI functionalities can be associated with one sub-model. As such, by training or updating one sub-model, multiple functionalities can be trained or updated. In this way, the learning latency can be reduced.

In some embodiments, a sub-model of an AI model may be defined based on the assigned task (s) . In some implementations, the sub-model is associated with an AI task among a plurality of AI tasks associated with the AI model based on a one-to-one correspondence. In an example, a sub-model is associated with one task of the associated AI model or the associated AI feature. For example, the big AI model may support multiple AI tasks, and a sub-model of the AI model is associated with one AI task within the supported AI tasks of the AI model.

FIG. 2C illustrates example distributed sub-models of an AI model according to some embodiments of the present disclosure. As shown in FIG. 2C, the AI model in block 250 may support Task-1, Task-2, …and Task-N. Model Part 0 (in the block 221) , Model Part 1, …, and Model Part k are sub-models of the AI model. The Model Part 0 in block 251 may be associated with the Task-1, the Model Part 1 may be associated with the Task-2, and the Model Part k may be associated with the Task-k.

In some implementations, the sub-model is associated with at least one AI task among a plurality of AI tasks associated with the AI model. In an example, a sub-model is associated with one or more AI tasks among a plurality of AI tasks of the associated AI model or the associated AI feature. In other words, the whole AI model supports multiple AI tasks, and a sub-model of the AI model may support more than one AI task within the multiple AI tasks. In this way, if the sub-model is trained or updated, multiple tasks can be trained or updated, and the learning latency can be reduced accordingly.

In some embodiments, the sub-model is associated with a predefined size requirement. In other words, a sub-model of an AI model may be defined based on the model size. In some implementations, a sub-model of the whole AI model may be of a model size that is based on the number of size parameters or a size level. For example, the sub-model may be associated with a model size. The model size may include, but limited to, the number of parameters, the number of layers of an AI model, the computational complexity of the model (e.g. float point computations, number of real-value operation) , storage for the model. In another example, a sub-model may be associated with a certain level of an AI model size. For example, level 0 corresponds to 0 to 10 thousand parameters, level 1 corresponds to 10 thousand to 1 million parameters, level 2 corresponds to 1 million to 1 billion parameters, and level 3 corresponds to larger than 1 billion parameters.

In some embodiments, the sub-model is associated with a predefined neuron network algorithm. In other words, a sub-model of an AI model may be defined based on the model structure, for example, the neuron network algorithm. In some embodiments, a sub-model of the whole AI model may be associated with an Neuron Network (NN) structure (or NN algorithm) . For example, the whole AI model may include multiple architectures (or called structures) . Example structures may include transformer neural network, Deep Neural Network (DNN) , Convolution Neural Network (CNN) , Recurrent Neural Network (RNN) , Long Short Term Memory networks (LSTM) , and so on. The sub-model of the whole AI model may be of one or more of these structures.

In view of the above, some example relationships between sub-models and the whole AI model are described. It is to be understood that the above relationships are only examples, and any other relationships between the sub-models of the AI model and the whole AI model may be possible, which is not limited in this disclosure. Once the whole AI model is distributed and trained on the massive devices in the network, managing of the whole AI model having these sub-models is needed. In some embodiments of this disclosure, the life cycle management of the distributed AI model is further discussed with respect to various aspects, e.g., configuring, activating, training/updating, validating and reporting the sub-models of the AI model.

FIG. 3 illustrates a signaling process 300 for training or updating sub-models of an AI model according to some embodiments of the present disclosure. For illustrative purposes, the process 300 will be described with reference to FIGS. 1A to 1F. Only as an example and without limitation, as shown in FIG. 3, the terminal device 310 may be the UE 110 or ED 110 as shown in FIGS. 1A to 1E, and the network device 370 may be the BS 170, T-TRP 170 or NT-TRP 172 as shown in FIGS. 1A to 1E. In addition, the network device 370 may include the aggregation node as mentioned above.

In the signaling process 300, the network device 370 transmits (301) a configuration 302 of a sub-model of an AI model associated with an AI feature. The configuration 302 may be transmitted to the terminal device 310 in any signaling message. For example, the configuration 302 may be carried in the DCI or higher layer signaling. Without any limitation, the configuration 302 may be carried in other signaling messages. The terminal device 310 receives (303) the configuration 302 of the sub-model. The terminal device 310 trains or updates (304) the sub-model and reports (305) information associated with the sub-model. Accordingly, the network device 370 receives (307) a report 306 of the information associated with the sub-model which is trained or updated by the terminal device 310.

Regarding the sub-model configuration, in some embodiments, the terminal device 310 (for example, UE) may report its learning capability to the network device 370 (for example, the aggregation node) . The learning capability may include one or multiple of supporting AI service, AI functionalities, AI models, AI model structures, AI model size, and so on. In an example, an AI/ML feature is associated with an index, and therefore the terminal device may report the index of its supporting AI/ML feature. If the AI/ML feature includes one or multiple AI/ML functionalities, the device may also report the index of its supporting functionalities/functionality. Then, the network device 370 (or the network) may configure one or more sub-models of the AI model (s) for one or more respective AI features on the terminal device based on the terminal device’s learning capability.

In some embodiments, the terminal device 310 may be configured with at least one sub-model including the sub-model. For example, for an AI feature (AI service) , one or more MoPs may be configured for the terminal device 310. Specifically, the configuration for the one or more MoPs may be unicast, multicast or groupcast via downlink control information (DCI) or a high layer signaling. In some implementations, when transmitting the configuration of the sub-model to the terminal device 310, the network device 370 may transmit at least one configuration of at least one sub-model which includes the sub-model to the terminal device 310. For example, for an AI feature (AI service) , one or more MoPs may be configured for the terminal device 310 via a signaling, e.g., a radio resource control (RRC) signaling. In some implementations, when transmitting the configuration of the sub-model to the terminal device 310, the network device 370 may first transmit a plurality of configurations of a plurality of candidate sub-models, and then transmit at least one configuration of at least one sub-model among the plurality of candidate sub-models, wherein the at least one sub-model includes the sub-model. In other words, a plurality of candidate sub-models may be preconfigured and at least one sub-model among the plurality of candidate sub-models may be further selected to be used by a further signaling. For example, the network device 370 may configure a plurality of candidate sub-models to the terminal device 310 by broadcast signaling. Then, the network device 370 may further indicate at least sub-model among the plurality of candidate sub-models by a RRC signaling. In an example, N (N is an integer, N>==1) candidate MoPs are configured by broadcast signalling, e.g. by system information. Then, one or more MoPs are indicated from the N candidate MoPs to the terminal device 310 by RRC signaling.

In some embodiments, individual sub-model configurations may be provided for different AI features. As mentioned above, the sub-model may be configured with an index. In some implementations, the sub-model is identified based on an index , wherein the index is unique in at least one set of sub-models of at least one AI models associated with one or more AI features. In an example, the MoP index may be unique among multiple AI features. In this case, by indicating the MoP index, the terminal device 310 may be aware which MoP is configured for an AI feature, and which AI feature the MoP is configured for. In some implementations, the sub-model is identified based on an index of the sub-model and an index of the AI feature. In an example, the MoP index is unique within one AI feature, but MoPs for different AI features may have the same MoP index. In this case, the network device 370 may need to indicate the respective AI feature index and the MoP index to the terminal device 310 to align the understanding on the indicated MoP between the network device 370 and the terminal device 310. Without any limitation, the MoP and/or the association between the MoP and the AI feature may be identified in any other manners, which is not limited in this disclosure.

In some embodiments, once configured with a sub-model, the terminal device 310 may train or update the sub-model. Alternatively, the terminal device 310 may train or update the configured sub-model when the configured sub-model is activated.

In some embodiments, in order to train or update the sub-model by the terminal device 310, the network device 370 may transmit an indication of activating the sub-model to the terminal device 310. Based on the indication of activating the sub-model, the terminal device 310 may activate the sub-model for the training or updating, or maintaining the at least one sub-model as active (for example, the currently active MoP is required to be monitored) . In an example, the network device 370 may first configure one or multiple MoPs for an AI/ML-enabled feature for the terminal device 310, and the network device 370 may then activate at least one MoPs to be trained or updated by the terminal device 310 within the configured MoPs. The term “activating a MoP” herein refers to enabling a MoP for a specific AI/ML-enabled feature. The term “deactivating a MoP” herein refers to disabling a MoP for a specific AI/ML-enabled feature.

In some embodiments, the sub-model is a first sub-model. The terminal device 310 may receive an indication of activating the first sub-model. If a second sub-model of the AI model is being activated when the indication of activating the first sub-model is received, the terminal device 310 may deactivate the second sub-model. In other words, the sub-models configured on the terminal device can be active one by one but not in parallel. That is, only one MoP for an AI feature may be active for a given time. The network device 370 may indicate the active MoP to the terminal device 310 via a DCI or a high layer signaling. For example, there may be a field called “MoP indicator” indicating the active MoP index in the signaling carrying the indication of activating the MoP. If the MoP indicated by the active MoP index is different from the currently active MoP for the same AI feature, the MoP switching may be performed by the terminal device 310. The MoP switching delay can be zero or non-zero depending on device capability. In addition, if there are more than one AI features, the network device 370 may indicate the respective active MoP index for each of these AI features.

In some embodiments, an AI feature can also be activated or deactivated. In some further embodiments, in order to train or update the sub-model by the terminal device 310, the network device 370 may transmit an indication of activating the AI feature, wherein the sub-model is a default sub-model of the AI feature. After receiving the indication of activating the AI feature, the terminal device 310 may activate the AI feature and determine the default sub-model of the activated AI feature to be active. In other words, if an AI feature is indicated to be activated, the default MoP associated with this AI feature may be activated. This default MoP for the AI feature can be indicated by the network device or may be pre-defined. For example, the default active MoP may be predefined as MoP 0.

In some embodiments, in order to train or update the sub-model by the terminal device 310, the network device 370 may transmit an indication of activating at least one sub-model of the AI model, wherein the at least one sub-model includes the sub-model. After receiving the indication of activating at least one sub-model of the AI model, the terminal device 310 may train or update the at least one sub-model. In other words, the network device 370 may indicate to activate one or multiple sub-models from the configured sub-models; accordingly, multiple sub-models configured on the terminal device can be active simultaneously, and thus may be trained or updated by the terminal device in parallel. In this way, multiple MoPs may be monitored simultaneously, so as to reduce the monitoring latency. In an example, the maximum number of active MoPs for monitoring of an AI feature is N, where N is an integer and N>1. For example, a number of the at least one sub-model of the AI model indicated to be activated is smaller than a pre-defined number.

In some embodiments, the activation/deactivation signaling (e.g., a signaling for activating/deactivating a MoP or a signaling for activating/deactivating an AI feature) may be a UE-specific signaling, or a UE-group specific signaling, or a broadcast signaling. When a MoP is activated at a terminal device, the terminal device starts to train/update the MoP. When MoP is deactivated at a terminal device, the training or updating of the MoP by the terminal device is ended. In an example, if the network device 370 indicates by a broadcast signaling that an AI feature learning is disabled, then all the MoPs associated with the AI feature may be disabled, i.e., deactivated.

In some embodiments, the terminal device 310 may transmit assistance information to the network device 370. The assistance information may include a computing capability of the terminal device. Alternatively or additionally, the assistance information may include a size of a dataset of the terminal device for model training or model updating. In other words, the terminal device 310 may report assistance information to the network device 370 to assist the network device 370 to make MoP activation decision. For example, the terminal device 310 may report its dynamic capability and/or dynamic dataset size to the network device 370. If the capability of the terminal device 310 is increased, the network device 370 may indicate the terminal device 310 to switch to train a larger sub-model, e.g. switch to another MoP with more parameters.

In some embodiments, the sub-model is a first sub-model. When training or updating the first sub-model, the terminal device 310 may train or update the first sub-model with input data associated with the AI model. The terminal device 310 may train or update a second sub-model of the AI model with the input data associated with the AI model. In an example, all the configured sub-models of the AI model may be trained, updated, monitored and/or inferenced using the same input data. For example, the network device 370 may indicate the input data for all configured MoPs and any one of the configured MoPs can use the same input data for training, monitoring or inferencing. In some embodiments, the terminal device 310 may receive an indication of at least one of a format of the input data for the AI model or a format of an output data for the AI model. In other words, the same input format may be used for all the configured sub-models of the AI model, and/or the same output format may be used for all the configured sub-models of the AI model. In an example, the network device 370 may indicate input/output format for the configured MoPs, e.g. by broadcast/multicast/unicast signaling

In some embodiments, the sub-model is a first sub-model. When training or updating the first sub-model, the terminal device 310 may train or update the first sub-model with first input data. The terminal device 310 may train or update a second sub-model of the AI model with second input data different from the first input data. In an example, the network device 370 may indicate the index of a MoP and the input data for the MoP to the terminal device 310. The MoP with the indicated MoP index can use the input data for the MoP for training, monitoring or inferencing. In some embodiments, the terminal device 310 may receive an indication of at least one of a format of the first input data or a format of an output data for the first sub-model. In other words, different input formats or different output formats may be used for different configured MoPs of an AI model. For example, different model structures may be used for different configured MoPs, so the input/output formats for different MoPs may be different. In some embodiments, the network device 370 may indicate the input/output format for a configured MoP, e.g. by broadcast/multicast/unicast signaling. Separate configuration signaling may be used for indicating the input/output formats for different MoPs.

In some embodiments, the sub-model is a first sub-model. When reporting the information associated with the first sub-model, the terminal device 310 may compress at least one parameter of the first sub-model in a first compression, and transmit the at least one compressed parameter to the network device 370. The terminal device 310 may also receive a configuration of a second sub-model of the AI model. In an example, the second sub-model may be trained or updated when activated. The terminal device 310 may compressing at least one parameter of the second sub-model in a second compression scheme, and transmit the at least one compressed parameter of the second sub-model. In some implementations, the first compression scheme is the same as the second compression scheme. In other words, the same compression scheme may be used for different configured MoPs. In an example, for a MoP configuration, Float32 quantization may be used for compressing parameters of the MoP. In some implementations, the first compression scheme is different from the second compression scheme. In other words, different compression schemes may be used for different configured MoPs. In an example, considering that different MoPs may have different priorities, different compression schemes (e.g. compression ratio) may be used for for different MoPs. In a more specific example, Float16 quantization may be used for MoP0 which is less important and Float32 quantization may be used for MoP1 which is more important.

With some embodiments of the present disclosure, distributed learning by multiple devices is enabled, and each device may train or update partial of the final big AI model. For example, after a sub-model of an AI model is trained/updated by a terminal device, the terminal device may report information of its trained/updated model part to the network device. In some scenarios, the performance of the sub-model might not be good enough after being trained/updated and it will lead to unnecessary UL reporting overhead for the trained/updated sub-model. Therefore, a sub-model shall be validated or tested before information of the sub-model is reported to the network device. In the present disclosure, the AI/ML model validation refers to a subprocess of training, to evaluate the quality of an AI/ML model using a dataset different from a dataset used for model training, that helps selecting model parameters that generalize beyond the dataset used for model training. In the present disclosure, the AI/ML model testing refers to a subprocess of training, to evaluate the performance of a final AI/ML model using a dataset different from a dataset used for model training and validation. Different from the AI/ML model validation, the AI/ML model testing does not assume subsequent tuning of the model.

In some embodiments, before reporting the information associated with the sub-model, the terminal device 310 may determine a performance of the sub-model based on a test dataset and transmit an indication of the performance of the sub-model to the network device 370. The network device 370 may determine whether to report the information associated with the sub-model by the terminal device 310 based on the performance of the sub-model. If the terminal device 310 receives an indication to report the information associated with the sub-model, the terminal device 310 may transmit at least one parameter of the sub-model. If the terminal device 310 receives an indication to not report the information associated with the sub-model, the terminal device 310 may continue training or updating the sub-model. Alternatively, if the terminal device 310 fails to receive an indication to report the information associated with the sub-model in a time duration after transmitting the indication of the performance of the sub-model to the network device 370, the terminal device 310 may continue training or updating the sub-model. In order to determine a performance of the sub-model, the terminal device 310 may receive the test dataset for the sub-model and receive an indication to perform a validation operation for the sub-model. The terminal device 310 may determine the performance of the sub-model based on determining that the indication to perform the validation operation is received.

In a more specific example, the network device 370 may configure a test dataset (e.g., including input data and ground-truth data) for a MoP #i. The configuration of the test data may be UE-specific, group-common, or via a broadcast signaling. The network device 370 may indicate the terminal device 310 to perform a validation operation for the MoP #i. The terminal device 310 with the MoP #i (e.g. configured with MoP #i or activated with MoP #i) will validate the model part performance of the MoP #i according to the test dataset. The terminal device 310 may report the performance of the MoP #i. Based on the reporting performance, the network device 370 may indicate whether the performance of validation is good enough. If the terminal device 310 is indicated that the performance is bad, the terminal device 310 shall continue to train the MoP #i; if the terminal device 310 is indicated that the performance is good, the terminal device 310 shall stop training the MoP #i and report the information associated with the trained MoP #i.

During the training/updating procedure of the distributed AI model, the network device 370 and the terminal device 310 may exchange the trained/updated MoP several times. For example, federated learning may be used for training/updating the MoP. In wireless federated learning (FL) , the network device 370 may initialize MoP#i, samples a group of terminal devices and group-cast parameters of the MoP#i to the selected terminal devices. Each terminal device may initialize its local sub-model using the received model parameters, and updates (trains) its local sub-model using its own data. Then each terminal device may report the updated local sub-model’s parameters to the network device 370. The network device 370 aggregates the updated parameters reported from the terminal devices and updates the MoP#i. The aforementioned procedure is one iteration of AI training procedure. The network device 370 and the terminal devices may perform multiple iterations until the MoP#i is finalized. In other words, the network device 370 and the selected terminal device (s) may perform a federated learning of the sub-model by iteratively receiving at least one parameter of the sub-model from the other one, training or updating the sub-model and transmitting the at least one trained or updated parameter to the other one.

One aspect is related to enabling the terminal devices selected for training/updating the MoP#i to receive the MoP#i parameters from the network device and/or transmit the trained/updated MoP#i parameters to the network device.

In some embodiments, the sub-model is associated with a dedicated radio network temporary identifier (RNTI) . In other words, a MoP is associated with a dedicated RNTI. FIG. 4 illustrates example distributed sub-models of an AI model associated with corresponding RNTIs according to some embodiments of the present disclosure. As shown in FIG. 4, RNTI-MoP0 is associated with MoP 0, RNTI-MoP1 is associated with MoP 1, and RNTI-MoP2 is associated with MoP 2. The RNTI value for a MoP may be configured by the network device 370.

In some implementations, the terminal device 310 may receive a DCI to schedule a resource for DL transmission of the sub-model. A cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI associated with the sub-model. The terminal device 310 may receive at least one parameter of the sub-model using the resource scheduled by the DCI. In an example, for DL transmission of MoP#i, the network device 370 may use a DCI to schedule the resources for the MoP transmission, where the DCI can be a terminal device-specific DCI or group-common DCI, and CRC of the DCI may be scrambled with the RNTI associated with MoP#i. Therefore, only the terminal devices configured with the MoP#i and the RNTI associated with the MoP#i can decode the DCI, and receive the MoP#i scheduled by the DCI. In addition, the PDSCH or PUSCH for MoP transmission can be scrambled by the RNTI associated with the MoP#i.

In some implementations, the terminal device 310 may receive a DCI to schedule a resource for reporting the information associated with the sub-model. A CRC of the DCI is scrambled with the dedicated RNTI associated with the sub-model. When reporting the information associated with the sub-model, the terminal device 370 may transmit at least one parameter of the sub-model using the resource scheduled by the DCI. In some embodiments, the terminal device 370 may transmit a scheduling request (SR) for reporting the information associated with the sub-model, wherein the scheduling request is associated with the sub-model. In an example, for UL transmission of MoP#i, the terminal device 310 may transmit a scheduling request (SR) to the network device 370. The SR resource for transmitting the SR may be associated with the UL transmission the MoP#i, which may be configured by the network device 370. Therefore, by the dedicated SR resource for MoP#i, the network device 370 is aware that the terminal device 310 is requesting the UL resources for MoP#i transmission, and thus may transit a DCI scrambled by the RNTI associated to the MoP#i for scheduling the UL transmission of the MoP#i.

In some embodiments, a group of sub-models is associated with a dedicated RNTI. In other words, a MoP group may be associated with a dedicated RNTI. For example, a first MoP group including MoP 0 and MoP 1 is associated with group-RNTI0 and a second MoP group including MoP 2 is associated with group-RNTI1. The RNTI value of a MoP group may be configured by the network device 370. In this way, the DCI blind detection complexity can be reduced, thus saving the resource overhead and power consumption of terminal devices.

In some implementations, the terminal device 310 may receive a DCI to schedule a resource for DL transmission of the sub-model. A CRC of the DCI is scrambled with the dedicated RNTI associated with the group of sub-models. The DCI may indicate a sub-model among the group of sub-models the DCI. The terminal device 310 may receive at least one parameter of the indicated sub-model using the resource scheduled by the DCI. In an example, for DL transmission of MoP#i, the network device 370 may use a DCI to schedule the resources for the MoP transmission, where the DCI can be a terminal device-specific DCI or group-common DCI, a CRC of the DCI may be scrambled with the RNTI associated with MoP group #j comprising the MoP#i and the DCI comprises an indication of the MoP#i. Therefore, only the terminal devices configured with a MoP among the MoP group #j and the RNTI associated with the MoP group #j can decode the DCI, and only the terminal devices configured with MoP group #j can receive the MoP#i scheduled by the DCI. In addition, the PDSCH or PUSCH for MoP transmission can be scrambled by the RNTI associated with the MoP group #j.

In some implementations, the terminal device 310 may receive a DCI to schedule a resource for reporting the information associated with the sub-model. A CRC of the DCI is scrambled with the dedicated RNTI associated with the group of sub-models. The DCI may indicate a sub-model among the group of sub-models the DCI. When reporting the information associated with the sub-model, the terminal device 370 may transmit at least one parameter of the sub-model using the resource scheduled by the DCI. In some embodiments, the terminal device 370 may transmit a scheduling request (SR) for reporting the information associated with the sub-model, wherein the scheduling request is associated with the sub-model. In an example, for UL transmission of MoP#i, the terminal device 310 may transmit a scheduling request (SR) to the network device 370. The SR resource for transmitting the SR may be associated with the UL transmission the MoP#i, which may be configured by the network device 370. Therefore, by the dedicated SR resource for MoP#i, the network device 370 is aware that the terminal device 310 is requesting the UL resources for MoP#i transmission, and thus may transit a DCI for scheduling the UL transmission of the MoP#i, where a CRC of the DCI may be scrambled with the RNTI associated with MoP group #j comprising the MoP#i and the DCI comprises an indication of the MoP#i.

It should be noted that the above scheme for MoP transmission can be also applied to other scenarios. For example, after training and validation of a MoP, the terminal device will report its trained MoP to the network device, where the MoP delivery procedure is similar to above implementations.

In the above embodiments, the life cycle management of the distributed AI model is further discussed with respect to various aspects, e.g., configuring, activating, training/updating, validating and reporting the sub-models of the AI model. In this way, the distributed training/updating performed by multiple devices in the network is enabled, and each device updates partial of the big model, to improve the learning efficiency.

FIG. 5 illustrates a flowchart of a method 500 of communication implemented at a terminal device in accordance with some embodiments of the present disclosure. The method 500 can be implemented at the terminal device 310 shown in FIG. 3 or the UE 110 or ED 110 as shown in FIGS. 1A to 1E. For the purpose of discussion, the method 500 will be described with reference to FIG. 3. It is to be understood that the method 500 may include additional acts not shown and/or may omit some shown acts, and the scope of the present disclosure is not limited in this regard.

At 510, the terminal device 310 receives a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature. At 520, the terminal device 310 trains or updates the sub-model. At 530, the terminal device 310 reports information associated with the sub-model. It should be noted that the method 500 may include various other operations which may be performed by the terminal device 310 as described above with reference to the signaling process 300 of FIG. 3.

FIG. 6 illustrates a flowchart of a method 600 of communication implemented at a network device in accordance with some embodiments of the present disclosure. The method 600 can be implemented at the network device 370 shown in FIG. 3 or the BS 170, T-TRP 170 or NT-TRP 172 as shown in FIGS. 1A to 1E. For the purpose of discussion, the method 600 will be described with reference to FIG. 3. It is to be understood that the method 600 may include additional acts not shown and/or may omit some shown acts, and the scope of the present disclosure is not limited in this regard.

At 610, the network device 370 transmits a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature. At 610, the network device 370 receives a report of information associated with the sub-model, wherein the sub-model is trained or updated by a terminal device. It should be noted that the method 600 may include various other operations which may be performed by the network device 370 as described above with reference to the signaling process 300 of FIG. 3.

FIG. 7 is a block diagram of a device 700 that may be used for implementing some embodiments of the present disclosure. In some embodiments, the device 700 may be an element of communications network infrastructure, such as a base station (for example, a NodeB, an evolved Node B (eNodeB, or eNB) , a next generation NodeB (sometimes referred to as a gNodeB or gNB) , a home subscriber server (HSS) , a gateway (GW) such as a packet gateway (PGW) or a serving gateway (SGW) or various other nodes or functions within a core network (CN) or a Public Land Mobility Network (PLMN) . In other embodiments, the device 700 may be a device that connects to the network infrastructure over a radio interface, such as a mobile phone, smart phone or other such device that may be classified as a User Equipment (UE) . In some embodiments, the device 700 may be a Machine Type Communications (MTC) device (also referred to as a machine-to-machine (M2M) device) , or another such device that may be categorized as a UE despite not providing a direct service to a user. In some embodiments, the device 700 may be a road side unit (RSU) , a vehicle UE (V-UE) , pedestrian UE (P-UE) or an infrastructure UE (I-UE) . In some scenarios, the device 700 may also be referred to as a mobile device, a term intended to reflect devices that connect to mobile network, regardless of whether the device itself is designed for, or capable of, mobility. Specific devices may utilize all of the components shown or only a subset of the components, and levels of integration may vary from device to device. Furthermore, the device 700 may contain multiple instances of a component, such as multiple processors, memories, transmitters, receivers, etc.

The device 700 typically includes a processor 702, such as a Central Processing Unit (CPU) , and may further include specialized processors such as a Graphics Processing Unit (GPU) or other such processor, a memory 704, a network interface 706 and a bus 708 to connect the components of the device 700. The device 700 may optionally also include components such as a mass storage device 710, a video adapter 712, and an I/O interface 716 (shown in dashed lines) .

The memory 704 may comprise any type of non-transitory system memory, readable by the processor 702, such as static random access memory (SRAM) , dynamic random access memory (DRAM) , synchronous DRAM (SDRAM) , read-only memory (ROM) , or a combination thereof. In an embodiment, the memory 704 may include more than one type of memory, such as ROM for use at boot-up, and DRAM for program and data storage for use while executing programs. The bus 708 may be one or more of any type of several bus architectures including a memory bus or memory controller, a peripheral bus, or a video bus.

The device 700 may also include one or more network interfaces 706, which may include at least one of a wired network interface and a wireless network interface. As illustrated in FIG. X, network interface 706 may include a wired network interface to connect to a network 722, and also may include a radio access network interface 720 for connecting to other devices over a radio link. When the device 700 is a network infrastructure element, the radio access network interface 720 may be omitted for nodes or functions acting as elements of the PLMN other than those at the radio edge (e.g., an eNB) . When the device 700 is infrastructure at the radio edge of a network, both wired and wireless network interfaces may be included. When the device 700 is a wirelessly connected device, such as a User Equipment, radio access network interface 720 may be present and it may be supplemented by other wireless interfaces such as WiFi network interfaces. The network interfaces 706 allow the device 700 to communicate with remote entities such as those connected to network 722.

The mass storage 710 may comprise any type of non-transitory storage device configured to store data, programs, and other information and to make the data, programs, and other information accessible via the bus 708. The mass storage 710 may comprise, for example, one or more of a solid state drive, hard disk drive, a magnetic disk drive, or an optical disk drive. In some embodiments, the mass storage 710 may be remote to the device 700 and accessible through use of a network interface such as interface 706. In the illustrated embodiment, the mass storage 710 is distinct from memory 704 where it is included, and may generally perform storage tasks compatible with higher latency, but may generally provide lesser or no volatility. In some embodiments, the mass storage 710 may be integrated with a heterogeneous memory 704.

The optional video adapter 712 and the I/O interface 716 (shown in dashed lines) provide interfaces to couple the device 700 to external input and output devices. Examples of input and output devices include a display 714 coupled to the video adapter 712 and an I/O device 718 such as a touch-screen coupled to the I/O interface 716. Other devices may be coupled to the device 700, and additional or fewer interfaces may be utilized. For example, a serial interface such as Universal Serial Bus (USB) (not shown) may be used to provide an interface for an external device. Those skilled in the art will appreciate that in embodiments in which the device 700 is part of a data center, I/O interface 716 and Video Adapter 712 may be virtualized and provided through network interface 706.

FIG. 8 is a schematic diagram of a structure of an apparatus 800 in accordance with some embodiments of the present disclosure. As shown in FIG. 8, the apparatus 800 includes a receiving unit 802, a training/updating unit 804 and a reporting unit 806. The apparatus 800 may be applied to the communication system as shown in FIGS. 1A to 1F, and may implement any of the methods provided in the foregoing embodiments. Optionally, a physical representation form of the apparatus 800 may be a communication device, for example, a UE. Alternatively, the apparatus 800 may be another apparatus that can implement a function of a communication device, for example, a processor or a chip inside the communication device. Specifically, the apparatus 800 may be some programmable chips such as a field-programmable gate array (field-programmable gate array, FPGA) , a complex programmable logic device (complex programmable logic device, CPLD) , an application-specific integrated circuit (application-specific integrated circuits, ASIC) , or a system on a chip (System on a chip, SOC) .

In some embodiments, the receiving unit 802 may be configured to receive a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature.

In some embodiments, the training/updating unit 804 may be configured to train or update the sub-model.

In some embodiments, the reporting unit 802 may be configured to report information associated with the sub-model.

In some other embodiments, the apparatus 800 can include various other units or modules which may be configured to perform various operations or functions as described in connection with the foregoing method embodiments. The details can be obtained referring to the detailed description of the foregoing method embodiments and are not described herein again.

It should be noted that division into the units or modules in the foregoing embodiments of the present disclosure is an example, and is merely logical function division. In actual implementation, there may be another division manner. In addition, function units in embodiments of the present disclosure may be integrated into one processing unit, or may exist alone physically, or two or more units may be integrated into one unit. The integrated unit may be implemented in a form of hardware, or may be implemented in a form of a software function unit.

When the integrated unit is implemented in a form of a software function unit and sold or used as an independent product, the integrated unit may be stored in a computer-readable storage medium. Based on such an understanding, the technical solutions of the present disclosure essentially, or all or some of the technical solutions may be implemented in a form of a software product. The computer software product is stored in a storage medium and includes several instructions for instructing a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to perform all or some of the steps of the methods described in embodiments of the present disclosure. The foregoing storage medium includes any medium that can store program code, such as a USB flash drive, a removable hard disk, a read-only memory (Read-Only Memory, ROM) , a random access memory (Random Access Memory, RAM) , a magnetic disk, or an optical disc.

FIG. 9 is a schematic diagram of a structure of an apparatus 900 in accordance with some embodiments of the present disclosure. As shown in FIG. 9, the apparatus 900 includes a transmitting unit 902 and a receiving unit 904. The apparatus 900 may be applied to the communication system as shown in FIGS. 1A to 1F, and may implement any of the methods provided in the foregoing embodiments. Optionally, a physical representation form of the apparatus 900 may be a communication device, for example, a network device. Alternatively, the apparatus 900 may be another apparatus that can implement a function of a communication device, for example, a processor or a chip inside the communication device. Specifically, the apparatus 900 may be some programmable chips such as a field-programmable gate array (field-programmable gate array, FPGA) , a complex programmable logic device (complex programmable logic device, CPLD) , an application-specific integrated circuit (application-specific integrated circuits, ASIC) , or a system on a chip (System on a chip, SOC) .

In some embodiments, the transmitting unit 902 may be configured to transmit a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature.

In some embodiments, the receiving unit 904 may be configured to receive a report of information associated with the sub-model, wherein the sub-model is trained or updated by a terminal device.

In some other embodiments, the apparatus 900 can include various other units or modules which may be configured to perform various operations or functions as described in connection with the foregoing method embodiments. The details can be obtained referring to the detailed description of the foregoing method embodiments and are not described herein again.

Based on the foregoing embodiments, an embodiment of the present disclosure further provides a computer program. When the computer program is run on a computer, the computer is enabled to perform any of the methods provided in the foregoing embodiments.

Based on the foregoing embodiments, an embodiment of the present disclosure further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a computer, the computer is enabled to perform the any of the methods provided in the foregoing embodiments. The storage medium may be any usable medium that can be accessed by a computer. By way of example and not limitation, the computer-readable medium may include a RAM, a ROM, an EEPROM, a CD-ROM or another optical disk storage, a magnetic disk storage medium or another magnetic storage device, or any other medium that can be used to carry or store expected program code in a form of an instruction or a data structure and that can be accessed by a computer.

Based on the foregoing embodiments, an embodiment of the present disclosure further provides a chip. The chip is configured to read a computer program stored in a memory, to implement any of the methods provided in the foregoing embodiments.

Based on the foregoing embodiments, an embodiment of the present disclosure provides a chip system. The chip system includes a processor, configured to support a computer apparatus in implementing functions related to communication devices in the foregoing embodiments. In a possible design, the chip system further includes a memory, and the memory is configured to store a program and data that are necessary for the computer apparatus. The chip system may include a chip, or may include a chip and another discrete component.

Based on the foregoing embodiments, an embodiment of the present disclosure provides an apparatus/chipset system comprising means (e.g., at least one processor) to implement a method implemented by (or at) a UE of the present disclosure. The apparatus/chipset system may be the UE (that is, a terminal device) or a module/component in the UE. In details, the at least one processor may execute instructions stored in a computer-readable medium to implement the method.

Based on the foregoing embodiments, an embodiment of the present disclosure provides an apparatus/chipset system comprising means (e.g., at least one processor) to implement the method implemented by (or at) a network device (e.g., base station) of the present disclosure. The apparatus/chipset system may be the network device or a module/component in the network device. In details, the at least one processor may execute instructions stored in a computer-readable medium to implement the method. In some aspects of the present disclosure, there is provided a system comprising at least one of an apparatus in (or at) a UE of the present disclosure, or an apparatus in (or at) a network device of the present disclosure.

It will be appreciated that any module, component, or device disclosed herein that executes instructions may include, or otherwise have access to, a non-transitory computer/processor readable storage medium or media for storage of information, such as computer/processor readable instructions, data structures, program modules and/or other data. A non-exhaustive list of examples of non-transitory computer/processor readable storage media includes magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, optical disks such as compact disc read-only memory (CD-ROM) , digital video discs or digital versatile discs (i.e., DVDs) , Blu-ray Disc^TM, or other optical storage, volatile and non-volatile, removable and non-removable media implemented in any method or technology, random-access memory (RAM) , read-only memory (ROM) , electrically erasable programmable read-only memory (EEPROM) , flash memory or other memory technology. Any such non-transitory computer/processor storage media may be part of a device/apparatus or accessible or connectable thereto. Computer/processor readable/executable instructions to implement a method, an application or a module described herein may be stored or otherwise held by such non-transitory computer/processor readable storage media.

A person skilled in the art should understand that embodiments of the present disclosure may be provided as a method, a system, or a computer program product. Therefore, the present disclosure may be in a form of a hardware-only embodiment, a software-only embodiment, or an embodiment combining software and hardware aspects. In addition, the present disclosure may be in a form of a computer program product implemented on one or more computer-usable storage media (including but not limited to a magnetic disk memory, a CD-ROM, an optical memory, and the like) including computer-usable program code.

The present disclosure is described with reference to the flowcharts and/or block diagrams of the method, the device (system) , and the computer program product according to the present disclosure. It should be understood that computer program instructions may be used to implement each process and/or each block in the flowcharts and/or the block diagrams and a combination of a process and/or a block in the flowcharts and/or the block diagrams. These computer program instructions may be provided for a general-purpose computer, a dedicated computer, an embedded processor, or a processor of another programmable data processing device to generate a machine, so that the instructions executed by a computer or a processor of another programmable data processing device generate an apparatus for implementing a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.

These computer program instructions may alternatively be stored in a computer-readable memory that can indicate a computer or another programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate an artifact that includes an instruction apparatus. The instruction apparatus implements a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.

These computer program instructions may alternatively be loaded onto a computer or another programmable data processing device, so that a series of operations and steps are performed on the computer or the another programmable device, to generate computer-implemented processing. Therefore, the instructions executed on the computer or the another programmable device provide steps for implementing a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.

The solutions described in the disclosure is applicable to a next generation (e.g. sixth generation (6G) or later) network, or a legacy (e.g. 5G, 4G, 3G or 2G) network.

It is clear that a person skilled in the art may make various modifications and variations to the present disclosure without departing from the protection scope of the present disclosure. Thus, the present disclosure is intended to cover these modifications and variations, provided that they fall within the scope of the claims of the present disclosure and their equivalent technologies. Although this disclosure refers to illustrative embodiments, this is not intended to be construed in a limiting sense. Various modifications and combinations of the illustrative embodiments, as well as other embodiments of the disclosure, will be apparent to persons skilled in the art upon reference to the description. When combining two or more embodiments, not all the features in the embodiments to be combined are necessary for the combination.

Features disclosed herein in the context of any particular embodiments may also or instead be implemented in other embodiments. Method embodiments, for example, may also or instead be implemented in apparatus, system, and/or computer program product embodiments. In addition, although embodiments are described primarily in the context of methods and apparatus, other implementations are also contemplated, as instructions stored on one or more non-transitory computer-readable media, for example. Such media could store programming or instructions to perform any of various methods consistent with the present disclosure.

Claims

A method comprising:

receiving, by a terminal device, a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature;

training or updating the sub-model; and

reporting information associated with the sub-model.
The method of claim 1, wherein the sub-model is associated with one of the following:

at least one parameter among a plurality of parameters associated with the AI model;

a functionality among a plurality of functionalities associated with the AI model based on a one-to-one correspondence;

at least one functionality among a plurality of functionalities associated with the AI model;

an AI task among a plurality of AI tasks associated with the AI model based on a one-to-one correspondence;

at least one AI task among a plurality of AI tasks associated with the AI model;

a predefined size requirement; or

a predefined neuron network algorithm.
The method of claim 1, wherein the sub-model is associated with a first subset of parameters among a plurality of parameters associated with the AI model, the sub-model is a first sub-model, and the method further comprises:

receiving a configuration of a second sub-model of the AI model, wherein the second sub-model is associated with a second subset of parameters among the plurality of parameters;

wherein:

the first subset of parameters and the second subset of parameters are non-overlapped; or

the first subset of parameters and the second subset of parameters are overlapped.
The method of any of claims 1-3, wherein receiving the configuration of the sub-model comprises:

receiving at least one configuration of at least one sub-model, wherein the at least one sub-model comprises the sub-model.
The method of any of claims 1-3, wherein receiving the configuration of the sub-model comprises:

receiving a plurality of configurations of a plurality of candidate sub-models; and

receiving at least one configuration of at least one sub-model among the plurality of candidate sub-models, wherein the at least one sub-model comprises the sub-model.
The method of claim any of claims 1-5, wherein the sub-model is identified based on an index, wherein the index is unique in at least one set of sub-models of at least one AI models associated with one or more AI features.
The method of claim any of claims 1-5, wherein the sub-model is identified based on an index of the sub-model and an index of the AI feature.
The method of any of claims 1-7, wherein training or updating the sub-model comprises:

receiving an indication of activating the sub-model.
The method of claim 8, wherein the sub-model is a first sub-model, and the method further comprises:

upon determining that a second sub-model of the AI model is being activated when the indication is received, deactivating the second sub-model.
The method of any of claims 1-7, wherein training or updating the sub-model comprises:

receiving an indication of activating the AI feature, wherein the sub-model is a default sub-model of the AI feature.
The method of any of claims 1-7, wherein training or updating the sub-model comprises:

receiving an indication of activating at least one sub-model of the AI model, wherein the at least one sub-model comprises the sub-model; and

training or updating the at least one sub-model.
The method of claim 11, wherein a number of the at least one sub-model is smaller than a pre-defined number.
The method of any of claims 1-12, further comprising:

transmitting assistance information, wherein the assistance information comprises at least one of the following:

a computing capability of the terminal device; or

a size of a dataset of the terminal device for model training or model updating.
The method of any of claims 1-13, wherein the sub-model is a first sub-model, and training or updating the sub-model comprises:

training or updating the first sub-model with input data associated with the AI model; and

wherein the method further comprises:

training or updating a second sub-model of the AI model with the input data associated with the AI model.
The method of claim 14, further comprising:

receiving an indication of at least one of a format of the input data for the AI model or a format of an output data for the AI model.
The method of any of claims 1-13, wherein the sub-model is a first sub-model, and training or updating the sub-model comprises:

training or updating the first sub-model with first input data; and

wherein the method further comprises:

training or updating a second sub-model of the AI model with second input data different from the first input data.
The method of claim 16, further comprising:

receiving an indication of at least one of a format of the first input data or a format of an output data for the sub-model.
The method of any of claims 1-17, wherein reporting the information associated with the sub-model comprises:

compressing at least one parameter of the sub-model in a first compression; and

transmitting the at least one compressed parameter; and

wherein the sub-model is a first sub-model, and wherein the method further comprises:

receiving a configuration of a second sub-model of the AI model;

compressing at least one parameter of the second sub-model in a second compression scheme; and

transmitting the at least one compressed parameter of the second sub-model.
The method of claim 18, wherein:

the first compression scheme is the same as the second compression scheme; or

the first compression scheme is different from the second compression scheme.
The method of any of claims 1-19, further comprising:

determining a performance of the sub-model based on a test dataset;

transmitting an indication of the performance of the sub-model; and

wherein reporting the information associated with the sub-model comprise: transmitting at least one parameter of the sub-model upon determining that an indication to report the information associated with the sub-model is received; and

wherein the method further comprises: continuing training or updating the sub-model upon determining that an indication to not report the information associated with the sub-model is received.
The method of claim 20, wherein determining a performance of the sub-model comprises:

receiving the test dataset for the sub-model;

receiving an indication to perform a validation operation for the sub-model; and

determining the performance of the sub-model based on determining that the indication to perform the validation operation is received.
The method of any of claims 1-21, wherein the sub-model is associated with a dedicated radio network temporary identifier (RNTI) .
The method of claim 22, further comprising:

receiving downlink scheduling information (DCI) to schedule a resource for the sub-model, wherein a cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI; and

receiving at least one parameter of the sub-model using the resource scheduled by the DCI.
The method of claim 22, further comprising:

receiving downlink scheduling information (DCI) to schedule a resource for reporting the information associated with the sub-model, wherein a cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI; and

wherein reporting the information associated with the sub-model comprises: transmitting at least one parameter of the sub-model using the resource scheduled by the DCI.
The method of claim 24, further comprising:

transmitting a scheduling request for reporting the information associated with the sub-model, wherein the scheduling request is associated with the sub-model.
The method of any of claims 1-25, further comprising:

performing a federated learning of the sub-model by iteratively receiving at least one parameter of the sub-model, training or updating the sub-model and transmitting the at least one trained or updated parameter.
A method comprising:

transmitting, by a network device, a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature; and

receiving a report of information associated with the sub-model, wherein the sub-model is trained or updated by a terminal device.
The method of claim 27, wherein the sub-model is associated with one of the following:

at least one parameter among a plurality of parameters associated with the AI model;

a functionality among a plurality of functionalities associated with the AI model based on a one-to-one correspondence;

at least one functionality among a plurality of functionalities associated with the AI model;

an AI task among a plurality of AI tasks associated with the AI model based on a one-to-one correspondence;

at least one AI task among a plurality of AI tasks associated with the AI model;

a predefined size requirement; or

a predefined neuron network algorithm.
The method of claim 27, wherein the sub-model is associated with a first subset of parameters among a plurality of parameters associated with the AI model, the sub-model is a first sub-model, and the method further comprises:

transmitting a configuration of a second sub-model of the AI model, wherein the second sub-model is associated with a second subset of parameters among the plurality of parameters;

wherein:

the first subset of parameters and the second subset of parameters are non-overlapped; or

the first subset of parameters and the second subset of parameters are overlapped.
The method of any of claims 27-29, wherein transmitting the configuration of the sub-model comprises:

transmitting at least one configuration of at least one sub-model, wherein the at least one sub-model comprises the sub-model.
The method of any of claims 27-29, wherein transmitting the configuration of the sub-model comprises:

transmitting a plurality of configurations of a plurality of candidate sub-models; and

transmitting at least one configuration of at least one sub-model among the plurality of candidate sub-models, wherein the at least one sub-model comprises the sub-model.
The method of claim any of claims 27-31, wherein the sub-model is identified based on an index, wherein the index is unique in at least one set of sub-models of at least one AI models associated with one or more AI features.
The method of claim any of claims 27-31, wherein the sub-model is identified based on an index of the sub-model and an index of the AI feature.
The method of any of claims 27-33, further comprising:

transmitting an indication of activating the sub-model.
The method of any of claims 27-33, further comprising:

transmitting an indication of activating the AI feature, wherein the sub-model is a default sub-model of the AI feature.
The method of any of claims 27-33, further comprising:

transmitting an indication of activating at least one sub-model of the AI model, wherein the at least one sub-model comprises the sub-model.
The method of claim 36, wherein the number of the at least one sub-model is smaller than a pre-defined number.
The method of any of claims 27-37, further comprising:

receiving assistance information, wherein the assistance information comprises at least one of the following:

a computing capability of the network device; or

a size of a dataset of the network device for model training or model updating.
The method of any of claims 27-38, further comprising:

transmitting an indication of at least one of a format of input data for the AI model or a format of an output data for the AI model.
The method of any of claims 27-38, further comprising:

transmitting an indication of at least one of a format of input data for the sub-model or a format of an output data for the sub-model.
The method of any of claims 27-40, wherein receiving the report of the information associated with the sub-model comprises:

receiving at least one parameter of the sub-model, wherein the at least one parameter of the sub-model is compressed in a first compression scheme, the sub-model is a first sub-model, and

wherein the method further comprises:

receiving at least one parameter of a second sub-model of the AI model, wherein the at least one parameter of the second sub-model is compressed in a second compression scheme.
The method of claim 41, wherein:

the first compression scheme is the same as the second compression scheme; or

the first compression scheme is different from the second compression scheme.
The method of any of claims 27-42, further comprising:

receiving an indication of a performance of the sub-model;

determining whether to report the information associated with the sub-model based on the performance of the sub-model; and

transmitting an indication to report the information associated with the sub-model or an indication not to report the information associated with the sub-model based on the determination.
The method of any of claims 27-42, further comprising:

transmitting a test dataset for determining a performance of the sub-model; and

transmitting an indication to perform a validation operation for the sub-model.
The method of any of claims 27-44, wherein the sub-model is associated with a dedicated radio network temporary identifier (RNTI) .
The method of claim 45, further comprising:

transmitting downlink scheduling information (DCI) to schedule a resource for the sub-model, wherein a cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI; and

transmitting at least one parameter of the sub-model using the resource scheduled by the DCI.
The method of claim 45, further comprising:

transmitting downlink scheduling information (DCI) to schedule a resource for the report of the information associated with the sub-model, wherein a cyclic redundancy check (CRC) of the DCI is scrambled with the dedicated RNTI.
The method of claim 47, further comprising:

receiving a scheduling request for the report of the information associated with the sub-model, wherein the scheduling request is associated with the sub-model.
The method of any of claims 27-48, further comprising:

performing a federated learning of the sub-model by iteratively receiving at least one parameter of the sub-model, training or updating the sub-model and transmitting the at least one trained or updated parameter.
A terminal device comprising:

a transceiver; and

a processor communicatively coupled with the transceiver,

wherein the processor is configured to:

receive, via the transceiver, a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature;

train or update the sub-model; and

report information associated with the sub-model.
A network device comprising:

a transceiver; and

a processor communicatively coupled with the transceiver,

wherein the processor is configured to:

transmit, via the transceiver, a configuration of a sub-model of an artificial intelligence (AI) model associated with an AI feature; and

receive, via the transceiver, a report of information associated with the sub-model, wherein the sub-model is trained or updated by a terminal device.
A non-transitory computer readable medium comprising computer program stored thereon, the computer program, when executed on at least one processor, causing the at least one processor to perform the method of any of claims 1-49.
A chip comprising at least one processing circuit configured to perform the method of any of claims 1-49.
A computer program product tangibly stored on a computer-readable medium and comprising computer-executable instructions which, when executed, cause an apparatus to perform the method of any of claims 1-49.