Communication method and apparatus, terminal device and network device

Abstract

The present application relates to the technical field of communications. Disclosed are a communication method and apparatus, a terminal device and a network device. The network device sends first information, the first information being used for indicating modulation enabling information corresponding to data and / or a modulation mode corresponding to the data, the modulation enabling information corresponding to the data being AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data being an AI modulation mode or a non-AI modulation mode; and correspondingly, the terminal device receives the first information. Hence, when AI technology is applied to channel, information or signal processing of wireless communication systems, the present application considers the combination of AI and modulation modes. In the combination of AI technology and modulation, the present application uses the first information to implement network indication for enabling or disabling AI-based modulation and / or network indication for AI modulation modes, and then flexibly configures modulation and demodulation of data and / or modulation modes of data on the basis of communication requirements and the like.

Description

Communication method and apparatus, terminal device and network device

[0001] The present application claims priority to the prior application No. 202410842040.4 entitled "Communication method and apparatus, terminal device and network device" filed on June 26, 2024, the contents of which are incorporated herein by reference. TECHNICAL FIELD

[0002] The present application relates to the field of communication technology, and in particular to a communication method and apparatus, a terminal device and a network device. BACKGROUND

[0003] Artificial intelligence (AI) technology can solve some problems that are difficult to solve by traditional modeling methods, such as some nonlinear problems, problems with too many parameters, etc. Through the training of a large amount of data, some problem solving patterns are established, and more accurate prediction results are provided. Among them, AI can include machine learning (ML), deep learning (DL), etc., and AI models can include various linear and nonlinear network models, such as linear regression, vector machines, convolutional neural networks (CNN), deep neural networks (DNN), etc.

[0004] With the continuous development of AI technology and the continuous evolution of wireless communication systems, the 3rd generation partnership project (3GPP) is currently discussing the application of AI technology to wireless communication systems. SUMMARY

[0005] The present application provides a communication method and apparatus, a terminal device and a network device, in order to realize AI-based modulation when AI technology is applied to the processing of channels, information or signals of wireless communication systems.

[0006] In a first aspect, the present application provides a communication method, comprising:

[0007] receiving first information, the first information being used to indicate modulation enabling information corresponding to data and / or a modulation mode corresponding to the data;

[0008] Among them, the modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode.

[0009] It can be seen that, when the AI technology is applied to the processing of the channel, information or signal of the wireless communication system, the application considers the combination of the AI and the modulation mode. In the combination of the AI and the modulation mode, the application introduces the first information. If the first information is used to indicate the modulation enabling information corresponding to the data, it indicates that the first information is used to indicate the AI-based modulation or the non-AI-based modulation, so that the network indication or the network configuration enables or disables the AI-based modulation based on the first information, so as to flexibly configure whether the AI technology needs to be used for the modulation of the data according to the communication demand and the like.

[0010] If the first information is used to indicate the modulation mode corresponding to the data, it indicates that the first information is used to indicate the AI modulation mode or the non-AI modulation mode, so that the network indication or the network configuration AI modulation mode is realized based on the first information, so as to flexibly configure whether the AI technology needs to be used for the modulation mode of the data according to the communication demand and the like.

[0011] If the first information is used to indicate the modulation enabling information corresponding to the data and the modulation mode corresponding to the data, it indicates that the first information is used to indicate the AI-based modulation or the non-AI-based modulation and the AI modulation mode or the non-AI modulation mode, so that the network indication or the network configuration enables or disables the AI-based modulation and the network indication or the network configuration AI modulation mode are realized based on the first information, so as to flexibly configure whether the AI technology needs to be used for the modulation of the data and the modulation mode of the data according to the communication demand and the like.

[0012] In some possible examples, the first information is carried by the RRC signaling or the MAC signaling. That is, the RRC signaling or the MAC signaling is used to indicate the AI-based modulation or the non-AI-based modulation.

[0013] In this way, the RRC signaling or the MAC signaling is used to realize the enabling or disabling of the AI-based modulation.

[0014] In some possible examples, the first information is carried by the DCI, and the first information is a field in the DCI, which is used to indicate the modulation enabling information corresponding to the data and / or the modulation mode corresponding to the data. That is, the DCI is used to indicate the AI-based modulation or the non-AI-based modulation.

[0015] In this way, the DCI is used to realize the enabling or disabling of the AI-based modulation.

[0016] In some possible examples, the AI-based modulation corresponds to the AI modulation mode, and the non-AI-based modulation corresponds to the non-AI modulation mode.

[0017] It should be noted that the AI-based modulation corresponds to the AI modulation manner. That is, the AI-based modulation has a corresponding relationship with the AI modulation manner. In this way, if there is only one AI modulation manner, when the first information is used to indicate the AI-based modulation, the one AI modulation manner can be determined or implicitly indicated through the corresponding relationship.

[0018] The non-AI-based modulation corresponds to the non-AI modulation manner. That is, the non-AI-based modulation has a corresponding relationship with the non-AI modulation manner. In this way, if there is only one non-AI modulation manner, when the first information is used to indicate the non-AI-based modulation, the one non-AI modulation manner can be determined or implicitly indicated through the corresponding relationship.

[0019] In some possible examples, the data includes one or more codewords, and modulation enabling information corresponding to each of the plurality of codewords is the same or different.

[0020] It should be noted that the modulation enabling information corresponding to the plurality of codewords is the same, which can be understood as that the plurality of codewords are all AI-based modulation or the plurality of codewords are all non-AI-based modulation. The modulation enabling information corresponding to the plurality of codewords is different, which can be understood as that one of the plurality of codewords is AI-based modulation and another of the plurality of codewords is non-AI-based modulation.

[0021] In some possible examples, the plurality of codewords includes a first codeword and a second codeword; the first information includes a first field and a second field in the DCI, the first field is used to indicate a modulation manner corresponding to the first codeword, and the second field is used to indicate a modulation manner corresponding to the second codeword.

[0022] In this way, the first field and the second field in the DCI enable the network to indicate the modulation manner corresponding to the first codeword and the modulation manner corresponding to the second codeword.

[0023] In some possible examples, interpretation or parsing of the second field depends on the first field.

[0024] It should be noted that the interpretation or parsing of the second field depends on the first field, which can be understood as that the content indicated by the second field depends on or is associated with the content indicated by the first field, or the bit width of the second field depends on or is associated with the content indicated by the first field, or the length of the second field depends on or is associated with the content indicated by the first field.

[0025] For example, under the condition that the modulation enabling information corresponding to the first codeword is same as the modulation enabling information corresponding to the second codeword, the interpretation or analysis of the second field depends on the first field. This is because, under the condition that the modulation enabling information corresponding to the first codeword is same as the modulation enabling information corresponding to the second codeword, since the content indicated by the first field is the modulation mode corresponding to the first codeword, and the modulation mode corresponding to the first codeword corresponds to the modulation enabling information corresponding to the first codeword, the modulation enabling information corresponding to the second codeword can be known through the modulation mode corresponding to the first codeword. In this way, the content indicated by the second codeword can be one of the one or more modulation modes corresponding to the content indicated by the first field, and the content indicated by the second field can determine the bit width or length of the second field.

[0026] In a second aspect, a communication method is provided, including:

[0027] sending first information, the first information being used to indicate modulation enabling information corresponding to data and / or a modulation mode corresponding to the data;

[0028] The modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode.

[0029] In a third aspect, a communication method is provided, including:

[0030] monitoring performance of AI-based modulation.

[0031] It can be seen that, in the combination of AI technology and modulation mode, AI technology can bring certain advantages, such as realizing adaptive modulation and improving performance. However, AI technology can also bring certain disadvantages, such as that the processing process of the AI model occupies a large amount of computing power or there is a large error between the inference result of the AI model and the actual result. Therefore, the terminal device or the network device needs to monitor the performance of AI-based modulation, so as to timely make corresponding adjustments according to the performance monitoring result, such as adjusting the parameters, architecture or training data of the AI model, or using non-AI-based modulation, so as to avoid the deficiencies brought by the AI model as much as possible and improve the inference result of the AI model.

[0032] In some possible examples, monitoring the performance of AI-based modulation includes:

[0033] monitoring the performance of AI-based modulation according to comparison result information between an input of a first AI model and an output of a second AI model, the first AI model being used for AI-based modulation, and the second AI model being used for AI-based demodulation.

[0034] It can be seen that, for the bilateral AI model composed of the first AI model and the second AI model, the terminal device or the network device takes the input of the first AI model as the actual truth and takes the output of the second AI model as the inference result of the bilateral AI model, and determines the inference accuracy of the bilateral AI model according to the comparison result information between the inference result of the bilateral AI model and the actual truth, so as to monitor the performance of the AI-based modulation through the inference accuracy of the bilateral AI model.

[0035] In some possible examples, monitoring the performance of the AI-based modulation according to the comparison result information between the input of the first AI model and the output of the second AI model includes:

[0036] Obtaining unmodulated information;

[0037] Processing the unmodulated information through the first AI model to obtain the output of the first AI model, and processing the output of the first AI model through the second AI model to obtain the output of the second AI model;

[0038] Monitoring the performance of the AI-based modulation according to the comparison result information between the unmodulated information and the output of the second AI model.

[0039] It can be seen that, for the bilateral AI model composed of the first AI model and the second AI model, the terminal device or the network device takes the unmodulated information as the actual truth and takes the output of the second AI model as the inference result of the bilateral AI model, and determines the inference accuracy of the bilateral AI model according to the comparison result information between the inference result of the bilateral AI model and the actual truth, so as to monitor the performance of the AI-based modulation through the inference accuracy of the bilateral AI model.

[0040] In some possible examples, monitoring the performance of the AI-based modulation includes:

[0041] Monitoring the performance of the AI-based modulation according to the comparison result information between the performance indicator information and the preset threshold.

[0042] It should be noted that the performance indicator information can be a performance indicator of a wireless communication system under AI-based modulation under certain conditions, which is an evaluation indicator considering various factors such as channel quality, signal quality or data distribution of the model. The preset threshold can be a threshold specified in a standard protocol or a preset threshold. Therefore, the terminal device or the network device can monitor the performance of the AI-based modulation according to the comparison result information between the performance indicator information and the preset threshold.

[0043] In some possible examples, the performance indicator information includes at least one of the following:

[0044] The signal-to-noise ratio or the signal-to-interference-plus-noise ratio corresponding to the AI-based modulation, the data distribution feature corresponding to the AI-based modulation, the bit error rate corresponding to the AI-based modulation, the block error rate corresponding to the AI-based modulation, or the probability of the AI-based modulation corresponding to the HARQ NACK.

[0045] In some possible examples, the performance of the AI-based modulation is monitored, including:

[0046] The performance of the AI-based modulation is monitored according to comparison result information between first repetition transmission information and second repetition transmission information, the first repetition transmission information being obtained based on the AI-based modulation, and the second repetition transmission information being obtained based on the non-AI-based modulation.

[0047] It should be noted that the first repetition transmission information can represent related information corresponding to repetition transmission of data, signals or information obtained based on the AI-based modulation. The second repetition transmission information can represent related information corresponding to repetition transmission of data, signals or information obtained based on the non-AI-based modulation. In this way, the performance of the AI-based modulation and the performance of the non-AI-based modulation are compared according to the comparison result information between the first repetition transmission information and the second repetition transmission information, so as to monitor the performance of the AI-based modulation.

[0048] In a fourth aspect, a communication apparatus is provided, and the communication apparatus includes:

[0049] The receiving unit is configured to receive first information, the first information being used to indicate modulation enabling information corresponding to data and / or a modulation mode corresponding to the data.

[0050] The modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode.

[0051] It can be seen that when the AI technology is applied to the processing of channels, information or signals of a wireless communication system, the present application considers the combination of the AI and the modulation mode. In the combination of the AI and the modulation mode, the present application introduces the first information. If the first information is used to indicate the modulation enabling information corresponding to the data, it indicates that the first information is used to indicate the AI-based modulation or the non-AI-based modulation, so as to enable or disable the AI-based modulation through the first information indicated or configured by the network, so as to flexibly configure whether the AI technology needs to be used for the modulation of the data according to the communication demand.

[0052] If the first information is used for indicating the modulation mode corresponding to the data, it indicates that the first information is used for indicating the AI modulation mode or the non-AI modulation mode, so that the network indicates or configures the AI modulation mode through the first information, so as to flexibly configure whether the modulation mode of the data needs to adopt the AI technology according to the communication demand and the like.

[0053] If the first information is used for indicating the modulation enabling information corresponding to the data and the modulation mode corresponding to the data, it indicates that the first information is used for indicating the AI-based modulation or the non-AI-based modulation and the AI modulation mode or the non-AI modulation mode, so that the network indicates or configures the enabling or disabling of the AI-based modulation and the network indicates or configures the AI modulation mode through the first information, so as to flexibly configure the modulation of the data and whether the modulation mode of the data needs to adopt the AI technology according to the communication demand and the like.

[0054] In a fifth aspect, a communication apparatus is provided, and the apparatus includes:

[0055] a sending unit configured to send first information, the first information being used for indicating modulation enabling information corresponding to data and / or a modulation mode corresponding to the data;

[0056] The modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode.

[0057] In a sixth aspect, a communication apparatus is provided, and the apparatus includes:

[0058] a monitoring unit configured to monitor the performance of AI-based modulation.

[0059] It can be seen that in the combination of the AI technology and the modulation mode, the AI technology can bring certain advantages, such as realizing adaptive modulation and improving performance. However, the AI technology can also bring certain disadvantages, such as that the processing process of the AI model occupies a large amount of computing power or that there is a large error between the inference result of the AI model and the actual result. Therefore, the terminal device or the network device needs to monitor the performance of the AI-based modulation, so as to timely make corresponding adjustments according to the performance monitoring result, such as adjusting the parameters, architecture or training data of the AI model, or adopting non-AI-based modulation, so as to avoid the deficiencies brought by the AI model as much as possible and improve the inference result of the AI model.

[0060] In a seventh aspect, the steps in the method designed in the first aspect or the third aspect are applied to a terminal device.

[0061] In an eighth aspect, the steps in the method designed in the second aspect or the third aspect are applied to a network device.

[0062] A ninth aspect is a terminal device of the present application, comprising a processor, a memory, and a computer program or instructions stored in the memory, wherein the processor executes the computer program or instructions to implement the steps in the method designed in the first aspect or the third aspect.

[0063] A tenth aspect is a network device of the present application, comprising a processor, a memory, and a computer program or instructions stored in the memory, wherein the processor executes the computer program or instructions to implement the steps in the method designed in the second aspect or the third aspect.

[0064] An eleventh aspect is a chip of the present application, comprising a processor, wherein the processor executes the steps in the method designed in the first aspect, the second aspect, or the third aspect.

[0065] A twelfth aspect is a chip module of the present application, comprising a transceiver component and a chip, wherein the chip comprises a processor, and the processor executes the steps in the method designed in the first aspect, the second aspect, or the third aspect.

[0066] A thirteenth aspect is a computer readable storage medium of the present application, wherein the computer readable storage medium stores a computer program or instructions, and the computer program or instructions are executed to implement the steps in the method designed in the first aspect, the second aspect, or the third aspect.

[0067] A fourteenth aspect is a computer program product of the present application, comprising a computer program or instructions, and the computer program or instructions are executed to implement the steps in the method designed in the first aspect, the second aspect, or the third aspect. For example, the computer program product can be a software installation package.

[0068] It should be understood that the beneficial effects brought by the technical solutions of the second aspect, the fourth aspect, the fifth aspect, the seventh aspect to the fourteenth aspect can refer to the technical effects brought by the technical solutions of the first aspect, which will not be repeated here. The beneficial effects brought by the technical solutions of the sixth aspect, the seventh aspect to the fourteenth aspect can refer to the technical effects brought by the technical solutions of the third aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0069] FIG. 1 is a schematic diagram of an architecture of a communication system according to an embodiment of the present application;

[0070] FIG. 2 is a schematic diagram of another architecture of a communication system according to an embodiment of the present application;

[0071] FIG. 3 is a schematic diagram of a flow of a communication method according to an embodiment of the present application;

[0072] FIG. 4 is a structure diagram of a MAC CE according to an embodiment of the present application;

[0073] FIG. 5 is a structure diagram of another MAC CE according to an embodiment of the present application;

[0074] FIG. 6 is a flow diagram of another communication method according to an embodiment of the present application;

[0075] FIG. 7 is a functional unit constituent block diagram of a communication apparatus according to an embodiment of the present application;

[0076] FIG. 8 is a functional unit constituent block diagram of another communication apparatus according to an embodiment of the present application;

[0077] FIG. 9 is a functional unit constituent block diagram of another communication apparatus according to an embodiment of the present application;

[0078] FIG. 10 is a structure diagram of a terminal device according to an embodiment of the present application;

[0079] FIG. 11 is a structure diagram of a network device according to an embodiment of the present application. DETAILED DESCRIPTION

[0080] It should be understood that the terms "first", "second" and the like in the description and in the claims of the present application are used for distinguishing between similar objects and not necessarily for describing a specific sequential or chronological order. The terms "comprises", "comprising", "includes", "including" and the like are to be construed open- ended, meaning that they are used to describe various embodiments and that not all embodiments include or are limited to the recited elements, steps or claims. Moreover, ordinal indicators, such as first, second or the like, cannot be construed as indicating a specific sequential order or chronology.

[0081] "Embodiments" in the present application mean that the specific features, structures or characteristics described in connection with the embodiments can be included in at least one embodiment of the present application. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it mutually exclusive or alternative to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0082] "at least one" or "at least one" in the present application means one or more, and multiple means two or more.

[0083] "and / or" in the present application describes the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can represent the following three cases: A exists alone, A and B exist together, and B exists alone. Wherein, A, B can be singular or plural. The character " / " can represent that the associated objects before and after are in an "or" relationship.

[0084] The "at least one of" or similar expressions in the embodiments of the present application refer to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can represent the following seven cases: a, b, c, a and b, a and c, b and c, a, b and c. Wherein, each of a, b, c can be an element or a set containing one or more elements.

[0085] The "of", "corresponding", "corresponding", "associated", "associated", "mapped" in the embodiments of the present application can be mixed sometimes. It should be pointed out that the concept or meaning to be expressed is consistent when no distinction is emphasized.

[0086] The "network" in the embodiments of the present application can be the same concept as "system" and the like, and the communication system is a communication network.

[0087] The "connection" in the embodiments of the present application refers to various connection modes such as direct connection or indirect connection to realize communication between devices, which is not limited specifically.

[0088] The related content involved in the technical solutions of the embodiments of the present application is specifically introduced as follows.

[0089] The communication system of the present embodiment is specifically described as follows.

[0090]

Communication system

[0091] The technical solutions of the embodiments of the present application can be applied to various wireless communication systems, for example: a long term evolution (LTE) system, an advanced long term evolution (LTE-A) system, a New Radio (NR) system, an evolved system of the NR system, an LTE-based access to unlicensed spectrum (LTE-U) system, an NR-based access to unlicensed spectrum (NR-U) system, a non-terrestrial network (NTN) system, a universal mobile telecommunication system (UMTS), a 6th-Generation (6G) communication system, or other communication systems in the future, and the like.

[0092] It should be noted that the number of user connections supported by the conventional communication system is limited and easy to implement. With the development of communication technology, the communication system of the present application can not only support the conventional communication system, but also support device to device (D2D) communication, machine to machine (M2M) communication, machine type communication (MTC), vehicle to vehicle (V2V) communication, vehicle to everything (V2X) communication, narrow band internet of things (NB-IoT) communication, and the like. Therefore, the technical solutions of the embodiments of the present application can also be applied to the above communication systems.

[0093] For example, the embodiments of the present application can be applied to beamforming, carrier aggregation (CA), dual connectivity (DC), or standalone (SA) deployment scenarios, and the like.

[0094] For another example, the embodiments of the present application can be applied to a communication scenario of unlicensed spectrum. In the embodiments of the present application, the unlicensed spectrum can also be considered as a shared spectrum. Alternatively, the embodiments of the present application can also be applied to licensed spectrum. The licensed spectrum can also be considered as a non-shared spectrum.

[0095] Exemplarily, a network architecture of a communication system according to an embodiment of the present application can be referred to FIG. 1. As shown in FIG. 1, the communication system 10 can include a network device 110 and a terminal device 120. The terminal device 120 can communicate with the network device 110 in a wireless manner.

[0096] It should be understood that FIG. 1 is only an example of the network architecture of the communication system, and does not limit the network architecture of the communication system according to the embodiments of the present application. For example, the communication system 10 can further include a server or other devices, or the communication system 10 can include other network devices in addition to the network device 110, or the communication system 10 can include other terminal devices in addition to the terminal device 120.

[0097] The terminal device and the network device mentioned in the embodiments are described below.

[0098]

Terminal device

[0099] The terminal device can be a device with transceiving function, which can also be referred to as a terminal, a user equipment (UE), a remote terminal device, a relay device, an access terminal device, a subscriber unit, a subscriber station, a mobile station, a mobile station, a remote station, a mobile device, a user terminal device, a smart terminal device, a wireless communication device, a user agent or a user apparatus. It should be noted that the relay device is a terminal device capable of providing relay forwarding service for other terminal devices (including remote terminal devices).

[0100] For example, the terminal device can be a mobile phone, a tablet computer (Pad), a computer with wireless transceiving function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in unmanned automatic driving, a wireless terminal device in remote medical treatment, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, or a wireless terminal device in smart home, etc.

[0101] For example, the terminal device can be a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a handheld device having wireless communication function, a computing device, or other processing device connected to a wireless modem, an in-vehicle device, a wearable device, a terminal device in a next-generation communication system (e.g., an NR communication system, a 6G communication system), or a terminal device in a future evolved public land mobile network (PLMN), etc., without specific limitation.

[0102] Optionally, the terminal device can be deployed on land, or can be deployed on water (such as a ship, etc.), or can be deployed in the air (such as an airplane, a balloon, a satellite, etc.).

[0103] Optionally, the terminal device includes a device with wireless communication function, such as a chip system, a chip, a chip module, a device, or a unit. For example, the chip system can include a chip and can also include other discrete devices.

[0104] Optionally, the terminal device includes an AI model. The AI model can be a software unit and / or a hardware unit for processing a channel, information, or a signal using AI technology. For example, the AI model can be used for AI-based modulation.

[0105] It should be noted that AI can include ML, DL, etc., and the AI model can have certain model parameters and model architectures, etc., and the AI model can include various linear and nonlinear network models, such as linear regression, vector machines, CNN, DNN, etc. In addition, the "AI model" appearing in the present embodiment can be understood as an ML model, a DL model, an intelligent model, an AI interface, an AI device, an AI unit, an AI module, an AI algorithm, or an intelligent module, etc., without specific limitation.

[0106] Optionally, the AI module includes a chip, a chip module, etc.

[0107]

Network device

[0108] The network device can be a device with transceiving function, and can be used for communication with the terminal device.

[0109] The network device can include a device with a wireless communication function, such as a chip system, a chip, or a chip module. For example, the chip system can include a chip or other discrete devices. The network device serves a cell, and a terminal device in the cell can communicate with the network device through a transmission resource, such as a frequency spectrum resource. The cell can be a macro cell, a small cell, a metro cell, a micro cell, a pico cell, a femto cell, or the like.

[0110] Optionally, the network device can have a mobile feature, for example, the network device can be a mobile device. Optionally, the network device can be a satellite, a balloon station. For example, the satellite can be a low earth orbit (LEO) satellite, a medium earth orbit (MEO) satellite, a geostationary earth orbit (GEO) satellite, a high elliptical orbit (HEO) satellite, or the like. Optionally, the network device can also be a base station arranged at a location on land, water, or the like.

[0111] Optionally, the network device includes a device with a wireless communication function, such as a chip system, a chip, a chip module, a device, or a unit. For example, the chip system can include a chip and other discrete devices.

[0112] Optionally, the network device includes an AI model. The AI model can be a software unit and / or a hardware unit for processing a channel, information, or a signal using an AI technology. For example, the AI model can be used for AI-based modulation.

[0113] Optionally, the AI model includes a chip, a chip module, or the like.

[0114] Optionally, the network device can include an access network device and / or a device in a core network (CN).

[0115] The access network device and the core network device are described below.

[0116]

Access network device

[0117] The access network device can be referred to as a radio access network (RAN). The RAN can be composed of a network of multiple 5G-RAN nodes, implementing radio physical layer functions, resource scheduling and radio resource management, radio access control, and mobility management functions. The 5G-RAN is connected to the UPF through the user plane interface N3 and is used to transmit data of the terminal device; the 5G-RAN establishes a control plane signaling connection through the control plane interface N2 and the access and mobility management function (AMF), and is used to implement radio access bearer control and other functions. The RAN can be any device with wireless transceiver functions, including but not limited to a 5G node base (gNB), an evolved node base (eNB), a wireless access point (WiFi AP), a world interoperability for microwave access base station (WiMAX BS), a transmission receiving point (TRP), a wireless relay node, a wireless backhaul node, a master node (MN) in a dual connectivity architecture, a second node or secondary node (SN) in a dual connectivity architecture, and the like.

[0118] In addition, the access network device can refer to a device used for communication with the terminal device. For example, the access network device can be a base transceiver station (BTS) in a global system of mobile communication (GSM) system or a code division multiple access (CDMA) system, can be a base station (nodeB, NB) in a wideband code division multiple access (WCDMA) system, can be an evolved node base (eNB) in an LTE system, can be a radio controller in a cloud radio access network (CRAN) scenario, or can be a relay station, an access point, a vehicle-mounted device, a wearable device, and an access network device in a future 5G network or an access network device in a future evolved PLMN network, and the like. The embodiments of the present application are not limited.

[0119] In 5G NR, the functions of an access network device are divided into two parts, referred to as centralized unit (CU)-distributed unit (DU) separation. From the perspective of the protocol stack, the CU includes the Radio Resource Control (RRC) layer and the packet data convergence protocol (PDCP) layer of the LTE base station, and the DU includes the radio link control (RLC) layer, the media access control (MAC) layer, and the physical (PHY) layer of the LTE base station. In a common 5G base station deployment, the CU and the DU can be connected by an optical fiber in a physical manner, and a specially defined F1 interface exists logically between the CU and the DU for communication between the CU and the DU. From the perspective of functions, the CU is mainly responsible for radio resource control and configuration, cross-cell mobility management, bearer management, etc. The DU is mainly responsible for scheduling, physical signal generation and transmission.

[0120] Optionally, the access network device can be a macro base station, a micro base station, a pico base station, a small station, a relay station, a balloon station, etc.

[0121]

Core network device

[0122] The core network device can include network elements that provide various functions. Among them, the "network element" can also be referred to as an entity, a device, an apparatus, or a module, etc., and no specific limitation is made. In addition, in order to facilitate understanding and description, the description of "network element" is omitted in part of the description, for example, the network exposure function (NEF) network element is simply referred to as NEF, in this case, the "NEF" should be understood as the NEF network element or the NEF entity, and the following, the same or similar cases are omitted. Description.

[0123] For example, the core network device can include a mobility management entity (MME), a broadcast multicast service center (BMSC), etc., or can include corresponding functional entities in the 5G system, such as core network control plane (CP) or user plane (UP) network functions, etc., and the core network control plane can also be understood as a core network control plane function (CPF) entity.

[0124] The following describes various network elements included in the core network device.

[0125] The session management function (SMF) can be responsible for the control plane function of the session management of the terminal device, including the selection and control of the user plane function (UPF), the allocation of the internet protocol (IP) address, the QoS management of the session, the acquisition of the policy and charging control (PCC) policy, and the like.

[0126] The user plane function (UPF) can be an anchor point of the protocol data unit (PDU) session connection, and is responsible for the data packet filtering, data transmission / forwarding, rate control, generation of charging information, and the like of the terminal device, and provides the connection with the data network (DN).

[0127] The policy control function (PCF) can provide the configuration policy information for the terminal device, and provide the policy information for controlling the terminal device for the control plane network element (such as the SMF) of the network; and generate the terminal device access policy and the QoS flow control policy.

[0128] The AF can interact with the network element of the core network to provide some services. For example, the AF interacts with the PCF to perform service policy control, interacts with the NEF to acquire some network capability information or provide some application information to the network, and provides some data network access point information to the PCF to generate the routing information of the corresponding data service.

[0129] The NEF can be responsible for providing some state information related to the network to the application service.

[0130] The authentication server function (AUSF) can implement 3GPP and non-3GPP access authentication.

[0131] The unified data management (UDM) is a unified data management function, and is responsible for 3GPP AKA authentication, user identification, access authorization, registration, mobility, subscription, short message management, and the like.

[0132] The network slice selection function (NSSF) can determine the network slice instance that the terminal device is allowed to access according to the slice selection assistance information, subscription information, and the like of the terminal device.

[0133] The network repository function (NRF) can be a new function that provides registration and discovery functions, and enables network functions (NFs) to discover each other and communicate through API interfaces.

[0134] The unified data management (UDM) can be responsible for user identification, subscription data, authentication data management, and user service network element registration management.

[0135] The unified data repository (UDR) can be used for UDM to store and read subscription data, and for PCF to store and read policy data.

[0136] The network data analytics function (NWDAF) can provide network analysis services according to the request data of network services.

[0137] The network slice specific authentication and authorization function (NSSAAF) can be used to provide authentication and authorization for specific network slices.

[0138] It should be noted that the terminal device is connected to the access network device in a wireless manner, and the access network device is connected to the core network device in a wireless or wired manner. The access network device and the core network device can be independent and different physical devices, or the functions of the core network device and the logical functions of the access network device can be integrated on the same physical device, or a physical device can integrate part of the functions of the core network device and part of the functions of the access network device.

[0139] For example, FIG. 2 is a schematic diagram of another communication system architecture according to an embodiment of the present application. In FIG. 2, the names of the various network elements included in FIG. 2 are only names, and the names do not limit the functions of the network elements. In 5G networks and future other networks, the above-mentioned various network elements can also be other names, which are not limited specifically. For example, in a 6G network, part or all of the above-mentioned various network elements can use the terms in 5G, or other names, etc. This is uniformly described below, and the following will not be described again.

[0140] In addition, the network elements in FIG. 2 are not necessarily present at the same time, and which network elements are needed can be determined according to requirements. The connection relationship between the network elements in FIG. 2 is also not uniquely determined, and can be adjusted according to requirements. It can be understood that the above network elements or functions can be network elements in a hardware device, or software functions running on a dedicated hardware, or virtualized functions instantiated on a platform (for example, a cloud platform).

[0141] It should be understood that FIG. 2 is only an illustration of a network architecture of a communication system, and does not constitute a limitation on the network architecture of the communication system of the embodiments of the present application.

[0142] The communication system has been described above, and the combination of the AI technology and the processing of channels, information or signals of the embodiments will be specifically described below.

[0143] The processing flow of channels, information or signals is an important link to ensure reliable transmission of data in a wireless communication system. The processing flow of channels, information or signals can include multiple steps to improve the reliability, efficiency and performance of data transmission.

[0144] For example, the processing flow of channels, information or signals includes at least one of the following steps: cyclic redundancy check (CRC), channel coding, rate matching, scrambling, modulation and layer mapping, etc. It should be understood that there is no set order between these steps, and the processing flow of channels, information or signals can also include other steps, which are not specifically limited.

[0145] It is worth noting that in the processing of channels, information or signals, modulation and demodulation is a key technology. Modulation is the process of converting the signal to be transmitted (called the baseband signal) into a form suitable for transmission over the channel at the sending end, which is mainly achieved by changing some properties (such as frequency, amplitude, phase, etc.) of the carrier signal. For example, the modulation method can include binary phase shift keying (BPSK), quadrature phase shift keying (QPSK), 16 quadrature amplitude modulation (QAM), 64QAM, etc., which can combine multiple bits into a carrier signal / complex number, and represent different data states by changing its frequency, phase or amplitude.

[0146] Demodulation is the process of recovering the original baseband signal from the modulated signal transmitted through the channel at the receiving end, which is the inverse process of modulation. Therefore, according to the different modulation methods, the demodulation methods will also be different.

[0147] It should be noted that the "sending end" mentioned in this embodiment refers to a device that sends data, and the "receiving end" refers to a device that receives data. Among them, for the scenario of network device sending data and terminal device receiving data, the sending end is the network device and the receiving end is the terminal device. For the scenario of terminal device sending data and network device receiving data, the sending end is the terminal device and the receiving end is the network device.

[0148] However, traditional modulation techniques mainly rely on fixed rules and algorithms, which are designed according to pre-set channel conditions and expected transmission requirements. Therefore, traditional modulation techniques may have certain deficiencies.

[0149] For example, traditional modulators usually assume specific channel conditions and environmental parameters during design, and cannot automatically adapt to changes in channel noise, multipath interference, etc. This makes traditional modulators may not perform well in complex and dynamic channel environments.

[0150] For another example, traditional modulators usually rely on mathematical modeling and empirical rules to design modulator parameters and algorithms. These methods may not accurately capture complex channel characteristics and nonlinear distortions, resulting in limitations on system performance.

[0151] For another example, traditional modulators may have limited performance in handling nonlinear distortions. For example, in high-power amplifiers or nonlinear channels, traditional modulators may not be able to accurately model and compensate for nonlinear distortions, thereby affecting system performance.

[0152] For another example, for new modulation schemes or complex channel conditions, traditional modulators may need to spend a lot of time and resources for simulation and optimization. This limits the efficiency and flexibility of traditional modulators in exploring and optimizing new modulation schemes.

[0153] Since AI technology can solve some problems that are difficult to solve with traditional modeling methods, and realize the intelligentization of communication, this embodiment can consider AI-based modulation and non-AI-based modulation.

[0154] AI-based modulation can refer to a method that uses AI technology to achieve the purpose of modulation, which can be a method that uses machine learning and data-driven methods to optimize and enhance the traditional modulation process, and can be implemented through an AI model.

[0155] The non-AI-based modulation can refer to a method that does not use AI technology to achieve the purpose of modulation (i.e., traditional modulation) and can be implemented by a traditional modulator.

[0156] It should be noted that traditional modulation is usually based on theoretical models and simulation analysis, while AI-based modulation focuses more on using a large amount of real data and measured data to optimize the design and parameter settings of the modulator. Through machine learning algorithms, complex channel characteristics and modulation effects can be learned from the data, thereby improving the performance and robustness of the system.

[0157] For example, AI-based modulation can involve the following steps: data preprocessing, feature extraction, AI model training, AI model testing, and AI model deployment.

[0158] It should be noted that data preprocessing can be understood as preprocessing input data, such as quantization, encoding, etc. Feature extraction can be understood as extracting key features of input data, which are crucial for determining the best modulation method.

[0159] AI model training can be understood as using a large amount of sample data to train the AI model so that it can accurately identify and simulate different modulation methods. The training objectives of the AI model can be diverse, such as maximizing spectral efficiency, minimizing bit error rate, or reducing power consumption. By defining appropriate loss functions and training processes, the AI model can learn how to achieve these training objectives.

[0160] Model testing can be understood as evaluating the performance of the AI model on an independent test set. AI model deployment can be understood as deploying the trained AI model to an actual communication system for the purpose of data modulation.

[0161] In addition, compared with traditional modulation, AI-based modulation can achieve at least one of the following items:

[0162] One is that compared with traditional modulators, AI models can have higher adaptability and can adjust their working parameters in real time according to channel conditions and environmental changes. For example, through neural networks or reinforcement learning algorithms, AI models can learn and optimize their working strategies to adapt to changes in channel noise, multipath interference, etc.

[0163] One is that AI models can be used to adaptively learn channel characteristics, predict interference and attenuation, and dynamically adjust the modulation used according to changes in channel conditions, achieving adaptive modulation. This enables the communication system to more flexibly cope with changing environments and needs. For example, AI models can dynamically adjust modulation schemes according to current channel conditions to achieve different modulation purposes to ensure transmission quality and efficiency. Or, high-order modulation is used when channel conditions are good to increase data transmission rate, and low-order modulation is switched to when channel conditions are poor to ensure communication reliability.

[0164] One is that traditional modulators may have limited effect in dealing with nonlinear and complex channel conditions, while AI models can better handle these situations by identifying and compensating for nonlinear distortion in the channel through deep learning and other techniques to improve the robustness and reliability of the system.

[0165] One is that AI models can help explore and optimize new modulation methods that may be difficult to implement or optimize in traditional modulators. For example, generative adversarial networks (GAN) models can generate new modulation methods or optimize existing modulation methods to meet future high-efficiency transmission needs.

[0166] It should be noted that the demodulation corresponding to the AI-based modulation of the present embodiment can be based on AI (i.e., AI-based demodulation), or can be based on non-AI (i.e., non-AI-based demodulation or traditional demodulation), depending on the implementation of the network device or terminal device, or on the indication of the network device / terminal device, or on the capability of the network device / terminal device.

[0167] Similarly, the demodulation corresponding to the non-AI-based modulation of the present embodiment can be based on AI (i.e., AI-based demodulation), or can be based on non-AI (i.e., non-AI-based demodulation or traditional demodulation), depending on the implementation of the network device or terminal device, or on the indication of the network device / terminal device, or on the capability of the network device / terminal device.

[0168] AI-based demodulation can refer to a method that uses AI technology to achieve the purpose of demodulation, which can be a method that uses machine learning and data-driven methods to optimize and enhance the traditional demodulation process, and can be the inverse process of AI-based modulation, which can be implemented through an AI model. Among them, AI-based demodulation can refer to the above-mentioned related content about AI-based modulation, and will not be repeated here.

[0169] The non-AI-based demodulation can refer to a method for achieving the purpose of demodulation without using AI technology (i.e., traditional demodulation), can be the inverse process of the traditional modulation method, and can be implemented by a traditional demodulator.

[0170] In addition, in the combination of AI and modulation methods, the present embodiment also relates to non-AI modulation methods and AI modulation methods.

[0171] The non-AI modulation method can refer to a method for achieving the purpose of modulation without using an AI model (i.e., a traditional modulation method) and can be implemented by a traditional modulator. For example, the traditional modulation method can include BPSK, QPSK, or QAM (such as 16QAM, 64QAM, or 256QAM, etc.).

[0172] The AI modulation method can refer to a method for achieving the purpose of modulation using an AI model and can provide higher flexibility and performance than the traditional modulation method. It is worth noting that different AI models can have different AI model parameters, AI model architectures, training data, or training targets, and thus different AI models can achieve different modulation purposes. For example, some AI models can achieve the purpose of QPSK, some AI models can achieve the purpose of 16QAM, some AI models can achieve the purpose of 64QAM, and some AI models can achieve the purpose of a new modulation method, etc.

[0173] That is, different AI modulation methods can be achieved according to the differences in AI model parameters, AI model architectures, training data, or training targets of the AI model. Therefore, by selecting appropriate AI model architectures, AI model parameters, or training data and defining appropriate training targets, AI modulation methods suitable for different application scenarios can be achieved.

[0174] In addition, the AI modulation method of the present embodiment can achieve the purpose of a traditional modulation method (such as QPSK, 16QAM, etc.) or can achieve the purpose of a new modulation method other than the traditional modulation method. For example, as shown in Table 1. It should be understood that the AI modulation method of the present embodiment is not limited to that shown in Table 1, and can also include other AI modulation methods, which are not specifically limited.

[0175] Table 1

[0176] It is worth noting that the non-AI modulation mode can implicitly represent the non-AI-based modulation, and the AI modulation mode can implicitly represent the AI-based modulation. In this way, the network device or the terminal device knowing the modulation mode is the non-AI modulation mode can be equivalent to the network device or the terminal device knowing that the non-AI-based modulation is enabled, and the network device or the terminal device knowing the modulation mode is the AI modulation mode can be equivalent to the network device or the terminal device knowing that the AI-based modulation is enabled.

[0177] In addition, the present embodiment can involve one or more non-AI modulation modes and one or more AI modulation modes. For example, the non-AI modulation modes include BPSK, QPSK, and QAM. For another example, in Table 1, the AI modulation modes include AI modulation mode 1, AI modulation mode 2, AI modulation mode 3, AI modulation mode 4, and AI modulation mode 5.

[0178] Therefore, if there is only one non-AI modulation mode and one AI modulation mode, when the non-AI-based modulation is enabled (or the AI-based modulation is disabled), the network device or the terminal device can know the non-AI modulation mode; and when the AI-based modulation is enabled (or the non-AI-based modulation is disabled), the network device or the terminal device can know the AI modulation mode.

[0179] If there are multiple non-AI modulation modes and multiple AI modulation modes, when the non-AI-based modulation is enabled (or the AI-based modulation is disabled), the network device or the terminal device can need to determine one non-AI modulation mode from the multiple non-AI modulation modes; and when the AI-based modulation is enabled (or the non-AI-based modulation is disabled), the network device or the terminal device can need to determine one AI modulation mode from the multiple AI modulation modes.

[0180] In summary, the present embodiment will now specifically describe how to indicate / configure the enabling / disabling of the AI-based modulation and / or the AI modulation mode.

[0181] As shown in FIG. 3, FIG. 3 is a flow diagram of a communication method according to an embodiment of the present disclosure, which specifically includes the following steps:

[0182] S310. The network device sends first information, the first information being used to indicate modulation enabling information corresponding to data and / or modulation mode corresponding to the data; wherein the modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is AI modulation mode or non-AI modulation mode.

[0183] Correspondingly, the terminal device receives the first information.

[0184] It can be seen that, when the AI technology is applied to the processing of the channel, information or signal of the wireless communication system, the application considers the combination of the AI and the modulation mode. In the combination of the AI and the modulation mode, the application introduces the first information. If the first information is used to indicate the modulation enabling information corresponding to the data, it indicates that the first information is used to indicate the AI-based modulation or the non-AI-based modulation, so that the network indication or network configuration enables or disables the AI-based modulation is realized through the first information, so as to flexibly configure whether the AI technology needs to be used for the modulation of the data according to the communication demand and the like.

[0185] If the first information is used to indicate the modulation mode corresponding to the data, it indicates that the first information is used to indicate the AI modulation mode or the non-AI modulation mode, so that the network indication or network configuration AI modulation mode is realized through the first information, so as to flexibly configure whether the AI technology needs to be used for the modulation mode of the data according to the communication demand and the like.

[0186] If the first information is used to indicate the modulation enabling information corresponding to the data and the modulation mode corresponding to the data, it indicates that the first information is used to indicate the AI-based modulation or the non-AI-based modulation and the AI modulation mode or the non-AI modulation mode, so that the network indication or network configuration enables or disables the AI-based modulation and the network indication or network configuration AI modulation mode is realized through the first information, so as to flexibly configure whether the AI technology needs to be used for the modulation of the data and the modulation mode of the data according to the communication demand and the like.

[0187] The first information is specifically described from different embodiments as follows.

[0188]

Embodiment 1

[0189] In the "embodiment 1", the first information is used to indicate the modulation enabling information corresponding to the data, and the modulation enabling information corresponding to the data is the AI-based modulation or the non-AI-based modulation. That is, the first information is used to indicate the AI-based modulation or the non-AI-based modulation.

[0190] It should be noted that, when the first information is used to indicate the AI-based modulation, it indicates that the network device informs the terminal device to enable the AI-based modulation through the first information, or informs the terminal device to disable the non-AI-based modulation through the first information. That is, enabling the AI-based modulation can be equivalent to disabling the non-AI-based modulation.

[0191] In some possible examples, the embodiment can implement AI-based modulation by means of an AI model. The AI model can be a software unit and / or a hardware unit for modulation and demodulation by means of AI technology. The AI model can be understood as an ML model, a DL model, an intelligent model, an AI interface, an AI device, an AI unit, an AI module, an AI algorithm, or an intelligent module, without specific limitation.

[0192] In this way, when AI-based modulation is enabled, the sending end can modulate data based on an AI model (such as an AI modulator) and send the modulated data to the receiving end, and the receiving end can demodulate the received data. The demodulation can be AI-based demodulation or non-AI-based demodulation. It should be understood that when the sending end is a network device, the receiving end can be a terminal device; when the sending end is a terminal device, the receiving end can be a network device.

[0193] When the first information is used to indicate non-AI-based modulation, it means that the network device informs the terminal device to enable non-AI-based modulation through the first information, or to disable AI-based modulation through the first information. That is, enabling non-AI-based modulation can be equivalent to disabling AI-based modulation.

[0194] In some possible examples, the embodiment can implement non-AI-based modulation by means of a modulator. The modulator can include a conventional modulator. In addition, the modulator can be understood as a modulation unit or a modulation device, or a modulation circuit, without specific limitation.

[0195] In this way, when non-AI-based modulation is enabled, the sending end can modulate data based on a modulator and send the modulated data to the receiving end, and the receiving end can demodulate the received data. The demodulation can be AI-based demodulation or non-AI-based demodulation. It should be understood that when the sending end is a network device, the receiving end can be a terminal device; when the sending end is a terminal device, the receiving end can be a network device.

[0196] The signaling type of the first information is described below.

[0197] In some possible examples, the first information is carried by RRC signaling or MAC signaling. That is, the RRC signaling or the MAC signaling is used to indicate AI-based modulation or non-AI-based modulation. In this way, the RRC signaling or the MAC signaling is used to enable or disable AI-based modulation.

[0198] For example, for the first information being carried by RRC signaling or MAC signaling, a specific implementation manner is that the first information is a field or information in the RRC signaling or the MAC signaling, which is used to indicate the modulation enabling information corresponding to the data.

[0199] It should be noted that the field or information can be a special field or special information, or can be an existing field or existing information. The special field or special information can refer to adding a new field or information in the existing RRC signaling or MAC signaling, which is specially used to indicate the modulation enabling information corresponding to the data. In addition, the existing field or existing information can refer to a field or information that already exists and is used in the existing RRC signaling or MAC signaling, and the modulation enabling information corresponding to the data is indicated by multiplexing the existing field or existing information.

[0200] For example, taking the MAC signaling as a MAC control element (MAC CE) as an example, as shown in FIG. 4, the MAC CE is identified by a MAC subheader, and the logical channel ID (LCID) thereof is a specific value. The MAC CE includes the following fields:

[0201] BWP ID field: This field can be used to indicate the BWP to which the MAC CE applies. The length of this field is 2 bits, and this field is located in Oct 1;

[0202] AI-E / D field, which can be used to indicate AI-based modulation or non-AI-based modulation. The length of this field is 1 bit, and this field is located in Oct 1; for example, if the value of the 1-bit of this field is 0, it indicates non-AI-based modulation; if the value of the 1-bit of this field is 1, it indicates AI-based modulation;

[0203] R field: This field represents a reservation, and is located in Oct 1.

[0204] In some possible examples, the first information is carried by DCI. That is, the DCI is used to indicate AI-based modulation or non-AI-based modulation. In this way, the AI-based modulation is enabled or disabled through the DCI.

[0205] For example, for the first information being carried by DCI, a specific implementation manner can be that the first information is a field in the DCI, which is used to indicate the modulation enabling information corresponding to the data.

[0206] It should be noted that the field can be a special field or an existing field. The special field can refer to adding a new field in the existing DCI, and the new field is specifically used to indicate the modulation enabling information corresponding to the data. In addition, the existing field can refer to a field or information that already exists and is used in the existing DCI, and the modulation enabling information corresponding to the data is indicated by multiplexing the existing field, for example, the existing field is a modulation and coding scheme (MCS) field.

[0207] In some possible examples, the AI-based modulation corresponds to an AI modulation manner, and the non-AI-based modulation corresponds to a non-AI modulation manner.

[0208] It should be noted that the AI-based modulation corresponds to an AI modulation manner. That is, the AI-based modulation has a corresponding relationship with the AI modulation manner. In this way, if there is only one AI modulation manner, when the first information is used to indicate the AI-based modulation, the one AI modulation manner can be determined or implicitly indicated through the corresponding relationship.

[0209] The non-AI-based modulation corresponds to a non-AI modulation manner. That is, the non-AI-based modulation has a corresponding relationship with the non-AI modulation manner. In this way, if there is only one non-AI modulation manner, when the first information is used to indicate the non-AI-based modulation, the one non-AI modulation manner can be determined or implicitly indicated through the corresponding relationship.

[0210] Optionally, the AI-based modulation corresponding to the AI modulation manner can be network configuration or network indication, predefinition, preconfiguration, default, or standard protocol specification, and the non-AI-based modulation corresponding to the non-AI modulation manner can be network configuration or network indication, predefinition, preconfiguration, default, or standard protocol specification.

[0211] Optionally, the non-AI-based modulation can correspond to one or more non-AI modulation manners.

[0212] For example, taking that the non-AI-based modulation corresponds to one non-AI modulation manner as an example, as shown in Table 2, the non-AI-based modulation corresponds to 16QAM. Table 2 can be network configuration or network indication, predefinition, preconfiguration, default, or standard protocol specification.

[0213] Table 2

[0214] For another example, taking that the non-AI-based modulation corresponds to multiple non-AI modulation manners as an example, as shown in Table 3, the non-AI-based modulation corresponds to BPSK, QPSK, 16QAM, and 64QAM. Table 3 can be network configuration or network indication, predefinition, preconfiguration, default, or standard protocol specification.

[0215] Table 3

[0216] It should be noted that when the non-AI-based modulation corresponds to one non-AI modulation mode, the terminal device or the network device can determine the one non-AI modulation mode according to the non-AI-based modulation, and modulate the data in the one non-AI modulation mode. When the non-AI-based modulation corresponds to multiple non-AI modulation modes, the terminal device or the network device can determine the multiple non-AI modulation modes according to the non-AI-based modulation, and then determine one non-AI modulation mode from the multiple non-AI modulation modes to modulate the data. The one non-AI modulation mode is determined from the multiple non-AI modulation modes by means of predefinition, pre-configuration, network configuration or network indication, default mode, or standard protocol specification.

[0217] For example, taking network configuration or network indication as an example, for the non-AI-based modulation corresponding to multiple non-AI modulation modes, the network device sends first information; correspondingly, the terminal device receives the first information. The first information is used to indicate the non-AI-based modulation, and the first information is also used to indicate one non-AI modulation mode from the multiple non-AI modulation modes. In this way, the first information realizes network configuration or network indication of one non-AI modulation mode.

[0218] For example, taking network configuration or network indication as an example, for the non-AI-based modulation corresponding to multiple non-AI modulation modes, the network device sends first information and second information; correspondingly, the terminal device receives the first information and the second information. The first information is used to indicate the non-AI-based modulation, and the second information is used to indicate one non-AI modulation mode from the multiple non-AI modulation modes. The first information and the second information can be in the same signaling or different signaling. In this way, the second information realizes network configuration or network indication of one non-AI modulation mode.

[0219] For example, taking predefinition, default mode or standard protocol specification as an example, for the non-AI-based modulation corresponding to multiple non-AI modulation modes, the network device sends first information; correspondingly, the terminal device receives the first information. The first information is used to indicate the non-AI-based modulation. Then, the network device or the terminal device selects a first non-AI modulation mode or any non-AI modulation mode from the multiple non-AI modulation modes according to the predefinition, the default mode or the standard protocol specification.

[0220] Optionally, the AI-based modulation can correspond to one or more AI modulation modes.

[0221] For example, taking the implementation of BPSK through an AI model as an example, as shown in Table 4, the AI-based modulation corresponds to AI modulation mode 1. Table 4 can be network configuration or network indication, predefinition, preconfiguration, default, or standard protocol specification.

[0222] Table 4

[0223] For example, taking the implementation of BPSK through an AI model as an example, as shown in Table 4, the AI-based modulation corresponds to AI modulation mode 1. Table 4 can be network configuration or network indication, predefinition, preconfiguration, default, or standard protocol specification.

[0224] Table 5

[0225] It should be noted that when the AI-based modulation corresponds to one AI modulation mode, the terminal device or the network device can determine the one AI modulation mode according to the AI-based modulation, and modulate the data using the one AI modulation mode. When the AI-based modulation corresponds to multiple AI modulation modes, the terminal device or the network device can determine the multiple AI modulation modes according to the AI-based modulation, and then determine one AI modulation mode from the multiple AI modulation modes to modulate the data. The one AI modulation mode can be determined from the multiple AI modulation modes by predefinition, preconfiguration, network configuration or network indication, default mode, or standard protocol specification.

[0226] For example, taking network configuration or network indication as an example, for the AI-based modulation corresponding to multiple AI modulation modes, the network device sends first information; correspondingly, the terminal device receives the first information. The first information is used to indicate the AI-based modulation, and the first information is also used to indicate one AI modulation mode from the multiple AI modulation modes. In this way, the first information realizes network configuration or network indication of one AI modulation mode.

[0227] For example, taking network configuration or network indication as an example, for the AI-based modulation corresponding to multiple AI modulation modes, the network device sends first information; correspondingly, the terminal device receives the first information. The first information is used to indicate the AI-based modulation, and the first information is also used to indicate one AI modulation mode from the multiple AI modulation modes. In this way, the first information realizes network configuration or network indication of one AI modulation mode.

[0228] For example, the first information is used to indicate modulation enabling information corresponding to one or more codewords (CWs) or one or more transport blocks (TBs).

[0229] The following describes the first information used to indicate modulation enabling information corresponding to one or more codewords (CWs) or the first information used to indicate modulation enabling information corresponding to one or more transport blocks (TBs).

[0230] In some possible examples, the data includes one or more codewords, or the data includes one or more transport blocks. Therefore, the first information can be used to indicate modulation enabling information corresponding to the one or more codewords, or the first information can be used to indicate modulation enabling information corresponding to the one or more transport blocks. That is, the first information is used to indicate that the one or more codewords are based on AI modulation or non-AI modulation, or the first information is used to indicate that the one or more transport blocks are based on AI modulation or non-AI modulation.

[0231] Optionally, when the data includes a plurality of codewords or a plurality of transport blocks, modulation enabling information corresponding to the plurality of codewords or the plurality of transport blocks is the same or different.

[0232] It should be noted that the modulation enabling information corresponding to the plurality of codewords is the same, which can be understood as that the plurality of codewords are all based on AI modulation or the plurality of codewords are all based on non-AI modulation.

[0233] For example, the plurality of codewords includes a first codeword and a second codeword. The first codeword and the second codeword are both based on AI modulation, or the first codeword and the second codeword are both based on non-AI modulation. The modulation enabling information corresponding to the plurality of codewords is different, which can be understood as that one of the plurality of codewords is based on AI modulation and the other is based on non-AI modulation.

[0234] For example, the plurality of codewords includes a first codeword and a second codeword. The first codeword is based on AI modulation, and the second codeword is based on non-AI modulation.

[0235] The modulation enabling information corresponding to the plurality of transport blocks is the same, which can be understood as that the plurality of transport blocks are all based on AI modulation or the plurality of transport blocks are all based on non-AI modulation.

[0236] For example, the plurality of transport blocks includes a first transport block and a second transport block. Wherein, the first transport block and the second transport block are both based on AI modulation, or the first transport block and the second transport block are both based on non-AI modulation.

[0237] The modulation enabling information corresponding to the plurality of transport blocks is different. It can be understood that one transport block in the transport block code word is based on AI modulation and another transport block is based on non-AI modulation.

[0238] For example, the plurality of transport blocks includes a first transport block and a second transport block. Wherein, the first transport block is based on AI modulation, and the second transport block is based on non-AI modulation.

[0239] Optionally, the embodiment can determine that the modulation enabling information corresponding to the plurality of code words or the plurality of transport blocks is the same or different through network configuration or network indication, predefinition, preconfiguration, default mode, standard protocol, etc.

[0240] Optionally, the first information is also used to indicate the modulation enabling information corresponding to each code word in the plurality of code words, or the first information is used to indicate the modulation enabling information corresponding to each transport block in the plurality of transport blocks. In this way, the first information can be used to determine that the modulation enabling information corresponding to the plurality of code words or the plurality of transport blocks is the same or different.

[0241] For example, taking the plurality of code words including a first code word and a second code word as an example, the first information includes a third field and a fourth field in the DCI, the third field is used to indicate the modulation enabling information corresponding to the first code word, and the fourth field is used to indicate the modulation enabling information corresponding to the second code word. Wherein, the third field and the fourth field can be a special field or an existing field. For example, the third field is the MCS field corresponding to the first code word, and the fourth field is the MCS field corresponding to the first code word.

[0242] Optionally, when the network device configures the terminal device before sending the first information that the modulation enabling information corresponding to the plurality of code words or the plurality of transport blocks is the same, or when it is determined through preconfiguration, default mode, predefinition or standard protocol that the modulation enabling information corresponding to the plurality of code words or the plurality of transport blocks is the same, the network device can only need to indicate the modulation enabling information corresponding to a certain code word in the plurality of code words or a certain transport block in the plurality of transport blocks through the first information, without the need to indicate the modulation enabling information corresponding to each code word or each transport block through the first information, thereby facilitating the saving of signaling overhead.

[0243] For example, taking that the plurality of code words include a first code word and a second code word, and modulation enabling information corresponding to the first code word and the second code word is the same as an example, the first information includes a third field in the DCI, and the third field is used to indicate modulation enabling information corresponding to the first code word. In this way, modulation enabling information corresponding to the second code word can be determined through the third field, without the need to separately indicate modulation enabling information corresponding to the second code word, thereby saving signaling overhead.

[0244] Embodiment 2

[0245] In “Embodiment 2”, the first information is used to indicate a modulation mode corresponding to the data, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode. That is, the first information is used to indicate an AI modulation mode or a non-AI modulation mode.

[0246] It should be noted that the related content of the AI modulation mode and the non-AI modulation mode can be seen from the above content, and will not be described again. When the first information indicates the AI modulation mode, it means that the network device informs the terminal device of the AI modulation mode through the first information. In this embodiment, the AI modulation mode can be implemented through an AI model. In this way, the sender can use the AI modulation mode to modulate the data, and the receiver can demodulate the received data, which can be AI-based demodulation or non-AI-based demodulation. When the sender is a network device, the receiver can be a terminal device; when the sender is a terminal device, the receiver can be a network device.

[0247] Table 6

[0248] For example, taking that the first information includes 2 bits as an example, on the basis of the above table 3 and table 5, as shown in table 6. Table 6 can be network configuration or network indication, predefinition, preconfiguration, default or standard protocol. In this way, when the value of the 2 bits is 10, the first information is used to indicate AI modulation mode 2; when the value of the 2 bits is 11, the first information is used to indicate AI modulation mode 3. That is, at this time, the first information is used to indicate an AI modulation mode.

[0249] When the first information indicates the non-AI modulation mode, it means that the network device informs the terminal device of the non-AI modulation mode through the first information. In this embodiment, the non-AI modulation mode can be implemented through a modulator. In this way, the sender can use the non-AI modulation mode to modulate the data, and the receiver can demodulate the received data, which can be AI-based demodulation or non-AI-based demodulation. When the sender is a network device, the receiver can be a terminal device; when the sender is a terminal device, the receiver can be a network device.

[0250] For example, as shown in Table 6, when the value of the 2 bits is 00, the first information is used to indicate BPSK; when the value of the first information is 01, the first information is used to indicate 16QAM. That is, at this time, the first information is used to indicate a non-AI modulation mode.

[0251] The signaling type of the first signaling is exemplarily described below.

[0252] In some possible examples, the first information is carried by RRC signaling or MAC signaling. That is, the RRC signaling or the MAC signaling is used to indicate the modulation mode corresponding to the data. In this way, the AI modulation mode or the non-AI modulation mode is indicated by the RRC signaling or the MAC signaling.

[0253] For example, for the first information being carried by the RRC signaling or the MAC signaling, a specific implementation manner can be that the first information is a field or information in the RRC signaling or the MAC signaling, and the field or information is used to indicate the modulation mode corresponding to the data.

[0254] It should be noted that the field or information can be a special field or special information, or can be an existing field or existing information. The special field or special information can be that a new field or information is added in the existing RRC signaling or MAC signaling, and the new field or information is specially used to indicate the modulation mode corresponding to the data. In addition, the existing field or existing information can be that a field or information already exists and is used in the existing RRC signaling or MAC signaling, and the modulation mode corresponding to the data is indicated by multiplexing the existing field or existing information.

[0255] For example, taking the MAC signaling as a MAC CE as an example, as shown in FIG. 5, the MAC CE is identified by a MAC subheader, and the LCID is a specific value. The MAC CE includes the following fields.

[0256] A BWP ID field: the field can be used to indicate the BWP to which the MAC CE is applied, the length of the field is 2 bits, and the field is located in Oct 1 (byte 1);

[0257] An M field: the field can be used to indicate the AI modulation mode or the non-AI modulation mode, the length of the field is 2 bits, and the field is located in Oct 1 (byte 1);

[0258] An R field: the field indicates a reservation, and the field is located in Oct 1 (byte 1).

[0259] In some possible examples, the first information is carried by DCI. That is, the DCI is used to indicate the modulation mode corresponding to the data. In this way, the AI modulation mode or the non-AI modulation mode is indicated by the DCI.

[0260] For example, for the first information being carried by the DCI, one specific implementation is that the first information is a field in the DCI, which is used to indicate the modulation mode corresponding to the data.

[0261] It should be noted that the field can be a special field or an existing field. The special field can refer to adding a new field in the existing DCI, which is specially used to indicate the modulation mode corresponding to the data. In addition, the existing field can refer to a field or information that already exists and is used in the existing DCI, and the modulation mode corresponding to the data is indicated by multiplexing the existing field, for example, the existing field is the MCS field.

[0262] In some possible examples, the AI-based modulation corresponds to an AI modulation mode, and the non-AI-based modulation corresponds to a non-AI modulation mode.

[0263] It should be noted that the AI-based modulation corresponds to an AI modulation mode. That is, the AI-based modulation has a corresponding relationship with the AI modulation mode. In this way, when the first information is used to indicate the AI modulation mode, the enabling of the AI-based modulation can be determined or implicitly indicated through the corresponding relationship.

[0264] The non-AI-based modulation corresponds to a non-AI modulation mode. That is, the non-AI-based modulation has a corresponding relationship with the non-AI modulation mode. In this way, when the first information is used to indicate the non-AI modulation mode, the enabling of the non-AI-based modulation can be determined or implicitly indicated through the corresponding relationship.

[0265] Optionally, the AI-based modulation corresponding to the AI modulation mode can be network configured or network indicated, predefined, preconfigured, default, or standard protocol specified, and the non-AI-based modulation corresponding to the non-AI modulation mode can be network configured or network indicated, predefined, preconfigured, default, or standard protocol specified.

[0266] The following illustrates that the first information is used to indicate the modulation mode corresponding to one or more codewords or the first information is used to indicate the modulation mode corresponding to one or more transport blocks.

[0267] In some possible examples, the data includes one or more codewords, or the data includes one or more transport blocks. Therefore, the first information can be used to indicate the modulation mode corresponding to the one or more codewords, or the first information can be used to indicate the modulation mode corresponding to the one or more transport blocks. That is, the first information is used to indicate that the modulation mode corresponding to the one or more codewords is an AI modulation mode or a non-AI modulation mode, or the first information is used to indicate that the modulation mode corresponding to the one or more transport blocks is an AI modulation mode or a non-AI modulation mode.

[0268] Optionally, when the data comprises a plurality of code words or a plurality of transport blocks, the modulation modes corresponding to the plurality of code words or the plurality of transport blocks are the same or different.

[0269] It should be noted that the modulation modes corresponding to the plurality of code words are the same, which can be understood as that the plurality of code words all adopt the same AI modulation mode or the plurality of code words all adopt the same non-AI modulation mode.

[0270] For example, taking that the plurality of code words comprise a first code word and a second code word as an example, the first code word and the second code word all adopt AI modulation mode 1, or the first code word and the second code word all adopt 16QAM.

[0271] The modulation modes corresponding to the plurality of code words are different, which can be understood as that one code word in the plurality of code words adopts an AI modulation mode and another code word adopts a non-AI modulation mode, or one code word in the plurality of code words adopts an AI modulation mode and another code word adopts another AI modulation mode, or one code word in the plurality of code words adopts a non-AI modulation mode and another code word adopts another non-AI modulation mode.

[0272] For example, taking that the plurality of code words comprise a first code word and a second code word as an example, the first code word adopts AI modulation mode 1 and the second code word adopts 16QAM, or the first code word adopts AI modulation mode 1 and the second code word adopts AI modulation mode 2, or the first code word adopts QPSK and the second code word adopts 16QAM.

[0273] The modulation modes corresponding to the plurality of transport blocks are the same, which can be understood as that the plurality of transport blocks all adopt the same AI modulation mode or the plurality of transport blocks all adopt the same non-AI modulation mode.

[0274] For example, taking that the plurality of transport blocks comprise a first transport block and a second transport block as an example, the first transport block and the second transport block all adopt AI modulation mode 1, or the first transport block and the second transport block all adopt 16QAM.

[0275] The modulation modes corresponding to the plurality of transport blocks are different, which can be understood as that one transport block in the plurality of transport blocks adopts an AI modulation mode and another transport block adopts a non-AI modulation mode, or one transport block in the plurality of transport blocks adopts an AI modulation mode and another transport block adopts another AI modulation mode, or one transport block in the plurality of transport blocks adopts a non-AI modulation mode and another transport block adopts another non-AI modulation mode.

[0276] For example, the multiple transport blocks include a first transport block and a second transport block, the first transport block adopts AI modulation mode 1 and the second transport block adopts 16QAM, or the first transport block adopts AI modulation mode 1 and the second transport block adopts AI modulation mode 2, or the first transport block adopts QPSK and the second transport block adopts 16QAM.

[0277] Optionally, the embodiment can determine that the modulation modes corresponding to the multiple codewords or the multiple transport blocks are the same or different through network configuration or network indication, predefinition, preconfiguration, default mode, standard protocol, and the like.

[0278] Optionally, the embodiment can determine that the modulation enabling information corresponding to the multiple codewords is the same or different through network configuration or network indication, preconfiguration, predefinition, default mode, standard protocol, and the like.

[0279] Optionally, the first information is further used to indicate the modulation mode corresponding to each codeword in the multiple codewords, or the first information is used to indicate the modulation mode corresponding to each transport block in the multiple transport blocks. In this way, the first information can determine that the modulation modes corresponding to the multiple codewords or the multiple transport blocks are the same or different.

[0280] For example, the multiple codewords include a first codeword and a second codeword, the first information includes a first field and a second field in the DCI, the first field is used to indicate the modulation mode corresponding to the first codeword, and the second field is used to indicate the modulation mode corresponding to the second codeword. The first field and the second field can be a special field or an existing field. For example, the first field is the MCS field corresponding to the first codeword, and the second field is the MCS field corresponding to the second codeword.

[0281] It should be noted that, for the second field, the interpretation or analysis of the second field can not depend on the first field, or the interpretation or analysis of the second field can depend on the first field.

[0282] The interpretation or analysis of the second field does not depend on the first field, which can be understood as that the content indicated by the second field does not depend on the content indicated by the first field, or the bit width of the first field does not depend on the content indicated by the first field, or the length of the second field does not depend on the content indicated by the first field. For example, the first field and the second field are two bits, in Table 6, the value of the first field is 10 and the value of the second field is 01, or the value of the first field is 10 and the value of the second field is 11, or the value of the first field is 10 and the value of the second field is 10.

[0283] The interpretation or analysis of the second field depends on the first field. It can be understood that the content indicated by the second field depends on or is associated with the content indicated by the first field, or the bit width of the second field depends on or is associated with the content indicated by the first field, or the length of the second field depends on or is associated with the content indicated by the first field.

[0284] For example, under the condition that the modulation enabling information corresponding to the first codeword is the same as the modulation enabling information corresponding to the second codeword, the interpretation or analysis of the second field depends on the first field. This is because, under the condition that the modulation enabling information corresponding to the first codeword is the same as the modulation enabling information corresponding to the second codeword, since the content indicated by the first field is the modulation mode corresponding to the first codeword, and the modulation mode corresponding to the first codeword corresponds to the modulation enabling information corresponding to the first codeword, the modulation enabling information corresponding to the second codeword can be known through the modulation mode corresponding to the first codeword. In this way, the content indicated by the second field can be one of the one or more modulation modes corresponding to the content indicated by the first field, and the content indicated by the second field can determine the bit width or length of the second field.

[0285] Taking Table 6 as an example, under the condition that the modulation enabling information corresponding to the first codeword is the same as the modulation enabling information corresponding to the second codeword, if the value of the first field is 10, the content indicated by the first field is AI-based modulation, that is, the modulation enabling information corresponding to the first codeword is AI-based modulation, so the modulation enabling information corresponding to the second codeword is also AI-based modulation. Since the AI modulation mode corresponding to the AI-based modulation in Table 6 includes AI modulation mode 2 and AI modulation mode 3, the second field can only need 1 bit to indicate one from AI modulation mode 2 and AI modulation mode 3. It can be seen that the content indicated by the second field is one of the AI modulation mode 2 and AI modulation mode 3 associated with the content indicated by the first field, and the bit width or length of the second field is 1 bit.

[0286] Optionally, when the network device configures the terminal device before sending the first information that the modulation modes corresponding to the plurality of codewords or the plurality of transport blocks are the same, or when it is determined through pre-configuration, pre-definition, default mode or standard protocol that the modulation modes corresponding to the plurality of codewords or the plurality of transport blocks are the same, the network device can only need to indicate the modulation mode corresponding to a certain codeword or a certain transport block in the plurality of codewords or the plurality of transport blocks through the first information, without the need to indicate the modulation mode corresponding to each codeword or each transport block through the first information, thereby facilitating saving signaling overhead.

[0287] For example, taking the case that the plurality of codewords include a first codeword and a second codeword, and the modulation modes corresponding to the first codeword and the second codeword are the same, the first information includes a first field in the DCI, and the first field is used to indicate the modulation mode corresponding to the first codeword. In this way, the modulation mode corresponding to the second codeword can be determined through the first field, and there is no need to separately indicate the modulation mode corresponding to the second codeword, thereby saving signaling overhead.

[0288] Embodiment 3

[0289] In the “embodiment 3”, the first information is used to indicate the modulation enabling information corresponding to the data and the modulation mode corresponding to the data.

[0290] It should be noted that since the “embodiment 3” is equivalent to the combination of the above-mentioned “embodiment 1” and “embodiment 2”, the related content can be found in the above-mentioned “embodiment 1” and “embodiment 2”, and will not be repeated here.

[0291] The following embodiment specifically describes how to monitor the performance of AI-based modulation.

[0292] As shown in FIG. 6, FIG. 6 is a flowchart of a communication method according to an embodiment of the present application, which specifically includes the following steps:

[0293] S610. Monitor the performance of AI-based modulation.

[0294] It can be seen that in the combination of AI technology and modulation mode, AI technology can bring certain advantages, such as realizing adaptive modulation and improving performance. However, AI technology may also bring certain disadvantages, such as the processing process of the AI model occupying a large amount of computing power or the inference result of the AI model having a large error with the actual result. Therefore, the terminal device or network device of the embodiment needs to monitor the performance of AI-based modulation, so as to timely make corresponding adjustments according to the performance monitoring result, such as adjusting the parameters, architecture or training data of the AI model, or using non-AI-based modulation, so as to avoid the deficiencies brought by the AI model as much as possible and improve the inference result of the AI model.

[0295] For example, for monitoring the performance of AI-based modulation, a specific implementation manner can be: monitoring the performance of AI-based modulation according to comparison result information between the input of the first AI model and the output of the second AI model, the first AI model being used for AI-based modulation, and the second AI model being used for AI-based demodulation.

[0296] It should be noted that the related descriptions of the AI-based modulation and the AI-based demodulation can be seen from the above content, and will not be repeated. Since the AI-based demodulation is an inverse process of the AI-based modulation, the first AI model and the second AI model can be regarded as an AI model pair or a two-side AI model. The first AI model can be a software unit and / or a hardware unit for modulation based on AI, and the second AI model can be a software unit and / or a hardware unit for demodulation based on AI.

[0297] For the two-side AI model composed of the first AI model and the second AI model, the two-side AI model has two different stages of training and inference. The training of the two-side AI model refers to adjusting parameters of the two-side AI model using labeled data / sample data, so that the two-side AI model can better adapt to a data set, has good generalization ability, and meets specific requirements. The inference of the two-side AI model refers to processing new data using the trained two-side AI model.

[0298] Therefore, the embodiment can monitor the performance of the AI-based modulation according to the inference accuracy of the two-side AI model. For example, if the inference accuracy of the two-side AI model is high, it means that the performance of the AI-based modulation is good, and the modulation result is more accurate. If the inference accuracy of the two-side AI model is low, it means that the performance of the AI-based modulation is poor, and the modulation result is less accurate.

[0299] The inference accuracy of the two-side AI model can refer to comparison result information between the inference result of the two-side AI model and the actual ground truth. The comparison result information can reflect error size, similarity, or compliance / matching / similarity / approximation degree, etc. If the comparison result information is smaller, it means that the inference accuracy of the two-side AI model is higher. If the comparison result information is larger, it means that the inference accuracy of the two-side AI model is lower.

[0300] The actual ground truth refers to the real and accurate label / answer / target information used in training or inference, representing the real situation that the terminal device or the network device hopes the two-side AI model can accurately predict or classify.

[0301] Based on this, this embodiment takes the input of the bilateral AI model (i.e., the input of the first AI model) as the actual truth and the output of the bilateral AI model (i.e., the output of the second AI model) as the inference result of the bilateral AI model as an example. In this case, the comparison result between the inference result of the bilateral AI model and the actual truth is the comparison result between the input of the first AI model and the output of the second AI model. Thus, the sending or receiving end of this embodiment can monitor the performance of AI-based modulation based on the comparison result between the input of the first AI model and the output of the second AI model.

[0302] It is worth noting that the sending or receiving end needs to obtain the input of the first AI model and the output of the second AI model in order to monitor the performance of AI-based modulation based on the comparison results between the input of the first AI model and the output of the second AI model.

[0303] For the transmitting end, it includes a first AI model. The transmitting end can acquire first unmodulated information, and then process this first unmodulated information using the first AI model to obtain its output. The first unmodulated information can be information generated / obtained by the transmitting end and can serve as input to the first AI model, thus allowing the transmitting end to acquire the input of the first AI model. Then, the transmitting end acquires the output of a second AI model. The transmitting end can acquire the output of the second AI model in the following two ways:

[0304] Method 1: The transmitting end includes a second AI model, which can be used to simulate AI-based demodulation performed by the receiving end. However, the transmitting end processes the output of the first AI model using the second AI model to obtain the output of the second AI model. In this way, the transmitting end can monitor the performance of AI-based modulation based on the comparison between the first unmodulated information and the output of the second AI model.

[0305] Method 2: After the sending end feeds back the output of the first AI model to the receiving end, the receiving end processes the output of the first AI model through the second AI model to obtain the output of the second AI model, and then the receiving end feeds back the output of the second AI model to the sending end.

[0306] It should be understood that, since the second AI model at the sending end and the second AI model at the receiving end may be different (such as different model structure parameters, different number of model training times, or different model training data), the output of the second AI model in Method 1 may be different from the output of the second AI model in Method 2.

[0307] For the receiving end, the receiving end comprises a second AI model. Wherein, the receiving end can obtain the input of the first AI model and the output of the second AI model in the following two ways:

[0308] Method 1: The sending end can feed back the input of the first AI model and the output of the first AI model to the receiving end. In this way, the receiving end processes the output of the first AI model through the second AI model to obtain the output of the second AI model, thereby obtaining the input of the first AI model and the output of the second AI model.

[0309] Method 2: The receiving end comprises a first AI model, and the first AI model of the receiving end can be used to simulate the sending end to perform AI-based modulation. In this way, the receiving end can obtain second unmodulated information, and then process the second unmodulated information through the first AI model to obtain the output of the first AI model. Wherein, the second unmodulated information can be information generated / obtained by the receiving end, and can be used as the input of the first AI model of the receiving end, so that the receiving end obtains the input of the first AI model. Then, the receiving end processes the output of the first AI model through the second AI model to obtain the output of the second AI model. In this way, the receiving end can monitor the performance of AI-based modulation according to the comparison result information between the second unmodulated information and the output of the second AI model.

[0310] It should be understood that the first unmodulated information and the second unmodulated information can be different, and the first AI model of the sending end and the first AI model of the receiving end can also be different (such as different model structure parameters, different model training times or different model training data, etc.). In summary, according to the comparison result information between the input of the first AI model and the output of the second AI model, the performance of AI-based modulation is monitored, and a specific implementation manner is as follows:

[0311] Obtaining unmodulated information;

[0312] Processing the unmodulated information through the first AI model to obtain the output of the first AI model, and processing the output of the first AI model through the second AI model to obtain the output of the second AI model;

[0313] Monitoring the performance of AI-based modulation according to the comparison result information between the unmodulated information and the output of the second AI model.

[0314] It should be noted that for the sending end, the unmodulated information here is the first unmodulated information. For the receiving end, the unmodulated information here is the second unmodulated information.

[0315] For example, for monitoring the performance of the AI-based modulation, one specific implementation can be: monitoring the performance of the AI-based modulation according to comparison result information between the performance index information and the preset threshold.

[0316] It should be noted that the performance index information can be the performance and index of the wireless communication system under the AI-based modulation under certain conditions, which is an evaluation index considering various factors (such as channel quality, signal quality, or data distribution of the model, etc.). The preset threshold can be a threshold specified by a standard protocol or a preset threshold.

[0317] Based on this, the sending end or the receiving end of the embodiment can monitor the performance of the AI-based modulation according to the comparison result information between the performance index information and the preset threshold.

[0318] Optionally, the performance index information includes at least one of the following: a signal to noise ratio (SNR) or a signal to interference plus noise ratio (SINR) corresponding to the AI-based modulation, a data distribution feature corresponding to the AI-based modulation, a bit error rate (BER) corresponding to the AI-based modulation, a block error rate (BLER) corresponding to the AI-based modulation, or a probability of a hybrid automatic repeat request (HARQ) negative acknowledgement (NACK) corresponding to the AI-based modulation.

[0319] It should be noted that the SNR corresponding to the AI-based modulation can be understood as the SNR or SINR corresponding to the data, signal or information obtained by the AI-based modulation. The SNR or SINR can be used to measure the signal quality under the AI-based modulation. Therefore, the embodiment can measure the size of the SNR or SINR according to the comparison result information between the SNR or SINR and the preset threshold.

[0320] For example, if the comparison result information indicates that the SNR or SINR is greater than or equal to the preset threshold, it means that the signal is stronger relative to the noise, and it is easier to correctly demodulate the signal, etc., indicating that the performance of the AI-based modulation is good. If the comparison result information indicates that the SNR or SINR is less than the preset threshold, it can cause information transmission errors or loss, indicating that the performance of the AI-based modulation is poor.

[0321] The data distribution feature corresponding to the AI-based modulation can be understood as a data distribution feature corresponding to an AI model used for AI-based modulation. The data distribution feature corresponding to the AI model is represented by a data distribution feature index. For example, the data distribution feature index includes data drift detection, or matching detection of data distribution features when the AI model is reasoning and data distribution features when the AI model is trained. Therefore, the embodiment can measure the data distribution feature index according to the comparison result information between the data distribution feature and the preset threshold.

[0322] For example, if the result of data drift detection is less than the preset threshold, or the matching degree of data distribution features when the AI model is reasoning and data distribution features when the AI model is trained is less than the preset threshold, the data distribution feature index is better, indicating that the performance of AI-based modulation is better. If the result of data drift detection is greater than or equal to the preset threshold, or the matching degree of data distribution features when the AI model is reasoning and data distribution features when the AI model is trained is greater than or equal to the preset threshold, the data distribution feature index is worse, indicating that the performance of AI-based modulation is worse.

[0323] The BER corresponding to the AI-based modulation can be understood as the BER corresponding to the data, signal or information obtained by AI-based modulation. The BER can be used to measure the signal transmission quality under AI-based modulation. Therefore, the embodiment can measure the size of the BER according to the comparison result information between the BER and the preset threshold.

[0324] For example, if the comparison result information indicates that the BER is greater than or equal to the preset threshold, it indicates that the system has poor anti-interference ability and low signal transmission quality, indicating that the performance of AI-based modulation is poor. If the comparison result information indicates that the BER is less than the preset threshold, it indicates that the system has strong anti-interference ability and high signal transmission quality, indicating that the performance of AI-based modulation is good.

[0325] The BLER corresponding to the AI-based modulation can be understood as the BLER corresponding to the data, signal or information obtained by AI-based modulation. The BLER can be used to measure the signal transmission quality under AI-based modulation. Therefore, the embodiment can measure the size of the BLER according to the comparison result information between the BLER and the preset threshold.

[0326] For example, if the comparison result information indicates that the BLER is greater than or equal to the preset threshold, it indicates that the system has poor anti-interference ability and low signal transmission quality, indicating that the performance of AI-based modulation is poor. If the comparison result information indicates that the BLER is less than the preset threshold, it indicates that the system has strong anti-interference ability and high signal transmission quality, indicating that the performance of AI-based modulation is good.

[0327] The probability of the AI-based modulation corresponding to the HARQ NACK can be understood as the probability of the data, signal or information obtained by the AI-based modulation corresponding to the HARQ NACK. The probability of the HARQ NACK can represent the probability of the signal transmission being unsuccessful under the AI-based modulation. Therefore, the embodiment can measure the size of the probability of the HARQ NACK according to the comparison result information between the probability of the HARQ NACK and the preset threshold.

[0328] For example, if the comparison result information indicates that the probability of the HARQ NACK is greater than or equal to the preset threshold, the higher the probability of the signal transmission being unsuccessful, the poorer the performance of the AI-based modulation. If the comparison result information indicates that the probability of the HARQ NACK is less than the preset threshold, the lower the probability of the signal transmission being unsuccessful, the better the performance of the AI-based modulation.

[0329] For example, for monitoring the performance of the AI-based modulation, a specific implementation manner can be: monitoring the performance of the AI-based modulation according to the comparison result information between the first repetition transmission information and the second repetition transmission information of the data channel, the first repetition transmission information being obtained by the AI-based modulation, and the second repetition transmission information being obtained by the non-AI-based modulation.

[0330] It should be noted that the first repetition transmission information can represent the related information corresponding to the repetition transmission of the data, signal or information obtained by the AI-based modulation. The second repetition transmission information can represent the related information corresponding to the repetition transmission of the data, signal or information obtained by the non-AI-based modulation.

[0331] Optionally, the first repetition transmission information comprises at least one of the following: data, signal obtained based on AI-based modulation, repetition transmission times corresponding to the data, signal obtained based on AI-based modulation, SNR or SINR corresponding to repetition transmission of the data, signal obtained based on AI-based modulation, BER corresponding to repetition transmission of the data, signal obtained based on AI-based modulation, BLER corresponding to repetition transmission of the data, signal obtained based on AI-based modulation, or probability of HARQ NACK corresponding to repetition transmission of the data, signal obtained based on AI-based modulation. Optionally, the second repetition transmission information comprises at least one of the following: data, signal obtained based on non-AI-based modulation, repetition transmission times corresponding to the data, signal obtained based on non-AI-based modulation, SNR or SINR corresponding to repetition transmission of the data, signal obtained based on non-AI-based modulation, BER corresponding to repetition transmission of the data, signal obtained based on non-AI-based modulation, BLER corresponding to repetition transmission of the data, signal obtained based on non-AI-based modulation, or probability of HARQ NACK corresponding to repetition transmission of the data, signal obtained based on non-AI-based modulation.

[0332] It should be noted that if the first repetition transmission information comprises repetition transmission times corresponding to data, signal obtained based on AI-based modulation (referred to as "first repetition transmission times"), and the second repetition transmission information comprises repetition transmission times corresponding to data, signal obtained based on non-AI-based modulation (referred to as "second repetition transmission times"), the embodiment can monitor the performance of AI-based modulation according to the comparison result information between the first repetition transmission times and the second repetition transmission times. For example, when the comparison result information is that the first repetition transmission times are greater than the second repetition transmission times, it indicates that the data, signal obtained based on AI-based modulation needs to be transmitted more times, which indicates that the performance of AI-based modulation is worse than that of non-AI-based modulation. When the comparison result information is that the first repetition transmission times are less than the second repetition transmission times, it indicates that the performance of AI-based modulation is better than that of non-AI-based modulation.

[0333] If the first repeated transmission information includes the SNR or SINR (referred to as "first SNR / first SINR") corresponding to the repeated transmission of the data, signal or information obtained based on the AI-based modulation, and the second repeated transmission information includes the SNR or SINR (referred to as "second SNR or second SINR") corresponding to the repeated transmission of the data, signal or information obtained based on the AI-based modulation, the embodiment can monitor the performance of the AI-based modulation according to the comparison result information between the first SNR / first SINR and the second SNR / second SINR. For example, when the comparison result information is that the first SNR / first SINR is greater than the second SNR / second SINR, it indicates that the data, signal or information obtained based on the AI-based modulation is more easily correctly demodulated, etc., and the performance of the AI-based modulation is better than that of the non-AI-based modulation. When the comparison result information is that the first SNR / first SINR is less than the second SNR / second SINR, it indicates that the performance of the AI-based modulation is worse than that of the non-AI-based modulation.

[0334] If the first repeated transmission information includes the BER (referred to as "first BER") corresponding to the repeated transmission of the data, signal or information obtained based on the AI-based modulation, and the second repeated transmission information includes the BER (referred to as "second BER") corresponding to the repeated transmission of the data, signal or information obtained based on the AI-based modulation, the embodiment can monitor the performance of the AI-based modulation according to the comparison result information between the first BER and the second BER. For example, when the comparison result information is that the first BER is greater than the second BER, it indicates that the transmission quality of the data, signal or information obtained based on the AI-based modulation is low, and the performance of the AI-based modulation is worse than that of the non-AI-based modulation. When the comparison result information is that the first BER is less than the second BER, it indicates that the performance of the AI-based modulation is better than that of the non-AI-based modulation.

[0335] If the first repeated transmission information includes the BLER (referred to as "first BLER") corresponding to the repeated transmission of the data, signal or information obtained based on the AI-based modulation, and the second repeated transmission information includes the BLER (referred to as "second BLER") corresponding to the repeated transmission of the data, signal or information obtained based on the AI-based modulation, the embodiment can monitor the performance of the AI-based modulation according to the comparison result information between the first BLER and the second BLER. For example, when the comparison result information is that the first BLER is greater than the second BLER, it indicates that the transmission quality of the data, signal or information obtained based on the AI-based modulation is low, and the performance of the AI-based modulation is worse than that of the non-AI-based modulation. When the comparison result information is that the first BLER is less than the second BLER, it indicates that the performance of the AI-based modulation is better than that of the non-AI-based modulation.

[0336] If the first repeated transmission information includes the probability of a HARQ NACK corresponding to the repeated transmission of the data, signal or information obtained based on AI-based modulation (referred to as "the first probability of a HARQ NACK"), and the second repeated transmission information includes the probability of a HARQ NACK corresponding to the repeated transmission of the data, signal or information obtained based on AI-based modulation (referred to as "the second probability of a HARQ NACK"), the embodiment can monitor the performance of AI-based modulation according to the comparison result information between the first probability of a HARQ NACK and the second probability of a HARQ NACK. For example, when the comparison result information is that the first probability of a HARQ NACK is greater than the second probability of a HARQ NACK, it indicates that the probability of unsuccessful transmission of the data, signal or information obtained based on AI-based modulation is high, and that the performance of AI-based modulation is worse than that of non-AI-based modulation. When the comparison result information is that the first probability of a HARQ NACK is less than the second probability of a HARQ NACK, it indicates that the performance of AI-based modulation is better than that of non-AI-based modulation.

[0337] A communication device of the embodiment is described below as an example.

[0338] The above describes the scheme of the embodiment of the application mainly from the perspective of the method. The functional units of a communication device of the embodiment are described below as an example. It can be understood that, in order to implement the above functions, the terminal device comprises a hardware structure and / or software module for executing each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of the examples described in the embodiments disclosed herein, the embodiments can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is implemented in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments.

[0339] The embodiment of the application can divide the functional units of the terminal device according to the above method examples. For example, each functional unit can be divided corresponding to each function, or two or more functions can be integrated in one processing unit. The integrated unit can be implemented in the form of hardware or in the form of a software program module. It should be noted that the division of units in the embodiments of the application is illustrative, and is only a logical functional division, and there can be another division method when actually implemented.

[0340] In the case of integrated units, FIG. 7 is a functional unit composition block diagram of a communication device of the embodiment of the application. The communication device 700 comprises a receiving unit 701.

[0341] Optionally, the receiving unit 701 can be a module unit for receiving and processing signals, information, etc., and no specific limitation is made thereto.

[0342] Optionally, the communication apparatus 700 can further include a sending unit. The sending unit can be a module unit for sending and processing signals, information, etc., and no specific limitation is made thereto.

[0343] Optionally, the communication apparatus 700 can further include a storage unit for storing computer program codes or instructions executed by the communication apparatus 700. The storage unit can be a memory.

[0344] Optionally, the communication apparatus 700 can be a chip or a chip module.

[0345] Optionally, the receiving unit 701 can be integrated in a communication unit. The communication unit can be a communication interface, a transceiver, a transceiving circuit, etc.

[0346] Optionally, the receiving unit 701 can be integrated in a processing unit.

[0347] It is to be noted that the processing unit can be a processor or a controller, for example, a baseband processor, a baseband chip, a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a transistor logic device, a hardware component, or any combination thereof. The processing unit can implement or execute various exemplary logical blocks, modules, and circuits described in connection with the disclosure of the present embodiment. The processing unit can also be a combination implementing computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0348] Optionally, the communication apparatus 700 is configured to perform any step of the above method embodiments performed by a terminal device, a chip, or a chip module, etc.

[0349] In specific implementation, the receiving unit 701 is configured to perform any step of the above method embodiments, and when performing actions such as sending, other units can be optionally invoked to complete the corresponding operations. The following will be described in detail.

[0350] The receiving unit 701 is configured to receive first information, the first information being used to indicate modulation enabling information corresponding to data and / or a modulation mode corresponding to the data.

[0351] The modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode.

[0352] It can be seen that, when AI technology is applied to the processing of channels, signals or information of a wireless communication system, the present application considers the combination of AI and a modulation mode. In the process of combining AI and the modulation mode, the present application introduces the first information. If the first information is used to indicate the modulation enabling information corresponding to the data, it means that the first information is used to indicate AI-based modulation or non-AI-based modulation, so that the network indicates or network configures the enabling or disabling of AI-based modulation through the first information, so as to flexibly configure whether AI technology needs to be used for the modulation of data according to communication requirements and the like.

[0353] If the first information is used to indicate the modulation mode corresponding to the data, it means that the first information is used to indicate an AI modulation mode or a non-AI modulation mode, so that the network indicates or network configures the AI modulation mode through the first information, so as to flexibly configure whether AI technology needs to be used for the modulation mode of data according to communication requirements and the like.

[0354] If the first information is used to indicate the modulation enabling information corresponding to the data and the modulation mode corresponding to the data, it means that the first information is used to indicate AI-based modulation or non-AI-based modulation and an AI modulation mode or a non-AI modulation mode, so that the network indicates or network configures the enabling or disabling of AI-based modulation and the network indicates or network configures the AI modulation mode through the first information, so as to flexibly configure whether AI technology needs to be used for the modulation of data and the modulation mode of data according to communication requirements and the like.

[0355] It should be noted that the specific implementation of each operation in the embodiment of FIG. 7 can be found in the description of the method embodiments shown above, and will not be described in detail here.

[0356] In some possible examples, the first information is carried by RRC signaling or MAC signaling. That is, the RRC signaling or the MAC signaling is used to indicate AI-based modulation or non-AI-based modulation.

[0357] In some possible examples, the first information is carried by DCI, and the first information is a field in the DCI, the field being used to indicate the modulation enabling information corresponding to the data and / or the modulation mode corresponding to the data. That is, the DCI is used to indicate AI-based modulation or non-AI-based modulation.

[0358] In some possible examples, the AI-based modulation corresponds to an AI modulation manner, and the non-AI-based modulation corresponds to a non-AI modulation manner.

[0359] In some possible examples, the data includes one or more code words, and modulation enabling information corresponding to each code word in the plurality of code words is the same or different.

[0360] In some possible examples, the plurality of code words includes a first code word and a second code word, and the first information includes a first field and a second field in the DCI, the first field being used to indicate a modulation manner corresponding to the first code word, and the second field being used to indicate a modulation manner corresponding to the second code word.

[0361] In some possible examples, interpretation or parsing of the second field depends on the first field.

[0362] The following describes another communication apparatus of the embodiment.

[0363] The above describes the scheme of the embodiment of the present application mainly from the perspective of a method, and the following describes functional units of another communication apparatus of the embodiment. It can be understood that, in order to implement the above functions, the network device contains hardware structures and / or software modules corresponding to the functions. It can be easily realized by those skilled in the art that, in combination with the units and algorithm steps of the examples described in the embodiments disclosed in the present text, the embodiments can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is implemented in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical scheme. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the embodiments.

[0364] The embodiments of the present application can divide the network device into functional units according to the above method examples. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in the form of hardware or in the form of a software program module. It should be noted that the division of units in the embodiments of the present application is illustrative, and is only a logical function division, and another division manner can be used in actual implementation.

[0365] In the case of integrated units, FIG. 8 is a functional unit composition block diagram of another communication apparatus of the embodiments of the present application. The communication apparatus 800 includes a sending unit 801.

[0366] Optionally, the sending unit 801 can be a module unit for sending processing of signals, information, and the like, and no specific limitation is made in this regard.

[0367] Optionally, the communication apparatus 800 can further include a receiving unit. The receiving unit can be a module unit for receiving and processing signals, information, etc., and no specific limitation is made thereto.

[0368] Optionally, the communication apparatus 800 can further include a storage unit for storing computer program codes or instructions executed by the communication apparatus 800. The storage unit can be a memory.

[0369] Optionally, the communication apparatus 800 can be a chip or a chip module.

[0370] Optionally, the sending unit 801 can be integrated in a communication unit. The communication unit can be a communication interface, a transceiver, a transceiving circuit, etc.

[0371] Optionally, the communication apparatus 800 can further include a processing unit.

[0372] It is to be noted that the processing unit can be a processor or a controller, such as a baseband processor, a baseband chip, a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a transistor logic device, a hardware component, or any combination thereof. The processing unit can implement or execute various exemplary logical blocks, modules and circuits described in conjunction with the disclosure of the present embodiment. The processing unit can also be a combination implementing computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0373] Optionally, the communication apparatus 800 is configured to perform any step of the above method embodiments performed by a chip, a chip module, or a network device, etc.

[0374] In a specific implementation, the sending unit 801 is configured to perform any step of the above method embodiments, and when performing an action such as sending, the sending unit 801 can optionally call other units to complete the corresponding operation. Details are described below.

[0375] The sending unit 801 is configured to send first information, the first information being used to indicate modulation enabling information corresponding to data and / or a modulation mode corresponding to the data.

[0376] The modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode.

[0377] It can be seen that when AI technology is applied to the processing of channels, information or signals of a wireless communication system, the present application considers the combination of AI and modulation modes. In the combination of AI and modulation modes, the present application introduces first information. If the first information is used to indicate the modulation enabling information corresponding to the data, it means that the first information is used to indicate AI-based modulation or non-AI-based modulation, so that the network indicates or network configures the enabling or disabling of AI-based modulation through the first information, so as to flexibly configure whether AI technology needs to be used for the modulation of data according to communication requirements and the like.

[0378] If the first information is used for the modulation mode corresponding to the data, it means that the first information is used to indicate an AI modulation mode or a non-AI modulation mode, so that the network indicates or network configures the AI modulation mode through the first information, so as to flexibly configure whether AI technology needs to be used for the modulation mode of data according to communication requirements and the like.

[0379] If the first information is used to indicate the modulation enabling information corresponding to the data and the modulation mode corresponding to the data, it means that the first information is used to indicate AI-based modulation or non-AI-based modulation and an AI modulation mode or a non-AI modulation mode, so that the network indicates or network configures the enabling or disabling of AI-based modulation and the network indicates or network configures the AI modulation mode through the first information, so as to flexibly configure whether AI technology needs to be used for the modulation of data and the modulation mode of data according to communication requirements and the like.

[0380] In some possible examples, the first information is carried by RRC signaling or MAC signaling. That is, the RRC signaling or the MAC signaling is used to indicate AI-based modulation or non-AI-based modulation.

[0381] In some possible examples, the first information is carried by DCI, and the first information is a field in the DCI, which is used to indicate the modulation enabling information corresponding to the data and / or the modulation mode corresponding to the data. That is, the DCI is used to indicate AI-based modulation or non-AI-based modulation.

[0382] In some possible examples, the AI-based modulation corresponds to an AI modulation mode, and the non-AI-based modulation corresponds to a non-AI modulation mode.

[0383] In some possible examples, the data includes one or more code words, and the modulation enabling information corresponding to each code word in the plurality of code words is the same or different.

[0384] In some possible examples, the plurality of codewords includes a first codeword and a second codeword; and the first information includes a first field and a second field in the DCI, the first field being used to indicate a modulation mode corresponding to the first codeword, and the second field being used to indicate a modulation mode corresponding to the second codeword.

[0385] In some possible examples, the interpretation or parsing of the second field depends on the first field.

[0386] Another communication apparatus of the present embodiment is described below.

[0387] In the case of an integrated unit, FIG. 9 is a functional unit constituent diagram of another communication apparatus of the present embodiment. The communication apparatus 900 includes a monitoring unit 901.

[0388] Optionally, the monitoring unit 901 can be a module unit for monitoring, and no specific limitation is made thereto.

[0389] Optionally, the communication apparatus 900 can further include a communication unit. The communication unit can be a module unit for performing sending or receiving processing on signals, information, and the like, and no specific limitation is made thereto.

[0390] For example, the communication unit can be a communication interface, a transceiver, a transceiving circuit, or the like.

[0391] Optionally, the communication apparatus 900 can further include a storage unit for storing computer program codes or instructions executed by the communication apparatus 900. The storage unit can be a memory.

[0392] Optionally, the communication apparatus 900 can be a chip or a chip module.

[0393] Optionally, the communication apparatus 900 can further include a processing unit.

[0394] It should be noted that the processing unit can be a processor or a controller, for example, can be a baseband processor, a baseband chip, a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute various exemplary logical blocks, modules and circuits described in combination with the disclosure of the embodiment. The processing unit can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of DSP and microprocessor, etc.

[0395] Optionally, the communication apparatus 900 is configured to perform any step described in the above method embodiments by a chip, a chip module, a network device, or a terminal device.

[0396] In specific implementation, the monitoring unit 901 is configured to perform any step described in the above method embodiments, and when performing an action such as sending, the other units can be optionally called to complete the corresponding operation. The following will be described in detail.

[0397] The monitoring unit 901 is configured to monitor the performance of the AI-based modulation.

[0398] It can be seen that in the combination of AI technology and modulation mode, AI technology can bring certain advantages, such as realizing adaptive modulation, improving performance, etc. However, AI technology may also bring certain disadvantages, such as the processing process of the AI model occupying a large amount of computing power or the inference result of the AI model having a large error with the actual result. Therefore, the terminal device or network device of the embodiment needs to monitor the performance of the AI-based modulation, so as to timely make corresponding adjustment according to the performance monitoring result, such as adjusting the parameters, architecture or training data of the AI model, or using non-AI-based modulation, etc., so as to avoid the deficiencies brought by the AI model as much as possible and improve the inference result of the AI model.

[0399] It should be noted that the specific implementation of each operation in the embodiment of FIG. 9 can be described in detail in the above-described method embodiments, and will not be described in detail here.

[0400] In some possible examples, in terms of monitoring the performance of the AI-based modulation, the monitoring unit 901 is configured to:

[0401] According to comparison result information between input of the first AI model and output of the second AI model, the performance of the AI-based modulation is monitored, the first AI model being used for the AI-based modulation, and the second AI model being used for the AI-based demodulation.

[0402] In some possible examples, in the aspect of monitoring the performance of the AI-based modulation according to comparison result information between input of the first AI model and output of the second AI model, the monitoring unit 901 is configured to:

[0403] obtain unmodulated information;

[0404] perform processing on the unmodulated information by the first AI model to obtain output of the first AI model, and perform processing on the output of the first AI model by the second AI model to obtain output of the second AI model;

[0405] monitor the performance of the AI-based modulation according to comparison result information between the unmodulated information and the output of the second AI model.

[0406] In some possible examples, in the aspect of monitoring the performance of the AI-based modulation, the monitoring unit 901 is configured to:

[0407] monitor the performance of the AI-based modulation according to comparison result information between the performance indicator information and the preset threshold.

[0408] In some possible examples, the performance indicator information includes at least one of the following:

[0409] a signal-to-noise ratio or a signal-to-interference-and-noise ratio corresponding to the AI-based modulation, a data distribution feature corresponding to the AI-based modulation, a bit error rate corresponding to the AI-based modulation, a block error rate corresponding to the AI-based modulation, or a probability of a HARQ NACK corresponding to the AI-based modulation.

[0410] In some possible examples, in the aspect of monitoring the performance of the AI-based modulation, the monitoring unit 901 is configured to:

[0411] monitor the performance of the AI-based modulation according to comparison result information between first repeated transmission information of the data channel and second repeated transmission information of the data channel, the first repeated transmission information being obtained based on the AI-based modulation, and the second repeated transmission information being obtained based on non-AI-based modulation.

[0412] The structure of a terminal device in this embodiment is described below.

[0413] Referring to FIG. 10, FIG. 10 is a structural schematic diagram of a terminal device in this embodiment. The terminal device 1000 can include a processor 1010, a memory 1020, and a communication bus used to connect the processor 1010 and the memory 1020.

[0414] Optionally, the memory 1020 includes, but is not limited to, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or a compact disc read-only memory (CD-ROM). The memory 1020 is configured to store program codes executed by the terminal device 1000 and transmitted data.

[0415] Optionally, the terminal device 1000 further includes a communication interface configured to receive and send data.

[0416] Optionally, the terminal device 1000 can be the first terminal device.

[0417] Optionally, the processor 1010 can be one or more CPUs. When the processor 1010 is a CPU, the CPU can be a single-core CPU or a multi-core CPU.

[0418] Optionally, the processor 1010 can be a baseband chip, a chip, a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.

[0419] Optionally, the processor 1010 in the terminal device 1000 is configured to execute the computer program or instructions 1021 stored in the memory 1020 to perform the following operations:

[0420] receive first information, the first information being used to indicate modulation enabling information corresponding to data and / or a modulation mode corresponding to the data;

[0421] The modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode.

[0422] It can be seen that when AI technology is applied to the processing of channels, information, or signals in a wireless communication system, the present application considers the combination of AI and modulation modes. In the combination of AI and modulation modes, the present application introduces first information. If the first information is used to indicate modulation enabling information corresponding to data, it means that the first information is used to indicate AI-based modulation or non-AI-based modulation, so that the network indicates or network configures to enable or disable AI-based modulation through the first information, so as to flexibly configure whether AI technology needs to be used for the modulation of data according to communication requirements, etc.

[0423] If the first information is used for indicating the modulation mode corresponding to the data, it means that the first information is used for indicating the AI modulation mode or the non-AI modulation mode, so as to realize network indication or network configuration of the AI modulation mode through the first information, so as to flexibly configure whether the modulation mode of the data needs to adopt the AI technology according to the communication demand and the like.

[0424] If the first information is used for indicating the modulation mode corresponding to the data, it means that the first information is used for indicating the AI modulation mode or the non-AI modulation mode, so as to realize network indication or network configuration of the AI modulation mode through the first information, so as to flexibly configure whether the modulation mode of the data needs to adopt the AI technology according to the communication demand and the like.

[0425] Optionally, the processor 1010 in the terminal device 1000 is configured to execute the computer program or the instruction 1021 stored in the memory 1020, and perform the following operations:

[0426] Monitoring the performance of the AI-based modulation.

[0427] It can be seen that in the combination of the AI technology and the modulation mode, the AI technology can bring certain advantages, such as realizing adaptive modulation, improving performance, and the like. However, the AI technology can also bring certain disadvantages, such as the processing process of the AI model occupying large computing power or the inference result of the AI model having a large error with the actual result, and the like. Therefore, the terminal device or the network device of the embodiment needs to monitor the performance of the AI-based modulation, so as to timely make corresponding adjustment according to the performance monitoring result, such as adjusting the parameters, architecture or training data of the AI model, or adopting the non-AI-based modulation, and the like, so as to avoid the deficiencies brought by the AI model as much as possible and improve the inference result of the AI model.

[0428] It should be noted that the specific implementation of each operation can adopt the corresponding description of the method embodiment shown above, and the terminal device 1000 can be configured to execute the method embodiment described above, and details are not described herein.

[0429] The structure of a network device of the embodiment will be described below.

[0430] Please refer to FIG. 11, which is a structure schematic diagram of a network device of an embodiment of the present application. The network device 1100 includes a processor 1110, a memory 1120 and a communication bus for connecting the processor 1110 and the memory 1120.

[0431] Optionally, the memory 1120 includes, but is not limited to, a RAM, a ROM, an EPROM, or a CD-ROM, and is configured to store relevant instructions and data.

[0432] Optionally, the network device 1100 further includes a communication interface configured to receive and send data.

[0433] Optionally, the processor 1110 can be one or more CPUs, and in the case that the processor 1110 is a CPU, the CPU can be a single-core CPU or a multi-core CPU.

[0434] Optionally, the processor 1110 can be a baseband chip, a chip, a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof.

[0435] Optionally, the processor 1110 in the network device 1100 is configured to execute the computer program or instructions 1121 stored in the memory 1120 to perform the following operations:

[0436] send first information, the first information being used to indicate modulation enabling information corresponding to data and / or a modulation mode corresponding to the data;

[0437] The modulation enabling information corresponding to the data is AI-based modulation or non-AI-based modulation, and the modulation mode corresponding to the data is an AI modulation mode or a non-AI modulation mode.

[0438] It can be seen that when AI technology is applied to the processing of channels, information, or signals in a wireless communication system, the present application considers the combination of AI and modulation modes. In the combination of AI and modulation modes, the present application introduces first information. If the first information is used to indicate modulation enabling information corresponding to data, it means that the first information is used to indicate AI-based modulation or non-AI-based modulation, so that the network indicates or network configures to enable or disable AI-based modulation through the first information, so as to flexibly configure whether AI technology needs to be used for data modulation according to communication requirements, etc.

[0439] If the first information is used for the modulation mode corresponding to the data, it means that the first information is used to indicate an AI modulation mode or a non-AI modulation mode, so that the network indicates or network configures an AI modulation mode through the first information, so as to flexibly configure whether AI technology needs to be used for the modulation mode of the data according to communication requirements, etc.

[0440] If the first information is used to indicate modulation enabling information corresponding to the data and modulation modes corresponding to the data, it indicates that the first information is used to indicate AI-based modulation or non-AI-based modulation and AI modulation modes or non-AI modulation modes, so that network indication or network configuration enables or disables AI-based modulation and network indication or network configuration AI modulation modes are realized through the first information, so as to flexibly configure the modulation of the data and whether the modulation mode of the data needs to adopt AI technology according to communication needs and the like.

[0441] Optionally, the processor 1110 in the network device 1100 is configured to execute the computer program or instructions 1121 stored in the memory 1120 to perform the following operations:

[0442] Monitoring the performance of AI-based modulation.

[0443] It can be seen that in the combination of AI technology and modulation modes, AI technology can bring certain advantages, such as realizing adaptive modulation and improving performance. However, AI technology may also have certain disadvantages, such as the processing process of the AI model occupying a large amount of computing power or the inference result of the AI model having a large error from the actual result. Therefore, the terminal device or the network device of the embodiment needs to monitor the performance of AI-based modulation, so as to timely make corresponding adjustments according to the performance monitoring result, such as adjusting the parameters, architecture or training data of the AI model, or using non-AI-based modulation, so as to avoid the deficiencies brought by the AI model as much as possible and improve the inference result of the AI model.

[0444] It should be noted that the specific implementation of each operation can be implemented by the corresponding description of the method embodiment described above, and the network device 1100 can be used to execute the method embodiment described above, and details are not repeated.

[0445] The other related contents of the embodiment are exemplarily illustrated as follows.

[0446] Optionally, the above-mentioned method embodiments can be applied to a terminal device or in a terminal device. That is, the execution subject of the above-mentioned method embodiments can be a terminal device, a chip, a chip module or a module, etc., and no specific limitation is made.

[0447] Optionally, the above-mentioned method embodiments can be applied to a network device or in a network device. That is, the execution subject of the above-mentioned method embodiments can be a network device, a chip, a chip module or a module, etc., and no specific limitation is made.

[0448] The embodiment of the present application further provides a chip, comprising a processor, a memory and a computer program or instructions stored in the memory, wherein the processor executes the computer program or instructions to realize the steps described in the method embodiment.

[0449] The embodiment of the present application further provides a chip module, comprising a transceiver component and a chip, the chip comprising a processor, a memory and a computer program or instructions stored in the memory, wherein the processor executes the computer program or instructions to realize the steps described in the method embodiment.

[0450] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program or instructions, and the computer program or instructions are executed to realize the steps described in the method embodiment.

[0451] The embodiment of the present application further provides a computer program product, comprising a computer program or instructions, and the computer program or instructions are executed to realize the steps described in the method embodiment.

[0452] The embodiment of the present application further provides a communication system, comprising the terminal device and the network device.

[0453] It should be noted that, for the above-mentioned various embodiments, in order to simply describe, they are all expressed as a series of action combinations. Those skilled in the art should know that the present application is not limited to the action sequence described, because some steps in the embodiment of the present application can be performed in other sequences or simultaneously. In addition, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions, steps, modules or units involved are not necessarily required in the embodiment of the present application. In the above embodiments, the description of each embodiment of the present application has its own emphasis, and the part not described in detail in the embodiment can be referred to the related description of other embodiments.

[0454] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can reside in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, a removable media, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor. The processor and the storage medium can reside in an ASIC. Alternatively, the processor and the storage medium can be located in a terminal device or an access device. The processor and the storage medium can also be located in any other

[0455] Those skilled in the art should clearly understand that, in one or more examples described above, the functions described in the embodiments of the present application can be implemented entirely or partially by software, hardware, firmware, or any combination thereof. When implemented by software, the functions can be implemented in the form of a computer program product entirely or partially. The computer program product includes one or more computer instructions. When loaded and executed on a computer, the computer instructions entirely or partially generate the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer instructions can be transferred from one website, computer, server, or data center to another via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media sets. The available media can be a magnetic medium (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a digital video disc (DVD)), or a semiconductor medium (such as a solid state disk (SSD)), etc.

[0456] The various modules / units included in the various devices and products described in the above embodiments can be software modules / units or hardware modules / units, or can be partially software modules / units and partially hardware modules / units. For example, for the various devices and products applied to or integrated into chips, the various modules / units included therein can all be implemented in the form of hardware such as circuits, or at least some of the modules / units can be implemented in the form of software programs running on a processor integrated in the chip, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits.

[0457] The above description is merely intended for the purpose of illustrating the principles of the embodiments of the present application, the technical solutions and advantages thereof, and should not be used to limit the scope of protection of the embodiments of the present application. Any modification, equivalent replacement, improvement, etc. made on the basis of the technical solutions of the embodiments of the present application shall be included in the scope of protection of the embodiments of the present application.

Claims

1. A communication method, characterized in that, include: Receive first information, the first information being used to indicate modulation enable information corresponding to the data and / or modulation mode corresponding to the data; The modulation enable information corresponding to the data is either AI-based modulation or non-AI-based modulation, and the modulation method corresponding to the data is either AI modulation or non-AI modulation.

2. The method according to claim 1, characterized in that, The first information is carried by Radio Resource Control (RRC) signaling or Media Access Control (MAC) signaling.

3. The method according to claim 1, characterized in that, The first information is carried by downlink control information (DCI), and the first information is a field in the DCI. The field is used to indicate the modulation enable information and / or the modulation mode corresponding to the data.

4. The method according to claim 1, characterized in that, The AI-based modulation corresponds to the AI ​​modulation method, and the non-AI-based modulation corresponds to the non-AI modulation method.

5. The method according to claim 1, characterized in that, The data includes one or more codewords, and the modulation enable information corresponding to each of the multiple codewords may be the same or different.

6. The method according to claim 5, characterized in that, The plurality of codewords includes a first codeword and a second codeword; The first information includes a first field and a second field in DCI, wherein the first field is used to indicate the modulation scheme corresponding to the first codeword, and the second field is used to indicate the modulation scheme corresponding to the second codeword.

7. The method according to claim 6, characterized in that, The interpretation or parsing of the second field depends on the first field.

8. A communication method, characterized in that, include: Send a first message, which is used to indicate the modulation enable information corresponding to the data and / or the modulation method corresponding to the data; The modulation enable information corresponding to the data is either AI-based modulation or non-AI-based modulation, and the modulation method corresponding to the data is either AI modulation or non-AI modulation.

9. The method according to claim 8, characterized in that, The first information is carried by Radio Resource Control (RRC) signaling or Media Access Control (MAC) signaling.

10. The method according to claim 8, characterized in that, The first information is carried by downlink control information (DCI), and the first information is a field in the DCI. The field is used to indicate the modulation enable information and / or the modulation mode corresponding to the data.

11. The method according to claim 8, characterized in that, The AI-based modulation corresponds to the AI ​​modulation method, and the non-AI-based modulation corresponds to the non-AI modulation method.

12. The method according to claim 8, characterized in that, The data includes one or more codewords, and the modulation enable information corresponding to each of the multiple codewords may be the same or different.

13. The method according to claim 11, characterized in that, The plurality of codewords includes a first codeword and a second codeword; The first information includes a first field and a second field in DCI, wherein the first field is used to indicate the modulation scheme corresponding to the first codeword, and the second field is used to indicate the modulation scheme corresponding to the second codeword.

14. The method according to claim 13, characterized in that, The interpretation or parsing of the second field depends on the first field.

15. A communication method, characterized in that, include: Monitor the performance of modulation based on artificial intelligence (AI).

16. The method according to claim 15, characterized in that, The monitoring of AI-based modulation performance includes: The performance of AI-based modulation is monitored based on the comparison results between the input of the first AI model and the output of the second AI model, where the first AI model is used for AI-based modulation and the second AI model is used for AI-based demodulation.

17. The method according to claim 16, characterized in that, The monitoring of AI-based modulation performance based on the comparison results between the input of the first AI model and the output of the second AI model includes: Acquire unmodulated information; The unmodulated information is processed by a first AI model to obtain the output of the first AI model, and the output of the first AI model is processed by a second AI model to obtain the output of the second AI model; The performance of AI-based modulation is monitored based on the comparison results between the unmodulated information and the output of the second AI model.

18. The method according to claim 15, characterized in that, The monitoring of AI-based modulation performance includes: The performance of AI-based modulation is monitored based on the comparison results between performance metrics and preset thresholds.

19. The method according to claim 18, characterized in that, The performance metrics information includes at least one of the following: The signal-to-noise ratio or signal-to-interference-plus-noise ratio corresponding to AI-based modulation, the data distribution characteristics corresponding to AI-based modulation, the bit error rate corresponding to AI-based modulation, the block error rate corresponding to AI-based modulation, or the probability of HARQ negative confirmation (NACK) corresponding to AI-based modulation.

20. The method according to claim 15, characterized in that, The monitoring of AI-based modulation performance includes: The performance of AI-based modulation is monitored based on the comparison results between the first and second repeated transmission information of the data channel. The first repeated transmission information is obtained based on AI-based modulation, and the second repeated transmission information is obtained based on non-AI-based modulation.

21. A communication device, characterized in that, include: A transmitting unit is configured to transmit first information, wherein the first information is used to indicate modulation enable information corresponding to data and / or modulation mode corresponding to the data; The modulation enable information corresponding to the data is either AI-based modulation or non-AI-based modulation, and the modulation method corresponding to the data is either AI modulation or non-AI modulation.

22. The apparatus according to claim 21, characterized in that, The first information is carried by Radio Resource Control (RRC) signaling or Media Access Control (MAC) signaling.

23. The apparatus according to claim 21, characterized in that, The first information is carried by downlink control information (DCI), and the first information is a field in the DCI. The field is used to indicate the modulation enable information and / or the modulation mode corresponding to the data.

24. The apparatus according to claim 21, characterized in that, The AI-based modulation corresponds to the AI ​​modulation method, and the non-AI-based modulation corresponds to the non-AI modulation method.

25. The apparatus according to claim 21, characterized in that, The data includes one or more codewords, and the modulation enable information corresponding to each of the multiple codewords may be the same or different.

26. The apparatus according to claim 25, characterized in that, The plurality of codewords includes a first codeword and a second codeword; The first information includes a first field and a second field in DCI, wherein the first field is used to indicate the modulation scheme corresponding to the first codeword, and the second field is used to indicate the modulation scheme corresponding to the second codeword.

27. The apparatus according to claim 26, characterized in that, The interpretation or parsing of the second field depends on the first field.

28. A communication device, characterized in that, include: The receiving unit is configured to receive first information, wherein the first information is used to indicate modulation enable information corresponding to the data and / or modulation mode corresponding to the data; The modulation enable information corresponding to the data is either AI-based modulation or non-AI-based modulation, and the modulation method corresponding to the data is either AI modulation or non-AI modulation.

29. The apparatus according to claim 28, characterized in that, The first information is carried by Radio Resource Control (RRC) signaling or Media Access Control (MAC) signaling.

30. The apparatus according to claim 28, characterized in that, The first information is carried by downlink control information (DCI), and the first information is a field in the DCI. The field is used to indicate the modulation enable information and / or the modulation mode corresponding to the data.

31. The apparatus according to claim 28, characterized in that, The AI-based modulation corresponds to the AI ​​modulation method, and the non-AI-based modulation corresponds to the non-AI modulation method.

32. The apparatus according to claim 28, characterized in that, The data includes one or more codewords, and the modulation enable information corresponding to each of the multiple codewords may be the same or different.

33. The apparatus according to claim 31, characterized in that, The plurality of codewords includes a first codeword and a second codeword; The first information includes a first field and a second field in DCI, wherein the first field is used to indicate the modulation scheme corresponding to the first codeword, and the second field is used to indicate the modulation scheme corresponding to the second codeword.

34. The apparatus according to claim 33, characterized in that, The interpretation or parsing of the second field depends on the first field.

35. A communication device, characterized in that, include: The monitoring unit is used to monitor the performance of AI-based modulation.

36. The apparatus according to claim 35, characterized in that, Regarding the monitoring of the performance of AI-based modulation, the monitoring unit is used to: The performance of AI-based modulation is monitored based on the comparison results between the input of the first AI model and the output of the second AI model, where the first AI model is used for AI-based modulation and the second AI model is used for AI-based demodulation.

37. The apparatus according to claim 36, characterized in that, In monitoring the performance of AI-based modulation based on the comparison results between the input of the first AI model and the output of the second AI model, the monitoring unit is used to: Acquire unmodulated information; The unmodulated information is processed by a first AI model to obtain the output of the first AI model, and the output of the first AI model is processed by a second AI model to obtain the output of the second AI model; The performance of AI-based modulation is monitored based on the comparison results between the unmodulated information and the output of the second AI model.

38. The apparatus according to claim 35, characterized in that, Regarding the monitoring of the performance of AI-based modulation, the monitoring unit is used to: The performance of AI-based modulation is monitored based on the comparison results between performance metrics and preset thresholds.

39. The apparatus according to claim 38, characterized in that, The performance metrics information includes at least one of the following: The signal-to-noise ratio or signal-to-interference-plus-noise ratio corresponding to AI-based modulation, the data distribution characteristics corresponding to AI-based modulation, the bit error rate corresponding to AI-based modulation, the block error rate corresponding to AI-based modulation, or the probability of HARQ negative confirmation (NACK) corresponding to AI-based modulation.

40. The apparatus according to claim 35, characterized in that, Regarding the monitoring of the performance of AI-based modulation, the monitoring unit is used to: The performance of AI-based modulation is monitored based on the comparison results between the first and second repeated transmission information of the data channel. The first repeated transmission information is obtained based on AI-based modulation, and the second repeated transmission information is obtained based on non-AI-based modulation.

41. A terminal device, comprising a processor, a memory, and a computer program or instructions stored in the memory, characterized in that, The processor executes the computer program or instructions to implement the steps of the method according to any one of claims 1-7 or 15-20.

42. A network device, comprising a processor, a memory, and a computer program or instructions stored in the memory, characterized in that, The processor executes the computer program or instructions to implement the steps of the method according to any one of claims 8-14 or 15-20.

43. A chip, comprising a processor and a communication interface, characterized in that, The processor performs the steps of the method according to any one of claims 1-20 through the communication interface.

44. A computer-readable storage medium, characterized in that, It stores a computer program or instructions that, when executed, implement the steps of the method as described in any one of claims 1-20.

Images