WO2025055252A1

WO2025055252A1 - Controlling method and related apparatuses

Info

Publication number: WO2025055252A1
Application number: PCT/CN2024/074290
Authority: WO
Inventors: Yiqun Ge; Jianglei Ma; Wuxian Shi; Huazi ZHANG; Wen Tong
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2023-09-13
Filing date: 2024-01-26
Publication date: 2025-03-20
Anticipated expiration: 2026-03-13
Also published as: CN121753382A

Abstract

Provided are a controlling method and related apparatuses. The method includes: generating, by using a language model (LM), a controlling message in natural language; obtaining, according to the controlling message, semantic information of the controlling message; forming an information payload, wherein the information payload comprises the semantic information of the controlling message; and sending the information payload. With the controlling method and related apparatuses of the present disclosure, an open vocabulary can be supported in controlling, more effective communication can be achieved, backward and forward compatibility can be supported, and cross-modality functionality can be accommodated.

Description

CONTROLLING METHOD AND RELATED APPARATUSES

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims priority to U.S. Provisional Application No. 63/582,307, filed on September 13, 2023. The disclosure of the above patent application is incorporated herein by reference in its entirety.

TECHNICAL FIELD

The present disclosure relates to the field of wireless communication technologies, and in particular, to a controlling method and an apparatus, a device, a system and a storage medium.

BACKGROUND

Communication can be divided into three levels: bit communication (s) , semantic communication (s) , and effectiveness communication (s) . In the past 60 years, wireless communication has focused on the level of bit communication, the lowest level, which is called as a “technical problem” to ensure an absolute accuracy of bits transmitted from a transmitter to a receiver. Effectiveness communication, as the highest level, was proposed to achieve an ultimate goal of communications, which is called as an “effectiveness problem” to ensure a success of conveying a semantic meaning to the receiver leads to the desired conduct on the receiver side.

This background information is provided to reveal information believed by the applicant to be of possible relevance to the present disclosure. No admission is necessarily intended, nor should be construed, that any of the preceding information constitutes prior art against the present disclosure.

SUMMARY

In a first aspect, an embodiment of the present disclosure provides a controlling method, where the method includes:

generating, by using a language model (LM) , a controlling message in natural language;

obtaining, according to the controlling message, semantic information of the controlling message;

forming an information payload, wherein the information payload comprises the semantic information of the controlling message; and

sending the information payload.

A central device generates a controlling message in natural language, and forms an information payload including the semantic information of the controlling message to control a user device. In this way, an open vocabulary can be supported in controlling, more effective communication can be achieved, backward and forward compatibility can be supported, and cross-modality functionality can be accommodated.

In a possible implementation of the first aspect, the LM may be a language learning model (LLM) .

In a possible implementation of the first aspect, the method further includes: obtaining registering information of a function modality, where the registering information of the function modality includes a description of the function modality, where the description is in the natural language.

In a possible implementation of the first aspect, the registering information of the function modality further includes a list of arguments of the function modality.

In a possible implementation of the first aspect, the method further includes: registering the function modality to the LM.

Since the registering information of a function modality includes the description is in natural language or further includes a list of arguments of the function modality, the system can accommodate the cross-modality functionality and support backward and forward compatibility. What’s more, when new function modalities are added, they can be registered with their respective descriptions in natural language without requiring changes to the system architecture or communication protocols, which improves the universality of the system, and makes the system more flexible.

In a possible implementation of the first aspect, the method further includes: recording the registering information of the function modality. After obtaining the registering information, it can be stored locally for further use, thereby facilitating further maintenance.

In a possible implementation of the first aspect, the method further includes: recording information indicative of a user device identifier (ID) of a user device from which the registering information of the function modality is received. The central device may record information indicative of a user device ID of a user device together with the registering information for further use, to call respective function modalities of different user devices more directly without the need for additional communication or negotiation, thereby helping optimize the utilization of transmission resources.

In a possible implementation of the first aspect, the method further includes: obtaining, by using an embedder and according to the description of the function modality, semantic information of the function modality. The embedder may translate, embed, or tokenize the description in natural language into semantic information, and the semantic information is in a more structured and machine-readable format, and is usually small in size compared with the description in natural language.

In a possible implementation of the first aspect, the method further includes: storing the semantic information of the function modality. After obtaining the semantic information of the function modality, it can be stored locally for future use, and thus the description does not need to be translated every time it needs to be used, thereby facilitating subsequent use, decreasing the processing overhead of description translation, and reducing communication latency.

In a possible implementation of the first aspect, the method further includes: sending the semantic information of the function modality. The central device may send the semantic information of the function modality, especially when a user device is unable to run any transform-based embedder, which improves the compatibility and flexibility of the system.

In a possible implementation of the first aspect, before the forming the information payload, the method further includes: determining a relevance between the semantic information of the controlling message and semantic information of each of at least one function modality that has been registered, where semantic information of each of the at least one function modality is obtained according to a description of each of the at least one function modality.

In a possible implementation of the first aspect, the method further includes: determining that the relevance between the semantic information of the controlling message and the semantic information of each of the at least one function modality registered is lower than a threshold, and rejecting the controlling message.

The central device would determine a relevance between the semantic information of the controlling message and semantic information of each of at least one function modality that has been registered, to determine whether the controlling message is a registered one. When the relevance between the semantic information of the controlling message and the semantic information of each of the at least one function modality registered is lower than a threshold, it can be determined that the controlling message is not registered, and the central device would reject the non-registered controlling message. The central device may not need to encode or transmit the non- registered controlling message and the central device may multicast or unicast the controlling message to only those user devices that have registered for the corresponding function modality, thereby saving the central device’s and user device’s power, and saving transmission resources.

In a possible implementation of the first aspect, the obtaining, according to the controlling message, the semantic information of the controlling message includes: determining, from the controlling message, a portion without any arguments; and obtaining, by using an embedder, and according to the portion, the semantic information of the controlling message. The generated controlling message may include a list of arguments (values) , and the central device may divide the controlling message into two portions: a first portion including the list of arguments if any and a second portion including the rest part of the controlling message, and the central device only translate the second portion into the semantic information. Since the arguments (values) , which may vary, are not included in the second portion translated to the semantic information, the semantic information of the controlling message can be highly relevant to the semantic information of the corresponding function modality, thereby facilitating the central device to determine whether the controlling message is a registered one, and facilitating the user device to determine a corresponding function modality, and meanwhile, reducing the computational complexity and facilitating standardizing the length of the semantic information.

In a possible implementation of the first aspect, the forming the information payload includes:

forming a first information payload, where the first information payload includes starting position information of a second transmission opportunity and length information of the semantic information; and

forming a second information payload, where the second information payload includes the semantic information of the controlling message, and where the second transmission opportunity is for transmission of the second information payload information.

The central device may form a first information payload and a second information payload to transmit the controlling message, which allows for more efficient use of radio resources and provides greater flexibility in terms of error correction and retransmission. Since the starting position information of the second transmission opportunity and the length information of the semantic information are included in the first information payload, the receiving device can synchronize and correctly receive the second information payload.

In a possible implementation of the first aspect, the controlling message further includes a list of arguments, and the first information payload further includes the list of arguments of the controlling message.

In a possible implementation of the first aspect, the method further includes: encoding the second information payload by using a first encoding method; where the first information payload further includes information of the first encoding method. By including the first encoding method used to encode the second information payload in the first information payload, the user device can immediately identify the appropriate coding scheme required to decode the subsequent second information payload.

In a possible implementation of the first aspect, the method further includes: encoding the first information payload by using a second encoding method.

In a possible implementation of the first aspect, the first encoding method comprises a first modulation and coding scheme (MCS) , and the second encoding method comprises a second MCS.

In a possible implementation of the first aspect, the method further includes:

determining, from the semantic information of each of the at least one function modality that has been registered, and according to the relevance between the semantic information of the controlling message and the semantic information of each of at least one function modality that has been registered, semantic information most relevant to the semantic information of the controlling message; and

determining at least one user device ID or at least one group user device ID corresponding to the most relevant semantic information;

where the first information payload further includes: the at least one user device ID, or the at least one group user device ID; or the first information payload further includes information indicative of the at least one user device ID or information indicative of the at least one group user device ID.

The central device can determine the specific function modality that the controlling message is targeted to, and determine a corresponding user device or group user device that the controlling message is sent to, which ensures that the controlling message is delivered to the intended recipients, improving the efficiency and accuracy of the communication system. What’s more, since the user device ID or the group user device ID, or the information indicative of the user device ID or information indicative of the group user device ID is included in the first information payload, the receiving user device can determine whether it is a correct recipient according to the first information payload, thereby improving the efficiency and accuracy of the communication system, and when the receiving user device is not an intended recipient, the user device may stop decoding the second information payload, thereby saving the energy of the user device.

In a possible implementation of the first aspect, the information indicative of the at least one user device ID includes a first code generated according to the at least one user device ID, where the first code includes a mask code, a spreading code, or an interleaving code; or the information indicative of the at least one group user device ID includes a second code generated according to the at least one group user device ID, where the second code includes a mask code, a spreading code, or an interleaving code.

In a possible implementation of the first aspect, the method further includes: generating the first code according to the at least one user device ID; or generating the second code according to the at least one group user device ID.

The user device ID or group user device ID could be indicated via a code generated based on them. Instead of transmitting the user device identifier or group user device identifier as a part of the first information payload, the user device identifier or group user device identifier can be used to generate a mask code, spreading code, or interleaving code, which provides more flexibility for the system, and supports blind detection and early termination to save the energy of the user device.

In a possible implementation of the first aspect, the sending the information payload includes:

sending the first information payload during a first transmission opportunity; and

sending the second information payload during the second transmission opportunity.

The central device may send the first information payload and the second information payload during the first transmission opportunity and the second transmission opportunity respectively. Thus the central device can allocate the radio resource more effectively by assigning different transmission opportunities to each payload, and can help prevent congestion and ensure that the information is transmitted smoothly.

In a possible implementation of the first aspect, the first transmission opportunity is in a controlling physical channel, and the second transmission opportunity is in a data physical channel; or both the first transmission opportunity and the second transmission opportunity are in a controlling physical channel; or both the first transmission opportunity and the second transmission opportunity are in a data physical channel. These different configurations allow for flexibility in radio resource management and enable efficient transmission of control and data information in the communication system.

In a possible implementation of the first aspect, the registering the function modality to the LM includes: registering the function modality to at least one LMs by providing the registering information of the function modality to the at least one LMs, where the at least one LM includes the LM.

In a possible implementation of the first aspect, the method further includes: sending configuration information of at least one embedder, where the embedder belongs to the at least one embedder.

In a possible implementation of the first aspect, the configuration information of the at least one embedder is sent via a broadcast message, a multicast message, or a unicast message.

There may be one or more LMs, and each LM may include one or more embedders, the function modality may be registered to at least one of the one or more LMs. Different LMs or embedders may be compatible with different types of communication systems or used for different application scenarios. By allowing for multiple LMs to be deployed on the central device, the system becomes more flexible and customizable. When different embedders are built on different domains, specialization on the vocabulary can greatly reduce the size and cost of the embedder but also improve the accuracy of the relevance computation.

In a possible implementation of the first aspect, the configuration information of the at least one embedder includes an architecture and parameters of each of the at least one embedder.

In a possible implementation of the first aspect, the configuration information of the at least one embedder further includes an embedder ID of each of the at least one embedder.

The configuration information of the embedder may include an architecture and parameters of each embedder, and the configuration information of the embedder may further include an embedder ID of each embedder, to facilitate the configuration of each embedder to the user device.

In a possible implementation of the first aspect, the first information payload further includes: information indicative of an embedder ID of an embedder used to obtain the semantic information of the controlling message. The central device may include the information indicative of an embedder ID used to obtain the semantic information of the controlling message, thereby improving the accuracy of the decision-making by the user device.

In a possible implementation of the first aspect, the information payload is sent via a multicast message or a unicast message.

In a possible implementation of the first aspect, the method further includes:

in response to generating, by the LM, at least two consecutive controlling messages more than a preset number of times, combining two function modalities corresponding to the two consecutive controlling messages to generate a combined function modality;

sending configuration information for the combined function modality.

In a possible implementation of the first aspect, the method further includes: registering the combined function modality to the LM.

If the central device consistently generates at least two consecutive controlling messages for a user device, it is possible to combine function modalities corresponding to the at least two controlling message into one function modality, thereby enhancing effectiveness by streamlining the execution of multiple functions within a single calling of a combined function modality.

In a possible implementation of the first aspect, there are multiple LMs, and the multiple LMs include the LM, where each of the multiple LMs has at least one embedder.

In a possible implementation of the first aspect, at least one function modality is registered in one or more LMs of the multiple LMs, but not registered in all the multiple LMs.

There may be multiple LMs, and different LMs may be developed by different companies, or may be for different specific purposes, which can reduce the size and cost by the LMs, improve the effectiveness of the generated controlling messages, enhancing resources utilization of the central device and improve overall efficiency.

In a possible implementation of the first aspect, the multiple LMs include at least one LM compatible for an LTE system.

In a possible implementation of the first aspect, the multiple LMs include at least one LM compatible for a 5G system.

In a possible implementation of the first aspect, a first LM of the multiple LMs has a first embedder and a second LM of the multiple LM has a second embedder, and the first embedder and a second embedder built on vocabularies of different domains.

In a possible implementation of the first aspect, the LM has a first embedder and a second embedder built on vocabularies of different domains.

The multiple LMs may be compatible for different systems, or an LM may have multiple embedders each of which may be built on vocabularies of different domains. The specialization on the vocabulary can reduce the size and cost of the embedder, but also improve the accuracy of the relevance computation.

In a possible implementation of the first aspect, the method further includes: registering the LM.

In a possible implementation of the first aspect, the method further includes: registering at least one embedder of the LM.

The registration for the LM or the embedder of the LM enables the system to recognize and utilize the LM and its embedder, allowing for effective natural language processing and improved system performance.

In a second aspect, an embodiment of the present disclosure provides a controlling method, where the method includes:

obtaining an information payload, where the information payload includes semantic information of a controlling message;

determining, according to the semantic information of the controlling message, target semantic information from at least one piece of semantic information, where each piece of the at least one piece of semantic information corresponds to a respective function modality in at least one function modality and is obtained according to a description of the respective function modality, and the description of the respective function modality is in natural language; and

executing a function modality corresponding to the target semantic information.

A user device may obtain an information payload including semantic information of a controlling message, may determine a target semantic information based on the controlling message, and may execute the corresponding function modality. In this way, open vocabulary can be supported in controlling, more effective communication can be achieved, backward and forward compatibility can be supported, and cross-modality functionality can be accommodated.

In a possible implementation of the second aspect, the method further includes: encapsulating each function modality of the at least one function modality into a calling function, where the calling function for the respective function modality includes the description of the respective function modality, or the calling function for the respective function modality includes a list of arguments of the respective function modality and the description of the respective function modality.

In a possible implementation of the second aspect, the calling function comprises an application programming interface (API) calling function.

Since each function modality can be encapsulated into a calling function, such as the API calling function, the calling function can be called to execute the function modality. The calling function can be implemented in various programming languages, by encapsulating each function modality of the at least one function modality into a calling function, it becomes easier to integrate and use with different programming languages, and promote interoperability between different systems or components that may be developed using different technologies.

In a possible implementation of the second aspect, the method further includes: sending registering information of each function modality of the at least one function modality, where the registering information of each function modality includes the description of the respective function modality, or, the registering information of each function modality includes a list of arguments of the respective function modality and the description of the respective function modality.

Since the registering information of a function modality includes the description in natural language or further includes a list of arguments of the function modality, the system can accommodate the cross-modality functionality and support backward and forward compatibility. What’s more, when new function modalities are added, they can be registered with their respective descriptions in natural language without requiring changes to the system architecture or communication protocols, which improves the universality of the system, and makes the system more flexible.

In a possible implementation of the second aspect, the obtaining the information payload includes:

obtaining a first information payload during a first transmission opportunity, where the first information payload includes starting position information of a second transmission opportunity and length information of the semantic information of the controlling message;

decoding the first information payload to obtain the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message;

obtaining a second information payload during the second transmission opportunity according to the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message, where the second information payload includes the semantic information of the controlling message; and

decoding the second information payload to obtain the semantic information of the controlling message.

The user device may obtain the first information payload and the second information payload during the first transmission opportunity and the second transmission opportunity respectively. Thus the radio resource can be allocated more effectively by assigning different transmission opportunities to each payload, and the process of obtaining and decoding the information payload ensures the reliable retrieval of the semantic information of the controlling message.

In a possible implementation of the second aspect, the first information payload further includes: a list of arguments of the controlling message.

In a possible implementation of the second aspect, the first information payload further includes information of a first encoding method, and the second information payload is decoded by using the first encoding method. By including the first encoding method used to encode the second information payload in the first information payload, the user device can immediately identify the appropriate coding scheme required to decode the subsequent second information payload.

In a possible implementation of the second aspect, the first information payload is decoded by using a second encoding method.

In a possible implementation of the second aspect, the first encoding method includes a first modulation and coding scheme (MCS) , and the second encoding method comprise a second MCS.

In a possible implementation of the second aspect, the first information payload further includes: information indicative of a user device identifier (ID) or information indicative of a group user device ID.

Since the information indicative of the user device ID or information indicative of the group user device ID is included in the first information payload, the user device can determine whether it is a correct recipient according to the first information payload, thereby improving the efficiency and accuracy of the communication system.

In a possible implementation of the second aspect, the method further includes: obtaining a first code generated according to information indicative of a user device ID, where the code includes a mask code, a spreading code, or an interleaving code; or obtaining a second code generated according to information indicative of a group user device ID; where the code includes a mask code, a spreading code, or an interleaving code. Instead of transmitting the user device identifier or group user device identifier as a part of the first information payload, the user device identifier or group user device identifier can be used to obtain a mask code, spreading code, or interleaving code, which supports blind detection and early termination to save the energy of the user device.

In a possible implementation of the second aspect, the method further includes: in response to determining that the user device ID or the group user device ID does not match a user device ID or a group user device ID of a user device, stopping decoding the second information payload. When the user device identifier is indicated in the first information payload, the user device may stop decoding the second information payload, in a case that the user device finds that the decoded user device ID or the group user device ID doesn’ t match that of itself, thereby saving the energy and processing resources of the user device.

In a possible implementation of the second aspect, the method further includes: obtaining configuration information of at least one embedder. Through the configuration information of at least one embedder, the at least one embedder can be configured to the user device and then the user device can use the at least one embedder to translate the description of the function modality to semantic information.

In a possible implementation of the second aspect, the semantic information of the controlling message is obtained by using a first embedder of the at least one embedder, and the at least one piece of semantic information of the at least one function modality is obtained by using the first embedder. Since the user device may compare the semantic information of the controlling message with the candidate semantic information obtained by using the same embedder, accuracy and reliability of the comparing result would be improved.

In a possible implementation of the second aspect, the configuration information of the at least one embedder includes an embedder ID of each of the at least one embedder. By including the embedder ID of each of the at least one embedder in the configuration, it becomes simpler to track and manage the specific embedder that is used or that needs updates or maintenance.

In a possible implementation of the second aspect, the information payload includes information indicative of the embedder ID of the first embedder.

In a possible implementation of the second aspect, the information indicative of the embedder ID of the first embedder is carried in the first information payload.

In a possible implementation of the second aspect, the method further includes:

obtaining, by using the at least one embedder and according to the description of each of the at least one function modality, one or more pieces of semantic information, where for each of the at least one function modality, each of the at least one embedder is used to obtain a respective piece of semantic information; and

storing the one or more pieces of semantic information.

The user device may use each of the at least one embedder to obtain a respective piece of semantic information for each of the at least one function modality, and may store the obtained semantic information, which enables the user device to access and utilize the semantic information when needed quickly.

In a possible implementation of the second aspect, the method further includes: obtaining, from the stored one or more pieces of semantic information and according to the information indicative of the embedder ID of the first embedder, the at least one piece of semantic information of the at least one function modality. When the semantic information of each function modality has been stored in the user device, the user device can obtain the candidate semantic information from the stored semantic information according to the embedder ID information, so as to determine the target semantic information.

In a possible implementation of the second aspect, after obtaining the information payload, the method further includes: obtaining, by using the first embedder and according to the description of each of the at least one function modality, the at one piece of semantic information of the at least one function modality. In some cases, the candidate semantic information is not stored in the user device, for example, when the user device has limited storage, or the embedder has just been configured or updated, etc. The user device may obtain the candidate semantic information by using a corresponding embedder.

In a possible implementation of the second aspect, the method further includes: receiving the at least one piece of semantic information of the at least one function modality. The user device may receive the semantic information from other devices, especially when the user device is unable to run any transform-based embedder, which improves the compatibility and flexibility of the system.

In a possible implementation of the second aspect, the determining the target semantic information includes: determining, from the at least one piece of semantic information, a piece of semantic information most relevant to the semantic information of the controlling message as the target semantic information.

In a possible implementation of the second aspect, more than one function modality exists, and the method further comprises:

obtaining configuration information for a combined function modality, where the combined function modality is combined from at least two function modalities of the more than one function modality; and

encapsulating the combined function modality into a first calling function.

If the central device consistently generates at least two consecutive controlling messages for a user device, it is possible to combine the function modalities corresponding to the at least two controlling message into one function modality, thereby enhancing effectiveness by streamlining the execution of multiple functions within a single calling of a combined function modality.

In a possible implementation of the second aspect, the configuration information for the combined function modality includes a calling order, and the first calling function calls calling functions corresponding to the at least two function modalities in the calling order.

The combined function calls the multiple calling function successively. Along with the running time, combined function modality that encapsulates a plurality of old function modalities in some order would be created so that the effectiveness gets improved.

In a third aspect, an embodiment of the present disclosure provides a first apparatus including various modules configured to execute the controlling method according to the first aspect or any possible implementation of the first aspect.

In a fourth aspect, an embodiment of the present disclosure provides a second apparatus including various modules configured to execute the controlling method according to the second aspect or any possible implementation of the second aspect.

In a fifth aspect, an embodiment of the present disclosure provides a third apparatus including processing circuitry for executing the controlling method according to the first aspect or any possible implementation of the first aspect.

In a sixth aspect, an embodiment of the present disclosure provides a fourth apparatus including processing circuitry for executing the controlling method according to the second aspect or any possible implementation of the second aspect.

In a seventh aspect, an embodiment of the present disclosure provides a communication system, including a first apparatus according to the third aspect or a third apparatus according to the fifth aspect, and a second apparatus according to the fourth aspect or a fourth apparatus according to the sixth aspect.

In an eighth aspect, an embodiment of the present disclosure provides a computer-readable medium storing computer execution instructions which, when executed by a processor, causes the processor to execute the controlling method according to the first aspect or any possible implementation of the first aspect or according to the second aspect or any possible implementation of the second aspect.

In a ninth aspect, an embodiment of the present disclosure provides a computer program product including computer execution instructions which, when executed by a processor, causes the processor to execute the controlling method according to the first aspect or any possible implementation of the first aspect or according to the second aspect or any possible implementation of the second aspect.

The present disclosure provides a controlling method and related apparatuses. The controlling message in natural language is generated by using an LM; semantic information of the controlling message is obtained according to the controlling message; an information payload is formed, where the information payload includes the semantic information of the controlling message; and the information payload is sent. In this way, open vocabulary can be support in controlling, more effective communication can be achieved, backward and forward compatibility can be supported, and cross-modality functionality can be accommodated.

BRIEF DESCRIPTION OF DRAWINGS

Reference will now be made, by way of example, to the accompanying drawings which show example embodiments of the present disclosure, and in which:

FIG. 1 is a simplified schematic illustration of a communication system according to one or more embodiments of the present disclosure.

FIG. 2 is a schematic illustration of an example communication system according to one or more embodiments of the present disclosure.

FIG. 3 is a schematic illustration of a basic component structure of a communication system according to one or more embodiments of the present disclosure.

FIG. 4 illustrates a block diagram of a device in a communication system according to one or more embodiments of the present disclosure.

FIG. 5 is a schematic diagram of a communication system.

FIG. 6 is a schematic diagram showing communications between a central device and a first user device.

FIG. 7 is a schematic diagram of an LLM model according one or more embodiments of the present disclosure.

FIG. 8 shows a schematic flowchart of a controlling method according to one or more example embodiments of the present disclosure.

FIG. 9 shows a schematic flowchart of a controlling method according to one or more example embodiments of the present disclosure.

FIG. 10 shows a schematic flowchart of a registering process according to one or more embodiments of the present disclosure.

FIG. 11 is a schematic diagram of a central device according one or more embodiments of the present disclosure.

FIG. 12 is a schematic diagram of configuration of an embedder according one or more embodiments of the present disclosure.

FIG. 13 is a schematic diagram of a function modality according one or more embodiments of the present disclosure.

FIG. 14 is a schematic diagram of a function modality according one or more embodiments of the present disclosure.

FIG. 15 is a schematic diagram of an example of obtaining a semantic vector according one or more embodiments of the present disclosure.

FIG. 16 is a schematic diagram of an example of informing a central device of registering information of a function modality by a user device according one or more embodiments of the present disclosure.

FIG. 17 is a schematic diagram of an example of registering a function modality by a central device according one or more embodiments of the present disclosure.

FIG. 18 is a schematic diagram of an example of recording a second user device by a central device according one or more embodiments of the present disclosure.

FIG. 19 is a schematic diagram of an example of generating a controlling message according one or more embodiments of the present disclosure.

FIG. 20 is a schematic diagram of an example of a relevance comparing process according one or more embodiments of the present disclosure.

FIG. 21 is a schematic diagram of an example of encoding and transmitting an open-voc controlling message in two transmission opportunity by a central device according one or more embodiments of the present disclosure.

FIG. 22 is a schematic diagram of an example of decoding and receiving an open-voc controlling message in two transmission opportunity by a user device according one or more embodiments of the present disclosure.

FIG. 23 is a schematic diagram of an example of executing process by a user device according one or more embodiments of the present disclosure.

FIG. 24 shows a schematic structural diagram of a first apparatus according to one or more example embodiments of the present disclosure.

FIG. 25 shows a schematic structural diagram of a second apparatus according to one or more example embodiments of the present disclosure

DESCRIPTION OF EMBODIMENTS

In the following description, reference is made to the accompanying figures, which form part of the present disclosure, and which show, by way of illustration, specific aspects of embodiments of the present disclosure or specific aspects in which embodiments of the present disclosure may be used. It is understood that embodiments of the present disclosure may be used in other aspects and include structural or logical changes not depicted in the figures. The following detailed description, therefore, is not to be taken in a limiting sense, and the scope of the present disclosure is defined by the appended claims.

To describe the technical solutions in embodiments of the present invention or in the prior art more clearly, the following briefly introduces the accompanying drawings needed for describing the embodiments or the prior art.

Wireless communication system heavily relies on pre-negotiated (standardized) contents and formats of controlling messages, which may cause backward and/or forward compatibility issue among different communication system.

For illustrative purposes, specific example embodiments will now be explained in greater detail in conjunction with the figures.

The embodiments set forth herein represent information sufficient to practice and illustrate ways of practicing such subject matter. Upon reading the following description in light of the accompanying figures, those of skill in the art will understand the concepts of the claimed subject matter and will recognize applications of these concepts not particularly addressed herein. It should be understood that these concepts and applications fall within the scope of the disclosure and the accompanying claims.

Moreover, 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.

Further, in the disclosure, the word “a” or “an” when used in conjunction with the term “comprising” or “including” in the claims and/or the specification may mean “one” , but it is also consistent with the meaning of “one or more” , “at least one” , and “one or more than one” unless the content clearly dictates otherwise. Similarly, the word “another” may mean at least a second or more unless the content clearly dictates otherwise.

The terms “coupled” , “coupling” or “connected” as used herein can have several different meanings depending on the context in which these terms are used. For example, as used herein, the terms coupled, coupling, or connected can indicate that two elements or devices are directly connected to one another or connected to one another through one or more intermediate elements or devices via a mechanical element depending on the particular context. The term “and/or” herein when used in association with a list of items means any one or more of the items comprising that list.

A two-stage controlling messaging system method, and apparatus for a wireless system is provided in this disclosure. The wireless system contains one center device (also called central device, e.g., one apparatus such as a base station, a core network) and at least one user device (e.g., another apparatus such as a UE) . In the following, the scenario in which the center device transmits some controlling messages to the user device is taken as an example. Please note that, the scenario in which the user device transmits some controlling messages to the center device could be included in the same way in this disclosure.

Referring to FIG. 1, as an illustrative example without limitation, a simplified schematic illustration of a communication system is provided. The communication system 100 (which may be the wireless system in FIG. 1) 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 a legacy (e.g. 5G, 4G, 3G or 2G) 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.

The uplink messages/data transmitted between the central device (e.g., the network node 170) and the sensing device (e.g., ED 180) could be carried in higher layer signaling, such as RRC signaling, or MAC layer signaling. Or, they could be carried in physical layer signaling, e.g., UCI. Or they could be carried in the combination of the higher layer signaling and the physical signaling. It could be noted that the message in the present disclosure could be replaced with information, which may be carried in one single message, or be carried in more than one separate message. The downlink messages/data transmitted between the central device and the ED 110 could be carried in higher layer signaling, such as RRC signaling, or MAC layer signaling. Or, they could be carried in physical layer signaling, e.g., UCI. Or they could be carried in the combination of the higher layer signaling and the physical signaling. It could be noted that the message in the present disclosure could be replaced with information, which may be carried in one single message, or be carried in more than one separate message.

Custom network power savings implementations can be considered where an operator performs power saving manually based on heuristics, traffic load management and balancing.

FIG. 2 illustrates an example communication system 100. In general, the communication system 100 enables multiple wireless or wired elements to communicate data and other content. The purpose of the communication system 100 may be to provide content, such as voice, data, video, and/or text, via broadcast, multicast, groupcast, unicast, etc. The communication system 100 may operate by sharing resources, such as carrier spectrum bandwidth, between its constituent elements. The communication system 100 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. 2, 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. As may be surmised on the basis of similarity in reference numerals, the non-terrestrial communication network 120c may be considered to be a radio access network, with operational aspects in common with the RANs 120a, 120b. The non-terrestrial communication network 120c may include at least one non-terrestrial network (NTN) device and at least one corresponding terrestrial network device, wherein the at least one non-terrestrial network device works as a transport layer device and the at least one corresponding terrestrial network device works as a radio access network node, which communicates with the ED via the non-terrestrial network device.

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) , or single-carrier FDMA (SC-FDMA, also known as discrete Fourier transform spread OFDMA, DFT-s-OFDMA) 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 172 for 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 to support such.

FIG. 3 is a schematic illustration of a basic component structure of a communication system according to one or more example embodiments of the present disclosure.

FIG. 3 illustrates another example of an ED 110 and a base station 170a, 170b and/or 170c. The ED 110 is used to connect persons, objects, machines, etc. The ED 110 may be widely used in various scenarios including, 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, a pair of glasses, head mounted equipment, etc. ) , an industrial device, or an apparatus in (e.g. communication module, modem, or chip) or comprising the forgoing devices, among other possibilities. Future generation EDs 110 may be referred to using other terms. The base station 170a and 170b is a T-TRP and will hereafter be referred to as T-TRP 170. Also shown in FIG. 3, 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 a transmitter 201 and a receiver 203 coupled to one or more antennas 204. Only one antenna 204 is illustrated to avoid congestion in the drawing. One, some, or all of the antennas 204 may alternatively be panels. The transmitter 201 and the receiver 203 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 204 or network interface controller (NIC) . The transceiver is also configured to demodulate data or other content received by the at least one antenna 204. 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 204 includes any suitable structure for transmitting and/or receiving wireless or wired signals.

The ED 110 includes at least one memory 208. The memory 208 stores instructions and data used, generated, or collected by the ED 110. For example, the memory 208 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 210) . Each memory 208 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. 1) . The input/output devices or interfaces permit interaction with a user or other devices in the network. Each input/output device or interface includes any suitable structure for providing information to or receiving information from a user, and/or for network interface communications. Suitable structures include, for example, a speaker, microphone, keypad, keyboard, display, touch screen, etc.

The ED 110 includes the processor 210 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 203, possibly using receive beamforming, and the processor 210 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 210 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 210 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 210 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 210 may form part of the transmitter 201 and/or part of the receiver 203. Although not illustrated, the memory 208 may form part of the processor 210.

The processor 210, the processing components of the transmitter 201, and the processing components of the receiver 203 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 208) . Alternatively, some or all of the processor 210, the processing components of the transmitter 201, and the processing components of the receiver 203 may each be implemented using dedicated circuitry, such as a programmed field-programmable gate array (FPGA) , an application-specific integrated circuit (ASIC) , or a hardware accelerator such as a graphics processing unit (GPU) or an artificial intelligence (AI) accelerator.

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 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 256 for the T-TRP 170, and may be coupled to the equipment that houses the antennas 256 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 256 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 252 and at least one receiver 254 coupled to one or more antennas 256. Only one antenna 256 is illustrated to avoid congestion in the drawing. One, some, or all of the antennas 256 may alternatively be panels. The transmitter 252 and the receiver 254 may be integrated as a transceiver. The T-TRP 170 further includes a processor 260 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 260 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 260 also generates an indication of beam direction, e.g. BAI, which may be scheduled for transmission by a scheduler 253. The processor 260 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 260 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 260 is sent by the transmitter 252. Note that “signaling” , as used herein, may alternatively be called control signaling. Signaling may be transmitted in a physical layer control channel, e.g. a physical downlink control channel (PDCCH) , in which case the signaling may be known as dynamic signaling. Signaling transmitted in a downlink physical layer control channel may be known as Downlink Control Information (DCI) . Siganling transmitted in an uplink physical layer control channel may be known as Uplink Control Information (UCI) . Signaling transmitted in a sidelink physical layer control channel may be known as Sidelink Control Information (SCI) . Signaling may be included in a higher-layer (e.g., higher than physical layer) packet transmitted in a physical layer data channel, e.g. in a physical downlink shared channel (PDSCH) , in which case the signaling may be known as higher-layer signaling, static signaling, or semi-static signaling. Higher-layer signaling may also refer to Radio Resource Control (RRC) protocol signaling or Media Access Control -Control Element (MAC-CE) signaling.

The scheduler 253 may be coupled to the processor 260. The scheduler 253 may be included within or operated separately from the T-TRP 170. The scheduler 253 may schedule uplink, downlink, sidelink, and/or backhaul transmissions, including issuing scheduling grants and/or configuring scheduling-free (e.g., “configured grant” ) resources. The T-TRP 170 further includes a memory 258 for storing information and data. The memory 258 stores instructions and data used, generated, or collected by the T-TRP 170. For example, the memory 258 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 260.

Although not illustrated, the processor 260 may form part of the transmitter 252 and/or part of the receiver 254. Also, although not illustrated, the processor 260 may implement the scheduler 253. Although not illustrated, the memory 258 may form part of the processor 260.

The processor 260, the scheduler 253, the processing components of the transmitter 252, and the processing components of the receiver 254 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 258. Alternatively, some or all of the processor 260, the scheduler 253, the processing components of the transmitter 252, and the processing components of the receiver 254 may be implemented using dedicated circuitry, such as a programmed FPGA, a hardware accelerator (e.g., a GPU or AI accelerator) , 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 satellites and high altitude platforms, including international mobile telecommunication base stations and unmanned aerial vehicles, for example. 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 272 and a receiver 274 coupled to one or more antennas 280. Only one antenna 280 is illustrated to avoid congestion in the drawing. One, some, or all of the antennas may alternatively be panels. The transmitter 272 and the receiver 274 may be integrated as a transceiver. The NT-TRP 172 further includes a processor 276 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 276 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 276 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 278 for storing information and data. Although not illustrated, the processor 276 may form part of the transmitter 272 and/or part of the receiver 274. Although not illustrated, the memory 278 may form part of the processor 276.

The processor 276, the processing components of the transmitter 272, and the processing components of the receiver 274 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 278. Alternatively, some or all of the processor 276, the processing components of the transmitter 272, and the processing components of the receiver 274 may be implemented using dedicated circuitry, such as a programmed FPGA, a hardware accelerator (e.g., a GPU or AI accelerator) , 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.

One or more steps of the embodiment methods provided herein may be performed by corresponding units or modules, according to FIG. 4. FIG. 4 is a block diagram of a device in a communication system according to one or more example embodiments of the present disclosure. FIG. 4 illustrates units or modules in a device or apparatus, 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 artificial intelligence (AI) or machine learning (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 a circuit such as an integrated circuit. Examples of an integrated circuit includes a programmed FPGA, a GPU, or an ASIC. For instance, one or more of the units or modules may be logical such as a logical function performed by a circuit, by a portion of an integrated circuit, or by software instructions executed by a processor. 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.

In the past decades, industrial priority has been paid on bit communications by optimizing how to transmit more bits at lowest radio cost as a primary aspect of communication, i.e. transmission efficiency. In this aspect, channel coding and source coding theories have been well studied to guide how to design FEC (forward error correction) coding scheme, modulation scheme, and compression scheme, either lossy or lossless in the engineering domains. However, the priority on this aspect is overshadowing another aspect of communications, i.e. transmission effectiveness (sender triggers a desired conduct on the receiver) .

In order for a receiver to conduct as a sender expects, receivers and senders should pre-negotiate some protocols between to map desired actions (observation or actuation) to bits. In modern wireless industry, these pre-negotiated protocols become a great portion of wireless standards that every equipment device follows. For example, in a wireless system, a new desired action mapped into several bits of a controlling message (or information in the message) cannot be used until the standardization is made and hardware equipment device compatible with the standardization is deployed.

The controlling messaging system (which is a system using pre-negotiated protocols for the controlling messages) in current communication system e.g., 4^th generation (4G) and 5^th generation (5G) system is a closed-vocabulary system available and accessible only to the transmission-related controlling messages, downlink control information (DCI) and uplink control information (UCI) . A great portion of 5G wireless standard is dedicated to contents and formats of DCI and UCI. A pre-negotiated function (actions and arguments) at a receiver side is carefully mapped to some bits of a controlling message with some format. When a new transmission technology appeared (e.g., from 4G to 5G) , some new and relevant controlling messages should be added into DCI and UCI or are defined as new DCI and UCI formats after years of standardization meetings and implementations. Similarly, when an old transmission technology got retired, the old controlling messages should be removed from DCI and UCI after years of meetings and implementations too. In the future that as the accelerating evolution of transmission technology requests faster updates in DCI and UCI, the current closed-vocabulary controlling messaging system would become an obstacle to any agile replacement.

Since 5G, Ultra Reliable Low Latency Communication (URLLC) has been discussed as one of the important application features for wireless systems. The recent AI technology has been speeding up driverless car and robotics and demanding more low-latency and high reliable communications. Unfortunately, these applications cannot enjoy the most reliable and fastest radio channels, because the controlling messages in these applications is not open to the radio technology (e.g., 5G) and can only be taken as data and will be transmitted only in data channel.

Every desired action is specifically defined and mapped on several bits of a controlling message. The definition and mapping should be well pre-negotiated and written into wireless standards and then be implemented on all the equipment users (UEs) and network node, e.g., base stations. A desired action includes a specific observation or actuation function modality. For example, a controlling message to request a receiver to conduct an action to measure the current “receiving signal power (e.g. reference signal received power (RSRP) ) ” is a specific “observation” function modality (that doesn’ t change the state of the receiver) ; another controlling message to request a receiver to conduct an action to configure its “carrier frequency” is a specific “actuation” function modality (that alters the state of the receiver) . It could be understood that, a communication protocol associates one controlling message to a function modality and another controlling message to another function modality. Controlling messages are cross modality.

A modern wireless cellular system like 5G contains a central device, e.g. (BTS, gNodeB, eNodeB) , and at least one user device, e.g., (UE, user, terminal, IoT device and so on) . In 5G, the central device is responsible for managing and maintaining the overall communication quality of all user devices associated to the central device, referring to FIG. 5 and FIG. 6. FIG. 5 is a schematic diagram of a communication system. FIG. 6 is a schematic diagram showing communications between a central device and a first user device.

The central device performs the management and maintaining jobs mostly through a controlling plane of the wireless system. The controlling plane includes a physical layer controlling channel (e.g., PDCCH for downlink and PUCCH for uplink) and a part of physical layer data channel (e.g., PDSCH for downlink and PUCCH for uplink) , and a number of carefully standardized controlling messages. Usually, most controlling messages are robustly encoded (with lower coding rate and modulation scheme) without HARQ (for shorter latency purpose) and transmitted on the physical controlling channels. The content, format, and position &length of a standardized controlling message are well defined and specified so that all the equipment may exactly conform to. Usually, a standardized controlling message is short enough to form a small block to fit into DCI. Optionally, a standardized controlling message, if long enough, can be alternatively transmitted via physical data channels too.

Some bits of a standardized controlling message indicate a desired function modality, which is either observation (measurement) action and/or actuation (configuration or setup) action. Which bits are associated to which function modality is well specified (akind of offline pre-negotiation) in the standard documents, to which both the central device and user device are aligned.

Some bits of a standardized controlling message indicate argument values for a function modality. Which bits and how many bits are associated to which arguments of a function modality is well specified (akind of offline pre-negotiation) in the standard, to which both the central device and user device are aligned.

Adding a new function modality means updating some bits of a standardized controlling message, which takes years to be agreed in the standardization body and implemented by all the new equipment. For example, as the number of bits allocated for an argument value is fixed, the quantization of the value cannot be updated until the change to the number of bits is agreed by the standardization body and implemented by all the new equipment.

The central device generates a standardized controlling message by filling the bits exactly according to the pre-negotiated standard. Then the central device encodes the controlling message by some pre-negotiated coding and modulation scheme (MCS) and transmits it on a transmission opportunity within a physical radio allocation (controlling channel or data channel) to the user device.

The user device receives the controlling message from the transmission opportunity, decodes them according to the pre-negotiated MCS. The user device may conduct the function modality with the argument values in terms of the decoded bits from the controlling message.

Natural language is one of few carriers that not only support open-vocabulary and backward &forward compatibility but also accommodate cross-modality functionality. So far, all theory, rule, law, devices, knowledge, gadget, and even wireless standards are developed and stored in natural languages. Nevertheless, direct embedding a sentence in a natural language, for example by American Standard Code for Information Interchange (ASIC) letter-by-letter, is sensitive to order of words and formats of a sentence. For example, “Set the carrier frequency to 3.5GHz” would be considered as different from “the carrier frequency should be set as 3.5GHz” , even though both have the similar semantic meaning. Or “what is current received signal power? ” and “Received signal power is high” makes a pair of a query and an answer close to the same semantic. During the past decades, no machine had been able to understand flexible and open sentence in natural languages or there had been no heuristic way to map and compare sentences in natural language in a common semantic domain.

Large-language-model (LLM) enables machine to “understand” sentences in natural languages. Although the primary goal of a LLM is to make extraction or predication (generation) in sentences, every LLM may contain some components that embeds a sentence in natural language into a semantic domain. Usually, these embedding components are realized by transformer-based deep neural network and trained with contrastive learning with a great amount of training samples. This paves a way for a machine to “understand” and “compare” different sentences in semantic domain.

There are at least one of several following issues in modern wireless system. Firstly, contents and formats of the controlling messages in wireless communication standards are pre-negotiated (standardized) to form a closed vocabulary in a transmission-only semantic context. The controlling messages are responsible for managing and maintaining an entire wireless communication system, given the most reliable and fastest physical channels. But this resource or capability is not open to other applications, especially ultra-reliable low-latency communication (URLLC) applications including driverless car or robotics, for example, a controlling message such as “emergency brake” for a driverless car can only be transmitted as data via physical data channel. Secondly, pre-negotiated (standardized) controlling messages in the wireless communication standards suffer from forward and backward compatibility. While some controlling messages that become “outdated” would still “occupy” some resource room in the physical channels, some new controlling messages (for new technology) couldn’ t be added until a new standard is released (e.g. years later) . It has undermined the wireless system to benefit from cutting-edge transmission technology.

The current wireless system heavily relies on pre-negotiated (standardized) contents and formats of the controlling messages, mainly because a machine was unable to understand semantic meaning. For long, effectiveness communication had been considered as a propaganda theory, as human is only receiver who can catch semantic meaning of a sentence and conduct a desired action.

In the disclosure of the present invention, a method, apparatus and system to use the AI technology in the domain of natural language model to improve controlling messaging system in communication system from a closed vocabulary to open vocabulary to address the at least one of two problems above: providing an openness for other applications, and supporting a backward &forward compatibility.

The central device generates a controlling message in a natural language sentence. This controlling message contains a description of a function modality and one or more relevant argument values to that function modality. The function modality is a desired action implemented at the user device.

The central device embeds the description of the function modality from a natural language sentence into a semantic embedding or token by a transformer-based deep neural network.

The central device may update the transformer-based deep neural network to the user device in a pre-negotiation period. Usually, the pre-negotiation doesn’ t happen frequently.

The central device encodes the argument values and the semantic embedding respectively. The central device transmits controlling message indicating the encoded argument values in a first stage of wireless physical channel and then transmits the encoded semantic embedding in a second stage of wireless physical channel to that user device. In the first stage transmission, the central device transmits an indicator together with the encoded argument values (e.g., in the controlling message) . The indicator indicates the starting positions and lengths of the semantic embedding in the second stage transmission.

The user device implements at least one function modality. The function modality contains an API function that triggers a desired physical action and a description in natural language. The user device embeds the description of the function modality from a natural language sentence into a semantic embedding or token, as functional semantic embedding, by the transformer-based deep neural network obtained in the pre-negotiation period. The user device may generate the semantic embedding once and store it for future use, or may generate the semantic embedding every time that it receives at least a controlling message.

When the user device receives a part of the controlling message from the central device in the first stage, the user will decode it to obtain the indicator and argument values. Then, based on the indicator, the user device will receive the other part of the controlling message transmitted in the second stage. The use device decodes it to obtain the query semantic embedding. The user device may measure the correlation (or relevance) between the query semantic embedding and its functional semantic embeddings. If the user device finds a strong relevance between the query semantic embedding and one of its functional embeddings, then the user device may call the API function associated to the most relevant function embedding with the decoded argument values associated to the query semantic embedding. If the user device doesn’ t find any strong enough relevance between the query semantic embedding and anyone of its functional embeddings, the user device does nothing.

Embedding Model of LLM

The central device schedules and manages user devices mostly through a controlling plane of the wireless system. The controlling plane includes a number of physical layer channels and a number of controlling messages. Unlike 5G, the disclosure of this invention comprises some open-vocabulary controlling messages in a natural language for backward and forward compatibility in wireless communication.

These controlling messages are called as open-vocabulary controlling messages in a natural language. The major reasons to employ a natural language rather than an artificial language (like 5G standardization) are as at least one of following:

Natural language is the best de-facto semantic “standard” for nearly everything. Wireless systems (e.g., 4G, 5G) have been discussed, developed, written, and standardized in a natural language (English) .

Inference over a powerful LLM model can generate a chain or sequence of open-vocabulary controlling messages in natural language. If a modern wireless system took advantage of an ever-growing LLM model, it may be open for controlling messages in a natural language.

It is widely accepted to keep “man-in-the-loop” for LLM models in both training and inference periods. During a training period, “man-in-the-loop” can help regularize LLM model’s behavior for safe and moral codes; during an inference period, “man-in-the-loop” can help retrace and accumulate real samples for human inspection. Both of the periods request a system to support controlling messages in a natural language.

In this disclosure, the central device is supposed to use a LLM to generate a chain of controlling messages in a natural language. How to build a LLM is open for individual implementation. A company, an operator, can gain its efficiency advantage by its LLM model. The company can keep improving its LLM models by “man-in-the-loop” reinforcement learning to update the LLM’s generative models.

A typical LLM model may comprise several components: an encoder that translates, embeds, or tokenizes a sentence in a natural language into a semantic embedding, a generator or predictor that generates a sequence of semantic embeddings, and a decoder that translates generative semantic embeddings back to sentences or messages in a natural language. In some LLM models, a generator and a decoder are closely coupled together.

There is some randomness in a LLM model. For example, two semantic embeddings from the same input sentence in a natural language are not exactly same. But the two semantic embeddings are very relevant in the sense that their inner product is high.

There are many names for semantic embedding such as semantic vector, semantic token, token, and so on.In some LLM models, semantic embedding is in tensor; in other LLM models, semantic embedding is in matrix. Often, longer sentence in natural language would result into longer semantic embedding. In order to fit a semantic embedding into a wireless physical layer, an embedder is proposed:

Regularize a semantic embedding into a vector (semantic vector) ; for example, simply by vectorizing a tensor or matrix;

Pre-define a finite number of fixed lengths and then regularize the semantic vector into one of the fixed lengths. In general, longer sentence is regularized into a longer semantic vector;

Call the semantic vector of one of the pre-defined lengths as “semantic vector” in the following discussion;

Call the encoder component of a LLM and the regularizations above as “embedder” .

An embedding model is a component of a LLM model, referring to FIG. 7, which is a schematic diagram of a LLM model. In fact, it is the smallest, simplest, least energy-consuming one in a LLM model than generator and decoder that usually take billions of neurons.

In FIG. 7, a typical LLM model is illustrated. A sentence in a natural language is input into a LLM model. This sentence is firstly embedded into a semantic embedding by a deep neural network (for example, a transform-based deep neural network-1 in the figure) . Then, the semantic embedding is input into a generator or predictor that will generate a chain of semantic embeddings following this input semantic embedding. The chain of semantic embeddings are translated to a chain of sentences in a natural language by a decoder. Optionally the generator and decoder may be combined together by a deep neural network (for example, another transform-based deep neural network-2 in the figure) .

A function modality is an action conducted by a user device. An action includes observation and/or actuation. When a user device executes an observation action, the user device would measure some of its states or status. When a user device executes an actuation action, the user device would change some of its states or status. A function modality can be either observation action, or actuation observation, or both simultaneously.

FIG. 8 shows a schematic flowchart of a controlling method according to one or more example embodiments of the present disclosure. The method can be implemented by a first apparatus. Optionally, the first apparatus can be a central device or other device that has similar function (for example, the first apparatus could be a chip) , which is not limited herein. As shown in FIG. 8, the method can include the following steps.

S810, generate a controlling message in natural language by using an LM.

In the embodiment, the central device generates a controlling message by using an LM, and the controlling message is in natural language. In a possible implementation, the controlling message may be an open-vocabulary controlling message, and the controlling message is used to control a function modality of a user device. In a possible implementation, the LM may be a language learning model (LLM) .

In a possible implementation, the controlling message may further include a list of arguments. The arguments may be values or parameters necessary for the proper execution of a function modality, such as specific parameters, settings, or data inputs.

S820, obtain semantic information of the controlling message according to the controlling message.

In the embodiment, the central device obtains semantic information of the controlling message according to the controlling message. The semantic information also refers to the semantic vector, semantic embedding, semantic token, etc. in the present disclosure. The semantic information may be a numerical representation of the meaning or semantics of words, phrases, sentences, or documents.

In a possible implementation, the central device may determine a portion without any arguments from the controlling message; and obtain the semantic information of the controlling message according to the portion by using an embedder. Considering that the controlling message may include a list of arguments, the central device may divide the controlling message into two portions: a first portion including the list of arguments if any and a second portion including the rest part of the controlling message. The central device may only translate the second portion into the semantic information. Since the arguments in the controlling message are typically variable in length and content, and the length of the semantic information would depend on the length and complexity of the input message, which can vary widely and would result in pieces of semantic information of different lengths, obtaining the semantic information of the controlling message according to the portion without any arguments helps standardizing the length of the semantic information. What’s more, if the arguments is included in the translation processing, noise and variability of the semantic representation may be added, making it harder to compare and analyze the semantic relationships between different pieces of semantic information.

In order to obtain the semantic information, the central device may choose a length and use the embedder to translate the portion without any arguments into the semantic information.

In an implementation, the central device may determine a relevance between the semantic information of the controlling message and semantic information of each of at least one function modality that has been registered, where the semantic information of each of the at least one function modality is obtained according to a description of each of the at least one function modality. The semantic information of each of the at least one function modality may be stored in the central device, or may be translated from a description of a function modality when using.

In an implementation, the central device may determine the relevance between the semantic information of the controlling message and the semantic information of each of the at least one function modality registered, and when the relevance is determined to be lower than a threshold, and the central device may reject the controlling message. When the relevance between the semantic information of the controlling message and the semantic information of each of the at least one function modality registered is lower than a threshold, it can be determined that the controlling message is not registered, and the central device would reject the non-registered controlling message. The central device may not need to encode or transmit the non-registered controlling message and the central device may multicast or unicast the controlling message to only those user devices that have registered for the corresponding function modality, thereby saving the central device’s and user device’s power, and saving transmission resources.

S830, form an information payload, where the information payload includes the semantic information of the controlling message.

In the embodiment, in order to transmit the controlling message, the central device forms an information payload, and the information payload includes the semantic information of the controlling message., The semantic information is in a more structured and machine-readable format, and is usually small in size compared with the controlling message, thereby making the communication more effective and saving transmission resources.

S840, send the information payload.

In the embodiment, after forming the information payload, the central device sends the information payload. The central device may send the first information payload during the first transmission opportunity, and send the second information payload during the second transmission opportunity. In an implementation, the central device may multicast or unicast to the user devices whom are registered with the relevant function modality, saving the user device’s power.

In an implementation, the central device may form a first information payload, where the first information payload includes starting position information of a second transmission opportunity and length information of the semantic information, and the central device may form a second information payload, where the second information payload includes the semantic information of the controlling message, and where the second transmission opportunity is for transmission of the second information payload information.

In an implementation, the central device may send the first information payload during a first transmission opportunity, and send the second information payload during a second transmission opportunity.

The central device allocates a trunk of radio resource for controlling messages. A first part of the radio resource is reversed for standardized controlling messages. The second part of the radio resource is allocated for open-vocabulary controlling messages that at least includes the first open-vocabulary controlling messages.

By reserving a trunk of radio resource for controlling messages and allocating separate parts for standardized and open-vocabulary messages, the central device can ensure that these messages are transmitted efficiently and without interference from other data transmissions.

The central device may allocate specific time slots or time intervals for different transmission opportunities. This could be done using a predetermined schedule or by dynamically assigning slots based on resource availability and demand.

The first information payload and the second information payload may be encoded and transmitted during different transmission opportunities, which allows for more efficient use of radio resources and provides greater flexibility in terms of error correction and retransmission.

Since the starting position information of the second transmission opportunity and the length information of the semantic information are included in the first information payload, the receiving device can synchronize and correctly receive the second information payload.

In an implementation, the controlling message further includes a list of arguments, and the first information payload further includes the list of arguments of the controlling message. If the controlling message further includes a list of arguments, the list of arguments would be included in the first information payload to transmit.

In an implementation, the central device may encode the second information payload by using a first encoding method, where the first information payload further includes information of the first encoding method. The encoding method used on the second information payload may be explicitly or implicitly indicated by the first information payload. By including the first encoding method used to encode the second information payload in the first information payload, the user device can immediately identify the appropriate coding scheme required to decode the subsequent second information payload.

In an implementation, the central device may encode the first information payload by using a second encoding method. The second encoding method and the first encoding method can be the same or different, which is not limited here.

In an implementation, the first encoding method includes a first modulation and coding scheme (MCS) , and the second encoding method includes a second MCS. The first and the second encoding method may be MCS. In a possible implementation, the first encoding method and the second encoding method may be other encoding methods that can be used for wireless transmission, such as pulse amplitude modulation (PAM) , frequency shift keying (FSK) , phase shift keying (PSK) , quadrature amplitude modulation (QAM) , and so on.

In an implementation, the central device may determine semantic information most relevant to the semantic information of the controlling message from the semantic information of each of the at least one function modality that has been registered, according to the relevance between the semantic information of the controlling message and the semantic information of each of at least one function modality that has been registered; and may determine at least one user device ID or at least one group user device ID corresponding to the most relevant semantic information; and the first information payload may further include: the at least one user device ID, or at least one group user device ID; or the first information payload may further include information indicative of the at least one user device ID or information indicative of the at least one group user device ID.

The central device may store a mapping relationship between each registered function modality and its corresponding information indicative of the at least one user device ID or information indicative of the at least one group user device ID. In a possible implementation, the central device may store a mapping relationship between each registered function modality and its user device ID or group user device ID. The central device may compute relevance between the semantic information of the controlling message and the semantic information of each of at least one function modality that has been registered, and then determine semantic information most relevant to the semantic information of the controlling message. When the most relevant semantic information is determined, the central device can retrieve the corresponding information indicative of the at least one user device ID or information indicative of the at least one group user device ID, or user device ID or group user device ID from its database or associated records.

The central device can determine the specific function modality that the controlling message is targeted to, and determine a corresponding user device or group user device that should receive the controlling message according to the relevance between the semantic information of the controlling message and the semantic information of each of at least one function modality that has been registered, which ensures that the controlling message is delivered to the intended recipients, improving the efficiency and accuracy of the communication system.

The at least one user device ID, or at least one group user device ID, or information indicative of the at least one user device ID or information indicative of the at least one group user device ID may be included in the first information payload. Therefore, the receiving user device can determine whether it is a correct recipient according to the first information payload, thereby improving the efficiency and accuracy of the communication system. When the receiving user device is not an intended recipient, it may stop decoding the second information payload, thereby saving the energy of the user device.

In an implementation, the information indicative of the at least one user device ID may include a first code generated according to the at least one user device ID, where the first code includes a mask code, a spreading code, or an interleaving code; or the information indicative of the at least one group user device ID includes a second code generated according to the at least one group user device ID, where the second code includes a mask code, a spreading code, or an interleaving code.

In an implementation, the central device may generate the first code according to the at least one user device ID; or generating the second code according to the at least one group user device ID. The central device may apply a code generation algorithm that takes at least one user device ID or the at least one group user device ID as input and produces a desired mask code, spreading code, or interleaving code. The algorithm typically involves mathematical operations or encoding techniques to transform an identifier into a code suitable for the intended purpose. The code generation algorithm may be at least one of a hashing algorithm, a pseudorandom number generator (PRNG) , a Reed-Solomon coding or convolutional coding, a symmetric key block cipher (such as, advanced encryption standard (AES) ) , a simple bitwise operation, and so on. The hashing algorithm can be SHA-256, MD5, or other cryptographic hash functions can be used to generate a fixed-size hash value from the user device identifier.

In an implementation, the first transmission opportunity is in a controlling physical channel, and the second transmission opportunity is in a data physical channel; or both the first transmission opportunity and the second transmission opportunity are in a controlling physical channel; or both the first transmission opportunity and the second transmission opportunity are in a data physical channel. The data physical channel may be a physical downlink shared channel (PDSCH) . The controlling physical channel may be physical downlink control channel (PDCCH) . In case that the first transmission opportunity is in a controlling physical channel, such as PDCCH, while the second transmission opportunity is in a data physical channel, such as PDSCH, the first transmission opportunity may reach to the user device before the second opportunity.

In a case that both the first and second transmission opportunities are in the controlling physical channel, the order in which the first transmission opportunity and the second transmission opportunity reach the user device depends on the specific scheduling and transmission procedures defined by a cellular communication standard being used (e.g., LTE, 5G NR) .

In a case that both the first and second transmission opportunities are in a data physical channel, such as PDSCH, the first and second transmission opportunities could be scheduled simultaneously or in quick succession on the same data physical channel.

From radio resource view, it is not mandatory that the transmission opportunity may reach to the first user device before the second opportunity. There are several possible ways:

The first transmission opportunity is in the controlling physical channel such as PDCCH; and the second transmission opportunity is in the data physical channel such as PDSCH;

Both the first transmission opportunity and the second transmission opportunity are in the controlling physical channel such as PDCCH.

Both the first transmission opportunity and the second transmission opportunity are in the data physical channel such as PDSCH.

From the top level, the central device transmits the controlling message in a natural language to the user device who executes the function modality.

The central device generates a controlling message in natural language, and form an information payload including the semantic information of the controlling message to control a user device. In this way, open vocabulary can be supported in controlling, more efficient communication can be achieved, backward and forward compatibility can be supported, and cross-modality functionality can be accommodated.

FIG. 9 shows a schematic flowchart of a controlling method according to one or more example embodiments of the present disclosure. The method can be implemented by a second apparatus. Optionally, the second apparatus can be a user device or other device that has similar function (for example, the second apparatus could be a chip) , which is not limited herein. As shown in FIG. 9, the method can include the following steps.

S910, obtain an information payload, where the information payload includes semantic information of a controlling message.

In the embodiment, the user device obtains an information payload from a central device, and the information payload includes semantic information of the controlling message. The controlling message is in natural language, and is used to instruct a receiver to conduct a specific action. The semantic information of the controlling message may be a numerical representation of meaning or semantics of the controlling message.

In an implementation, the information payload may include a first information payload and a second information payload. The first information payload may be transmitted during a first transmission opportunity, and the second information payload may be transmitted during a second transmission opportunity. The first information payload may include starting position information of the second transmission opportunity and length information of the semantic information of the controlling message. The user device may obtain the first information payload during the first transmission opportunity, and decode the first information payload to obtain the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message. The user device may obtain the second information payload during the second transmission opportunity according to the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message, where the second information payload includes the semantic information of the controlling message. After obtaining the second information payload, the user device may decode the second information payload to obtain the semantic information of the controlling message.

The first transmission opportunity and the second transmission opportunity do not indicate order in which the first information payload and the second payload reach to the user device. In other words, from radio resource view, it is not mandatory that the transmission opportunity may reach to the user device before the second opportunity. The first information payload and the second information payload may be encoded and transmitted during different transmission opportunities, which allows for more efficient use of radio resources and provides greater flexibility in terms of error correction and retransmission.

The first information payload includes starting position information of the second transmission opportunity and length information of the semantic information of the controlling message. Thus the receiving device can synchronize and correctly receive the second information payload according to the first information payload.

In an implementation, the first information payload further includes a list of arguments of the controlling message. The controlling message generated by the central device may include a list of arguments, and when the central device only translate a portion without any arguments of the controlling message to the semantic information, the list of arguments may be included in the first information payload to transmit.

In an implementation, the first information payload further includes information of a first encoding method, and the second information payload is decoded by using the first encoding method. In an implementation, the first information payload is decoded by using a second encoding method. The second encoding method and the first encoding method can be the same or different, which is not limited here.

In an implementation, the first encoding method includes a first modulation and coding scheme (MCS) , and the second encoding method includes a second MCS. In a possible implementation, the first encoding method and the second encoding method may be other encoding methods that can be used for wireless transmission, such as pulse amplitude modulation (PAM) , frequency shift keying (FSK) , phase shift keying (PSK) , quadrature amplitude modulation (QAM) , and so on.

The user device may decode the first information payload by using the second encoding method, to obtain the list of arguments of the controlling message, the starting position of the second transmission opportunity, the length of the semantic information, the first encoding method. With them, the user device can obtain the second information payload and decode the second information payload to obtain the semantic information.

S920, determine target semantic information from at least one piece of semantic information according to the semantic information of the controlling message, where each piece of the at least one piece of semantic information corresponds to a respective function modality in at least one function modality and is obtained according to a description of the respective function modality, and the description of the respective function modality is in natural language.

In the embodiment, the user device determines target semantic information by comparing the semantic vector obtained from the network device and all its candidate semantic information. Each piece of candidate semantic information corresponds to a respective function modality in at least one function modality of the user device. The candidate semantic information is obtained according to a description of the respective function modality, and the description of the respective function modality is in natural language. The candidate semantic information may be stored in the user device, or may be obtained every time the user device needs. The description of a respective function modality is used to describe the respective function modality in a natural language.

In an implementation, the user device may determine a piece of semantic information most relevant to the semantic information of the controlling message as the target semantic information from the at least one piece of semantic information. The user device may compute the relevance or similarity between the semantic information obtained from the central device and all its candidate semantics. Based on the relevance computation, the target semantic information is identified as the most relevant to the semantic information.

In an implementation, the first information payload further includes: information indicative of a user device identifier (ID) or information indicative of a group user device ID. Since the information indicative of the user device ID or information indicative of the group user device ID is included in the first information payload, the user device can determine whether it is a correct recipient according to the first information payload, thereby improving the efficiency and accuracy of the communication system.

In an implementation, the user device may obtain a first code generated according to information indicative of a user device ID where the first code includes a mask code, a spreading code, or an interleaving code; or the user device may obtain a second code generated according to information indicative of a group user device ID, where the second code includes a mask code, a spreading code, or an interleaving code. The user device may obtain the information indicative of a user device ID or information indicative of a group user device ID from the obtained code by decoding the obtained code.

In an implementation, in response to determining that the user device ID or the group user device ID does not match a user device ID or a group user device ID of the user device, the user device may stop decoding the second information payload. When the user device identifier is indicated in the first information payload, the user device may stop decoding the second information payload, in a case that the user device finds that the decoded user device ID or the decoded group user device ID doesn’ t match that of itself, thereby saving the energy and processing resources of the user device.

In an implementation, after obtaining the information payload, the user device may obtain the at least one piece of semantic information of the at least one function modality by using a first embedder and according to the description of each of the at least one function modality. The first embedder is the embedder which is used to obtain the semantic information of the controlling message included in the information payload. That is to say, the user device may compare the semantic information of the controlling message with the candidate semantic information obtained by using the same embedder, and accuracy and reliability of the comparing result would be improved. In some cases, the candidate semantic information is not stored in the user device, for example, when the user device has limited storage, or the embedder has just been configured or updated, etc. The user device may obtain the candidate semantic information by using a corresponding embedder.

In an implementation, the information indicative of the embedder ID of the first embedder may be carried in the information payload. In an implementation, when the information payload is transmitted via a first information payload and a second information payload, the information indicative of the embedder ID of the first embedder may be carried in the first information payload. The first embedder is the one which is used to obtain the semantic information included in the information payload. In an implementation, the first information payload further includes: information indicative of an embedder ID of an embedder used to obtain the semantic information of the controlling message. By including the information indicative of an embedder ID of an embedder used to obtain the semantic information of the controlling message in the information payload, for example the first information payload, the accuracy of the decision-making by the user device about the semantic information can be improved.

In an implementation, the semantic information of the controlling message is obtained by using a first embedder of the at least one embedder, and correspondingly, the at least one piece of semantic information of the at least one function modality is obtained by using the first embedder. That is to say, the target semantic information may be determined from the candidate semantic information which is obtained by the same embedder as the embedder used to obtain the semantic information included in the information payload. Since the user device may compare the semantic information with the candidate semantic information obtained by using the same embedder, accuracy and reliability of the comparing result would be improved.

In an implementation, the user device may obtain one or more pieces of semantic information by using at least one embedder and according to the description of each of the at least one function modality, where for each of the at least one function modality, each of the at least one embedder is used to obtain a respective piece of semantic information, and the user device may store the one or more pieces of semantic information. The user device may use each of the at least one embedder to obtain a respective piece of semantic information for each of the at least one function modality, and may store the obtained semantic information, which enables the user device to access and utilize the semantic information when needed quickly.

In an implementation, the user device may obtain the at least one piece of semantic information of the at least one function modality from the stored one or more pieces of semantic information according to the information indicative of the embedder ID of the first embedder. When the semantic information of each function modality has been stored in the user device, the user device can obtain the candidate semantic information from the stored semantic information according to the embedder ID information, so as to determine the target semantic information.

In an implementation, the central device may send the semantic information of the function modality. Correspondingly, the user device may obtain at least one piece of semantic information of at least one function modality. Especially when the user device is unable to run any transform-based embedder, the user device may obtain the semantic information of the function modality from the central device instead of translating the description of the function modality by itself. This method can improve the compatibility and flexibility of the system.

S930, execute a function modality corresponding to the target semantic information.

In the embodiment, after determining the target semantic information, the user device executes the function modality corresponding to the target semantic information.

A user device may obtain an information payload including semantic information of a controlling message, may determine a target semantic information based on the controlling message, and may execute the corresponding function modality. In this way, open vocabulary can be supported in controlling, more efficient communication can be achieved, backward and forward compatibility can be supported, and cross-modality functionality can be accommodated.

FIG. 10 shows a schematic flowchart of a registering process according to one or more embodiments of the present disclosure. As shown in FIG. 10, the method can include the following steps.

S1010, a user device sends registering information of a function modality, where the registering information of the function modality includes a description of the function modality, where the description is in natural language.

In the embodiment, the user device sends registering information of a function modality to the central device. The registering information of the function modality includes a description of the function modality and the description is in natural language. In a possible implementation, the registering information may further include a list of arguments of the function modality.

A user device in real-world comprises a number of states and a number of observation functions and actuation functions, a function modality comprises at least an observation function to read at least one state or at least an actuation function to change at least one state.

In an implementation, the user device may encapsulate each function modality of the at least one function modality into a calling function, where the calling function for the respective function modality includes the description of the respective function modality, or the calling function for the respective function modality includes a list of arguments of the respective function modality and the description of the respective function modality. The calling function is used to interact with a corresponding function by sending requests and processing the data returned by the corresponding function.

In a possible implementation, the calling function includes an application programming interface (API) calling function. The API is a set of rules, protocols, and tools that allows different software applications to communicate and interact with each other. The API defines methods and data formats that applications can use to exchange information and request services from each other.

Since the calling function, such as the API calling function, can be implemented in various programming languages, by encapsulating each function modality of the at least one function modality into a calling function, it becomes easier to integrate and use within different programming languages, and promote interoperability between different systems or components that may be developed using different technologies.

In a possible implementation, the user device may use other technologies to achieve the functionality of interacting with a function modality. For example, the system could directly call the function modality using an appropriate programming language’s syntax, and this approach may be suitable for smaller applications or when the function modality is tightly coupled with the rest of the application’s logic. For another example, if the function modality is provided as a library or software development kit (SDK) , the system can directly integrate and use the library functions within an application code. For still another example, in a distributed system, the function modality may be implemented as a standalone service accessible through a network. The system can interact with the function modality using standard communication protocols such as representational state transfer (REST) , gRPC, or message queues.

In an implementation, the user device may send registering information of each function modality of the at least one function modality, where the registering information of each function modality includes the description of the respective function modality, or, the registering information of each function modality includes a list of arguments of the respective function modality and the description of the respective function modality. The user device may send registering information to the central device, to register the function modality to LM. In a possible implementation, the LM may be an LLM.

Since the registering information of a function modality just includes the description in natural language or further includes a list of arguments of the function modality, the system can accommodate the cross-modality functionality and support backward and forward compatibility. What’s more, when new function modalities are added, they can be registered with their respective descriptions in natural language without requiring changes to the system architecture or communication protocols, which improves the generalizability of the system, and makes the system more flexible.

In the embodiment, correspondingly, the central device may obtain the registering information of the function modality from the user device.

S1020, the central device registers the function modality to an LM.

In the embodiment, after obtaining the registering information from the user device, the central device registers the function modality to an LM. In an implementation, the registering information of the function modality may further include a list of arguments of the function modality.

In an implementation, the central device may register the function modality to at least one LMs by providing the registering information of the function modality to the at least one LMs, where the at least one LM includes the LM. The system may have one or more LMs. The central device may register the function modality to at least one of the one or more LMs by providing the registering information of the function modality to the at least one LMs. That is to say, when the system has multiple LMs, the central device may register the function modality to one LM of the multiple LMs, or may register the function modality to more than one LMs of the multiple LMs, which can be determined according to actual needs.

In an implementation, the central device may record the registering information of the function modality. After obtaining the registering information, the central device may store the registering information locally for further use, thereby facilitating further maintenance.

In an implementation, the central device may further record information indicative of a user device ID of a user device from which the registering information of the function modality is received. The central device may associate information indicative of a user device ID of a user device, from which the registering information of the function modality is received, with the corresponding the function modality, and may record the association relationship. That is to say, the central device may record the information indicative of a user device ID of a user device along with the description and the list of arguments (if any) of the function modality that the user device has.

In a possible implementation, the central device may maintain a data structure or database that maps information indicative of a user device ID (for example, user device identifiers) to registered function modalities. Whenever a user device registers a new function modality or sends information related to a function modality, the central device can update the mapping accordingly.

The central device may record information indicative of a user device ID of a user device together with the registering information for further use, to call the corresponding function modality of different user devices more directly without the need for additional communication or negotiation, thereby helping optimize the utilization of transmission resources.

In an implementation, the central device may further obtain semantic information of the function modality by using an embedder and according to the description of the function modality. The embedder is included in the LM and is register in the system. The central device may translate, embed, or tokenize the description in natural language of the function modality to semantic information by using the embedder. The semantic information is in a more structured and machine-readable format.

In an implementation, the central device may further store the semantic information of the function modality. After the central device obtain semantic information of the function modality, the central device may store the semantic information of the function modality for further use, thereby facilitating further maintenance.

In a possible implementation, the central device may translate the description of the function modality into the semantic information each time that it needs instead of storing the semantic information.

In a possible implementation, if other user device (for example, a second user device) has the same function modality (for example, the first function modality) with the user device (for example, the first user device) and the second user device also sends the registering information of the first function modality to the central device, the central device may add information indicative of a user device ID of the second user device to an existing association for the first function modality.

In an implementation, in a case that there are multiple LMs, the user device may register the function modality to one or some of the multiple LMs, or may register the function modality to all of the multiple LMs. After the function modality has registered to one of the multiple LMs, the central device may inform other LM of the function modality by providing the registering information of the function modality to other LM according to actual needs. By allowing for multiple LMs to be deployed on the central device, the system becomes more flexible and customizable.

In an implementation, the central device may send configuration information of at least one embedder, where the embedder belongs to the at least one embedder. The LM may register its embedder into the system, which may include the central device and the user device. The central device may send configuration information of the embedder to the user device, to register the embedder to the user device. When there is an embedder introduced or updated, the central device may send the configuration information, including, such as architecture and parameters (neuron values) of the embedder to the user device. In a possible implementation, the configuration information of the at least one embedder is sent via a broadcast message, a multicast message, or a unicast message.

Correspondingly, the user device may obtain configuration information of at least one embedder. The user device may register or update the at least one embedder after obtaining the configuration information.

In an implementation, multiple embedders (e.g. including a first embedder and a second embedder) may be registered in the system. The multiple embedders may be included in the same LM or in different LMs.

For example, when there is a second embedder and the second embedder is introduced or updated, the central device may transmit the configuration of the second embedder to all the user devices. In some cases, the first LM model may have the first embedder and the second embedder; in other cases, the first LM model has the first embedder and a second LM model has the second embedder.

In an implementation, the configuration information of the at least one embedder further includes an embedder ID of each of the at least one embedder. The central device may inform the user device of an embedder ID of each of the at least one embedder, especially when multiple embedders exist. By including the embedder ID of each of the at least one embedder in the configuration information, it becomes simpler to track and manage the specific embedders that need updates or maintenance.

In an implementation, the central device and the user device may use the same embedder to obtain the semantic information of the function modality according to the description of the function modality included in the registering information. The semantic information of the function modality obtained by the central device and the user device may not be exactly the same, due to the inherent randomness or variability in the embedder. However, despite this variability, there is still a strong semantic relevance between the semantic information obtained by the central device and the semantic information obtained by the user device. The randomness in the embedding process can introduce slight variations in the output semantic information, even when the input description remains the same. These variations can be attributed to factors such as the specific initialization of the embedder’s parameters or the stochastic nature of certain neural network operations.

When the system has at least two embedders, there may be several scenarios. In a possible implementation, there are multiple LMs, and the multiple LMs include the LM, where each of the multiple LMs has at least one embedder. The system may have multiple LMs, and each of the multiple LMs has at least one embedder. In another possible implementation, at least one LM in the system has multiple embedders.

In an implementation, at least one function modality is registered in one or more LMs of the multiple LMs, but not registered in all the multiple LMs. When there are multiple LMs in the system, the function modality may be registered in one or more of the multiple LMs, but may not registered in all of the multiple LMs. Therefore, the size of the overall system can be reduced and optimize resources of the central device.

In an implementation, the multiple LMs may include at least one LM compatible for an LTE system. In an implementation, the multiple LMs include at least one LM compatible for a 5G system. The multiple LMs may include different LMs compatible for different communication systems. That is to say, different LMs may have different registered function modalities.

In a possible implementation, a first LM of the multiple LMs may have a first embedder and a second LM of the multiple LM may have a second embedder, and the first embedder and a second embedder may be built on vocabularies of different domains. In another implementation, the LM may have a first embedder and a second embedder built on vocabularies of different domains. That is to say, the system may have multiple embedders, and at least two of them may be built on vocabularies of different domains. The specialization on the vocabulary can reduce the size and cost of the embedder, but also improve the accuracy of the relevance computation.

For example, the system has a first embedder and a second embedder. There are several advantageous scenarios:

Scenario 1: system can have two LM models: the first LM model and the second LM model. For example, the first LM model is developed to be compatible for LTE system. The open-vocabulary controlling messages generated by the first LM model can generate the function modalities supported by LTE standard. The second LM model is developed to be compatible for 5G system. The open-vocabulary controlling messages generated by the second LM model can generate the function modalities supported by 5G standard. Some function modalities such as Massive-MIMO and Polar Code are NOT comprised by the first LM model. Accordingly, the first LM model has the first embedder and the second LM model has the second embedder.

Scenario 2: the system has one LM model but has two embedders. For example, the first embedder is built on the vocabulary of IoT industrial domain and the second embedder is built on city security domain. Specialization on the vocabulary can greatly reduce the size and cost of the embedder but also improve the accuracy of the relevance computation.

In an implementation, in response to generating at least two consecutive controlling messages more than a preset number of times by the LM, the central device may combine function modalities corresponding to the at least two consecutive controlling messages to generate a combined function modality, and may send configuration information for the combined function modality. In an implementation, after generating the combined function modality, the central device may register the combined function modality to the LM. In practical application, the central device may generate at least two consecutive controlling messages all the time by the LM. When the central device generates at least two consecutive controlling messages more than a preset number of times by the LM, the central device may negotiate with the user device to generate a combined function modality. That is to say, at least two function modalities corresponding to the at least two consecutive controlling messages can be combined into one function modality, thereby enhancing effectiveness by streamlining the execution of multiple functions within a single calling of a combined function modality

In an implementation, when more than one function modality exists, the user device may obtain configuration information for a combined function modality, where the combined function modality is combined from at least two function modalities of the more than one function modality, and the user device may encapsulate the combined function modality into a first calling function. In an implementation, the configuration information for the combined function modality may include a calling order, and the first calling function calls calling functions corresponding to the at least two function modalities in the calling order. The combined function calls the multiple calling function successively. Along with the running time, combined function modality that encapsulates a plurality of old function modalities in some order would be created so that the effectiveness gets improved.

An example scenario of the controlling method of the present disclosure will be described in the following description for ease of understanding. It should be noted that the example scenario is for illustration purpose only, which should not be construed as limitations to the solution of the present disclosure.

In FIG. 7, a typical LLM model is illustrated. A sentence in a natural language is input into a LLM model. This sentence is firstly embedded into a semantic embedding by a deep neural network (for example, a transform-based deep neural network-1 in the figure) . Then, the semantic embedding is input into a generator or predictor that will generate a chain of semantic embeddings following this input semantic embedding. The chain of semantic embeddings is translated to a chain of sentences in a natural language by a decoder. Optionally the generator and decoder may be combined together by a deep neural network (for example, another transform-based deep neural network-2 in the figure) .

As illustrated in scenario 1 of the following FIG. 11, a first LLM model may register its first embedder into the system, in which a central device and at least a first user device will register the first embedder. When the first embedder is introduced or updated, the central device configures the first embedder, including the architecture and the parameters (neuron values) , to all the user devices in a broadcast, multicast, or even unicast way.

When there is a second embedder and the second embedder is introduced or updated, the central device may transmit the second embedder to all the user devices. In some cases, the first LLM model may have the first embedder and the second embedder; in other cases, the first LLM model has the first embedder and a second LLM model has the second embedder. For example, the first embedder is English and the second embedder is Chinese, or, the first embedder is a general one and the second embedder is limited to IoT (much smaller and faster) .

As LLMs are open for individual implementation, a central device may have more than one LLM model as illustrated in scenario 2 of FIG. 11. For example, the first LLM model is built by a company-A for green-energy equipment management; the second LLM model is built by a company-B for smart-city traffic sensors. The details for each LLM model could refer to above description.

The central device may inform all the user devices of the embedder identifier, if there are more than one embedder, when configuring an embedder, see the following FIG. 12.

Registering Function Modality into System

A function modality is an action conducted by a user device. An action includes observation and/or actuation. When a user device executes an observation action, the user device would measure some of its states or status. When a user device executes an actuation action, the user device would change some of its states or status. A function modality can be either observation action, or actuation observation, or both simultaneously. In FIG. 13, an example of two states, state #1 and state #2 is given. To access the states (state #1 and state #2) , a function is needed. Suppose that a function can access to only one state. This function can be either to change the state (actuation function, for example, an actuation function can be: “SetState#1 (value) ” ) or to read the state (observation function, for example, an observation function can be: “Result = ReadState#1 (void) ” , “IsState#1HigherThan (value) ” , or “IsState#1InRangeBetween (value1, value2) ” ) , or read and change the state (actuation &observation function) simultaneously.

At least a first function modality is enabled, equipped, or implemented on at least a first user device. In order for the central device to make use of the first function modality enabled on the first user device, the system registers the first function modality into the system if the first function modality is new or updated to the system.

On the first user device, the first function modality may be encapsulated into at least a first API (application programming interface) calling function, see FIG. 14. The first API calling function may be in various programming languages such as C, Java, Python, C++, assemble etc. The first API calling function may have a first list of arguments. The first API calling function may have a first description that describes the first function modality in a natural language. Please note that a function modality may be achieved by a sequence of API functions, though one function modality by one API function is preferred. A function modality shown in FIG. 14 comprises API, list of arguments, and description in a natural language.

Optionally, as illustrated in FIG. 15, the first user device may use the first embedder to translate the first description into a first semantic vector (or a batch of first semantic vectors with different lengths) and then store it.Or the first user device may use the second embedder to translate the first description into a second semantic vector (or a batch of second semantic vectors with different lengths) and then store it. Or the first user device may use the first embedder to translate the first description into a first semantic vector each time that it needs instead of storing it. Due to some randomness in the first embedder, the first semantic vectors generated by the first embedder may not be exactly the same but very close on the semantic domain.

As illustrated in FIG. 16, the first user device may inform the central device of the first function modality by transmitting the first description and the first list of arguments to the central device. The first user device transmits the first description of the first function modality to the central device in order that the central device registers the first function modality into the system.

After receiving the first description and the first list of arguments, the central device may register the first function modality into the first LLM model by providing the first description and the first list of arguments. Optionally, the central device may register the first function modality into the second LLM model by providing the first description and the first list of arguments, if there is the second LLM model.

Optionally, as shown in FIG. 17, the central device may use the first embedder to translate the first description into a third semantic vector (or a third batch of semantic vectors with different lengths) , and then store it.Preferably, the central device may associate the first user device identifier to the first function modality. Or the central device may use the second embedder to translate the first description into a fourth semantic vector (or a fourth batch of semantic vectors with different lengths) and then store it, if there is the second embedder. Or the central device may use the first embedder to translate the first description into a third semantic vector each time that it needs instead of storing it. Due to some randomness in the first embedder, the third semantic vectors generated by the first embedder may not be exactly the same but very close on the semantic domain.

As illustrated in FIG. 18, if a second user device has the same first function modality, the second user device may provide the first description and the first list of arguments to the central device. The central device may add the second user device identifier with the first function modality. That is to say, if the first function modality has been registered, the central device may record the second user device that supports the first function modality.

If a second user device has a second function modality (new or updated) , the second user device may provide a second description and a second list of arguments to the central device. The central device may record the second user device identifier with the second function modality. In turn, the central device may inform the first LLM model provider of the second function modality by providing the second description and a second list of arguments to the first LLM model provider. Optionally, the central device may inform the second LLM model provider of the second function modality by providing the second description and a second list of arguments to the second LLM model provider, if there are more than one LLM model provider.

Although the first user device and central device use the same first embedder to translate the same first description, the first semantic vector and the third semantic vector may not be exactly the same due to some randomness of the first embedder. However, the first semantic vector and the third semantic vector have a strong semantic relevance.

If the central device generates at least two consecutive open-vocabulary controlling messages for a user device all the time, it means that the two function modalities can be combined into one function modality. The central device and the user device may negotiate to generate a third function modality that has a third function API. The third API function calls the two function API successively. Along with the running time, new function modality that encapsulates a plurality of old function modalities in some order would be created so that the effectiveness gets improved.

As illustrated in FIG. 19, the central device uses the first LLM model to generate at least a first open-vocabulary controlling message. The first controlling message in natural language may include a list of arguments (values) and is targeted at the first function modality on the first user device.

The central device divides the first open-vocabulary (Open-Voc) controlling message into two portions: the first portion comprising the list of arguments if any and the second portion comprising the rest part of the first open-vocabulary controlling message.

The central device may choose a first length and use the first embedder to translate the second portion of the first open-vocabulary controlling message into a fifth semantic vector.

The central device may use the first embedder to translate description of the first function modality into a third semantic vector.

Although the fifth semantic vector and the third semantic vector may not be exactly the same due to some randomness of the first embedding model, the fifth semantic vector and the third semantic vector indicate a strong semantical relevance.

Optionally, as shown in FIG. 20, the central device may compare the relevance between the fifth semantic vector and the third semantic vector to a first threshold. If the relevance between the fifth semantic vector and the semantic vector of each registered function modality is lower than the first threshold, the central device may judge that the first controlling message is non-registered one; otherwise, the central device may judge that the first controlling message is registered one. The central device may dismiss a non-registered one.

One of the advantages is that the central device may multicast or unicast to the user devices whom are registered with the relevant function modality, thereby saving the user device’s power.

As shown in FIG. 21, the central device may encode and transmit the list of arguments of the first open-vocabulary controlling message during a first transmission opportunity, and encode and transmit the fifth semantic vector during a second transmission opportunity.

That is to say, the central device may encode and transmit an open-voc controlling message in two transmission opportunity.

The central device allocates a trunk of radio resource for controlling messages. The first part of the radio resource is reversed for standardized controlling messages. The second part of the radio resource is allocated for open-vocabulary controlling messages that at least includes the first open-vocabulary controlling messages.

The central device may form a first information payload that comprises the list of arguments of the first open-vocabulary controlling message, a starting position of the second transmission opportunity, the first length of the fifth semantic vector, a first MCS that encodes the fifth semantic vector, optionally the first user device identifier. The central device uses a second MCS to encode the first information payload. The central device grants the first transmission opportunity in the second part of the radio resource to transmit the first information payload.

The central device may form the second information payload that comprises the fifth semantic vector. The central device uses the first MCS to encode the second information payload. The central device grants the second transmission opportunity with the starting position and the first length in the second part of the radio resource to transmit the second information payload.

As illustrated in FIG. 22 the first user device receives the first transmission opportunity (i.e., receives some information on the first transmission opportunity) . The first user device uses the second MCS to decode the first information payload. Optionally if the user device identifier is transmitted in the first information payload, the first user device may stop decoding the second information payload, if the first user device finds that the decoded user device identifier doesn’ t match.

As illustrated in FIG. 23, the first user device decodes the list of the arguments of the first open-vocabulary controlling message, the starting position of the second transmission opportunity, the first length of the fifth semantic vector, the first MCS. With them, the first user device receives the second transmission opportunity and decodes the fifth semantic vector by the first MCS.

The first user device computes the relevance between the fifth semantic and all its candidate semantic that includes at least the first semantic vector. The first user device may select the first semantic as the most relevant to the fifth semantic. Then, the first user device may call the first API calling function with the arguments that the first user device decodes in the first transmission opportunity. Finally, the first user device executes the first function modality with the right arguments.

From the top level, the central device transmits the first controlling message in a natural language to the first user device who executes the first function modality.

In addition to above examples, the following are some further description for this disclosure.

For some low-power IoT user device that is unable to run any transform-based embedder, the first user device can ask the central device to transmit the third semantic vector and then store it into the memory of the first user device.

Please note that, for V2X scenario, the DCI may be replayed by SCI, DCI field bundle may be replaced by SCI field bundle.

Please not that in the disclosure, “controlling message” , “control message” and “control signaling” could be used exchangeable with same meaning.

Please note that in this disclosure, “message” could represent “information” , or carry “information” . In some aspects of the present disclosure, there is provided a computer program comprising instructions. The instructions, when executed by a processor, may cause the processor to implement the method of the present disclosure.

Next, embodiments of products related to the wireless communication methods will be described.

FIG. 24 shows a schematic structural diagram of a first apparatus according to one or more example embodiments of the present disclosure. The first apparatus may be applied to the central device described above, or installed in or applied to a chip in the central device or any other equipment, module, circuit or unit that can implement the steps of the central device in above method embodiments. As shown in FIG. 24, a first apparatus 2400 may include:

a processing module 2401, configured to: generate a controlling message in natural language by using an LM, obtain semantic information of the controlling message according to the controlling message, and form an information payload, where the information payload includes the semantic information of the controlling message; and

a sending module 2402, configured to send the information payload.

In a possible implementation, the LM may be an LLM.

In a possible implementation, the first apparatus further includes an obtaining module 2403, configured to obtain registering information of a function modality, where the registering information of the function modality includes a description of the function modality, where the description is in the natural language.

In a possible implementation, the registering information of the function modality further includes a list of arguments of the function modality.

In a possible implementation, the processing module 2401 is further configured to register the function modality to the LM.

In a possible implementation, the processing module 2401 is further configured to record the registering information of the function modality.

In a possible implementation, the processing module 2401 is further configured to record information indicative of a user device ID of a user device from which the registering information of the function modality is received.

In a possible implementation, the processing module 2401 is further configured to obtain semantic information of the function modality by using an embedder according to the description of the function modality.

In a possible implementation, the processing module 2401 is further configured to store the semantic information of the function modality.

In a possible implementation, the sending module 2402 is further configured to send the semantic information of the function modality.

In a possible implementation, the processing module 2401 is further configured to determine a relevance between the semantic information of the controlling message and semantic information of each of at least one function modality that has been registered, where the semantic information of each of the at least one function modality is obtained according to a description of each of the at least one function modality.

In a possible implementation, the processing module 2401 is further configured to determine that the relevance between the semantic information of the controlling message and the semantic information of each of the at least one function modality registered is lower than a threshold, and reject the controlling message.

In a possible implementation, the processing module 2401 is further configured to determine a portion without any arguments from the controlling message; and obtain the semantic information of the controlling message according to the portion by using an embedder.

In a possible implementation, the processing module 2401 is further configured to form a first information payload, where the first information payload includes starting position information of a second transmission opportunity and length information of the semantic information; and form a second information payload, where the second information payload includes the semantic information of the controlling message, and where the second transmission opportunity is for transmission of the second information payload information.

In a possible implementation, the controlling message further includes a list of arguments, and the first information payload further includes the list of arguments of the controlling message.

In a possible implementation, the processing module 2401 is further configured to encode the second information payload by using a first encoding method; where the first information payload further includes information of the first encoding method

In a possible implementation, the processing module 2401 is further configured to encode the first information payload by using a second encoding method.

In a possible implementation, the first encoding method includes a first modulation and coding scheme (MCS) , and the second encoding method includes a second MCS.

In a possible implementation, the processing module 2401 is further configured to determine semantic information most relevant to the semantic information of the controlling message from the semantic information of each of the at least one function modality that has been registered according to the relevance between the semantic information of the controlling message and the semantic information of each of at least one function modality that has been registered; and determine at least one user device ID or at least one group user device ID corresponding to the most relevant semantic information; where the first information payload further includes: the at least one user device ID, or at least one group user device ID; or the first information payload further includes information indicative of the at least one user device ID or information indicative of the at least one group user device ID.

In a possible implementation, the information indicative of the at least one user device ID includes a first code generated according to the at least one user device ID, where the first code includes a mask code, a spreading code, or an interleaving code; or the information indicative of the at least one group user device ID includes a second code generated according to the at least one group user device ID, where the second code includes a mask code, a spreading code, or an interleaving code.

In a possible implementation, the processing module 2401 is further configured to generate the first code according to the at least one user device ID; or generate the second code according to the at least one group user device ID.

In a possible implementation, the sending module 2402 is further configured to send the first information payload during a first transmission opportunity; and send the second information payload during the second transmission opportunity.

In a possible implementation, the first transmission opportunity is in a controlling physical channel, and the second transmission opportunity is in a data physical channel; or both the first transmission opportunity and the second transmission opportunity are in a controlling physical channel; or both the first transmission opportunity and the second transmission opportunity are in a data physical channel.

In a possible implementation, the processing module 2401 is further configured to register the function modality to at least one LMs by providing the registering information of the function modality to the at least one LMs, where the at least one LM includes the LM.

In a possible implementation, the sending module 2402 is further configured to send configuration information of at least one embedder, where the embedder belongs to the at least one embedder.

In a possible implementation, the configuration information of the at least one embedder is sent via a broadcast message, a multicast message, or a unicast message.

In a possible implementation, the configuration information of the at least one embedder includes an architecture and parameters of each of the at least one embedder.

In a possible implementation, the configuration information of the at least one embedder further includes an embedder ID of each of the at least one embedder.

In a possible implementation, the first information payload further includes: information indicative of an embedder ID of an embedder used to obtain the semantic information of the controlling message.

In a possible implementation, the information payload is sent via a multicast message or a unicast message.

In a possible implementation, the processing module 2401 is further configured to in response to generating at least two consecutive controlling messages more than a preset number of times by the LM, combine two function modalities corresponding to the two consecutive controlling messages to generate a combined function modality; where the sending module 2402 is further configured to send configuration information for the combined function modality.

In a possible implementation, the processing module 2401 is further configured to register the combined function modality to the LM.

In a possible implementation, there are multiple LMs, and the multiple LMs include the LM, where each of the multiple LMs has at least one embedder.

In a possible implementation, at least one function modality is registered in one or more LMs of the multiple LMs, but not registered in all the multiple LMs.

In a possible implementation, the multiple LMs include at least one LM compatible for an LTE system.

In a possible implementation, the multiple LMs include at least one LM compatible for a 5G system.

In a possible implementation, a first LM of the multiple LMs has a first embedder and a second LM of the multiple LM has a second embedder, and the first embedder and a second embedder built on vocabularies of different domains

In a possible implementation, the LM has a first embedder and a second embedder built on vocabularies of different domains.

In a possible implementation, the processing module 2401 is further configured to register the LM.

In a possible implementation, the processing module 2401 is further configured to register at least one embedder of the LM.

FIG. 25 shows a schematic structural diagram of a second apparatus according to one or more example embodiments of the present disclosure. The second apparatus may be applied to the user device described above, or installed in or applied to a chip in the central device or any other equipment, module, circuit or unit that can implement the steps of the user device in above method embodiments. As shown in FIG. 25, a second apparatus 2500 may include:

an obtaining module 2501, configured to obtain an information payload, where the information payload includes semantic information of a controlling message;

a processing module 2502, configured to determine target semantic information from at least one piece of semantic information according to the semantic information of the controlling message, where each piece of the at least one piece of semantic information corresponds to a respective function modality in at least one function modality and is obtained according to a description of the respective function modality, and the description of the respective function modality is in natural language; and execute a function modality corresponding to the target semantic information.

In a possible implementation, the processing module 2502 is further configured to encapsulate each function modality of the at least one function modality into a calling function, where the calling function for the respective function modality includes the description of the respective function modality, or the calling function for the respective function modality includes a list of arguments of the respective function modality and the description of the respective function modality.

In a possible implementation, the calling function includes an application programming interface (API) calling function.

In a possible implementation, the second apparatus further includes a sending module 2503, and the sending module 2503 is configured to send registering information of each function modality of the at least one function modality, where the registering information of each function modality includes the description of the respective function modality, or, the registering information of each function modality includes a list of arguments of the respective function modality and the description of the respective function modality.

In a possible implementation, the obtaining module 2501 is configured to obtain a first information payload during a first transmission opportunity, where the first information payload includes starting position information of a second transmission opportunity and length information of the semantic information of the controlling message, and obtain a second information payload during the second transmission opportunity according to the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message, where the second information payload includes the semantic information of the controlling message. The processing module 2502 is further configured to decode the first information payload to obtain the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message; and decode the second information payload to obtain the semantic information of the controlling message.

In a possible implementation, the first information payload further includes: a list of arguments of the controlling message.

In a possible implementation, the first information payload further includes information of a first encoding method, and the second information payload is decoded by using the first encoding method.

In a possible implementation, the first information payload is decoded by using a second encoding method.

In a possible implementation, the first encoding method includes a first modulation and coding scheme (MCS) , and the second encoding method include a second MCS.

In a possible implementation, the first information payload further includes: information indicative of a user device identifier (ID) or information indicative of a group user device ID.

In a possible implementation, the obtaining module 2501 is configured to obtain a first code generated according to information indicative of a user device ID, where the first code includes a mask code, a spreading code, or an interleaving code; or obtain a second code generated according to information indicative of a group user device ID, where the second code includes a mask code, a spreading code, or an interleaving code.

In a possible implementation, the processing module 2502 is configured to, in response to determining that the user device ID or the group user device ID does not match a user device ID or a group user device ID of a user device, stop decoding the second information payload.

In a possible implementation, the obtaining module 2501 is configured to obtain configuration information of at least one embedder.

In a possible implementation, the semantic information of the controlling message is obtained by using a first embedder of the at least one embedder, and the at least one piece of semantic information of the at least one function modality is obtained by using the first embedder.

In a possible implementation, the configuration information of the at least one embedder includes an embedder ID of each of the at least one embedder.

In a possible implementation, the information payload includes information indicative of the embedder ID of the first embedder.

In a possible implementation, the information indicative of the embedder ID of the first embedder is carried in the first information payload.

In a possible implementation, the processing module 2502 is configured to obtain one or more pieces of semantic information by using the at least one embedder and according to the description of each of the at least one function modality, where for each of the at least one function modality, each of the at least one embedder is used to obtain a respective piece of semantic information; and store the one or more pieces of semantic information.

In a possible implementation, the processing module 2502 is configured to obtain the at least one piece of semantic information of the at least one function modality from the stored one or more pieces of semantic information and according to the information indicative of the embedder ID of the first embedder.

In a possible implementation, the processing module 2502 is configured to obtain the at one piece of semantic information of the at least one function modality by using the first embedder according to the description of each of the at least one function modality.

In a possible implementation, the obtaining module 2501 is configured to receive the at least one piece of semantic information of the at least one function modality.

In a possible implementation, the processing module 2502 is configured to determine a piece of semantic information most relevant to the semantic information of the controlling message as the target semantic information from the at least one piece of semantic information.

In a possible implementation, more than one function modality exists, and the obtaining module 2501 is further configured to obtain configuration information for a combined function modality, where the combined function modality is combined from at least two function modalities of the more than one function modality; and the processing module 2502 is further configured to encapsulate the combined function modality into a first calling function.

In a possible implementation, the configuration information for the combined function modality includes a calling order, and the first calling function calls calling functions corresponding to the at least two function modalities in the calling order.

It should be understood by a person skilled in the art that, the relevant description of the above modules in the embodiments of the present disclosure may be understood with reference to the relevant description of the data processing method in the embodiments of the present disclosure.

An embodiment of the present disclosure provides a third apparatus including processing circuitry for executing any of the above communication methods performed by the central device. It should be understood that the third apparatus can execute the steps performed by the central device in the above method embodiments, which will not be repeated here.

An embodiment of the present disclosure provides a fourth apparatus including processing circuitry for executing any of the above communication methods performed by the user device. It should be understood that the fourth apparatus can execute the steps performed by the user device in the above method embodiments, which will not be repeated here.

An embodiment of the present disclosure provides a communication system which includes at least one first apparatus and/or at least one third apparatus described above, at least one second apparatus and/or at least one fourth apparatus described above.

An embodiment of the present disclosure provides a communication system, including at least one first processing circuitry and at least one second processing circuitry. The first processing circuitry is configured to execute the steps executed by the central device in any of the above communication methods, and the second processing circuitry is configured to execute the steps executed by the user device in any of the above communication methods.

An embodiment of the present disclosure provides a computer-readable medium storing computer execution instructions which, when executed by a processor, causes the processor to execute any of the above communication methods.

An embodiment of the present disclosure provides a computer program product including computer execution instructions which, when executed by a processor, causes the processor to execute any of the above communication methods.

Although embodiments of the present disclosure describes methods and processes with steps in a certain order, one or more steps of the methods and processes may be omitted or altered as appropriate. One or more steps may take place in an order other than that in which they are described, as appropriate.

In some aspects of the present disclosure, there is provided a non-transitory computer-readable medium storing instructions, the instructions, when executed by a processor, may cause the processor to implement the method of the present disclosure.

In some aspects of the present disclosure, there is provided an apparatus/chipset system comprising means to implement the method implemented by the sensing device of the present disclosure.

In some aspects of the present disclosure, there is provided an apparatus/chipset system comprising means to implement the method implemented by the central device of the present disclosure.

In some aspects of the present disclosure, there is provided an apparatus/chipset system comprising means to implement the method implemented by the GPT device of the present disclosure.

In some aspects of the present disclosure, there is provided a system comprising at least two of an apparatus in the sensing device of the present disclosure, an apparatus in the central device of the present disclosure and an apparatus in the GPT device of the present disclosure.

In some aspects of the present disclosure, there is provided an apparatus/chipset system comprising at least one processor executing instructions stored in a computer-readable medium to implement the method implemented by the sensing device of the present disclosure.

In some aspects of the present disclosure, there is provided an apparatus/chipset system comprising at least one processor executing instructions stored in a computer-readable medium to implement the method implemented by the central device of the present disclosure.

In some aspects of the present disclosure, there is provided an apparatus/chipset system comprising at least one processor executing instructions stored in a computer-readable medium to implement the method implemented by the GPT device of the present disclosure.

The present disclosure encompasses various embodiments, including not only method embodiments, but also other embodiments such as apparatus embodiments and embodiments related to non-transitory computer readable storage media. Embodiments may incorporate, individually or in combinations, the features disclosed herein.

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.

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 controlling method, comprising:

generating, by using a language model (LM) , a controlling message in natural language;

obtaining, according to the controlling message, semantic information of the controlling message;

forming an information payload, wherein the information payload comprises the semantic information of the controlling message; and

sending the information payload.
The method according to claim 1, further comprising:

obtaining registering information of a function modality, wherein the registering information of the function modality comprises a description of the function modality, wherein the description is in the natural language.
The method according to claim 2, wherein the registering information of the function modality further comprises a list of arguments of the function modality.
The method according to claim 2 or 3, further comprising:

registering the function modality to the LM.
The method according to claim 4, further comprising:

recording the registering information of the function modality.
The method according to claim 5, further comprising:

recording information indicative of a user device identifier (ID) of a user device from which the registering information of the function modality is received.
The method according to any one of claims 2 to 6, further comprising:

obtaining, by using an embedder and according to the description of the function modality, semantic information of the function modality.
The method according to claim 7, further comprising:

storing the semantic information of the function modality.
The method according to claim 7 or 8, further comprising:

sending the semantic information of the function modality.
The method according to any one of claims 7 to 9, before the forming the information payload, the method further comprises:

determining a relevance between the semantic information of the controlling message and semantic information of each of at least one function modality that has been registered, wherein the semantic information of each of the at least one function modality is obtained according to a description of each of the at least one function modality.
The method according to claim 10, further comprising:

determining that the relevance between the semantic information of the controlling message and the semantic information of each of the at least one function modality registered is lower than a threshold, and

rejecting the controlling message.
The method according to any one of claims 1 to 11, wherein the obtaining, according to the controlling message, the semantic information of the controlling message comprises:

determining, from the controlling message, a portion without any arguments;

obtaining, by using an embedder, and according to the portion, the semantic information of the controlling message.
The method according to claim 12, wherein the forming the information payload comprises:

forming a first information payload, wherein the first information payload comprises starting position information of a second transmission opportunity and length information of the semantic information; and

forming a second information payload, wherein the second information payload comprises the semantic information of the controlling message, and wherein the second transmission opportunity is for transmission of the second information payload information.
The method according to claim 13, wherein the controlling message further comprises a list of arguments, and the first information payload further comprises the list of arguments of the controlling message.
The method according to claim 13, further comprising:

encoding the second information payload by using a first encoding method;

wherein the first information payload further comprises information of the first encoding method.
The method according to claim 15, further comprising:

encoding the first information payload by using a second encoding method.
The method according to claim 16, wherein the first encoding method comprises a first modulation and coding scheme (MCS) , and the second encoding method comprises a second MCS.
The method according to claim 13, further comprising:

determining, from the semantic information of each of the at least one function modality that has been registered, and according to the relevance between the semantic information of the controlling message and the semantic information of each of at least one function modality that has been registered, semantic information most relevant to the semantic information of the controlling message; and

determining at least one user device ID or at least one group user device ID corresponding to the most relevant semantic information;

wherein the first information payload further comprises: the at least one user device ID, or the at least one group user device ID; or the first information payload further comprises information indicative of the at least one user device ID or information indicative of the at least one group user device ID.
The method according to claim 18, wherein the information indicative of the at least one user device ID comprises a first code generated according to the at least one user device ID, wherein the first code comprises a mask code, a spreading code, or an interleaving code; or

the information indicative of the at least one group user device ID comprises a second code generated according to the at least one group user device ID, wherein the second code comprises a mask code, a spreading code, or an interleaving code.
The method according to claim 19, further comprising:

generating the first code according to the at least one user device ID; or

generating the second code according to the at least one group user device ID.
The method according to claim 13, wherein the sending the information payload comprises:

sending the first information payload during a first transmission opportunity; and

sending the second information payload during the second transmission opportunity.
The method according to claim 21, wherein the first transmission opportunity is in a controlling physical channel, and the second transmission opportunity is in a data physical channel; or

both the first transmission opportunity and the second transmission opportunity are in a controlling physical channel; or

both the first transmission opportunity and the second transmission opportunity are in a data physical channel.
The method according to claim 4, wherein the registering the function modality to the LM comprises:

registering the function modality to at least one LMs by providing the registering information of the function modality to the at least one LMs, wherein the at least one LM comprises the LM.
The method according to claim 7, further comprising:

sending configuration information of at least one embedder, wherein the embedder belongs to the at least one embedder.
The method according to claim 24, wherein the configuration information of the at least one embedder is sent via a broadcast message, a multicast message, or a unicast message.
The method according to claim 24 or 25, wherein the configuration information of the at least one embedder comprises an architecture and parameters of each of the at least one embedder.
The method according to claim 26, wherein the configuration information of the at least one embedder further comprises an embedder ID of each of the at least one embedder.
The method according to claim 27, wherein the first information payload further comprises: information indicative of an embedder ID of an embedder used to obtain the semantic information of the controlling message.
The method according to any one of claims 1 to 28, wherein the information payload is sent via a multicast message or a unicast message.
The method according to any one of claims 1 to 29, further comprising:

in response to generating, by the LM, at least two consecutive controlling messages more than a preset number of times, combining two function modalities corresponding to the two consecutive controlling messages to generate a combined function modality;

sending configuration information for the combined function modality.
The method according to claim 30, further comprising:

registering the combined function modality to the LM.
The method according to any one of claims 1 to 31, wherein there are multiple LMs, and the multiple LMs comprise the LM, wherein each of the multiple LMs has at least one embedder.
The method according to 32, wherein at least one function modality is registered in one or more LMs of the multiple LMs, but not registered in all the multiple LMs.
The method according to claim 33, wherein the multiple LMs comprise at least one LM compatible for an LTE system.
The method according to claim 32, wherein the multiple LMs comprise at least one LM compatible for a 5G system.
The method according to claim to 32, wherein a first LM of the multiple LMs has a first embedder and a second LM of the multiple LM has a second embedder, and the first embedder and a second embedder built on vocabularies of different domains.
The method according to claim 1, wherein the LM has a first embedder and a second embedder built on vocabularies of different domains.
The method according to claim 1, further comprising:

registering the LM.
The method according to claim 38, further comprising:

registering at least one embedder of the LM.
A controlling method, comprising:

obtaining an information payload, wherein the information payload comprises semantic information of a controlling message;

determining, according to the semantic information of the controlling message, target semantic information from at least one piece of semantic information, wherein each piece of the at least one piece of semantic information corresponds to a respective function modality in at least one function modality and is obtained according to a description of the respective function modality, and the description of the respective function modality is in natural language; and

executing a function modality corresponding to the target semantic information.
The method according to claim 40, further comprising:

encapsulating each function modality of the at least one function modality into a calling function, wherein the calling function for the respective function modality comprises the description of the respective function modality, or the calling function for the respective function modality comprises a list of arguments of the respective function modality and the description of the respective function modality.
The method according to claim 41, wherein the calling function comprises an application programming interface (API) calling function.
The method according to claim 40, further comprising:

sending registering information of each function modality of the at least one function modality, wherein the registering information of each function modality comprises the description of the respective function modality, or, the registering information of each function modality comprises a list of arguments of the respective function modality and the description of the respective function modality.
The method according to claim 40, wherein the obtaining the information payload comprises:

obtaining a first information payload during a first transmission opportunity, wherein the first information payload comprises starting position information of a second transmission opportunity and length information of the semantic information of the controlling message;

decoding the first information payload to obtain the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message;

obtaining a second information payload during the second transmission opportunity according to the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message, wherein the second information payload comprises the semantic information of the controlling message; and

decoding the second information payload to obtain the semantic information of the controlling message.
The method according to claim 44, wherein the first information payload further comprises: a list of arguments of the controlling message.
The method according to claim 44 or 45, wherein the first information payload further comprises information of a first encoding method, and the second information payload is decoded by using the first encoding method.
The method according to claim 46, wherein the first information payload is decoded by using a second encoding method.
The method according to claim 47, wherein the first encoding method comprises a first modulation and coding scheme (MCS) , and the second encoding method comprise a second MCS.
The method according to any of claims 44 to 48, wherein the first information payload further comprises: information indicative of a user device identifier (ID) or information indicative of a group user device ID.
The method according to any one of claims 44 to 47, further comprising:

obtaining a first code generated according to information indicative of a user device ID, wherein the first code comprises a mask code, a spreading code, or an interleaving code; or

obtaining a second code generated according to information indicative of a group user device ID wherein the second code comprises a mask code, a spreading code, or an interleaving code.
The method according to claim 49 or 50, further comprising:

in response to determining that the user device ID or the group user device ID does not match a user device ID or a group user device ID of a user device, stopping decoding the second information payload.
The method according to any one of claims 44 to 51, further comprising:

obtaining configuration information of at least one embedder.
The method according to claim 52, wherein the semantic information of the controlling message is obtained by using a first embedder of the at least one embedder, and the at least one piece of semantic information of the at least one function modality is obtained by using the first embedder.
The method according to claim 53, wherein the configuration information of the at least one embedder comprises an embedder ID of each of the at least one embedder.
The method according to claim 54, wherein the information payload comprises information indicative of the embedder ID of the first embedder.
The method according to claim 55, wherein the information indicative of the embedder ID of the first embedder is carried in the first information payload.
The method according to claim 55 or 54, further comprising:

obtaining, by using the at least one embedder and according to the description of each of the at least one function modality, one or more pieces of semantic information, wherein for each of the at least one function modality, each of the at least one embedder is used to obtain a respective piece of semantic information; and

storing the one or more pieces of semantic information.
The method according to claim 57, further comprising:

obtaining, from the stored one or more pieces of semantic information and according to the information indicative of the embedder ID of the first embedder, the at least one piece of semantic information of the at least one function modality.
The method according claim 55 or 56, after obtaining the information payload, further comprising:

obtaining, by using the first embedder and according to the description of each of the at least one function modality, the at one piece of semantic information of the at least one function modality.
The method according to any one of claims 40 to 51, further comprising:

receiving the at least one piece of semantic information of the at least one function modality.
The method according to any one of claims 40 to 60, wherein the determining the target semantic information comprises:

determining, from the at least one piece of semantic information, a piece of semantic information most relevant to the semantic information of the controlling message as the target semantic information.
The method according to any one of claim 40 to 61, wherein more than one function modality exists, and the method further comprises:

obtaining configuration information for a combined function modality, wherein the combined function modality is combined from at least two function modalities of the more than one function modality; and

encapsulating the combined function modality into a first calling function.
The method according to claim 62, wherein the configuration information for the combined function modality comprises a calling order, and the first calling function calls calling functions corresponding to the at least two function modalities in the calling order.
A first apparatus, comprising:

a processing module, configured to:

generate a controlling message in natural language by using a language model (LM) ;

obtain semantic information of the controlling message according to the controlling message; and

form an information payload, where the information payload comprises the semantic information of the controlling message; and

a sending module, configured to send the information payload.
The first apparatus according to claim 64, wherein the first apparatus further comprises an obtaining module, configured to obtain registering information of a function modality, where the registering information of the function modality comprises a description of the function modality, where the description is in the natural language.
The first apparatus according to claim 65, wherein the registering information of the function modality further comprises a list of arguments of the function modality.
The first apparatus according to claim 65 or 66, wherein the processing module is further configured to:

register the function modality to the LM.
The first apparatus according to claim 67, wherein the processing module is further configured to:

record the registering information of the function modality.
The first apparatus according to claim 68, wherein the processing module is further configured to:

record information indicative of a user device identifier (ID) of a user device from which the registering information of the function modality is received.
The first apparatus according to any one of claims 65 to 69, wherein the processing module is further configured to:

obtain semantic information of the function modality by using an embedder according to the description of the function modality.
The first apparatus according to claim 70, wherein the processing module is further configured to:

store the semantic information of the function modality.
The first apparatus according to claim 70 or 71, wherein the processing module is further configured to:

send the semantic information of the function modality.
The first apparatus according to any one of claims 70 to 72, wherein the processing module is further configured to:

determine a relevance between the semantic information of the controlling message and semantic information of each of at least one function modality that has been registered, wherein the semantic information of each of the at least one function modality is obtained according to a description of each of the at least one function modality.
The first apparatus according to claim 73, wherein the processing module is further configured to:

determine that the relevance between the semantic information of the controlling message and the semantic information of each of the at least one function modality registered is lower than a threshold, and

reject the controlling message.
The first apparatus according to any one of claims 64 to 74, wherein the processing module is further configured to:

determine a portion without any arguments from the controlling message;

obtain the semantic information of the controlling message according to the portion by using an embedder.
The first apparatus according to claim 75, wherein the processing module is further configured to:

form a first information payload, wherein the first information payload comprises starting position information of a second transmission opportunity and length information of the semantic information; and

form a second information payload, wherein the second information payload comprises the semantic information of the controlling message, and wherein the second transmission opportunity is for transmission of the second information payload information.
The first apparatus according to claim 76, wherein the controlling message further comprises a list of arguments, and the first information payload further comprises the list of arguments of the controlling message.
The first apparatus according to claim 76, wherein the processing module is further configured to:

encode the second information payload by using a first encoding method;

wherein the first information payload further comprises information of the first encoding method.
The first apparatus according to claim 78, wherein the processing module is further configured to:

encode the first information payload by using a second encoding method.
The first apparatus according to claim 79, wherein the first encoding method comprises a first modulation and coding scheme (MCS) , and the second encoding method comprises a second MCS.
The first apparatus according to claim 76, wherein the processing module is further configured to:

determine semantic information most relevant to the semantic information of the controlling message from the semantic information of each of the at least one function modality that has been registered according to the relevance between the semantic information of the controlling message and the semantic information of each of at least one function modality that has been registered; and

determine at least one user device ID or at least one group user device ID corresponding to the most relevant semantic information;

wherein the first information payload further comprises: the at least one user device ID, or the at least one group user device ID; or the first information payload further comprises information indicative of the at least one user device ID or information indicative of the at least one group user device ID.
The first apparatus according to claim 81, wherein the information indicative of the at least one user device ID comprises a first code generated according to the at least one user device ID, wherein the first code comprises a mask code, a spreading code, or an interleaving code; or

the information indicative of the at least one group user device ID comprises a second code generated according to the at least one group user device ID, wherein the second code comprises a mask code, a spreading code, or an interleaving code.
The first apparatus according to claim 82, wherein the processing module is further configured to:

generate the first code according to the at least one user device ID; or

generate the second code according to the at least one group user device ID.
The first apparatus according to claim 76, wherein the sending module is further configured to:

send the first information payload during a first transmission opportunity; and

send the second information payload during the second transmission opportunity.
The first apparatus according to claim 84, wherein the first transmission opportunity is in a controlling physical channel, and the second transmission opportunity is in a data physical channel; or

both the first transmission opportunity and the second transmission opportunity are in a controlling physical channel; or

both the first transmission opportunity and the second transmission opportunity are in a data physical channel.
The first apparatus according to claim 67, wherein the processing module is further configured to:

register the function modality to at least one LMs by providing the registering information of the function modality to the at least one LMs, wherein the at least one LM comprises the LM.
The first apparatus according to claim 70, wherein the sending module is further configured to:

send configuration information of at least one embedder, wherein the embedder belongs to the at least one embedder.
The first apparatus according to claim 87, wherein the configuration information of the at least one embedder is sent via a broadcast message, a multicast message, or a unicast message.
The first apparatus according to claim 87 or 88, wherein the configuration information of the at least one embedder comprises an architecture and parameters of each of the at least one embedder.
The first apparatus according to claim 89, wherein the configuration information of the at least one embedder further comprises an embedder ID of each of the at least one embedder.
The first apparatus according to claim 90, wherein the first information payload further comprises: information indicative of an embedder ID of an embedder used to obtain the semantic information of the controlling message.
The first apparatus according to any one of claims 64 to 91, wherein the information payload is sent via a multicast message or a unicast message.
The first apparatus according to any one of claims 64 to 92, wherein the processing module is further configured to:

in response to generating at least two consecutive controlling messages more than a preset number of times by the LM, combine two function modalities corresponding to the two consecutive controlling messages to generate a combined function modality;

wherein the sending module is further configured to send configuration information for the combined function modality.
The first apparatus according to claim 93, wherein the processing module is further configured to:

register the combined function modality to the LM.
The first apparatus according to any one of claims 64 to 94, wherein there are multiple LMs, and the multiple LMs comprise the LM, wherein each of the multiple LMs has at least one embedder.
The first apparatus according to claim 95, wherein at least one function modality is registered in one or more LMs of the multiple LMs, but not registered in all the multiple LMs.
The first apparatus according to claim 96, wherein the multiple LMs comprise at least one LM compatible for an LTE system.
The first apparatus according to claim 95, wherein the multiple LMs comprise at least one LM compatible for a 5G system.
The first apparatus according to claim 95, wherein a first LM of the multiple LMs has a first embedder and a second LM of the multiple LM has a second embedder, and the first embedder and a second embedder built on vocabularies of different domains.
The first apparatus according to claim 64, wherein the LM has a first embedder and a second embedder built on vocabularies of different domains.
The first apparatus according to claim 64, wherein the processing module is further configured to:

register the LM.
The first apparatus according to claim 101, wherein the processing module is further configured to: register at least one embedder of the LM.
A second apparatus, comprising:

an obtaining module, configured to obtain an information payload, wherein the information payload comprises semantic information of a controlling message;

a processing module, configured to:

determine target semantic information from at least one piece of semantic information according to the semantic information of the controlling message, wherein each piece of the at least one piece of semantic information corresponds to a respective function modality in at least one function modality and is obtained according to a description of the respective function modality, and the description of the respective function modality is in natural language; and

execute a function modality corresponding to the target semantic information.
The second apparatus according to claim 103, wherein the processing module is further configured to:

encapsulate each function modality of the at least one function modality into a calling function, wherein the calling function for the respective function modality comprises the description of the respective function modality, or the calling function for the respective function modality comprises a list of arguments of the respective function modality and the description of the respective function modality.
The second apparatus according to claim 104, wherein the calling function comprises an application programming interface (API) calling function.
The second apparatus according to claim 103, wherein the second apparatus further comprises a sending module, and the sending module is configured to:

send registering information of each function modality of the at least one function modality, wherein the registering information of each function modality comprises the description of the respective function modality, or, the registering information of each function modality comprises a list of arguments of the respective function modality and the description of the respective function modality.
The second apparatus according to claim 103, wherein the obtaining module is configured to:

obtain a first information payload during a first transmission opportunity, wherein the first information payload comprises starting position information of a second transmission opportunity and length information of the semantic information of the controlling message; and

obtain a second information payload during the second transmission opportunity according to the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message, wherein the second information payload comprises the semantic information of the controlling message;

wherein the processing module is further configured to:

decode the first information payload to obtain the starting position information of the second transmission opportunity and the length information of the semantic information of the controlling message; and

decode the second information payload to obtain the semantic information of the controlling message.
The second apparatus according to claim 107, wherein the first information payload further comprises: a list of arguments of the controlling message.
The second apparatus according to claim 107 or 108, wherein the first information payload further comprises information of a first encoding method, and the second information payload is decoded by using the first encoding method.
The second apparatus according to claim 109, wherein the first information payload is decoded by using a second encoding method.
The second apparatus according to claim 110, wherein the first encoding method comprises a first modulation and coding scheme (MCS) , and the second encoding method comprise a second MCS.
The second apparatus according to any of claims 107 to 111, wherein the first information payload further comprises: information indicative of a user device identifier (ID) or information indicative of a group user device ID.
The second apparatus according to any one of claims 107 to 110, wherein the obtaining module is configured to:

obtain a first code generated according to information indicative of a user device ID, wherein the first code comprises a mask code, a spreading code, or an interleaving code; or

obtain a second code generated according to information indicative of a group user device ID, wherein the second code comprises a mask code, a spreading code, or an interleaving code.
The second apparatus according to claim 112 or 113, wherein the processing module is configured to:

in response to determining that the user device ID or the group user device ID does not match a user device ID or a group user device ID of a user device, stop decoding the second information payload.
The second apparatus according to any one of claims 107 to 114, wherein the obtaining module is configured to:

obtain configuration information of at least one embedder.
The second apparatus according to claim 115, wherein the semantic information of the controlling message is obtained by using a first embedder of the at least one embedder, and the at least one piece of semantic information of the at least one function modality is obtained by using the first embedder.
The second apparatus according to claim 116, wherein the configuration information of the at least one embedder comprises an embedder ID of each of the at least one embedder.
The second apparatus according to claim 117, wherein the information payload comprises information indicative of the embedder ID of the first embedder.
The second apparatus according to claim 118, wherein the information indicative of the embedder ID of the first embedder is carried in the first information payload.
The second apparatus according to claim 118 or 117, wherein the processing module is configured to:

obtain one or more pieces of semantic information by using the at least one embedder and according to the description of each of the at least one function modality, wherein for each of the at least one function modality, each of the at least one embedder is used to obtain a respective piece of semantic information; and

store the one or more pieces of semantic information.
The second apparatus according to claim 120, wherein the processing module is configured to:

obtain the at least one piece of semantic information of the at least one function modality from the stored one or more pieces of semantic information and according to the information indicative of the embedder ID of the first embedder.
The method according claim 118 or 119, wherein the processing module is configured to:

obtain the at one piece of semantic information of the at least one function modality by using the first embedder according to the description of each of the at least one function modality.
The second apparatus according to any one of claims 103 to 114, wherein the obtaining module is configured to:

receive the at least one piece of semantic information of the at least one function modality.
The second apparatus according to any one of claims 103 to 123, wherein the processing module is configured to:

determine a piece of semantic information most relevant to the semantic information of the controlling message as the target semantic information from the at least one piece of semantic information.
The second apparatus according to any one of claim 103 to 124, wherein more than one function modality exists, and the obtaining module is further configured to:

obtain configuration information for a combined function modality, wherein the combined function modality is combined from at least two function modalities of the more than one function modality; and

the processing module is further configured to encapsulate the combined function modality into a first calling function.
The second apparatus according to claim 125, wherein the configuration information for the combined function modality comprises a calling order, and the first calling function calls calling functions corresponding to the at least two function modalities in the calling order.
A third apparatus, comprising a processing circuitry for executing the controlling method according to any one of claims 1 to 39.
A fourth apparatus, comprising a processing circuitry for executing the controlling method according to any one of claims 40 to 63.
A communication system, comprising:

a first apparatus according to any one of claims 64 to 102 or a third apparatus according to claim 127; and

a second apparatus according to any one of claims 103 to 126 or a fourth apparatus according to claim 128.
A communication system, comprising:

a first processing circuitry for executing the controlling method according to any one of claims 1 to 39; and

a second processing circuitry for executing the controlling method according to any one of claims 40 to 63.
A computer-readable storage medium storing computer execution instructions which, when executed by a processor, cause the processor to execute the controlling method according to any one of claims 1 to 39, or the controlling method according to any one of claims 40 to 63.
A computer program product including computer execution instructions which, when executed by a processor, cause the processor to execute the controlling method according to any one of claims 1 to 39, or the controlling method according to any one of claims 40 to 63.