Communication processing method and communication apparatus

Abstract

Disclosed in embodiments of the present disclosure are a communication processing method and a communication apparatus. The method may comprise: demodulating, on the basis of a first surrogate model, data to be demodulated corresponding to first data, so as to obtain a demodulation result, wherein modulation of the first data is performed on the basis of a first artificial intelligence model, the first surrogate model is used for simulating a second artificial intelligence model, and the second artificial intelligence model is used for demodulation; and acquiring a performance monitoring result of the first artificial intelligence model and / or the second artificial intelligence model on the basis of the demodulation result.

Description

Communication processing methods and communication devices

[0001] This disclosure claims priority to Chinese Patent Application No. 202411578923.5, filed on November 6, 2024, entitled "Communication Processing Method and Communication Apparatus", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure relates to the field of communication technology, and in particular to a communication processing method and a communication device. Background Technology

[0003] Artificial intelligence (AI) technology can solve problems that are difficult to address using traditional modeling methods, such as nonlinear problems and problems with overly complex parameters. It establishes problem-solving patterns through training on large amounts of data and provides relatively accurate predictions. AI can include machine learning (ML), deep learning (DL), and other technologies, while AI models can include various linear and nonlinear network models, such as linear regression, vector machines, convolutional neural networks (CNNs), and deep neural networks (DNNs).

[0004] With the continuous development of AI technology and the evolution of wireless communication systems, the introduction of AI into wireless communication is currently being widely discussed, such as modulation and demodulation of signals based on AI models. However, how to monitor the performance of AI models in modulation and demodulation still requires further research. Summary of the Invention

[0005] This disclosure provides a communication processing method and a communication device that can effectively monitor the modulation and / or demodulation performance of AI models.

[0006] In a first aspect, embodiments of this disclosure provide a communication processing method, which can be executed by a network device or by a device compatible with the network device, such as a processor, chip, or chip module. The method may include: demodulating data to be modulated corresponding to first data based on a first proxy model to obtain a demodulation result; the modulation of the first data is based on a first AI model, the first proxy model is used to simulate a second AI model, and the second AI model is used for demodulation; based on the demodulation result, obtaining performance monitoring results of the first AI model and / or the second AI model.

[0007] In this process, the network device uses a first proxy model to simulate a second AI model, demodulates the data to be demodulated corresponding to the first data, and obtains the demodulation result. The modulation of the first data is based on the first AI model. Based on the demodulation result, the performance monitoring results of the first AI model and / or the second AI model are obtained. This enables effective monitoring of the modulation and / or demodulation performance of the AI ​​model without the network device interacting with the terminal device, which is beneficial for optimizing the performance of the AI ​​model and improving communication performance.

[0008] Secondly, embodiments of this disclosure provide another communication processing method, which can be executed by a network device or by a device compatible with the network device, such as a processor, chip, or chip module. The method may include: transmitting first data, wherein the modulation of the first data is based on a first AI model; obtaining performance monitoring results of the first AI model and / or a second AI model based on demodulation performance information of the data to be demodulated corresponding to the first data; obtaining the demodulation performance information based on the demodulation result of the data to be demodulated, wherein the second AI model is used for demodulation.

[0009] In this process, the network device sends first data to the terminal device, and the modulation of the first data is based on a first AI model; the terminal device demodulates the data to be demodulated corresponding to the first data based on a second AI model, obtains the demodulation result, and returns the demodulation result or demodulation performance information of the data to be demodulated to the network device; the network device obtains the performance monitoring results of the first AI model and / or the second AI model based on the demodulation performance information of the data to be demodulated, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0010] Optionally, demodulation performance information may include at least one of the following: bit error rate, block error rate, probability of hybrid automatic repeat request negative response, and hybrid automatic repeat request acknowledgment information.

[0011] Thirdly, embodiments of this disclosure provide another communication processing method, which can be executed by a network device or by a device compatible with the network device, such as a processor, chip, or chip module. The method may include: transmitting first data, the modulation of which is based on a first AI model; receiving second data, the second data including partial or complete data obtained by demodulating the data to be demodulated corresponding to the first data based on a second AI model; and acquiring performance monitoring results of the first AI model and / or the second AI model based on the second data.

[0012] In this process, the network device sends first data to the terminal device, and the modulation of the first data is based on a first AI model; the terminal device demodulates the data to be demodulated corresponding to the first data based on a second AI model, and sends second data to the network device; the network device obtains the performance monitoring results of the first AI model and / or the second AI model based on the second data, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0013] Optionally, the method further includes: sending a first DCI, the first DCI being used to schedule the transmission of first data and the transmission of second data; wherein the transmission of second data includes transmitting the second data through the transmission resources scheduled by the first DCI. That is, the transmission of first data and the transmission of second data can be scheduled through the same DCI, which facilitates the coordinated scheduling of the transmission of first data and the transmission of second data, and enables the effective use of the transmission resources scheduled by the first DCI to transmit the second data after the terminal device receives the data to be demodulated.

[0014] Optionally, the first data and the second data are transmitted based on the same frequency domain resources; and / or, the first data and the second data are transmitted based on symbol resources at the same location in different time slots; and / or, the time slot for transmitting the second data has an offset of k time slots from the time slot for transmitting the first data, where k is an integer greater than or equal to 1, and k is predefined, obtained based on network configuration information, or based on the indication of the first DCI. That is, the first DCI may include scheduling resources for the data to be demodulated, and the scheduling resources for the second data may be determined based on the scheduling resources for the first data; or, the first DCI may include scheduling resources for the second data, and the scheduling resources for the first data may be determined based on the scheduling resources for the second data.

[0015] Optionally, the first DCI includes scheduling resources for the first data and scheduling resources for the second data.

[0016] Optionally, the method further includes: sending a second DCI; the second DCI includes first indication information, which indicates second data to be transmitted through the transmission resources scheduled by the second DCI. That is, the first indication information included in the second DCI indicates the second data to be fed back.

[0017] Optionally, the first indication information includes a first downlink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the first downlink hybrid automatic repeat request process identifier; and / or, the first indication information includes an uplink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the second downlink hybrid automatic repeat request process identifier, wherein the second downlink hybrid automatic repeat request process identifier has the same value as the uplink hybrid automatic repeat request process identifier; and / or, the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the third data, wherein the third data is data in the time domain unit in the first time domain unit that is closest to the time domain unit of the second DCI, and the first time domain unit is the time domain unit used to receive the first data.

[0018] Optionally, some data includes data obtained by demodulating the first code block by the terminal device, wherein the first data includes the first code block, which is a code block with demodulation errors; and / or, some data includes data obtained by demodulating the first code block group, wherein the data to be demodulated includes the first code block group, which is a code block group with demodulation errors. In other words, the terminal device can send the data with demodulation errors obtained by demodulating the data to be demodulated, or all the data obtained by demodulation, to the network device.

[0019] Optionally, the method further includes: receiving second indication information, the second indication information being used to indicate the first code block or first code block group corresponding to the partial data. That is, the terminal device uses the second indication information to indicate to the network device which code blocks or code block groups correspond to the demodulated partial data.

[0020] Fourthly, embodiments of this disclosure provide another communication processing method, which can be executed by a network device or by a device matched with the network device, such as a processor, chip, or chip module. The method may include: demodulating the data to be demodulated corresponding to the fourth data based on a fourth AI model to obtain a demodulation result; modulation of the fourth data based on a second proxy model, the second proxy model being used to simulate a third AI model, and the third AI model being used for modulation; and obtaining performance monitoring results of the third AI model and / or the fourth AI model based on the demodulation result.

[0021] In this system, the network device demodulates the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result. The modulation of the fourth data is based on the second proxy model, which is used to simulate the third AI model. The third AI model is used for modulation. Based on the demodulation result, the system obtains the performance monitoring results of the third AI model and / or the fourth AI model. This enables effective monitoring of the modulation and / or demodulation performance of the AI ​​model without information interaction between the network device and the terminal device. This facilitates the optimization of the AI ​​model's performance and improves communication performance.

[0022] Fifthly, embodiments of this disclosure provide another communication processing method, which can be executed by a network device or by a device compatible with the network device, such as a processor, chip, or chip module. The method may include: receiving fourth data, the modulation of which is based on a third AI model; obtaining the performance monitoring results of the third AI model and / or the fourth AI model based on demodulation performance information of the data to be demodulated corresponding to the fourth data; obtaining the demodulation performance information based on the demodulation result; and obtaining the demodulation result by demodulating the data to be demodulated based on the fourth AI model.

[0023] In this process, the terminal device sends fourth data to the network device, and the modulation of the fourth data is based on the third AI model. Subsequently, the network device demodulates the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result. Based on the demodulation result, it obtains the demodulation performance information of the data to be demodulated, and based on the demodulation performance information of the data to be demodulated, it obtains the performance monitoring results of the third AI model and / or the fourth AI model. This effectively monitors the performance of the AI ​​model in modulation and / or demodulation, which is conducive to the optimization of the AI ​​model's performance and improves communication performance.

[0024] Optionally, the demodulation performance information includes at least one of the following: the signal-to-noise ratio of the received signal, the distribution characteristics of the received data, the bit error rate, the block error rate, the probability of a negative response to a hybrid automatic repeat request, and the acknowledgment information for a hybrid automatic repeat request.

[0025] Sixthly, embodiments of this disclosure provide another communication processing method, which can be executed by a network device or by a device compatible with the network device, such as a processor, chip, or chip module. The method may include: sending a third DCI, the third DCI being used to indicate at least two repeated transmissions, the at least two repeated transmissions including at least one first transmission and at least one second transmission, wherein the modulation of the first transmission is based on an AI model, and the modulation of the second transmission differs from that of the first transmission; the performance monitoring result of the AI ​​model is obtained based on the demodulation results of the at least two repeated transmissions.

[0026] In this process, the network device sends a third DCI to the terminal device. The third DCI is used to indicate at least two repeated transmissions, which include at least one first transmission and at least one second transmission. The modulation of the first transmission is based on the AI ​​model, and the modulation of the second transmission is different from that of the first transmission. The terminal device performs at least two repeated transmissions with the network device based on the third DCI. Based on the demodulation results of the at least two repeated transmissions, the network device obtains the performance monitoring results of the AI ​​model, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0027] In a seventh aspect, embodiments of this disclosure provide yet another communication processing method, which can be executed by a terminal device or by a device matched with the terminal device, such as a processor, chip, or chip module. The method may include: receiving first data, wherein the modulation of the first data is based on a first AI model; sending a demodulation result or demodulation performance information of the data to be demodulated corresponding to the first data; obtaining the demodulation result by demodulating the data to be demodulated based on a second AI model; obtaining the demodulation performance information based on the demodulation result; and obtaining the performance monitoring results of the first AI model and / or the second AI model based on the demodulation performance information.

[0028] In this process, the network device sends first data to the terminal device, and the modulation of the first data is based on a first AI model; the terminal device demodulates the data to be demodulated corresponding to the first data based on a second AI model, obtains the demodulation result, and returns the demodulation result or demodulation performance information of the data to be demodulated to the network device; the network device obtains the performance monitoring results of the first AI model and / or the second AI model based on the demodulation performance information of the data to be demodulated, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0029] Optionally, demodulation performance information may include at least one of the following: bit error rate, block error rate, probability of hybrid automatic repeat request negative response, and hybrid automatic repeat request acknowledgment information.

[0030] Eighthly, embodiments of this disclosure provide another communication processing method, which can be executed by a terminal device or by a device matched with the terminal device, such as a processor, chip, or chip module. The method may include: receiving first data, the modulation of which is based on a first AI model; transmitting second data, the second data including partial or complete data obtained by demodulating the data to be demodulated corresponding to the first data based on a second AI model; and acquiring performance monitoring results of the first AI model and / or the second AI model based on the second data.

[0031] In this process, the network device sends first data to the terminal device, and the modulation of the first data is based on a first AI model; the terminal device demodulates the data to be demodulated corresponding to the first data based on a second AI model, and sends second data to the network device; the network device obtains the performance monitoring results of the first AI model and / or the second AI model based on the second data, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0032] Optionally, the method further includes: receiving a first DCI, the first DCI being used to schedule the transmission of first data and the transmission of second data; wherein the transmission of the second data includes transmitting the second data through the transmission resources scheduled by the first DCI. That is, the transmission of the first data and the transmission of the second data can be scheduled through the same DCI, which facilitates the coordinated scheduling of the transmission of the first data and the transmission of the second data, and enables the effective use of the transmission resources scheduled by the first DCI to transmit the second data after the terminal device receives the data to be demodulated.

[0033] Optionally, the first data and the second data are transmitted based on the same frequency domain resources; and / or, the first data and the second data are transmitted based on symbol resources at the same position in different time slots; and / or, the time slot for transmitting the second data has an offset of k time slots from the time slot for transmitting the first data, where k is an integer greater than or equal to 1, and k is predefined, obtained based on network configuration information, or based on the indication of the first DCI. That is, the first DCI may include the scheduling resources of the first data, and the scheduling resources of the second data may be determined based on the scheduling resources of the data to be demodulated; or, the first DCI may include the scheduling resources of the second data, and the scheduling resources of the first data may be determined based on the scheduling resources of the second data.

[0034] Optionally, the first DCI includes scheduling resources for the first data and scheduling resources for the second data.

[0035] Optionally, the method further includes: receiving a second DCI; the second DCI includes first indication information, the first indication information being used to indicate second data to be transmitted through the transmission resources scheduled by the second DCI. That is, the first indication information included in the second DCI indicates the second data to be fed back.

[0036] Optionally, the first indication information includes a first downlink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the first downlink hybrid automatic repeat request process identifier; and / or, the first indication information includes an uplink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the second downlink hybrid automatic repeat request process identifier, wherein the second downlink hybrid automatic repeat request process identifier has the same value as the uplink hybrid automatic repeat request process identifier; and / or, the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the third data, wherein the third data is data in the time domain unit in the first time domain unit that is closest to the time domain unit of the second DCI, and the first time domain unit is the time domain unit used to receive the first data.

[0037] Optionally, some data includes data obtained by demodulating the first code block, where the data to be demodulated includes the first code block, which is a code block with demodulation errors; and / or, some data includes data obtained by demodulating the first code block group, where the data to be demodulated includes the first code block group, which is a group of code blocks with demodulation errors. In other words, the terminal device can send the data with demodulation errors obtained by demodulating the data to be demodulated, or all the data obtained by demodulation, to the network device.

[0038] Optionally, the method further includes: sending second indication information, the second indication information being used to indicate the first code block or first code block group corresponding to the partial data. That is, the terminal device uses the second indication information to indicate to the network device which code blocks or code block groups correspond to the demodulated partial data.

[0039] Ninthly, embodiments of this disclosure provide another communication processing method, which can be executed by a terminal device or by a device matched with the terminal device, such as a processor, chip, or chip module. The method may include: transmitting fourth data; modulation of the fourth data based on a third AI model; obtaining the performance monitoring results of the third AI model and / or the fourth AI model based on demodulation performance information of the data to be demodulated corresponding to the fourth data; obtaining the demodulation performance information based on the demodulation result; and obtaining the demodulation result by demodulating the data to be demodulated based on the fourth AI model.

[0040] In this process, the terminal device sends fourth data to the network device, and the modulation of the fourth data is based on the third AI model. Subsequently, the network device demodulates the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result. Based on the demodulation result, it obtains the demodulation performance information of the data to be demodulated, and based on the demodulation performance information of the data to be demodulated, it obtains the performance monitoring results of the third AI model and / or the fourth AI model. This effectively monitors the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0041] Optionally, the demodulation performance information includes at least one of the following: the signal-to-noise ratio of the received signal, the distribution characteristics of the received data, the bit error rate, the block error rate, the probability of a negative response to a hybrid automatic repeat request, and the acknowledgment information for a hybrid automatic repeat request.

[0042] In a tenth aspect, embodiments of this disclosure provide yet another communication processing method, which can be executed by a terminal device or by a device compatible with the terminal device, such as a processor, chip, or chip module. The method may include: receiving a third DCI, the third DCI being used to indicate at least two repeated transmissions, the at least two repeated transmissions including at least one first transmission and at least one second transmission, the modulation of the first transmission being based on an AI model, and the modulation of the second transmission being different from that of the first transmission; performing at least two repeated transmissions; and obtaining the performance monitoring result of the AI ​​model based on the demodulation result of the at least two repeated transmissions.

[0043] In this process, the network device sends a third DCI to the terminal device. The third DCI is used to indicate at least two repeated transmissions, which include at least one first transmission and at least one second transmission. The modulation of the first transmission is based on the AI ​​model, and the modulation of the second transmission is different from that of the first transmission. The terminal device performs at least two repeated transmissions with the network device based on the third DCI. Based on the demodulation results of the at least two repeated transmissions, the network device obtains the performance monitoring results of the AI ​​model, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0044] Eleventhly, embodiments of this disclosure provide a communication device, the communication device comprising:

[0045] The demodulation unit is used to demodulate the data to be demodulated corresponding to the first data based on the first proxy model to obtain the demodulation result; the modulation of the first data is based on the first AI model, the first proxy model is used to simulate the second AI model, and the second AI model is used to perform demodulation.

[0046] The acquisition unit is used to acquire the performance monitoring results of the first AI model and / or the second AI model based on the demodulation results.

[0047] Alternatively, the communication device may include:

[0048] A communication unit is used to transmit first data, the modulation of which is based on a first AI model; the performance monitoring results of the first AI model and / or the second AI model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the first data; the demodulation performance information is obtained based on the demodulation result of the data to be demodulated, and the second AI model is used to perform demodulation.

[0049] Alternatively, the communication device may include:

[0050] The communication unit is used to transmit first data, the modulation of which is based on a first AI model; receive second data, which includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the first data based on the second AI model; and acquire the performance monitoring results of the first AI model and / or the second AI model based on the second data.

[0051] Alternatively, the communication device may include:

[0052] The demodulation unit is used to demodulate the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result; the modulation of the fourth data is based on the second surrogate model, which is used to simulate the third AI model, and the third AI model is used for modulation.

[0053] The acquisition unit is used to acquire the performance monitoring results of the third AI model and / or the fourth AI model based on the demodulation results.

[0054] Alternatively, the communication device may include:

[0055] The communication unit is used to receive fourth data, the modulation of which is based on the third AI model; the performance monitoring results of the third AI model and / or the fourth AI model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the first data; the demodulation performance information is obtained based on the demodulation result; the demodulation result is obtained by demodulating the data to be demodulated based on the fourth AI model.

[0056] Alternatively, the communication device may include:

[0057] The communication unit is used to send a third DCI, which is used to instruct the terminal device to perform at least two repeated transmissions. The at least two repeated transmissions include at least one first transmission and at least one second transmission. The modulation of the first transmission is based on the AI ​​model, and the modulation of the second transmission is different from that of the first transmission. The performance monitoring results of the AI ​​model are obtained based on the demodulation results of the at least two repeated transmissions.

[0058] Alternatively, the communication device may include:

[0059] The communication unit is used to receive first data, the modulation of which is based on a first AI model; send demodulation results or demodulation performance information of the data to be demodulated corresponding to the first data; the demodulation results are obtained by demodulating the data to be demodulated based on a second AI model; the demodulation performance information is obtained based on the demodulation results; and the performance monitoring results of the first AI model and / or the second AI model are obtained based on the transmission performance information.

[0060] Alternatively, the communication device may include:

[0061] The communication unit is used to receive first data, the modulation of which is based on a first AI model; to transmit second data, which includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the first data based on a second AI model; and to acquire the performance monitoring results of the first AI model and / or the second AI model based on the second data.

[0062] Alternatively, the communication device may include:

[0063] A communication unit is used to transmit fourth data; the modulation of the fourth data is based on the third AI model; the performance monitoring results of the third AI model and / or the fourth AI model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the fourth data; the demodulation performance information is obtained based on the demodulation result; the demodulation result is obtained by demodulating the data to be demodulated based on the fourth AI model.

[0064] Alternatively, the communication device may include:

[0065] A communication unit is used to receive a third DCI, which is used to indicate that at least two repeated transmissions are performed. The at least two repeated transmissions include at least one first transmission and at least one second transmission. The modulation of the first transmission is based on an AI model, and the modulation of the second transmission is different from that of the first transmission. The performance monitoring results of the AI ​​model are obtained based on the demodulation results of the at least two repeated transmissions.

[0066] In a twelfth aspect, embodiments of this disclosure provide a communication device including 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 of the methods involved in any one of the first to tenth aspects described above.

[0067] In a thirteenth aspect, embodiments of this disclosure provide a chip including a processor, wherein the processor performs the steps of the methods involved in any one of the first to tenth aspects described above.

[0068] In a fourteenth aspect, embodiments of this disclosure provide a chip module including a communication interface and a chip, the chip including a processor, wherein the processor performs the steps of the methods involved in any one of the first to tenth aspects described above.

[0069] In a fifteenth aspect, embodiments of this disclosure provide a computer-readable storage medium storing a computer program or instructions that, when executed, implement the steps of the methods involved in any one of the first to tenth aspects described above.

[0070] In a sixteenth aspect, embodiments of this disclosure provide a computer program product, including a computer program or instructions, wherein when the computer program or instructions are executed, they implement the steps of the methods involved in any one of the first to tenth aspects described above.

[0071] In a seventeenth aspect, embodiments of this disclosure provide a communication system that may include a network device performing the methods of the second aspect and a terminal device performing the methods of the seventh aspect; or, the communication system may include a network device performing the methods of the third aspect and a terminal device performing the methods of the eighth aspect; or, the communication system may include a network device performing the methods of the fifth aspect and a terminal device performing the methods of the ninth aspect; or, the communication system may include a network device performing the methods of the sixth aspect and a terminal device performing the methods of the tenth aspect. Attached Figure Description

[0072] Figure 1 is a schematic diagram of a system architecture applying an embodiment of the present disclosure;

[0073] Figure 2 is a schematic diagram of a communication process based on an AI model for modulation and demodulation provided in an embodiment of this disclosure;

[0074] Figure 3 is a flowchart illustrating a communication processing method provided in an embodiment of this disclosure;

[0075] Figure 4 is a flowchart illustrating another communication processing method provided in an embodiment of this disclosure;

[0076] Figure 5 is a flowchart illustrating another communication processing method provided in an embodiment of this disclosure;

[0077] Figure 6 is a flowchart illustrating another communication processing method provided in an embodiment of this disclosure;

[0078] Figure 7 is a flowchart illustrating another communication processing method provided in an embodiment of this disclosure;

[0079] Figure 8 is a flowchart illustrating another communication processing method provided in an embodiment of this disclosure;

[0080] Figure 9 is a schematic diagram of the structure of a communication device provided in an embodiment of this disclosure;

[0081] Figure 10 is a schematic diagram of another communication device provided in an embodiment of this disclosure;

[0082] Figure 11 is a schematic diagram of another communication device provided in an embodiment of this disclosure;

[0083] Figure 12 is a schematic diagram of the structure of a chip module provided in an embodiment of this disclosure. Detailed Implementation

[0084] In the specification, claims, and drawings of this disclosure, terms such as "first," "second," etc., may be used to distinguish similar objects without necessarily describing a specific order, sequence, and / or quantity. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders and / or quantities other than those illustrated or described herein.

[0085] "AND / OR" describes the relationship between related objects, indicating that there can be three relationships. For example, A AND / OR B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the objects before and after it are in an "OR" relationship.

[0086] It should be understood that in this disclosure, "at least one" refers to one or more; "multiple" refers to two or more. Furthermore, the word "equal to" in this disclosure can be used with either "greater than" or "less than". When "equal to" and "greater than" are used together, the technical solution using "greater than" is adopted; when "equal to" and "less than" are used together, the technical solution using "less than" is adopted.

[0087] In the embodiments of this disclosure, the terms "of," "corresponding (relevant)," "corresponding," "associated (related)," and "mapped" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction, the concepts or meanings expressed are consistent.

[0088] First, the system architecture involved in this disclosure will be described.

[0089] This disclosure can be applied to fifth-generation (5G) systems, also known as new radio (NR) systems; or to sixth-generation (6G) systems, or seventh-generation (7G) systems, or other future communication systems; or to device-to-device (D2D) systems, machine-to-machine (M2M) systems, vehicle-to-everything (V2X) systems, etc.

[0090] This disclosure can be applied to the system architecture shown in Figure 1. The system architecture shown in Figure 1 may include, but is not limited to, network device 110 and terminal device 120. The number and form of the devices in Figure 1 are for illustrative purposes only and do not constitute a limitation on the embodiments of this disclosure. For example, Figure 1 uses one network device and one terminal device as an example, but in actual applications, more network devices and / or more terminal devices may be included.

[0091] I. Terminal Equipment

[0092] Terminal equipment can be a device with transceiver capabilities, and can also be called a terminal, user equipment (UE), remote terminal equipment (relay UE), relay equipment (relay UE), access terminal equipment, user unit, user station, mobile station, mobile station, remote station, mobile device, user terminal equipment, smart terminal equipment, wireless communication equipment, user agent, or user device. It should be noted that relay equipment is a terminal device capable of providing relay forwarding services to other terminal equipment (including remote terminal equipment).

[0093] For example, terminal devices can be mobile phones, tablets, computers with wireless transceiver capabilities, virtual reality (VR) terminal devices, augmented reality (AR) terminal devices, wireless terminal devices in industrial control, wireless terminal devices in autonomous driving, wireless terminal devices in remote medical care, wireless terminal devices in smart grids, wireless terminal devices in transportation safety, wireless terminal devices in smart cities, or wireless terminal devices in smart homes, etc.

[0094] For example, terminal devices can also be cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices with wireless communication capabilities, computing devices or other processing devices connected to a wireless modem, in-vehicle devices, wearable devices, terminal devices in next-generation communication systems (such as NR communication systems and 6G communication systems), or terminal devices in future evolved public land mobile networks (PLMNs), etc., without specific limitations.

[0095] In some possible implementations, the terminal device can be deployed on land, including indoors or outdoors, handheld, wearable, or vehicle-mounted; it can be deployed on water (such as ships); or it can be deployed in the air (such as airplanes, balloons, and satellites).

[0096] In some possible implementations, the terminal device may include means for wireless communication functionality, such as a chip system, a chip, or a chip module. For example, the chip system may include a chip, and may also include other discrete devices.

[0097] In some possible implementations, the terminal device described in the embodiments of this disclosure may be a chip, chip module, device, unit, etc., and there is no limitation thereto. The embodiments of this disclosure do not limit the specific technology or specific device form used in the terminal device.

[0098] II. Network Equipment

[0099] A network device is a device with transceiver capabilities that can be used to communicate with terminal devices.

[0100] In some possible implementations, network devices can be responsible for radio resource management (RRM), quality of service (QoS) management, data compression and encryption, and data transmission and reception on the air interface side.

[0101] In some possible implementations, network devices may include base stations (BS) in a communication system or devices deployed in a radio access network (RAN) to provide wireless communication functions; that is, network devices may include devices in the RAN.

[0102] For example, devices in the RAN may include evolved node B (eNB or eNodeB) in the LTE communication system, next generation evolved node B (ng eNB) in the NR communication system, next generation node B (gNB) in the NR communication system, master node (MN) in the dual connectivity architecture, and secondary node (SN) in the dual connectivity architecture, etc., without specific restrictions.

[0103] In some possible implementations, network devices may include devices in the core network (CN).

[0104] For example, devices in a CN may include access and mobility management function (AMF), user plane function (UPF), session management function (SMF), etc.

[0105] In some possible implementations, network devices can also be access points (APs) in WLANs, relay stations, communication devices in future PLMN networks, communication devices in NTN networks, etc.

[0106] In some possible implementations, the network device may include means for providing wireless communication capabilities to terminal devices, such as a chip system, a chip, or a chip module. For example, the chip system may include a chip, or it may include other discrete devices.

[0107] In some possible implementations, network devices can communicate with Internet Protocol (IP) networks, such as the Internet, private IP networks, or other data networks.

[0108] In some possible implementations, the network device may include a single node to implement the functions of the aforementioned base station, or it may include two or more independent nodes to implement the functions of the aforementioned base station. For example, the network device includes a centralized unit (CU) and a distributed unit (DU), such as a gNB CU and a gNB DU. Furthermore, in some other embodiments of this disclosure, the network device may also include an active antenna unit (AAU). The CU implements some of the functions of the network device, and the DU implements other functions. For example, the CU is responsible for handling non-real-time protocols and services, implementing the functions of the radio resource control (RRC) layer, the service data adaptation protocol (SDAP) layer, and the packet data convergence protocol (PDCP) layer. The DU is responsible for handling physical layer protocols and real-time services, implementing the functions of the radio link control (RLC) layer, the medium access control (MAC) layer, and the physical (PHY) layer. Additionally, the AAU can implement some physical layer processing functions, radio frequency processing, and related functions of the active antenna. Since RRC layer information ultimately becomes PHY layer information, or is derived from PHY layer information, in this network deployment, higher-layer signaling (such as RRC signaling) can be considered to be generated by the CU and sent by the DU, or jointly sent by the DU and AAU. It is understood that network devices can include at least one of CU, DU, and AAU. Furthermore, the CU can be classified as a RAN device, or it can be classified as a core network device; there are no specific restrictions on this.

[0109] In some possible implementations, the network device can be any station in a multi-site coherent joint transmission (CJT) with the terminal device, or another station outside of that multi-site group, or other network devices communicating with the terminal device; no specific limitations are imposed. Multi-site coherent joint transmission can be multiple stations jointly transmitting coherently, or different data belonging to the same Physical Downlink Shared Channel (PDSCH) being sent from different stations to the terminal device, or multiple stations being virtually merged into one station for transmission. Names with the same meaning as those specified in other standards also apply to this disclosure; that is, this disclosure does not limit the names of these parameters. The stations in multi-site coherent joint transmission can be remote radio heads (RRHs), transmission and reception points (TRPs), network devices, etc.; no specific limitations are imposed.

[0110] In some possible implementations, the network device can be any of the multiple sites performing noncoherent cooperative transmission with the terminal device, or other sites outside of the multiple sites, or other network devices communicating with the terminal device; no specific limitations are imposed. The multi-site noncoherent cooperative transmission can be a joint noncoherent transmission by multiple sites, or different data belonging to the same PDSCH being sent to the terminal device from different sites, or different data belonging to the same PDSCH being sent to the terminal device from different sites. Names with the same meaning as those specified in other standards also apply to this disclosure; that is, this disclosure does not limit the names of these parameters. The sites in the multi-site noncoherent cooperative transmission can be RRHs, TRPs, network devices, etc., without specific limitations.

[0111] In some possible implementations, the network device can have mobility characteristics; for example, the network device can be a mobile device. Optionally, the network device can be a satellite or 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 highly elliptical orbit (HEO) satellite, etc. Optionally, the network device can also be a base station located on land, water, or other similar locations.

[0112] In some possible implementations, network devices can provide services to a cell, and terminal devices within that cell can communicate with the network devices via transmission resources (such as spectrum resources). This cell can be a macro cell, small cell, metro cell, micro cell, pico cell, or femto cell, etc.

[0113] In some possible implementations, the network device described in the embodiments of this disclosure may be a chip, chip module, device, unit, etc., and there are no specific limitations on it. The embodiments of this disclosure do not limit the specific technology or specific device form used in the network device.

[0114] In this embodiment of the disclosure, the terminal device may implement model functionality, or the terminal device may include a model module for implementing model functionality. The network device may implement model functionality, or the network device may include a model module for implementing model functionality. The model may include, but is not limited to, an artificial intelligence (AI) model. For example, the model module may implement modulation and / or demodulation, etc., enabling the AI ​​model.

[0115] Optionally, the model module can be a software unit and / or hardware unit that implements model functionality using algorithms such as convolutional neural network algorithms and deep neural network algorithms. Alternatively, the model module can be a chip, chip module, etc.

[0116] It is understood that the system architecture described in the embodiments of this disclosure is for the purpose of more clearly illustrating the technical solutions of the embodiments of this disclosure, and does not constitute a limitation on the technical solutions provided in the embodiments of this disclosure. Those skilled in the art will know that as the system architecture evolves and new business scenarios emerge, the technical solutions provided in the embodiments of this disclosure are also applicable to similar technical problems.

[0117] Secondly, to facilitate understanding of the embodiments of this disclosure, the relevant concepts involved in the embodiments of this disclosure will be explained.

[0118] I. Modulation and Demodulation

[0119] In a broad sense, modulation refers to the processing of information from a signal source to transform it into a form suitable for transmission over a channel. For example, modulation can include, but is not limited to, signal modulation and / or channel coding. As another example, modulation can include encoding bit-level information into bit-level information suitable for transmission, and / or modulation can include transforming bit-level information into symbol-level information (e.g., complex or real numbers). As yet another example, modulation can include encoding multiple bits into a single symbol in a signal based on a constellation diagram; this can be called constellation modulation. A constellation diagram, by plotting the possible states of a signal in one-dimensional or multi-dimensional space, illustrates the amplitude and phase information of the modulated signal. In other words, a constellation diagram can show the possible states of a signal under different modulation methods and their distribution in one-dimensional or multi-dimensional space. Each modulation scheme has its own specific constellation diagram. For example, the constellation diagram of 16 quadrature amplitude modulation (QAM) has 16 points, each point representing 4 bits of information, and each symbol carries 4 bits of information. Another example is the constellation diagram of quadrature phase shift keying (QPSK), which has 4 points, representing 00, 01, 10, and 11 respectively, and each symbol carries 2 bits of information.

[0120] In a narrow sense, modulation means using a baseband signal to control (modulate) one or more parameters of a single-frequency carrier signal, thereby embedding the desired transmitted information into the modulated signal. Based on the carrier signal (also called the modulated signal), modulation can be divided into three categories: sinusoidal modulation, pulse modulation, and intensity modulation, where the modulated carrier is a sine wave, a pulse, and an optical wave, respectively. Based on the baseband signal (also called the modulating signal), modulation can be divided into two categories: analog modulation and digital modulation. In analog modulation, the baseband signal is an analog signal, while in digital modulation, the baseband signal is a digital signal.

[0121] Demodulation, also known as demodulation, is the reverse process of modulation; in other words, it is the process of restoring modulated information to its original state before modulation.

[0122] Modulation can be achieved through a modulator, which can be viewed as a software and / or hardware unit in a computer device used to implement modulation functions. Modulators can have other names depending on their modulation function; for example, a modulator can also be called an encoder to encode the data to be modulated. Optionally, a modulator can be a chip, a chip module, etc.

[0123] Demodulation can be achieved using a demodulator, which can be viewed as a software and / or hardware unit in a computer device used to implement demodulation functionality. Demodulators may have other names depending on their demodulation function; for example, a demodulator can also be called a decoder to decode encoded data, without limitation. Optionally, a demodulator can be a chip, chip module, etc.

[0124] In this embodiment of the disclosure, modulation can be in the broad sense described above, or it can be in the narrow sense described above, and there is no limitation thereto.

[0125] II. Artificial Intelligence (AI) and AI Models

[0126] AI is a technology that simulates human intelligence, enabling machines to learn, think, and make decisions like humans, thus allowing them to autonomously perform various tasks. AI technology can solve problems that are difficult to address using traditional modeling methods, such as nonlinear problems and problems with overly complex parameters. It establishes problem-solving patterns through training on large amounts of data and provides relatively accurate predictions. AI can include, but is not limited to, machine learning (ML) and deep learning (DL).

[0127] An AI model is a model with specific parameters and architecture that can be used to implement AI functions. AI models can include, but are not limited to, various linear and nonlinear network models, such as linear regression, vector machines, convolutional neural networks (CNNs), and deep neural networks (DNNs). Optionally, the term "AI model" can also be used to refer to other concepts, such as, but not limited to, AI algorithms and AI modules. Optionally, an AI model can include, but is not limited to, software and / or hardware units within a computer device used to implement AI functions. Optionally, an AI model can be a chip, a chip module, etc., without limitation.

[0128] With the continuous development of AI technology and the continuous evolution of wireless communication systems, there is currently widespread discussion about introducing AI technology into wireless communication technology, such as modulation and demodulation of signals based on AI models.

[0129] Please refer to Figure 2, which is a schematic diagram of a communication process based on an AI model for modulation and demodulation according to an embodiment of this disclosure. As shown in Figure 2, at the transmitting end, the data to be transmitted is first encoded using a Low Density Parity Check Code (LDPC); then, the LDPC-encoded data is modulated based on an AI model, for example, but not limited to, encoding based on a trainable constellation using an encoder; subsequently, the data modulated by the AI ​​model is normalized, for example, but not limited to, layer normalization or batch normalization; and then the normalized data is transmitted to the receiving end. At the receiving end, the received data is demodulated based on an AI model, for example, but not limited to, using a decoder for decoding based on a neural demapper. Optionally, before demodulating the received data, the data can be further processed (not shown in Figure 2), for example, but not limited to, performing inverse processing corresponding to normalization, so that the processed data can be used for demodulation and / or restored to the modulated state. Subsequently, the demodulation result based on the AI ​​model is subjected to LDPC decoding, realizing the communication process of modulation and demodulation based on the AI ​​model. The AI ​​model used for modulation and the AI ​​model used for demodulation can be the same AI model or different AI models; there is no restriction on this. It should be noted that Figure 2 is only an example of the communication process of modulation and demodulation based on the AI ​​model. This communication process may have more or fewer communication processing steps. For example, but not limited to, after the transmitting end modulates the data, it also performs inverse fast Fourier transform (IFFT) processing on the modulated data, and the receiving end performs fast Fourier transform (FFT) processing on the data before demodulation; and / or this communication process may also have different communication processing steps, such as but not limited to not performing normalization processing and its inverse processing, etc. Figure 2 does not constitute a limitation.

[0130] However, how to monitor the performance of AI models in modulation and demodulation still requires further research.

[0131] In view of this, this disclosure proposes a communication processing method, which demodulates the data to be demodulated corresponding to the first data based on a first proxy model to obtain a demodulation result; the modulation of the first data is based on a first AI model, the first proxy model is used to simulate a second AI model, and the second AI model is used for demodulation; based on the demodulation result, the performance monitoring results of the first AI model and / or the second AI model are obtained, thereby realizing effective monitoring of the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0132] Based on the system architecture shown in Figure 1, the execution entity in this embodiment can be a terminal device and / or a network device. Alternatively, the execution entity in this embodiment can be a device compatible with the terminal device, such as a processor, chip, or chip module, and / or a device compatible with the network device, such as a processor, chip, or chip module. Alternatively, the execution entity in this embodiment can also be other computer devices or devices compatible with other computer devices, such as a processor, chip, or chip module; there are no limitations on this. The following description uses a terminal device and / or a network device as an example.

[0133] In one possible implementation, the network device or terminal device can obtain the performance monitoring results of the AI ​​model based on demodulation information. The AI ​​model is used to modulate the data to be modulated, and / or the AI ​​model can be used to demodulate the data to be demodulated. For example, the network device or terminal device can obtain the performance monitoring results of a first AI model and / or a second AI model based on demodulation information. The first AI model is used to modulate the data to be modulated, and the second AI model is used to demodulate the data to be demodulated.

[0134] Optionally, the demodulation information may include, but is not limited to, at least one of the following: demodulation result, demodulation performance information of the data to be demodulated, and second data. The second data is part or all of the data obtained by demodulating the data to be demodulated based on the second AI model. The demodulation performance of the data is related to the modulation, transmission, and / or demodulation of the data.

[0135] Optionally, the first AI model and the second AI model can be different AI models or the same AI model; there is no restriction on this.

[0136] For a detailed description of this implementation method, please refer to the subsequent implementation methods.

[0137] In one optional implementation, the network device can modulate the data to be modulated and demodulate the data to be demodulated based on an AI model to obtain a demodulation result; based on the demodulation result, obtain the performance monitoring result of the AI ​​model; wherein, the demodulation result is obtained by demodulating the data to be demodulated corresponding to the first data based on a first proxy model, the modulation of the first data is based on a first AI model, the first proxy model is used to simulate a second AI model, and the second AI model is used to perform demodulation.

[0138] Please refer to Figure 3, which is a flowchart illustrating a communication processing method provided in an embodiment of this disclosure. This method may include, but is not limited to, the following steps:

[0139] 1001. The network device demodulates the data to be demodulated corresponding to the first data based on the first proxy model to obtain the demodulation result; the modulation of the first data is based on the first AI model, the first proxy model is used to simulate the second AI model, and the second AI model is used for demodulation.

[0140] Optionally, before proceeding to step 1001, the network device may also modulate the data to be modulated based on the first AI model.

[0141] The data to be modulated is data that can be modulated. In specific examples, the data to be modulated can be data that needs to be transmitted; and / or, the data to be modulated can be test data of the first AI model; and / or, the data to be modulated can be data that has been processed to be suitable for modulation, such as data after LDPC encoding (see Figure 2); or, the data to be modulated can also be other data that can be modulated, without limitation.

[0142] The first AI model is used for modulation; that is, the first AI model is used to modulate the data to be modulated. The data to be modulated can be input into the first AI model, and correspondingly, the first AI model outputs modulated data based on the input data.

[0143] In some embodiments, the first AI model can be obtained by training the first preset model using a first sample dataset, or it can be obtained in other ways, without limitation. The first sample dataset may include data that needs to be modulated in uplink and / or downlink communication, as well as the data obtained after modulation of this data. The data in the first sample dataset can be obtained by collecting data during routine uplink and / or downlink communication, or by modulating a series of sample data using a modulation method not based on the AI ​​model (e.g., but not limited to, 16-orthogonal amplitude modulation, or quadrature phase shift keying, etc.), or by other methods, without limitation. The first preset model can be various linear or nonlinear network models, such as CNN, DNN, or recurrent neural network (RNN), or other models, without limitation.

[0144] Optionally, the network device may include a modulator for modulation, which can modulate the data to be modulated based on a first AI model. For example, the modulator may be the first AI model used for modulation, or it may have the first AI model built in, or it may simulate or invoke the first AI model, etc., without specific limitations.

[0145] In some embodiments, the modulator may be an encoder for encoding, and the network device may use the encoder to encode the data to be modulated.

[0146] Optionally, before performing step 1001 and after modulating the data to be modulated based on the first AI model, the modulated data may undergo a first processing. This first processing enables the processed data to be transmitted through transmission resources, i.e., it can be used for transmission on a channel, for example, but not limited to, downlink and / or uplink transmission, etc., without limitation.

[0147] In specific examples, the first process may include, but is not limited to, precoding, inverse fast Fourier transform (IFFT), normalization, and / or channel modulation. There are no restrictions on the order or content of the processes included in the first process. The first process may also include more or fewer processes, and / or may include processes different from those listed above.

[0148] It can be understood that the first data can be data obtained by modulating the data to be modulated based on the first AI model; or, the first data can be data obtained by performing a first process on the data modulated based on the first AI model. Therefore, the modulation of the first data is based on the first AI model.

[0149] Optionally, the data after the first processing can also undergo a second processing. This second processing is used to demodulate the data and / or restore it to its modulated state.

[0150] In specific examples, the second processing may include, but is not limited to, deprecoding processing, fast Fourier transform (FFT) processing, and / or the inverse processing corresponding to normalization processing. There are no restrictions on the processing order and processing steps included in the second processing. The first processing may also include more or fewer processing, and / or may include processing different from the processing listed above.

[0151] Here, the data to be demodulated is data that can be demodulated. It can be understood that the data to be demodulated corresponding to the first data can be the first data; or, the data to be demodulated corresponding to the first data can be data obtained by performing a first processing and / or a second processing on the data modulated based on the first AI model.

[0152] For example, the first data is the data obtained after modulating the data to be modulated based on the first AI model, and the corresponding demodulated data can be the first data. As another example, the first data is the data obtained after modulating the data to be modulated based on the first AI model, and the corresponding demodulated data can be the data obtained by performing a first processing and a second processing on the data modulated based on the first AI model. As yet another example, the first data is the data obtained by performing a first processing on the data modulated based on the first AI model, and the corresponding demodulated data can be the data obtained by performing a first processing and a second processing on the data modulated based on the first AI model. The above are merely examples of the first data and its corresponding demodulated data, and do not constitute a limitation.

[0153] The second AI model is used for demodulation; that is, it is an AI model used to demodulate the data to be demodulated. The data to be demodulated can be input into the second AI model, and the second AI model will output the demodulation result based on the input data.

[0154] In some embodiments, the second AI model can be obtained by training the second preset model using a second sample dataset, or by other methods, without limitation. The second sample dataset can include data that needs to be demodulated in uplink and / or downlink communication, as well as the data obtained after demodulation. The data in the second sample dataset can be obtained by collecting data during routine uplink and / or downlink communication, or by demodulating a series of sample data using non-AI model-based demodulation methods (e.g., but not limited to, demodulation methods corresponding to 16-orthogonal amplitude modulation, or demodulation methods corresponding to four-phase phase shift keying, etc.), or by other methods, without limitation. The second preset model can be various linear or nonlinear network models, such as CNN, DNN, or RNN, or other models, without limitation. Optionally, the first sample dataset and the second sample dataset can include different sample data, and / or can include the same sample data, without limitation. Optionally, the second preset model used for training and the first preset model used for training can be the same model or different models, without limitation. Optionally, the second AI model can be the same as the first AI model, or it can be a different AI model; there are no restrictions on this.

[0155] Optionally, the network device may not have a dedicated demodulator for demodulation deployed, but may have a first proxy model deployed. This demodulator can demodulate the data to be demodulated based on a second AI model to obtain the demodulation result. For example, the demodulator can be a second AI model used for demodulation, or it can have a built-in second AI model, or it can simulate or invoke a second AI model, etc., without specific limitations.

[0156] A surrogate model is used to simulate the original model; in other words, the surrogate model has the same or similar functions as the original model, and can achieve the same or similar functions as the original model based on the surrogate model.

[0157] In this context, the proxy model has the same input and / or output as the original model. Same input means that the data structure and / or data meaning of the input to the proxy model and the input to the original model are the same. Same output means that the data structure and / or data meaning of the output of the proxy model and the original model are the same. Therefore, it can be understood that the proxy model and the original model have the same function.

[0158] Optionally, the surrogate model may differ from the original model in at least one of the following aspects: model structure, number of layers, and parameter values ​​for each layer. Due to these differences, the performance of the surrogate model and the original model may also differ, for example, but not limited to, the precision of the output data and the time taken for the model to derive the output from the input. Therefore, it can be understood that the surrogate model and the original model have similar functionalities.

[0159] Optionally, the proxy model is identical to the original model in several of the aforementioned aspects. Therefore, it can be understood that the proxy model and the original model have the same functionality.

[0160] It should be noted that the surrogate model and the original model may differ or be the same in other aspects, giving the surrogate model the same or similar functionality as the original model; this is not limited here. Based on the fact that the surrogate model and the original model have the same or similar functionality, the surrogate model can simulate the original model.

[0161] The first proxy model and the second AI model have the same or similar demodulation functions and can be used to demodulate the data to be demodulated, thereby simulating the demodulation function of the second AI model. In other words, the second AI model is the original model of the first proxy model, and the first proxy model is used to simulate the second AI model.

[0162] In some embodiments, the first agent model can be an AI model. Optionally, the first agent model and the second AI model can be the same AI model. Optionally, the first agent model and the second AI model can be different AI models.

[0163] In other embodiments, the first surrogate model may also be other models, such as, but not limited to, a polynomial model, a radial basis function model, a kriging model, etc., without limitation.

[0164] Optionally, the first proxy model can be a software unit and / or hardware unit deployed in a network device to implement the aforementioned simulation function. Optionally, the first proxy model can be a chip, chip module, etc.

[0165] Optionally, the demodulator can be deployed in a terminal device that can communicate with network devices. Thus, it can be understood that the first agent model is used to simulate a second AI model deployed on the terminal device side for demodulation.

[0166] Optionally, the demodulator can be a decoder for decoding. The network device can use the first proxy model to perform decoding processing on the data to be demodulated based on the second AI model. The network device can also perform other processing on the decoded data, such as LDPC decoding processing, without limitation.

[0167] Network devices can use bits, code blocks (CBs), code block groups (CBGs), and / or transport blocks (TBs) as data units for modulation or demodulation, or other data units for modulation or demodulation; there are no restrictions on this. Each code block group includes at least one code block, and different code block groups can include different numbers or the same number of code blocks. In other words, the network device can modulate each bit of the data to be modulated; or the network device can divide the data to be modulated into several code blocks and modulate each code block; or the network device can divide the data to be modulated into several code block groups and modulate each code block group; or the network device can divide the data to be modulated into several transport blocks and modulate each transport block. Accordingly, when demodulating, the network device can demodulate the data corresponding to the aforementioned data units in the data to be demodulated.

[0168] In some possible implementations, the demodulation result may include demodulation errors and / or successful demodulation results. Optionally, the demodulation result may include some or all of the data units in the data to be demodulated that were demodulated incorrectly or correctly; in other words, the demodulation result may indicate the data units corresponding to demodulation errors and / or successful demodulation.

[0169] In some embodiments, the demodulation result may include a demodulation error in the first code block or the first code block group. Here, the data to be modulated includes the first code block and / or the data to be demodulated includes the first code block, where the first code block is a code block with a demodulation error, meaning that at least one data unit in the first code block is demodulated incorrectly; and / or, the data to be modulated includes the first code block group and / or the data to be demodulated includes the first code block group, where the first code block group is a code block group with a demodulation error, meaning that at least one data unit in the first code block group is demodulated incorrectly. A demodulation error in a data unit means that the data obtained by demodulating the data unit is different from the original data of the data unit before modulation. Conversely, a correct demodulation of a data unit means that the data obtained by demodulating the data unit is the same as the original data of the data unit before modulation.

[0170] For example, the first code block includes 64 bits of data. If 32 bits are demodulated incorrectly, but the other 32 bits are demodulated correctly, then the first code block is demodulated incorrectly. As another example, the first code block group includes 64 code blocks, each containing 128 bits. If 8 code blocks are demodulated incorrectly, but the remaining code blocks are demodulated correctly, then the first code block group is demodulated incorrectly. Alternatively, if 8 bits in the first code block group are demodulated incorrectly, but the remaining bits are demodulated correctly, then the first code block group is demodulated incorrectly. The understanding of code block demodulation errors is explained in the previous examples and will not be repeated here.

[0171] In other embodiments, the demodulation result may further include correct demodulation of the second code block or the second code block group. Wherein, the data to be modulated includes the second code block and / or the data to be demodulated includes the second code block; correct demodulation of the second code block may mean correct demodulation of all data units in the second code block, in other words, the second code block is a correctly demodulated code block in the data to be modulated and / or the data to be demodulated; and / or, the data to be modulated includes the second code block group and / or the data to be demodulated includes the second code block group; correct demodulation of the second code block group may mean correct demodulation of all data units in the second code block group, in other words, the second code block group is a correctly demodulated code block group in the data to be modulated and / or the data to be demodulated.

[0172] In some other possible implementations, the demodulation result may include the number of demodulated error and / or correctly demodulated data units. For example, the demodulation result may include 32 code block demodulation errors and / or 96 code block demodulations correctly. Yet another example is that the demodulation result may include 16 code block groups demodulation errors and / or 48 code block groups demodulations correctly.

[0173] In some other possible implementations, the demodulation result may include some or all of the demodulated data. Specifically, the demodulated partial data may include data obtained by demodulating the first code block or the first code block group, and / or the demodulated partial data may include data obtained by demodulating the second code block or the second code block group. For a description of the first code block, the first code block group, the second code block, and the second code block group, please refer to the aforementioned implementations, and they will not be repeated here.

[0174] In this embodiment of the disclosure, the demodulation result may also include other demodulation-related information or data, and there are no limitations on this. Furthermore, the demodulation result may include one or more types of information or data. For example, the demodulation result may include the result of demodulation errors. As another example, the demodulation result may include the result of demodulation errors and the number of data units with demodulation errors, etc.

[0175] 1002. Based on the demodulation results, the network device obtains the performance monitoring results of the first AI model and / or the second AI model.

[0176] The performance monitoring results of the first AI model and / or the second AI model may be used, but are not limited to, to represent at least one of the following performance characteristics:

[0177] The first type of performance refers to the performance of the first AI model in modulation.

[0178] The second type of performance is the performance of the second AI model in demodulation.

[0179] The third type of performance is the combined performance of the first and second AI models.

[0180] The combined performance of the first AI model and the second AI model refers to the combined performance of the first AI model for modulation and the second AI model for demodulation.

[0181] In a specific example, the performance monitoring result may include at least one of the above three performance characteristics failing to meet the modulation and demodulation performance requirements of the communication system; and / or may include at least one of the above three performance characteristics meeting the modulation and demodulation performance requirements of the communication system.

[0182] Optionally, performance monitoring results can be categorized, with different levels representing the relative merits of at least one of the three performance metrics mentioned above. For example, a first-level performance monitoring result is superior to a second-level result and is more suitable for application in communication systems to improve communication performance. This is merely an example and does not constitute a limitation; performance monitoring results can also take other forms.

[0183] In some possible implementations, network devices can obtain information such as whether the demodulated data obtained is the same as the data to be modulated, whether it is different, and / or the degree of difference, based on the demodulation results, thereby obtaining the performance monitoring results of the first AI model and / or the second AI model.

[0184] Optionally, if the demodulated data is the same as the modulated data, it indicates that at least one of the above three performance characteristics meets the performance requirements of the communication system for modulation and demodulation, or represents a second-level performance monitoring result indicating good communication performance. For example, if the demodulation result includes 128 correctly demodulated data blocks, and the modulated data contains only 128 code blocks and no other data, then the demodulated data is determined to be the same as the modulated data. Another example is that the demodulation result includes all the demodulated data, and the demodulated data is determined to be the same as the modulated data after comparison.

[0185] Optionally, if the demodulated data differs from the modulated data, it indicates that at least one of the three performance metrics mentioned above fails to meet the modulation and demodulation performance requirements of the communication system, or it represents a second-level performance monitoring result indicating poor communication performance. For example, if the demodulation result includes demodulation errors in the first code block or the first code block group, then it is determined that the demodulated data differs from the modulated data. Another example is if the demodulation result includes 32 code blocks with demodulated errors, then it is determined that the demodulated data differs from the modulated data. Yet another example is if the demodulation result includes all the demodulated data, and a comparison between the demodulated data and the modulated data confirms that they differ.

[0186] Optionally, if the demodulated data differs from the data to be modulated, the network device can obtain the performance monitoring results of the first AI model and / or the second AI model based on the degree of difference between the demodulated data and the data to be modulated. For example, if the demodulation result includes 32 code block demodulation errors and 96 code block demodulation successes, the probability of demodulation errors is 25%, which is greater than the performance threshold of 10%. Therefore, it is determined that at least one of the above three performance characteristics fails to meet the modulation and demodulation performance requirements of the communication system. As another example, the demodulation result may include all the demodulated data. By comparing all the demodulated data with the data to be modulated, and determining that the proportion of data units with demodulation errors in the data to be modulated is 25%, which is greater than the performance threshold of 10%, it is determined that the performance monitoring result of at least one of the above three performance characteristics is a second-level performance monitoring result indicating poor communication performance.

[0187] In some other possible implementations, the network device can determine the demodulation performance information of the data to be demodulated based on the demodulation results; and obtain the performance monitoring results of the first AI model and / or the second AI model based on the demodulation performance information of the data to be demodulated.

[0188] The demodulation performance information of the data to be demodulated may be used, but is not limited to, to represent the modulation performance of the data to be modulated and / or the demodulation performance of the data to be demodulated, and there are no restrictions on this.

[0189] Optionally, the demodulation performance information of the data to be demodulated may include, but is not limited to, at least one of the following: bit error rate (BER) and block error rate (BRR). Both BER and BRR can be used to represent the proportion of data units that have data transmission errors and / or demodulation errors. In a specific example, the BER of the data to be demodulated may be the proportion of data units (e.g., bits) that have demodulated errors after demodulation of the data to be demodulated. The BRR of the data to be demodulated is the proportion of data units (e.g., code blocks) that have demodulated errors after demodulation of the data to be demodulated. In other words, the BER and BRR of the data to be demodulated are the proportion of data units in the demodulated data that differ from the modulated data. The above are merely examples of the BER and BRR of the data to be demodulated and do not constitute a limitation; the BER and BRR of the data to be demodulated may also have other meanings or be obtained through other means.

[0190] Optionally, demodulation performance information can be compared with its corresponding performance threshold and / or performance threshold range. Based on the comparison results, performance monitoring results of the first AI model and / or the second AI model can be obtained. The performance threshold and / or performance threshold range corresponding to different demodulation performance information can be different or the same. The performance threshold and / or performance threshold range can be, but is not limited to, predefined, obtained based on network configuration information, or indicated based on downlink control information (DCI).

[0191] For example, if the bit error rate (BER) of the demodulated data is 25%, which is greater than the BER threshold of 10%, then at least one of the above three performance characteristics meets the performance requirements of the communication system for modulation and demodulation. As another example, if the block error rate (BOR) of the demodulated data is 6%, which is less than the BOR threshold of 10%, then at least one of the above three performance characteristics meets the performance requirements of the communication system for modulation and demodulation.

[0192] It should be noted that the above is merely an example of how to obtain performance monitoring results for the first AI model and / or the second AI model based on demodulation results, and does not constitute a limitation. Other methods can also be used to obtain performance monitoring results. Steps 1001-1002 can be performed by network devices or by other computer devices besides network devices, such as terminal devices or servers, without limitation. The method shown in Figure 3 can also include more or fewer steps to obtain performance monitoring results for the AI ​​model, without limitation.

[0193] In this embodiment of the disclosure, the network device simulates a second AI model based on a first proxy model to demodulate the data to be modulated corresponding to the first data, and obtains the demodulation result. The modulation of the first data is based on the first AI model. Based on the demodulation result, the network device obtains the performance monitoring results of the first AI model and / or the second AI model. This enables effective monitoring of the modulation and / or demodulation performance of the AI ​​model without the network device interacting with the terminal device, which is beneficial for optimizing the performance of the AI ​​model and improving communication performance.

[0194] In another alternative implementation, the network device can obtain the performance monitoring results of the AI ​​model based on the demodulation performance information of the demodulated data corresponding to the first data.

[0195] Optionally, the network device sends first data to the terminal device. Correspondingly, the terminal device receives the first data from the network device. The modulation of the first data is based on a first AI model; the performance monitoring results of the first AI model and / or the second AI model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the first data; the demodulation performance information of the data to be demodulated is obtained based on the demodulation result of the data to be demodulated, and the second AI model is used for demodulation.

[0196] Please refer to Figure 4, which is a flowchart illustrating another communication processing method provided in this embodiment of the present disclosure. This method may include, but is not limited to, the following steps:

[0197] In step 2001, the network device sends first data to the terminal device, the modulation of which is based on a first AI model. Correspondingly, the terminal device receives data to be demodulated from the network device.

[0198] For downlink transmission, the data to be demodulated can be carried through a downlink channel, such as a PDSCH. The description of the data to be demodulated corresponding to the first data and the modulation of the first data can be found in step 1001 of the method embodiment shown in Figure 3, and will not be repeated here.

[0199] In 2002, the terminal device demodulated the data to be demodulated corresponding to the first data based on the second AI model to obtain the demodulation result.

[0200] Optionally, the terminal device may be deployed with a demodulator for demodulation, which can demodulate the data to be demodulated based on a second AI model. The terminal device may use bits, code blocks, code block groups, and / or transport blocks as data units for modulation or demodulation, or other data units for modulation or demodulation; there are no restrictions on this. Each code block group includes at least one code block, and different code block groups may include different numbers or the same number of code blocks. In other words, the network device may modulate each bit of the data to be modulated; or the network device may divide the data to be modulated into several code blocks and modulate each code block; or the network device may divide the data to be modulated into several code block groups and modulate each code block group; or the network device may divide the data to be modulated into several transport blocks and modulate each transport block. Accordingly, when performing demodulation, the terminal device may demodulate the data corresponding to the aforementioned data units in the data to be demodulated.

[0201] Optionally, the second AI model can be the same as the first AI model, or it can be a different AI model; there is no restriction on this.

[0202] For a description of the second AI model and the demodulation results, please refer to the description of step 1001 in the method embodiment shown in Figure 3, which will not be repeated here.

[0203] In 2003, the terminal device sends the demodulation result or demodulation performance information of the data to be demodulated to the network device. Correspondingly, the network device receives the demodulation result or demodulation performance information of the data to be demodulated from the terminal device.

[0204] In a specific example, the terminal device sends a demodulation result to the network device, and the network device can obtain the demodulation performance information of the data to be demodulated based on the received demodulation result; and / or, the terminal device sends the demodulation performance information of the data to be demodulated to the network device, and before sending the demodulation performance information, the terminal device can obtain the demodulation performance information of the data to be demodulated based on the demodulation result.

[0205] The demodulation performance information of the data to be demodulated may, but is not limited to, being used to represent the modulation performance of the data to be modulated, the transmission performance of the first data, and / or the demodulation performance of the data to be demodulated, etc., and there are no restrictions on this.

[0206] Optionally, the demodulation performance information of the data to be demodulated includes, but is not limited to, at least one of the following: bit error rate, block error rate, probability of Hybrid Automatic Repeat Request Non-Acknowledge (HARQ NACK), and Hybrid Automatic Repeat Request Acknowledge (HARQ ACK) information. The descriptions of the bit error rate and block error rate of the data to be demodulated are given in step 1002 of the method embodiment shown in Figure 3, and will not be repeated here.

[0207] During the demodulation process, in response to a demodulation error in any data unit, the terminal device generates a HARQ NACK for that data unit and sends it to the network device. Conversely, in response to a successful demodulation of any data unit, it generates a HARQ ACK for that data unit and sends it to the network device. Thus, the terminal device can determine the probability of obtaining the HARQ NACK for the data to be demodulated based on the proportion of the generated HARQ NACKs among all generated HARQ NACKs and HARQ ACKs, and / or the network device can determine the probability of obtaining the HARQ NACK for the data to be demodulated based on the proportion of the received HARQ NACKs among all received HARQ NACKs and HARQ ACKs. In other words, the probability of obtaining the HARQ NACK for the data to be demodulated can refer to the proportion of the HARQ NACKs received as feedback from the demodulated data to all HARQ NACKs and HARQ ACKs, or it can be viewed as the probability of obtaining the HARQ NACK for the demodulated data obtained from the demodulation process. For example, if the terminal device fails to demodulate any first code block, it generates a HARQ NACK for that first code block and sends it to the network device. If the terminal device successfully demodulates any second code block, it generates a HARQ ACK for that second code block and sends it to the network device. For instance, if 32 HARQ NACKs and 96 HARQ ACKs are generated and sent after demodulation of the data to be demodulated, the probability of obtaining a HARQ NACK for the data to be demodulated is 25%.

[0208] HARQ ACK information may include HARQ ACK generated by the terminal device, and / or HARQ ACK information may include HARQ ACK received by the network device.

[0209] In 2004, network devices obtained performance monitoring results of the first AI model and / or the second AI model based on the demodulation performance information of the data to be demodulated.

[0210] In some possible implementations, the demodulation performance information of the data to be demodulated can be compared with its corresponding performance threshold and / or performance threshold range. Based on the comparison result, the performance monitoring results of the first AI model and / or the second AI model can be obtained. The performance threshold and / or performance threshold range corresponding to different demodulation performance information can be different or the same. The performance threshold and / or performance threshold range can be, but is not limited to, predefined, obtained based on network configuration information, or based on DCI indications.

[0211] For example, if the bit error rate (BER) of the demodulated data is 25%, which is greater than the BER threshold of 10%, then at least one of the above three performance characteristics fails to meet the modulation and demodulation performance requirements of the communication system. As another example, if the block error rate (BOR) of the demodulated data is 6%, which is less than the BOR threshold of 10%, then at least one of the above three performance characteristics meets the modulation and demodulation performance requirements of the communication system. As yet another example, if the HARQ NACK probability of the demodulated data is 4%, which is less than the probability threshold of 8%, then the performance monitoring result of at least one of the above three performance characteristics is determined to be the first-level performance monitoring result with optimal communication performance. As yet another example, if the demodulated data has 5 HARQ NACKs, exceeding the HARQ NACK threshold of 3, then the performance monitoring result of the first AI model and / or the second AI model is determined to be the second-level performance monitoring result with poor communication performance. For a description of at least one of the above three performance characteristics, please refer to the description of step 1002 in the method embodiment shown in Figure 3, which will not be repeated here.

[0212] It should be noted that the above is merely an example of how to obtain performance monitoring results of the first AI model and / or the second AI model based on demodulation performance information of the data to be demodulated, and does not constitute a limitation. Other methods can also be used to obtain performance monitoring results. The method shown in Figure 4 can also include more or fewer steps to obtain performance monitoring results of the AI ​​model, and there are no restrictions on this.

[0213] In this embodiment of the present disclosure, the network device sends the data to be demodulated corresponding to the first data to the terminal device. The modulation of the first data is based on the first AI model. The terminal device demodulates the data to be demodulated corresponding to the first data based on the second AI model to obtain the demodulation result, and returns the demodulation result or the transmission performance information of the data to be demodulated to the network device. Based on the demodulation performance information of the data to be demodulated, the network device obtains the performance monitoring results of the first AI model and / or the second AI model, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving the communication performance.

[0214] In another alternative implementation, the network device may obtain the performance monitoring results of the AI ​​model based on the second data.

[0215] Optionally, the network device sends first data to the terminal device, the modulation of which is based on a first AI model. Correspondingly, the terminal device receives the first data from the network device.

[0216] The terminal device sends second data to the network device. This second data includes partial or complete data obtained by demodulating the data to be demodulated corresponding to the first data based on a second AI model. Correspondingly, the network device receives the second data from the terminal device. The performance monitoring results of the first and / or second AI models are obtained based on the second data.

[0217] Please refer to Figure 5, which is a flowchart illustrating another communication processing method provided in this embodiment of the present disclosure. This method may include, but is not limited to, the following steps:

[0218] 3001, The network device sends first data to the terminal device, the modulation of the first data being based on a first AI model. Correspondingly, the terminal device receives the first data from the network device.

[0219] Step 3001 can be referred to the description of step 2001 in the method embodiment shown in Figure 4. The demodulated data corresponding to the first data and the modulation of the first data can be referred to the description of step 1001 in the method embodiment shown in Figure 3, and will not be repeated here.

[0220] 3002, The terminal device demodulates the data to be demodulated corresponding to the first data based on the second AI model.

[0221] Optionally, the second AI model can be the same as the first AI model, or it can be a different AI model; there is no restriction on this.

[0222] For a description of step 3002, please refer to the description of step 2002 in the method embodiment shown in Figure 4. For a description of the second AI model and demodulation results, please refer to the description of step 1001 in the method embodiment shown in Figure 3. They will not be repeated here.

[0223] 3003, the terminal device sends second data to the network device. The second data includes partial or complete data obtained by demodulating the data to be demodulated corresponding to the first data based on the second AI model. Accordingly, the network device receives the second data from the terminal device.

[0224] The second data includes part or all of the data obtained in step 3002 by demodulating the data to be demodulated corresponding to the first data based on the second AI model by the terminal device. Optionally, the transmission of the first data and / or the transmission of the second data can be implemented through, but is not limited to, dynamic scheduling, semi-static scheduling, and / or continuous scheduling. The transmission of the first data and the transmission of the second data can be scheduled in the same way or in different ways, without limitation.

[0225] Optionally, the second data may be different from the first data, and / or may be the same data, without limitation.

[0226] In a specific example, the transmission of the second data may be scheduled via a DCI, but is not limited to. Optionally, the DCI used to schedule the transmission of the second data (e.g., the first DCI or the second DCI mentioned below) may be sent to the terminal device at the same time as the first data is sent; or, sent to the terminal device before the first data is sent; or, sent to the terminal device after the first data is sent and before the second data is received.

[0227] In some embodiments, the transmission of first data and the transmission of second data can be scheduled using the same DCI. Before performing step 3001, the network device may send a first DCI to the terminal device. The first DCI is used to schedule the transmission of first data and the transmission of second data. The transmission of first data includes transmitting first data through the transmission resources scheduled by the first DCI, and the transmission of second data includes transmitting second data through the transmission resources scheduled by the first DCI. In other words, the transmission resources scheduled by the first DCI include transmission resources for transmitting first data and transmission resources for transmitting second data. For example, the first DCI may be a downlink (DL) DCI for scheduling downlink transmission resources (e.g., PDSCH), which further schedules uplink transmission resources (e.g., Physical Uplink Shared Channel (PUSCH)) to transmit first data through downlink transmission resources and second data through uplink transmission resources.

[0228] The first DCI may schedule the transmission of the first data and the transmission of the second data in at least one of the following ways:

[0229] Method 1, the first DCI includes scheduling resources for first data or scheduling resources for second data, and the scheduling resources for first data and the scheduling resources for second data satisfy at least one of the following relationships:

[0230] (1) The first data and the second data are transmitted based on the same frequency domain resources; and / or,

[0231] (2) The first data and the second data are transmitted based on symbol resources at the same location in different time slots; and / or,

[0232] (3) There is an offset of k time slots between the time slot for transmitting the second data and the time slot for transmitting the first data.

[0233] Where k is an integer greater than or equal to 1, and k can be predefined, obtained based on network configuration information, or based on the indication of the first DCI. For example, k can be predefined as 4; another example is that the network device sends configuration information to the terminal device, and this configuration information carries the specific value of k; yet another example is that the first DCI carries the specific value of k. These are just examples and do not constitute a limitation, and k can also be obtained in other ways.

[0234] Optionally, the relationship between the scheduling resources of the first data and the scheduling resources of the second data may be, but is not limited to, predefined, obtained through network configuration information or indicated by DCI (e.g., the first DCI), and there are no restrictions on this.

[0235] For example, the first DCI includes the scheduling resources of the first data, that is, the transmission resources of the first data. Based on the relationship between the scheduling resources of the first data and the second data, and the scheduling resources of the first data, the scheduling resources of the second data are determined, that is, the transmission resources of the second data are determined. For example, the scheduling resources of the first data include Resource Block (RB)i, symbol j, and time slot m. The scheduling resources of the second data are determined to include RB i, symbol j, and time slot m+k. In other words, the first data can be carried by symbol j in RB i and time slot m, and the second data can be carried by symbol j in RB i and time slot m+k. i, j, and m are integers greater than or equal to 0.

[0236] For example, the first DCI includes the scheduling resources for the second data. Based on the relationship between the scheduling resources of the first and second data, and the scheduling resources of the second data, the scheduling resources for the first data are determined. For instance, the scheduling resources for the second data include RB i, symbol j, and time slot n. The scheduling resources for the first data are determined to include RB i, symbol j, and time slot nk. In other words, the second data can be carried by symbol j in RB i and time slot n, and the first data can be carried by symbol j in RB i and time slot nk, where n is an integer greater than or equal to 0.

[0237] Method 2, the first DCI includes scheduling resources for the first data and scheduling resources for the second data.

[0238] For example, the scheduling resources of the first data included in the first DCI, that is, the transmission resources of the first data, are RB i, symbol j and time slot m, and the scheduling resources of the second data included in the first DCI, that is, the transmission resources of the second data, are RB p, symbol q and time slot n, where p and q are integers greater than or equal to 0.

[0239] In other embodiments, the transmission of second data is scheduled via a second DCI, but the second DCI is not used to schedule the transmission of first data. For example, the transmission of first data and the transmission of second data are scheduled separately via different DCIs. In a specific example, the network device sends a second DCI to the terminal device before performing step 3003. The second DCI is used to schedule the transmission of second data, which includes transmitting the second data through the transmission resources scheduled by the second DCI. In other words, the transmission resources scheduled by the second DCI include transmission resources used for transmitting the second data. For example, the second DCI may be an uplink (UL) DCI used to schedule uplink transmission resources (e.g., PUSCH) to transmit the second data through those uplink transmission resources.

[0240] It should be noted that the above is only an example of the scheduling method for the transmission of the first data and the transmission of the second data, and does not constitute a limitation. Other scheduling methods can also be used.

[0241] Optionally, the second DCI and the first DCI can be the same DCI, or they can be different DCIs; there is no restriction on this.

[0242] In one possible implementation, the second DCI further includes first indication information, which indicates the second data to be transmitted via the transmission resources scheduled by the second DCI. In other words, the first indication information indicates which second data the terminal device sends to the network device.

[0243] Optionally, the first indication information can indicate the correspondence between the second data and the data to be demodulated. In other words, the first indication information can indicate which data in the data to be demodulated, either partially or completely, are sent by the terminal device to the network device; that is, the first indication information can indicate the data before demodulation corresponding to the second data sent by the terminal device to the network device.

[0244] Optionally, the first indication information may indicate the correspondence between the second data and the first data. In other words, the first indication may indicate to the network device which data in the first data corresponds to which data to be demodulated, resulting in partial or complete data. In a specific example, the second data to be sent can be determined based on the correspondence between the second data and the first data, as well as the correspondence between the first data and the data to be demodulated.

[0245] The first indication information indicates that the second data is transmitted through the transmission resources scheduled by the second DCI, and may include, but is not limited to, at least one of the following three indication methods:

[0246] In the first method, the first indication information includes the first downlink hybrid automatic repeat request process ID (DL HARQ process ID), and the second data includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the first DL HARQ process ID.

[0247] One approach is to use a Hybrid Automatic Repeat Request (HARQ) retransmission mechanism. This involves using a stop-and-wait protocol to send data, with multiple threads (also known as HARQ processes) transmitting in parallel. For example, on thread 1, the network device sends data (i.e., PDSCH), then pauses to wait for a response from the terminal device confirming successful reception, before continuing to send the next data (i.e., PDSCH). Many threads like thread 1 can transmit in parallel, such as thread 1, thread 2, thread 3, and so on. The HARQ process ID (also called the HARQ process number) is used to identify different HARQ processes. The DL HARQ process ID identifies the downlink HARQ process, and the demodulated data corresponding to the DL HARQ process ID refers to the data transmitted on the HARQ process identified by that DL HARQ process ID. The demodulated data corresponding to the data transmitted on the HARQ process identified by the DL HARQ process ID includes the data transmitted on the HARQ process identified by the DL HARQ process ID, or data obtained by performing a second processing on the data transmitted on the HARQ process identified by the DL HARQ process ID. A description of the second processing is given in the method embodiment shown in Figure 3, and will not be repeated here.

[0248] The first data may include data transmitted on at least one HARQ process identified by a DL HARQ process ID. Based on the first indication information, the terminal device executes a process to demodulate part or all of the data obtained by demodulating the data to be demodulated corresponding to the first DL HARQ process ID, and transmits it to the network device through the transmission resources scheduled by the second DCI. For example, the first data includes data transmitted on HARQ processes with DL HARQ process IDs of 0, 1, 2, 3, or 4; the first indication information includes a first DL HARQ process ID of 2 or indicates a first DL HARQ process ID of 2; the terminal device demodulates the data to be demodulated corresponding to the data transmitted on the HARQ process with the first DL HARQ process ID of 2 according to the first indication information, and transmits part or all of the data obtained by demodulating this data through the transmission resources scheduled by the second DCI.

[0249] In the second method, the first indication information includes the uplink hybrid automatic repeat request process identifier (UL HARQ process ID), and the second data includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the second DL HARQ process ID. The second DL HARQ process ID has the same value as the UL HARQ process ID.

[0250] The UL HARQ process ID is used to identify the HARQ process of the uplink transmission. The first data may include data transmitted on at least one HARQ process identified by a DL HARQ process ID. Based on the first indication information, the terminal device executes the demodulation of part or all of the data to be demodulated corresponding to the second DL HARQ process ID, and sends the demodulated part or all of the data through the transmission resources scheduled by the second DCI. The second DL HARQ process ID has the same value as the UL HARQ process ID, so the second DL HARQ process ID can be determined by the UL HARQ process ID. For example, the data to be demodulated includes data transmitted on HARQ processes with DL HARQ process IDs of 0, 1, 2, 3, and 4. The first indication information includes a UL HARQ process ID of 2 or indicates a UL HARQ process ID of 2. The data to be demodulated corresponding to the data transmitted on the second HARQ process with DL HARQ process ID of 2 is demodulated according to the first indication information. Part or all of the data obtained from demodulating these data is sent through the transmission resources scheduled by the second DCI.

[0251] Optionally, the second DL HARQ process ID and the first DL HARQ process ID can be the same DL HARQ process ID, or they can be different DL HARQ process IDs; there is no restriction on this.

[0252] In the third method, the second data includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the third data. The third data is the data in the time domain unit in the first time domain unit that is closest to the time domain unit of the second DCI. The first time domain unit is the time domain unit used to receive the first data.

[0253] The time domain unit can be, but is not limited to, a slot or symbol; there are no restrictions on this. Data in a time domain unit refers to the data received by the terminal device in that time domain unit. The time domain unit of the second DCI refers to the time domain unit used to receive the second DCI. The terminal device receives first data in a first time domain unit, which includes at least one time domain unit. After receiving the first data, the terminal device receives the second DCI in the time domain unit of the second DCI. Based on the scheduling of the second DCI, it determines the transmission resources used to transmit the second data, and also determines the second data to be transmitted based on the first indication information included in the second DCI. In other words, the first indication information indicates that the second data to be transmitted includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the third data. For example, the terminal device receives first data in slots 0, 1, 2, and 3, and receives second DCI in slot 4. Slot 3 is the closest time-domain unit to slot 4. Therefore, based on the first indication information, the transmission resources scheduled through the second DCI are determined, and partial or complete data obtained by demodulating the first data (i.e., the third data) received in slot 3 are transmitted. It is evident that the first data includes the third data. The data to be demodulated corresponding to the third data can be the third data itself, or data obtained after performing a second processing on the third data. A description of the second processing can be found in the description of step 1001 in the method embodiment shown in Figure 3, and will not be repeated here.

[0254] The above are merely examples of the indication methods for the first indication information and do not constitute a limitation. The first indication information may have other indication methods.

[0255] Additionally, the demodulated data includes data obtained by the terminal device demodulating the first code block, where the data to be demodulated includes the first code block, which is a code block with demodulation errors; and / or, the demodulated data includes data obtained by demodulating the first code block group, where the data to be demodulated includes the first code block group, which is a code block group with demodulation errors. For example, if the terminal device uses code blocks as the data unit for demodulation, and performs demodulation processing on the data to be demodulated, and finds a demodulation error in the first code block of the data to be demodulated, then the second data includes the data obtained by demodulating the first code block. As another example, if the terminal device uses code block groups as the data unit for demodulation, and performs demodulation processing on the data to be demodulated, and finds a demodulation error in the first code block group of the data to be demodulated, then the second data includes the data obtained by demodulating the first code block group. For a description of demodulation errors in the first code block and the first code block group, please refer to the description of step 1001 in the method embodiment shown in Figure 3, which will not be repeated here.

[0256] In some embodiments, the terminal device may also send second indication information to the network device. The second indication information indicates which code block or group of code blocks in the demodulated partial data corresponds to. In other words, the second indication information indicates which code blocks or groups of code blocks in the data to be demodulated and / or modulated are included in the second data sent by the terminal device to the network device. For example, the second indication information may include, but is not limited to, an index of the first code block and / or an index of the first code block group; wherein the index of the code block is used to identify the code block to distinguish different code blocks, and the index of the code block group is used to identify the code block group to distinguish different code block groups. Optionally, the second indication information may be carried using the same transmission resources as the second data. For example, the second indication information may be packaged with the second data and transmitted through the transmission resources scheduled by the second DCI; or, the second indication information may be carried using different transmission resources than the second data. For example, the second indication information may be transmitted to the network device through the transmission resources scheduled by another DCI.

[0257] 3004. The network device obtains the performance monitoring results of the first AI model and / or the second AI model based on the second data.

[0258] For a description of the performance monitoring results of the first AI model and / or the second AI model, please refer to the description of step 1002 in the method embodiment shown in Figure 3, which will not be repeated here.

[0259] In some possible implementations, the network device can compare the second data with the data to be modulated to obtain information such as whether the second data and the data to be modulated are the same, different, and / or have different degrees of difference, thereby obtaining the performance monitoring results of the first AI model and / or the second AI model.

[0260] For example, the second data includes all the data obtained from demodulation. The second data is the same as the data to be modulated, indicating that at least one of the above three performance characteristics can meet the performance requirements of the communication system for modulation and demodulation.

[0261] As another example, the second data includes all the data obtained from demodulation. If the second data differs from the data to be modulated, it indicates that at least one of the above three performance characteristics cannot meet the performance requirements of the communication system for modulation and demodulation.

[0262] For another example, the second data includes all the demodulated data. Since the second data differs from the modulated data, the network device can obtain the proportion of the second data that differs from the modulated data out of the total modulated data. Based on this proportion and a performance threshold, the device determines the performance monitoring result of at least one of the three performance metrics. For instance, if the second data differs from the modulated data by 16 bits, and the modulated data is 128 bits, the proportion of the second data differing from the modulated data is 12.5%, which is greater than the performance threshold of 8%. Therefore, it is determined that at least one of the three performance metrics fails to meet the modulation and demodulation performance requirements of the communication system.

[0263] As another example, the second data includes the demodulated portion of the data, indicating that the second data differs from the data to be modulated, thus indicating that at least one of the above three performance characteristics cannot meet the performance requirements of the communication system for modulation and demodulation.

[0264] As another example, the second data includes a portion of the demodulated data. The network device can obtain the ratio between the amount of data in the second data and the amount of data to be modulated. Based on this ratio and a performance threshold, it determines the performance monitoring results of the first AI model and / or the second AI model. For example, if the amount of data in the second data is 16 bits and the amount of data to be modulated is 128 bits, the ratio between the amount of data in the second data and the amount of data to be modulated is 12.5%, which is greater than the performance threshold of 8%. Therefore, it is determined that at least one of the above three performance metrics fails to meet the modulation and demodulation performance requirements of the communication system.

[0265] It should be noted that the description of at least one of the above three performance characteristics can be found in step 1002 of the method embodiment shown in Figure 3, and will not be repeated here. The above is merely an example of how to obtain the performance monitoring results of the first AI model and / or the second AI model based on the second data, and does not constitute a limitation. Other methods can also be used to obtain the performance monitoring results. The method shown in Figure 5 may also include more or fewer steps to obtain the performance monitoring results of the AI ​​model, and there are no limitations on this.

[0266] In this embodiment of the disclosure, the network device sends the data to be demodulated corresponding to the first data to the terminal device. The modulation of the first data is based on the first AI model. The terminal device demodulates the data to be demodulated corresponding to the first data based on the second AI model and sends the second data to the network device. Based on the second data, the network device can conveniently obtain the performance monitoring results of the first AI model and / or the second AI model, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0267] In another alternative implementation, the network device can modulate the data to be modulated and demodulate the data to be demodulated based on the AI ​​model to obtain the demodulation result; based on the demodulation result, obtain the performance monitoring result of the AI ​​model; wherein, based on the fourth AI model, the data to be demodulated corresponding to the fourth data is demodulated to obtain the demodulation result; the modulation of the fourth data is based on the second proxy model.

[0268] Please refer to Figure 6, which is a flowchart illustrating another communication processing method provided in this embodiment of the present disclosure. This method may include, but is limited to, the following steps:

[0269] 4001. The network device demodulates the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result; the modulation of the fourth data is based on the second proxy model, which is used to simulate the third AI model, and the third AI model is used for modulation.

[0270] Optionally, before performing step 4001, the network device may also modulate the data to be modulated based on the second proxy model. A description of the data to be modulated can be found in the description of step 1001 in the method embodiment shown in Figure 3, and will not be repeated here.

[0271] The third AI model is used for modulation; that is, it is an AI model used to modulate the data to be modulated. The data to be modulated can be input into the third AI model, and the third AI model will output the modulated data based on the input data.

[0272] In some embodiments, the third AI model can be obtained by training the third preset model using a third sample dataset, or it can be obtained in other ways without limitation. The third sample dataset can include data that needs to be modulated during uplink and / or downlink communication, as well as the data obtained after modulation of this data. The data in the third sample dataset can be obtained by collecting data during routine uplink and / or downlink communication, or by modulating a series of sample data using non-AI model-based modulation methods (e.g., but not limited to, 16-orthogonal amplitude modulation, or quadrature phase shift keying, etc.), or by other methods without limitation. The third preset model can be various linear or nonlinear network models, such as CNN, DNN, or RNN, without limitation.

[0273] Optionally, the network device may not have a modulator deployed for modulation, but it may have a second proxy model deployed. Alternatively, the modulator may be deployed in a terminal device communicating with the network device. Therefore, the second proxy model can be understood as simulating a third AI model deployed on the terminal device side for modulation. This modulator can modulate the data to be modulated based on the third AI model. For example, the modulator may be a third AI model for modulation, or it may have a built-in third AI model, or it may simulate or invoke a third AI model, etc., without specific limitations.

[0274] In some embodiments, the modulator may be an encoder for encoding, and the network device may implement encoding processing of the data to be modulated based on a third AI model, based on a second proxy model.

[0275] For a description of the proxy model, please refer to the description of step 1001 in the method embodiment shown in Figure 3, which will not be repeated here.

[0276] Among them, the second proxy model and the third AI model have the same or similar modulation functions, which can be used to modulate the data to be modulated and realize the modulation function of the third AI model. In other words, the third AI model is the original model of the second proxy model, and the second proxy model is used to simulate the third AI model.

[0277] In some embodiments, the second agent model can be an AI model. Optionally, the second agent model and the third AI model can be the same AI model. Optionally, the second agent model and the third AI model can be different AI models.

[0278] In other embodiments, the second surrogate model may also be other models, such as, but not limited to, a polynomial model, a radial basis function model, a kriging model, etc., without limitation.

[0279] Optionally, the second proxy model can be a software unit and / or hardware unit deployed in a network device to implement the aforementioned simulation function. Optionally, the second proxy model can be a chip, chip module, etc.

[0280] Optionally, the second proxy model and the first proxy model can be different proxy models or the same proxy model; there is no restriction on this.

[0281] Optionally, before performing step 4001 and after modulating the data to be modulated based on the second proxy model, a first processing can be performed on the modulated data. A description of the first processing can be found in the description of step 1001 in the method embodiment shown in Figure 3, and will not be repeated here.

[0282] It can be understood that the fourth data can be data obtained by modulating the data to be modulated based on the second surrogate model or the third AI model; or, the fourth data can be data obtained by performing a first process on the data modulated based on the second surrogate model or the third AI model. Therefore, the modulation of the fourth data is based on the second surrogate model or the third AI model.

[0283] Optionally, the data after the first processing can also undergo a second processing. A description of the second processing can be found in the description of step 1001 in the method embodiment shown in Figure 3, and will not be repeated here.

[0284] Here, the data to be demodulated is data that can be demodulated. It can be understood that the data to be demodulated corresponding to the fourth data can be the fourth data; or, the data to be demodulated corresponding to the fourth data can be data obtained by performing the first and / or second processing on data modulated based on the second proxy model or the third AI model.

[0285] For example, the fourth data is data obtained by modulating the data to be modulated based on the second surrogate model or the third AI model, and the corresponding demodulated data can be the fourth data. As another example, the fourth data is data obtained by modulating the data to be modulated based on the second surrogate model or the third AI model, and the corresponding demodulated data can be data obtained by performing a first and a second processing on the data modulated based on the second surrogate model or the third AI model. As yet another example, the fourth data is data obtained by performing a first processing on the data modulated based on the second surrogate model or the third AI model, and the corresponding demodulated data can be data obtained by performing a first and a second processing on the data modulated based on the second surrogate model or the third AI model. The above are merely examples of the fourth data and its corresponding demodulated data, and do not constitute a limitation.

[0286] The fourth AI model is used for demodulation. It is an AI model specifically designed to demodulate the data to be demodulated. Network devices can input the data to be demodulated into the fourth AI model, which then outputs the demodulation result based on the input data.

[0287] In some embodiments, the fourth AI model can be obtained by training the fourth preset model using a fourth sample dataset, or it can be obtained in other ways, without limitation. The fourth sample dataset can include data that needs to be demodulated in uplink and / or downlink communication, as well as the demodulated data. The data in the fourth sample dataset can be obtained by collecting data during daily uplink and / or downlink communication, or by demodulating a series of sample data using non-AI model-based demodulation methods (e.g., but not limited to, demodulation methods corresponding to 16-orthogonal amplitude modulation, or demodulation methods corresponding to four-phase phase shift keying, etc.), or by other methods, without limitation. The fourth preset model can be various linear or nonlinear network models, such as CNN, DNN, or RNN, without limitation. Optionally, the fourth AI model and the third AI model can be different AI models, or they can be the same AI model, without limitation. Optionally, the fourth sample dataset and the third sample dataset can include different sample data, and / or include the same sample data, without limitation. Optionally, the fourth preset model used for training can be the same model as the third preset model used for training, or it can be a different model, without restriction.

[0288] Optionally, the network device may be deployed with a demodulator for demodulation, which can demodulate the data to be demodulated based on the fourth AI model to obtain the demodulation result. For example, the demodulator may be the fourth AI model used for demodulation, or it may have the fourth AI model built in, or it may simulate or call the fourth AI model, etc., without specific restrictions.

[0289] Optionally, the demodulator can be a decoder used for decoding, and the network device can also perform other processing on the decoded data, such as LDPC decoding, without limitation.

[0290] Network devices can use bits, code blocks, code block groups, and / or transport blocks as data units for modulation or demodulation, or they can use other data units for modulation or demodulation without limitation. Each code block group includes at least one code block, and different code block groups can include different numbers or the same number of code blocks. In other words, a network device can modulate each bit of data in the data to be modulated; or it can divide the data to be modulated into several code blocks and modulate each code block; or it can divide the data to be modulated into several code block groups and modulate each code block group; or it can divide the data to be modulated into several transport blocks and modulate each transport block. Accordingly, when demodulating, the network device can demodulate the data in the data to be demodulated that corresponds to the aforementioned data units.

[0291] In some possible implementations, the demodulation result may include demodulation errors and / or successful demodulation results. Optionally, the demodulation result may include some or all data units in the data to be demodulated that were demodulated incorrectly or correctly; in other words, the demodulation result may indicate the data units corresponding to demodulation errors and / or successful demodulation. In some embodiments, the demodulation result may include a demodulation error for the first code block or the first code block group. A description of the first code block and the first code block group can be found in the description of step 1001 in the method embodiment shown in FIG3, and will not be repeated here.

[0292] For example, the first code block includes 64 bits of data. If 32 bits are demodulated incorrectly, but the other 32 bits are demodulated correctly, then the first code block is demodulated incorrectly. As another example, the first code block group includes 64 code blocks, each containing 128 bits. If 8 code blocks are demodulated incorrectly, but the remaining code blocks are demodulated correctly, then the first code block group is demodulated incorrectly. Alternatively, if 8 bits in the first code block group are demodulated incorrectly, but the remaining bits are demodulated correctly, then the first code block group is demodulated incorrectly. The understanding of code block demodulation errors is explained in the previous examples and will not be repeated here.

[0293] In other embodiments, the demodulation result may further include correct demodulation of the second code block or the second code block group. Wherein, the data to be modulated includes the second code block and / or the data to be demodulated includes the second code block; correct demodulation of the second code block may mean correct demodulation of all data units in the second code block, in other words, the second code block is a correctly demodulated code block in the data to be modulated and / or the data to be demodulated; and / or, the data to be modulated includes the second code block group and / or the data to be demodulated includes the second code block group; correct demodulation of the second code block group may mean correct demodulation of all data units in the second code block group, in other words, the second code block group is a correctly demodulated code block group in the data to be modulated and / or the data to be demodulated.

[0294] In some other possible implementations, the demodulation result may include the number of demodulated error and / or correctly demodulated data units. For example, the demodulation result may include 32 code block demodulation errors and / or 96 code block demodulations correctly. Yet another example is that the demodulation result may include 16 code block groups demodulation errors and / or 48 code block groups demodulations correctly.

[0295] In some other possible implementations, the demodulation result may include some or all of the demodulated data. Specifically, the demodulated partial data may include data obtained by demodulating the first code block or the first code block group, and / or the demodulated partial data may include data obtained by demodulating the second code block or the second code block group. For a description of the first code block, the first code block group, the second code block, and the second code block group, please refer to the aforementioned implementations, and they will not be repeated here.

[0296] In this embodiment of the disclosure, the demodulation result may also include other demodulation-related information or data, and there are no limitations on this. Furthermore, the demodulation result may include one or more types of information or data. For example, the demodulation result may include the result of demodulation errors. As another example, the demodulation result may include the result of demodulation errors and the number of data units with demodulation errors, etc.

[0297] 4002. Based on the demodulation results, the network device obtains the performance monitoring results of the third AI model and / or the fourth AI model.

[0298] The performance monitoring results of the third AI model and / or the fourth AI model may be used, but are not limited to, to represent at least one of the following performance characteristics:

[0299] Performance 1, the performance of modulation by the third AI model;

[0300] Performance 2, the performance of demodulation by the fourth AI model;

[0301] Performance 3: The combined performance of the third and fourth AI models.

[0302] The combined performance of the third AI model and the fourth AI model refers to the combined performance of the third AI model for modulation and the fourth AI model for demodulation.

[0303] In a specific example, the performance monitoring result may include at least one of the above three performance characteristics failing to meet the modulation and demodulation performance requirements of the communication system; or it may include at least one of the above three performance characteristics meeting the modulation and demodulation performance requirements of the communication system.

[0304] Optionally, performance monitoring results can be categorized, with different levels representing the performance superiority or inferiority of at least one of the three performance metrics mentioned above. For example, a first-level performance monitoring result is superior to a second-level result and is more suitable for application in communication systems to improve communication performance. This is merely an example and does not constitute a limitation; performance monitoring results can also take other forms.

[0305] In some possible implementations, network devices can obtain information such as whether the demodulated data obtained is the same as the data to be modulated, whether it is different, and / or the degree of difference, based on the demodulation results, thereby obtaining the performance monitoring results of the third AI model and / or the fourth AI model.

[0306] Optionally, the demodulated data being identical to the modulated data indicates that at least one of the three performance characteristics described above meets the modulation and demodulation performance requirements of the communication system. For example, if the demodulation result includes 128 correctly demodulated data blocks, and the modulated data contains only 128 code blocks and no other data, then the demodulated data is determined to be identical to the modulated data. Another example is that the demodulation result includes all the demodulated data, and the demodulated data is determined to be identical to the modulated data after comparison.

[0307] Optionally, if the demodulated data differs from the modulated data, it indicates that at least one of the three performance metrics mentioned above fails to meet the modulation and demodulation performance requirements of the communication system, or it represents a second-level performance monitoring result indicating poor communication performance. For example, if the demodulation result includes demodulation errors in the first code block or first code block group, then it is determined that the demodulated data differs from the modulated data. As another example, if the demodulation result includes 32 code blocks with demodulated errors, then it is determined that the demodulated data differs from the modulated data. As yet another example, if the demodulation result includes all the demodulated data, and a comparison between all the demodulated data and the modulated data confirms that they differ, then the demodulated data may also differ from the modulated data.

[0308] Optionally, if the demodulated data differs from the data to be modulated, the network device can obtain the performance monitoring results of the third AI model and / or the fourth AI model based on the degree of difference between the demodulated data and the data to be modulated. For example, if the demodulation result includes 32 code block demodulation errors and 96 code block demodulation successes, the probability of demodulation errors is 25%, which is greater than the performance threshold of 10%. Therefore, it is determined that at least one of the above three performance characteristics fails to meet the modulation and demodulation performance requirements of the communication system. As another example, the demodulation result may include all the demodulated data. By comparing all the demodulated data with the data to be modulated, and determining that the proportion of data units with demodulation errors in the data to be modulated is 25%, which is greater than the performance threshold of 10%, it is determined that at least one of the above three performance characteristics is a second-level performance monitoring result indicating poor communication performance.

[0309] In some other possible implementations, the network device can determine the demodulation performance information of the data to be demodulated based on the demodulation results; and obtain the performance monitoring results of the third AI model and / or the fourth AI model based on the demodulation performance information of the data to be demodulated.

[0310] The demodulation performance information of the data to be demodulated may be used, but is not limited to, to represent the modulation performance of the data to be modulated and / or the demodulation performance of the data to be demodulated, and there are no restrictions on this.

[0311] Optionally, the demodulation performance information of the data to be demodulated may include, but is not limited to, at least one of the following: bit error rate (BER) and block error rate (BRR). Both BER and BRR can be used to represent the proportion of data units that have transmission errors and / or demodulation errors. In a specific example, the BER of the data to be demodulated may be the proportion of data units (e.g., bits) that have demodulated errors after demodulation among all data units. The BRR of the data to be demodulated is the proportion of data units (e.g., code blocks) that have demodulated errors after demodulation among all data units. In other words, the BER and BRR of the data to be demodulated are the proportion of data units in the demodulated data that differ from the modulated data. The above are merely examples of the BER and BRR of the data to be demodulated and do not constitute a limitation; the BER and BRR of the data to be demodulated may have other meanings or be obtained through other means.

[0312] Optionally, demodulation performance information can be compared with its corresponding performance thresholds and / or performance threshold ranges. Based on the comparison results, performance monitoring results of the third AI model and / or the fourth AI model can be obtained. The performance thresholds and / or performance threshold ranges corresponding to different demodulation performance information can be different or the same. The performance thresholds and / or performance threshold ranges can be, but are not limited to, predefined, obtained based on network configuration information, or based on DCI indications.

[0313] For example, if the bit error rate (BER) of the demodulated data is 25%, which is greater than the BER threshold of 10%, then at least one of the above three performance characteristics fails to meet the modulation and demodulation performance requirements of the communication system. As another example, if the block error rate (BOR) of the demodulated data is 6%, which is less than the BOR threshold of 10%, then at least one of the above three performance characteristics meets the modulation and demodulation performance requirements of the communication system.

[0314] It should be noted that the above is merely an example of how to obtain performance monitoring results for the third and / or fourth AI models based on demodulation results, and does not constitute a limitation. Other methods can also be used to obtain performance monitoring results. Steps 4001-4002 can be performed by network devices or by other computer devices besides network devices, such as terminal devices or servers, without limitation. The method shown in Figure 6 can also include more or fewer steps to obtain performance monitoring results for the AI ​​models, without limitation.

[0315] In this embodiment, the network device demodulates the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result. The modulation of the fourth data is based on the second proxy model, which is used to simulate the third AI model. The third AI model is used for modulation. Based on the demodulation result, the performance monitoring results of the third AI model and / or the fourth AI model are obtained. This enables effective monitoring of the performance of the AI ​​model in modulation and / or demodulation without the network device interacting with the terminal device. This facilitates the optimization of the AI ​​model's performance and improves communication performance.

[0316] In another alternative implementation, the network device can obtain the performance monitoring results of the AI ​​model based on the demodulation performance information of the demodulated data corresponding to the fourth data.

[0317] Optionally, the terminal device sends fourth data to the network device, and the modulation of the fourth data is based on the third AI model. Correspondingly, the network device receives the fourth data from the terminal device. The performance monitoring results of the third AI model and / or the fourth AI model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the fourth data; the demodulation performance information is obtained based on the demodulation results; and the demodulation results are obtained by demodulating the data to be demodulated using the fourth AI model.

[0318] Please refer to Figure 7, which is a flowchart illustrating another communication processing method provided in this embodiment of the present disclosure. This method may include, but is not limited to, the following steps:

[0319] 5001. The terminal device sends fourth data to the network device. The modulation of the fourth data is based on the third AI model. Correspondingly, the network device receives the fourth data from the terminal device.

[0320] For uplink transmission, the data to be demodulated can be carried through the uplink channel, for example, through PUSCH. The demodulated data corresponding to the fourth data and the modulation of the fourth data can be found in the description of step 4001 in the method embodiment shown in Figure 6, and will not be repeated here.

[0321] 5002. The network device demodulates the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result.

[0322] Optionally, the fourth AI model and the third AI model can be the same AI model, or they can be different AI models; there is no restriction on this.

[0323] Step 5002 can be referred to the description of step 4001 in the method embodiment shown in Figure 6, and will not be repeated here.

[0324] 5003. Based on the demodulation results, the network device obtains the demodulation performance information of the data to be demodulated.

[0325] The demodulation performance information of the data to be demodulated may, but is not limited to, being used to represent the modulation performance of the data to be modulated, the transmission performance of the first data, and / or the demodulation performance of the data to be demodulated, etc., and there are no restrictions on this.

[0326] Optionally, the transmission performance information of the data to be demodulated may include, but is not limited to, at least one of the following: the signal-to-noise ratio of the received signal, the distribution characteristics of the received data, the bit error rate, the block error rate, the probability of HARQ NACK, and HARQ ACK information. For a description of the bit error rate and block error rate of the data to be demodulated, please refer to the description of step 4002 in the method embodiment shown in Figure 6, which will not be repeated here.

[0327] The signal-to-noise ratio (SNR) of a received signal refers to the ratio of the signal energy to the noise energy of the received signal. Optionally, the received signal may include the signal corresponding to the first data received by the network device; alternatively, the received signal may also include other signals. These other signals may include, but are not limited to, reference signals, interference signals, and / or other noise signals. The signal energy used to obtain the SNR of the received signal may include the signal energy corresponding to the demodulated data corresponding to the first data. The noise energy used to obtain the SNR of the received signal may include the signal energy of other signals, or the signal energy of noise signals, or the signal energy of interference signals and noise signals, etc., without limitation.

[0328] Optionally, the signal-to-noise ratio of the received signal can be obtained before the network device demodulates the received signal, or it can be obtained after the network device demodulates the received signal.

[0329] The distribution characteristics of the data are characterized by data distribution characteristic indicators, which include, but are not limited to, data drift detection. Data drift detection can be used to detect differences between the distribution of the data and the distribution of the training data for the AI ​​model. Received data includes, but is not limited to, the first data received by the network device. Optionally, the received data may also include other data received by the network device.

[0330] Optionally, the distribution characteristics of the received data can be obtained before the network device demodulates the received data, or after the network device demodulates the received data.

[0331] During the demodulation process, in response to a demodulation error in any data unit, the network device generates a HARQ NACK for that data unit and can send it to the terminal device. Conversely, in response to a successful demodulation of any data unit, it generates a HARQ ACK for that data unit and can send it to the terminal device. Thus, the network device can obtain the probability of obtaining the HARQ NACK of the data to be demodulated based on the proportion of the generated HARQ NACKs among all generated HARQ NACKs and HARQ ACKs. In other words, the probability of obtaining the HARQ NACK of the data to be demodulated can refer to the proportion of the HARQ NACKs generated during demodulation to all HARQ NACKs and HARQ ACKs, or it can be viewed as the probability of obtaining the HARQ NACK of the demodulated data. For example, if the network device fails to demodulate any first block or any first block group, the network device generates a HARQ NACK for that first block or first block group and can send the HARQ NACK to the terminal device. If the network device successfully demodulates any second block or any second block group, the network device generates a HARQ ACK for that second block or second block group and can send the HARQ ACK to the terminal device. For instance, after demodulating the data to be demodulated, 32 HARQ NACKs and 96 HARQ ACKs can be generated and sent, and the probability of obtaining the HARQ NACK for the data to be demodulated is 25%.

[0332] HARQ ACK information may include HARQ ACKs generated by network devices.

[0333] 5004. The network device obtains the performance monitoring results of the third AI model and / or the fourth AI model based on the demodulation performance information.

[0334] In some possible implementations, the demodulation performance information of the data to be demodulated can be compared with its corresponding performance threshold and / or performance threshold range. Based on the comparison result, the performance monitoring results of the third AI model and / or the fourth AI model can be obtained. The performance thresholds and / or performance threshold ranges corresponding to different demodulation performance information can be different or the same. The performance thresholds and / or performance threshold ranges can be, but are not limited to, predefined, obtained based on network configuration information, or based on DCI indications.

[0335] For example, if the bit error rate (BER) of the demodulated data is 25%, which is greater than the BER threshold of 10%, then the third AI model and / or the fourth AI model cannot meet the performance requirements of the communication system for modulation and demodulation. As another example, if the block error rate (BOR) of the demodulated data is 6%, which is less than the BOR threshold of 10%, then the third AI model and / or the fourth AI model can meet the performance requirements of the communication system for modulation and demodulation. Yet another example, if the HARQ NACK probability of the demodulated data is 4%, which is less than the probability threshold of 8%, then the performance monitoring result of the third AI model and / or the fourth AI model is determined to be the first-level performance monitoring result with optimal communication performance. As yet another example, if the demodulated data has 5 HARQ NACK entries, exceeding the HARQ NACK threshold of 3, then the performance monitoring result of the third AI model and / or the fourth AI model is determined to be the second-level performance monitoring result with poor communication performance.

[0336] It should be noted that the above is merely an example of how to obtain performance monitoring results for the third and / or fourth AI models based on demodulation performance information of the data to be demodulated, and does not constitute a limitation. Other methods can also be used to obtain performance monitoring results. The method shown in Figure 7 can also include more or fewer steps to obtain performance monitoring results for the AI ​​models, and there are no restrictions on this.

[0337] In this embodiment of the disclosure, the terminal device sends fourth data to the network device, and the modulation of the fourth data is based on the third AI model; subsequently, the network device demodulates the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result, and obtains the demodulation performance information of the data to be demodulated based on the demodulation result, and obtains the performance monitoring results of the third AI model and / or the fourth AI model based on the demodulation performance information of the data to be demodulated, thereby effectively monitoring the performance of the AI ​​model in modulation / demodulation, which is conducive to optimizing the performance of the AI ​​model and improving communication performance.

[0338] In another alternative implementation, the network device can obtain the performance monitoring results of the AI ​​model based on the demodulation results of at least two repeated transmissions between the network device and the terminal device.

[0339] Optionally, the network device sends a third DCI, which indicates at least two repeated transmissions. These at least two repeated transmissions include at least one first transmission and at least one second transmission. The modulation of the first transmission is based on an AI model, and the modulation of the second transmission differs from that of the first transmission. The performance monitoring results of the AI ​​model are obtained based on the demodulation results of the at least two repeated transmissions. Accordingly, the terminal device receives the third DCI and performs at least two repeated transmissions based on it.

[0340] Please refer to Figure 8, which is a flowchart illustrating another communication processing method provided in this embodiment of the present disclosure. This method may include, but is not limited to, the following steps:

[0341] 6001, the network device sends a third DCI to the terminal device. The third DCI indicates that at least two repeated transmissions are to be performed, including at least one first transmission and at least one second transmission. The modulation of the first transmission is based on an AI model, and the modulation of the second transmission is different from that of the first transmission. Accordingly, the terminal device receives the third DCI from the network device.

[0342] In one possible implementation, the third DCI can be used to schedule transmission resources that undergo at least two repeated transmissions. For example, the third DCI is a UL DCI used to schedule uplink transmission resources. While scheduling transmission resources that undergo at least two repeated transmissions, the third DCI can also instruct the terminal device to perform at least two repeated transmissions.

[0343] Optionally, the fourth AI model can be a different AI model from the third AI model, or it can be the same AI model; there are no restrictions on this.

[0344] 6002, The terminal device performs at least two repeated transmissions.

[0345] The at least two repeated transmissions include at least one first transmission and at least one second transmission. The first transmission modulates the data to be modulated based on an AI model. Optionally, the first transmission may include modulating the data to be modulated based on a third AI model to obtain fourth data; optionally, the first transmission may also include demodulating the data to be demodulated corresponding to the fourth data based on a fourth AI model. The modulation of the data to be modulated based on the third AI model and the demodulation of the data to be demodulated corresponding to the fourth data based on the fourth AI model are described in steps 5001-5002 of the method embodiment shown in Figure 7, and will not be repeated here.

[0346] The modulation method of the second transmission differs from that of the first transmission. Optionally, the modulation of the second transmission may not be based on an AI model. In other words, the second transmission may include modulation of the data to be modulated using a non-AI method. For example, the data to be modulated may be modulated using a first modulation method that does not apply an AI model. The first modulation method may be, for example, but not limited to, 16-quadrature amplitude modulation (QAM) and / or quadrature phase shift keying (QPSK), etc. No limitation is placed on the first modulation method here. Optionally, the second transmission may also include demodulation of the data to be demodulated corresponding to the data to be modulated using a non-AI method. For example, the data to be demodulated may be demodulated using a first demodulation method corresponding to the first modulation method. The first demodulation method does not apply an AI model. The first demodulation method may be, for example, but not limited to, the demodulation method corresponding to 16-quadrature amplitude modulation (QAM) and / or QPSK, etc. No limitation is placed on the first demodulation method here. Optionally, before demodulating the data to be demodulated corresponding to the data to be modulated using a non-AI method, the modulated data may undergo a first processing and / or a second processing. For a description of the first and second processing, please refer to the description of step 1001 in the method embodiment shown in Figure 3. For specific modulation and demodulation methods, please refer to the explanation of the relevant concepts above, which will not be repeated here.

[0347] It is evident that the difference between the first transmission and the second transmission lies in whether the modulation is based on an AI model; and the modulation data corresponding to the first transmission is the same as that corresponding to the second transmission.

[0348] The terminal device performs at least two repeated transmissions based on the third DCI. In a specific example, the terminal device can send fourth data and fifth data to the network device based on the transmission resources scheduled by the third DCI and the at least two repeated transmissions indicated by the third DCI. The fourth data can be referred to the description of the method embodiment shown in Figure 7, and will not be repeated here. The fifth data is the data sent in the second transmission. Optionally, the fifth data can be data obtained by modulating the data to be modulated based on the first modulation method, or the fifth data can be data after the aforementioned modulated data has undergone the first processing and / or the second processing. There is no limitation on the fifth data here. Accordingly, after receiving the fourth data, the network device demodulates the fifth data based on the fourth AI model to obtain the demodulation result of the fourth data, that is, the demodulation result of the first transmission; after receiving the fifth data, it demodulates the fifth data in a non-AI manner to obtain the demodulation result of the fifth data, that is, the demodulation result of the second transmission.

[0349] Optionally, the fourth and fifth data may include the same data, and / or may include different data, without limitation.

[0350] 6003. Network devices obtain performance monitoring results of AI models based on demodulation results from at least two repeated transmissions.

[0351] For a description of the demodulation results, please refer to the description of step 4001 in the method embodiment shown in Figure 6, which will not be repeated here.

[0352] In some possible implementations, the network device can determine whether the demodulation results of the first and second transmissions are the same, different, or have a certain degree of difference based on the demodulation results of the two transmissions, thereby obtaining the performance monitoring results of the AI ​​model. The performance monitoring results of the AI ​​model may include the monitoring results of at least one of the three performance metrics described in step 4002 of the method embodiment shown in Figure 6, which will not be elaborated further.

[0353] Optionally, if the demodulation results of the two transmissions are the same, it can be indicated that at least one of the above three performance characteristics meets the modulation and demodulation performance requirements of the communication system. For example, if the demodulation results of both transmissions include correct demodulation of the second code block or second code block group, then the demodulation results of the two transmissions are determined to be the same. As another example, if the demodulation results of both transmissions include 128 code blocks of correctly demodulated data units, then the demodulation results of the two transmissions are determined to be the same.

[0354] Optionally, the difference in demodulation results between the two transmissions indicates that at least one of the three performance characteristics mentioned above fails to meet the modulation and demodulation performance requirements of the communication system. For example, if the demodulation result of the first transmission includes correct demodulation of code block a, while the demodulation result of the second transmission includes incorrect demodulation of code block a, then the demodulation results of the two transmissions are determined to be different. As another example, if the demodulation result of the first transmission includes 96 correctly demodulated data blocks, while the demodulation result of the second transmission includes 128 correctly demodulated data blocks, then the demodulation results of the two transmissions are determined to be different.

[0355] Optionally, if the demodulation results of the two transmissions differ, the network device can further obtain the performance monitoring results of the AI ​​model based on the degree of difference between the two transmission results. For example, the demodulation result of the first transmission includes 96 correctly demodulated code blocks, and the demodulation result of the second transmission includes 128 correctly demodulated code blocks. The two results differ by 32 correctly demodulated code blocks, determining a difference of 25%, which is greater than the performance threshold of 10%. Therefore, it is determined that at least one of the above three performance characteristics fails to meet the modulation and demodulation performance requirements of the communication system. As another example, the two demodulation results may include all the data obtained from each demodulation. Comparing all the data obtained from the first transmission demodulation with all the data obtained from the second transmission demodulation, it is determined that there are 32 bits different, and the data to be modulated consists of 128 bits. Therefore, it is determined that the difference between the two demodulation results is 25%, which is greater than the performance threshold of 10%. Therefore, it is determined that at least one of the above three performance characteristics fails to meet the modulation and demodulation performance requirements of the communication system.

[0356] Therefore, the demodulation results of the second transmission can be used as the evaluation standard for the demodulation results of the first transmission, in order to determine the performance monitoring results of the third and fourth AI models applied in the first transmission.

[0357] It should be noted that the above is merely an example of obtaining AI model performance monitoring results based on demodulation results from at least two repeated transmissions, and does not constitute a limitation. Other methods can also be used to obtain performance monitoring results. The method shown in Figure 8 can also include more or fewer steps to obtain AI model performance monitoring results, and there are no restrictions on this.

[0358] In this embodiment of the disclosure, the network device sends a third DCI to the terminal device. The third DCI is used to indicate at least two repeated transmissions, which include at least one first transmission and at least one second transmission. The modulation of the first transmission is based on the AI ​​model, and the modulation of the second transmission is different from that of the first transmission. The terminal device performs at least two repeated transmissions with the network device based on the third DCI. Based on the demodulation results of the at least two repeated transmissions, the network device obtains the performance monitoring results of the AI ​​model, thereby effectively monitoring the performance of the AI ​​model in modulation and / or demodulation, which is beneficial for optimizing the performance of the AI ​​model and improving communication performance.

[0359] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this disclosure is not limited to the described order of actions, because according to this disclosure, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this disclosure.

[0360] In the above embodiments, the descriptions of each embodiment have their own emphasis, and any multiple embodiments can be used in combination. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0361] The foregoing primarily describes the solutions of the embodiments of this disclosure from a methodological perspective. It is understood that, in order to achieve the aforementioned functions, terminal devices and network devices include corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this disclosure can be implemented in hardware or a combination of hardware and computer software. Whether a particular function is executed in a hardware or computer software-driven hardware manner 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 this disclosure.

[0362] This disclosure embodiment can divide terminal devices and network devices into functional units according to the above method examples. For example, each function can be divided into a separate functional unit, or two or more functions can be integrated into one processing unit. The integrated unit can be implemented in hardware or as a software program module. It should be noted that the unit division in this disclosure embodiment is illustrative and only represents a logical functional division, while other division methods may be used in actual implementation.

[0363] Please refer to Figure 9, which is a schematic diagram of the structure of a communication device provided in an embodiment of this disclosure. The communication device 90 can be a network device or a device matched with a network device, such as a processor, chip, or chip module; alternatively, the communication device 90 can be a terminal device or a device matched with a terminal device, such as a processor, chip, or chip module. As shown in Figure 9, the communication device 90 includes a demodulation unit 901 and an acquisition unit 902. The demodulation unit 901 and the acquisition unit 902 can be module units for processing signals, data, information, etc., and there are no specific limitations on their use.

[0364] The communication device 90 may further include a storage unit for storing computer program code or instructions executed by the communication device 90. The storage unit may be a memory.

[0365] Additionally, it should be noted that the communication device 90 can be a chip or a chip module.

[0366] The demodulation unit 901 and the acquisition unit 902 can be integrated into the processing unit. The processing unit can be a processor or controller, such as 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 the various exemplary logic blocks, modules, and circuits described in conjunction with this disclosure. The processing unit can also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0367] In specific implementation, the demodulation unit 901 and the acquisition unit 902 are used to perform any step executed by the network device in the method embodiment shown in Figure 3 above, or to perform any step executed by the network device in the method embodiment shown in Figure 6 above. A detailed explanation follows.

[0368] When the demodulation unit 901 and the acquisition unit 902 are used to perform any step executed by the network device in the method embodiment shown in FIG3 above:

[0369] The demodulation unit 901 is used to demodulate the data to be demodulated corresponding to the first data based on the first proxy model to obtain the demodulation result; the modulation of the first data is based on the first AI model, the first proxy model is used to simulate the second AI model, and the second AI model is used to perform demodulation.

[0370] The acquisition unit 902 is used to acquire the performance monitoring results of the first AI model and / or the second AI model based on the demodulation results.

[0371] When the demodulation unit 901 and the acquisition unit 902 are used to perform any step executed by the network device in the method embodiment shown in FIG6 above:

[0372] The demodulation unit 901 is used to demodulate the data to be demodulated corresponding to the fourth data based on the fourth AI model to obtain the demodulation result; the modulation of the fourth data is based on the second proxy model, which is used to simulate the third AI model and the third AI model is used for modulation.

[0373] The acquisition unit 902 is used to acquire the performance monitoring results of the third AI model and / or the fourth AI model based on the demodulation results.

[0374] The relevant details of this implementation method can be found in the method embodiments shown in Figures 3 or 6 above. Further details are omitted here. This disclosure and the above method embodiments are based on the same concept and achieve the same technical effects. For specific principles, please refer to the descriptions of the above method embodiments; they will not be repeated here.

[0375] Please refer to Figure 10, which is a schematic diagram of another communication device provided in an embodiment of this disclosure. The communication device 100 can be a network device or a device compatible with a network device, such as a processor, chip, or chip module; alternatively, the communication device 100 can be a terminal device or a device compatible with a terminal device, such as a processor, chip, or chip module. As shown in Figure 10, the communication device 100 includes a communication unit 1010. The communication unit 1010 can be a module unit for processing signals, data, information, etc., and there are no specific limitations on this.

[0376] The communication device 100 may further include a storage unit for storing computer program code or instructions executed by the communication device 100. The storage unit may be a memory.

[0377] Additionally, it should be noted that the communication device 100 can be a chip or a chip module.

[0378] The communication unit 1010 can be integrated into the processing unit. The processing unit can be a processor or controller, such as 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 the various exemplary logic blocks, modules, and circuits described in conjunction with this disclosure. The processing unit can also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0379] In specific implementation, the communication unit 1010 is used to execute any step performed by the network device or the terminal device in any of the method embodiments shown in Figures 4, 5, 7, and 8 above. A detailed description follows.

[0380] In the case where the communication unit 1010 is used to perform any step performed by the network device in the method embodiment shown in FIG4 above:

[0381] The communication unit 1010 is used to transmit first data, the modulation of the first data is based on a first AI model; the performance monitoring result of the first AI model / or the second AI model is obtained based on the demodulation performance information of the data to be demodulated corresponding to the first data; the demodulation performance information is obtained based on the demodulation result of the data to be demodulated, and the second AI model is used for demodulation.

[0382] Optionally, demodulation performance information may include at least one of the following: bit error rate, block error rate, probability of hybrid automatic repeat request negative response, and hybrid automatic repeat request acknowledgment information.

[0383] In the case where the communication unit 1010 is used to perform any step performed by the network device in the method embodiment shown in FIG5 above:

[0384] The communication unit 1010 is used to transmit first data, the modulation of which is based on a first AI model; receive second data, which includes part or all of the data obtained by the terminal device demodulating the data to be demodulated corresponding to the first data based on the second AI model; and the performance monitoring results of the first AI model and / or the second AI model are obtained based on the second data.

[0385] Optionally, the communication unit 1010 is further configured to send a first DCI, which is used to schedule the transmission of first data and the transmission of second data; wherein the transmission of second data includes transmitting second data through the transmission resources scheduled by the first DCI.

[0386] Optionally, the first data and the second data are transmitted based on the same frequency domain resources; and / or, the first data and the second data are transmitted based on symbol resources at the same location in different time slots; and / or, the time slot for transmitting the second data is offset from the time slot for transmitting the first data by k time slots, where k is an integer greater than or equal to 1, and k is predefined, obtained based on network configuration information, or based on the first DCI indication.

[0387] Optionally, the first DCI includes scheduling resources for the first data and scheduling resources for the second data.

[0388] Optionally, the communication unit 1010 is also used to send a second DCI; the second DCI includes first indication information, which is used to indicate second data transmitted through the transmission resources scheduled by the second DCI.

[0389] Optionally, the first indication information includes a first downlink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the first downlink hybrid automatic repeat request process identifier; and / or, the first indication information includes an uplink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the second downlink hybrid automatic repeat request process identifier, wherein the second downlink hybrid automatic repeat request process identifier has the same value as the uplink hybrid automatic repeat request process identifier; and / or, the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the third data, wherein the third data is data in the time domain unit in the first time domain unit that is closest to the time domain unit of the second DCI, and the first time domain unit is the time domain unit used by the terminal device to receive the first data.

[0390] Optionally, some data includes data obtained by demodulating the first code block by the terminal device, the data to be demodulated includes the first code block, the first code block is a code block with demodulation error; and / or, some data includes data obtained by demodulating the first code block group, the data to be demodulated includes the first code block group, the first code block group is a code block group with demodulation error.

[0391] Optionally, the communication unit 1010 is also used to receive second indication information, which is used to indicate the first code block or first code block group corresponding to the partial data.

[0392] In the case where the communication unit 1010 is used to perform any step performed by the network device as shown in the method embodiment of FIG7 above:

[0393] The communication unit 1010 is used to receive fourth data, the modulation of which is based on the third AI model; the performance monitoring results of the third AI model and / or the fourth AI model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the fourth data; the demodulation performance information is obtained based on the demodulation result; the demodulation result is obtained by demodulating the data to be demodulated based on the fourth AI model.

[0394] Optionally, the demodulation performance information includes at least one of the following: the signal-to-noise ratio of the received signal, the distribution characteristics of the received data, the bit error rate, the block error rate, the probability of a negative response to a hybrid automatic repeat request, and the acknowledgment information for a hybrid automatic repeat request.

[0395] In the case where the communication unit 1010 is used to perform any step performed by the network device in the method embodiment shown in FIG8 above:

[0396] The communication unit 1010 is used to send a third DCI, which is used to instruct the terminal device to perform at least two repeated transmissions. The at least two repeated transmissions include at least one first transmission and at least one second transmission. The modulation of the first transmission is based on the AI ​​model, and the modulation of the second transmission is different from that of the first transmission. The performance monitoring result of the AI ​​model is obtained based on the demodulation result of the at least two repeated transmissions.

[0397] When the communication unit 1010 is used to perform any step executed by the terminal device in the method embodiment shown in FIG4 above:

[0398] The communication unit 1010 is used to receive first data, the modulation of which is based on a first AI model; send demodulation results or transmission performance information of the data to be demodulated corresponding to the first data; obtain the demodulation results by demodulating the data to be demodulated based on a second AI model; obtain demodulation performance information based on the demodulation results; and obtain the performance monitoring results of the first AI model and / or the second AI model based on the demodulation performance information.

[0399] Optionally, demodulation performance information may include at least one of the following: bit error rate, block error rate, probability of hybrid automatic repeat request negative response, and hybrid automatic repeat request acknowledgment information.

[0400] When the communication unit 1010 is used to perform any step executed by the terminal device in the method embodiment shown in FIG5 above:

[0401] The communication unit 1010 is used to receive first data, the modulation of which is based on a first AI model; to transmit second data, which includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the first data based on a second AI model; and to obtain the performance monitoring results of the first AI model and / or the second AI model based on the second data.

[0402] Optionally, the communication unit 1010 is further configured to receive a first DCI, which is used to schedule the transmission of first data and the transmission of second data; wherein the transmission of second data includes transmitting second data through the transmission resources scheduled by the first DCI.

[0403] Optionally, the first data and the second data are transmitted based on the same frequency domain resources; and / or, the first data and the second data are transmitted based on symbol resources at the same location in different time slots; and / or, the time slot for transmitting the second data is offset from the time slot for transmitting the first data by k time slots, where k is an integer greater than or equal to 1, and k is predefined, obtained based on network configuration information, or based on the first DCI indication.

[0404] Optionally, the first DCI includes scheduling resources for the first data and scheduling resources for the second data.

[0405] Optionally, the communication unit 1010 is also configured to receive a second DCI; the second DCI includes first indication information, which indicates second data transmitted through the transmission resources scheduled by the second DCI.

[0406] Optionally, the first indication information includes a first downlink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the first downlink hybrid automatic repeat request process identifier; and / or, the first indication information includes an uplink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the second downlink hybrid automatic repeat request process identifier, wherein the second downlink hybrid automatic repeat request process identifier has the same value as the uplink hybrid automatic repeat request process identifier; and / or, the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the third data, wherein the third data is data in the time domain unit in the first time domain unit that is closest to the time domain unit of the second DCI, and the first time domain unit is the time domain unit used to receive the first data.

[0407] Optionally, some data includes data obtained by demodulating the first code block, the data to be demodulated includes the first code block, and the first code block is a code block with demodulation errors; and / or, some data includes data obtained by demodulating the first code block group, the data to be demodulated includes the first code block group, and the first code block group is a code block group with demodulation errors.

[0408] Optionally, the communication unit 1010 is also used to send second indication information, which is used to indicate the first code block or first code block group corresponding to the partial data.

[0409] When the communication unit 1010 is used to perform any step executed by the terminal device in the method embodiment shown in FIG7 above:

[0410] The communication unit 1010 is used to transmit fourth data; the modulation of the fourth data is based on the third AI model; the performance monitoring results of the third AI model and the fourth AI model are obtained based on the demodulation performance information of the data to be demodulated; the demodulation performance information is obtained based on the demodulation result; the demodulation result is obtained by demodulating the data to be demodulated based on the fourth AI model.

[0411] Optionally, the demodulation performance information includes at least one of the following: the signal-to-noise ratio of the received signal, the distribution characteristics of the received data, the bit error rate, the block error rate, the probability of a negative response to a hybrid automatic repeat request, and the acknowledgment information for a hybrid automatic repeat request.

[0412] When the communication unit 1010 is used to perform any step executed by the terminal device in the method embodiment shown in FIG8 above:

[0413] The communication unit 1010 is used to receive a third DCI, which is used to indicate that at least two repeated transmissions are performed. The at least two repeated transmissions include at least one first transmission and at least one second transmission. The modulation of the first transmission is based on an AI model, and the modulation of the second transmission is different from that of the first transmission. The performance monitoring results of the AI ​​model are obtained based on the demodulation results of the at least two repeated transmissions.

[0414] The relevant details of this implementation method can be found in the method embodiments shown in any of Figures 4, 5, 7, and 8 above. Further details are omitted here. This disclosure and the above method embodiments are based on the same concept and achieve the same technical effects. For specific principles, please refer to the descriptions of the above method embodiments; they will not be repeated here.

[0415] Please refer to Figure 11, which is a schematic diagram of another communication device provided in an embodiment of this disclosure. The communication device 110 can be a network device or a device compatible with a network device, such as a processor, chip, or chip module; or it can be a terminal device or a device compatible with a terminal device, such as a processor, chip, or chip module. The communication device 110 may include a processor 1101. Optionally, the communication device 110 may also include a memory 1102 and a computer program or instructions (not shown in Figure 11) stored in the memory 1102. The processor 1101 and the memory 1102 are interconnected. Optionally, the communication device 110 may also include a transceiver 1103. The processor 1101, memory 1102, and transceiver 1103 can be connected via a bus 1104 or other means. The bus is represented by thick lines in Figure 11. The connection methods between other components are only illustrative and not intended to be limiting. The bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used in Figure 11, but this does not mean that there is only one bus or one type of bus.

[0416] The coupling in this disclosure is an indirect coupling or communication connection between devices, units, or modules, which can be electrical, mechanical, or other forms, used for information exchange between devices, units, or modules. This disclosure does not limit the specific connection medium between the processor 1101, memory 1102, and transceiver 1103.

[0417] Memory 1102 may include read-only memory and random access memory, and provides instructions and data to processor 1101. A portion of memory 1102 may also include non-volatile random access memory.

[0418] Processor 1101 can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor; optionally, processor 1101 can also be any conventional processor.

[0419] Transceiver 1103 is used to receive or send data.

[0420] In one implementation, memory 1102 is used to store computer programs or instructions; processor 1101 is used to call the computer programs or instructions stored in memory 1102 to execute the steps performed by the network device or terminal device in any of the method embodiments shown in Figures 3 to 8.

[0421] In the embodiments of this disclosure, the methods provided in the embodiments of this disclosure can be implemented by running a computer program (including program code or instructions) capable of performing the steps involved in the above-described methods on a general-purpose computing device, such as a computer, which includes processing elements and storage elements such as a CPU, random access memory (RAM), and read-only memory (ROM). The computer program or instructions can be recorded on, for example, a computer-readable recording medium, loaded into the aforementioned computing device through the computer-readable recording medium, and executed therein.

[0422] Based on the same inventive concept, the principle and beneficial effects of the communication device 110 provided in the embodiments of this disclosure in solving the problem are similar to those in the embodiments shown in any of Figures 3 to 8 of this disclosure. For reference, please refer to the principle and beneficial effects of the implementation of the method. For the sake of brevity, they will not be repeated here.

[0423] The aforementioned communication device may be, for example, a chip or a chip module.

[0424] This disclosure also provides a chip including a processor, which can execute the relevant steps of the network device or terminal device in the foregoing method embodiments. The specific implementation of the network device or terminal device can be found in the description of the relevant content in the foregoing method embodiments, and will not be repeated here.

[0425] In some possible implementations, the chip further includes at least one first memory and at least one second memory; the at least one first memory and the processor are interconnected via a circuit, and the first memory stores instructions; the at least one second memory and the processor are interconnected via a circuit, and the second memory stores data that needs to be stored in the above method embodiments.

[0426] Please refer to Figure 12, which is a schematic diagram of the structure of a chip module provided in an embodiment of this disclosure. The chip module 120 can perform the relevant steps of the network device or terminal device in the foregoing method embodiments. The chip module 120 includes: a communication interface 1201 and a chip 1202.

[0427] The communication interface 1201 is used for internal communication within the chip module or for communication between the chip module and external devices. The communication interface 1201 can also be described as a communication module. The chip 1202 includes a processor (not shown in Figure 12). The chip 1202 is used to implement the functions of the network device or terminal device in the embodiments of this disclosure; that is, the processor of the chip 1202 is used to execute the relevant steps of the network device or terminal device in the foregoing method embodiments. The specific implementation of the network device or terminal device can be found in the description of the relevant content in the foregoing method embodiments, and will not be repeated here.

[0428] Optionally, the chip 1202 may further include a memory (not shown in FIG12) and a computer program or instructions (not shown in FIG12) stored in the memory. The processor executes the computer program or instructions to implement the relevant steps performed by the network device or terminal device as described in the above method embodiments. The specific implementation of the network device or the terminal device can be referred to the description of the relevant content in the foregoing method embodiments, and will not be repeated here.

[0429] Optionally, the chip 1202 is interconnected with the communication interface 1201 via a line; through the communication interface 1201, the chip module 120 can exchange data with other chip modules, other terminals, servers, and other modules or devices.

[0430] Optionally, the chip module 120 may also include a storage module 1203 and a power module 1204. The storage module 1203 is used to store data and instructions. The power module 1204 is used to provide power to the chip module.

[0431] For various devices and products applied to or integrated into chip modules, each of its modules can be implemented using hardware methods such as circuits. Different modules can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module. Alternatively, at least some modules can be implemented using software programs that run on the processor integrated inside the chip module, while the remaining (if any) modules can be implemented using hardware methods such as circuits.

[0432] This disclosure also provides a computer-readable storage medium storing a computer program or instructions. When the computer program or instructions are executed, for example, when executed by a processor or computer, the method flow of the method embodiment executed by the aforementioned network device or terminal device will be implemented. Specific implementations of the network device or terminal device can be found in the descriptions of the relevant content in the foregoing embodiments, and will not be repeated here. It is understood that the computer storage medium here may include the built-in storage medium in the network device or terminal device, or it may include extended storage media supported by the network device or terminal device. The computer storage medium provides storage space that stores the operating system of the terminal device or network device. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions may be one or more computer programs (including program code). It should be noted that the computer storage medium here may be a high-speed RAM memory, or a non-volatile memory, such as at least one disk storage device, or Flash memory; optionally, it may also be at least one computer storage medium located remotely from the aforementioned processor. The specific implementation of the network device or the terminal device can be found in the description of the relevant content in the foregoing method embodiments, and will not be repeated here.

[0433] This disclosure also provides a computer program product, including a computer program or instructions, which, when executed, such as when the computer program or instructions are executed by a processor or computer, cause the processor or computer to perform a method flow of the method embodiment executed by the network device or the terminal device described above.

[0434] This disclosure provides a communication system that may include a network device that performs the methods described in the above method embodiments, and a terminal device that performs the methods described in the above method embodiments.

[0435] It should be noted that, for the sake of simplicity, the above embodiments are all described as a series of actions. Those skilled in the art should understand that this disclosure is not limited to the described order of actions, as some steps in the embodiments of this disclosure can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions, steps, modules, or units involved are not necessarily essential to the embodiments of this disclosure.

[0436] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0437] The steps of the methods or algorithms described in this disclosure can be implemented in hardware or by a processor executing software instructions. The software instructions can consist of corresponding software modules, which can be stored in RAM, flash memory, ROM, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, hard disks, portable hard disks, read-only optical discs (CD-ROMs), or any other form of storage medium well known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Additionally, the ASIC can reside in a network device or terminal device. Alternatively, the processor and storage medium can exist as discrete components in the network device or terminal device.

[0438] Those skilled in the art will recognize that, in one or more of the examples above, the functions described in the embodiments of this disclosure can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. This computer program product includes one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0439] Regarding the modules / units included in the various devices and products described in the above embodiments, they can be software modules / units, hardware modules / units, or a combination of both. For example, for various devices and products applied to or integrated into a chip, all of their modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits; for various devices and products applied to or integrated into a chip module, all of their modules / units can be implemented using hardware methods such as circuits, and different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using hardware methods such as circuits. The components can be implemented using software programs that run on the processor integrated within the chip module. The remaining (if any) modules / units can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into the terminal, each of its components / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or in different components within the terminal. Alternatively, at least some modules / units can be implemented using software programs that run on the processor integrated within the terminal, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits.

[0440] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the embodiments of this disclosure. It should be understood that the above descriptions are merely specific embodiments of the embodiments of this disclosure and are not intended to limit the protection scope of the embodiments of this disclosure. Any modifications, equivalent substitutions, improvements, etc., made on the basis of the technical solutions of the embodiments of this disclosure should be included within the protection scope of the embodiments of this disclosure.

Claims

1. A communication processing method, characterized in that, The method includes: Based on the first proxy model, the demodulated data corresponding to the first data is demodulated to obtain the demodulation result; the modulation of the first data is based on the first artificial intelligence model, the first proxy model is used to simulate the second artificial intelligence model, and the second artificial intelligence model is used for demodulation. Based on the demodulation results, obtain the performance monitoring results of the first artificial intelligence model and / or the second artificial intelligence model.

2. A communication processing method, characterized in that, The method includes: First data is transmitted, the modulation of the first data is based on a first artificial intelligence model; the performance monitoring results of the first artificial intelligence model and / or the second artificial intelligence model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the first data; the demodulation performance information is obtained based on the demodulation result of the data to be demodulated, and the second artificial intelligence model is used to perform demodulation.

3. The method as described in claim 2, characterized in that, The demodulation performance information includes at least one of the following: Bit error rate, block error rate, probability of negative response to hybrid automatic repeat request (HARQ) and acknowledgment information for hybrid automatic repeat request (HARQ).

4. A communication processing method, characterized in that, The method includes: First data is transmitted, and the modulation of the first data is based on a first artificial intelligence model; Receive second data, which includes partial or complete data obtained by demodulating the data to be demodulated corresponding to the first data based on the second artificial intelligence model; the performance monitoring results of the first artificial intelligence model and / or the second artificial intelligence model are obtained based on the second data.

5. The method as described in claim 4, characterized in that, The method further includes: Send first downlink control information, which is used to schedule the transmission of the first data and the transmission of the second data; wherein, the transmission of the second data includes transmitting the second data through the transmission resources scheduled by the first downlink control information.

6. The method as described in claim 5, characterized in that, The first data and the second data are transmitted based on the same frequency domain resources; and / or, The first data and the second data are transmitted based on symbol resources at the same location in different time slots; and / or, The time slot for transmitting the second data has an offset of k time slots from the time slot for transmitting the first data, where k is an integer greater than or equal to 1, and k is predefined, obtained based on network configuration information, or indicated based on the first downlink control information.

7. The method as described in claim 5, characterized in that, The first downlink control information includes the scheduling resources of the first data and the scheduling resources of the second data.

8. The method as described in claim 4, characterized in that, The method further includes: Send a second downlink control message; the second downlink control message includes a first indication message, which is used to indicate the second data to be transmitted through the transmission resources scheduled by the second downlink control message.

9. The method as described in claim 8, characterized in that, The first indication information includes a first downlink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the first downlink hybrid automatic repeat request process identifier; and / or, The first indication information includes an uplink hybrid automatic repeat request process identifier, and the second data includes partial or complete data obtained by demodulating the data to be demodulated corresponding to the second downlink hybrid automatic repeat request process identifier, wherein the second downlink hybrid automatic repeat request process identifier has the same value as the uplink hybrid automatic repeat request process identifier; and / or, The second data includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the third data. The third data is the data in the time domain unit in the first time domain unit that is closest to the time domain unit of the second downlink control information. The first time domain unit is the time domain unit used to receive the first data.

10. The method according to any one of claims 4-9, characterized in that, The partial data includes data obtained by demodulating a first code block, the data to be demodulated includes the first code block, the first code block being a code block with demodulation errors; and / or, the partial data includes data obtained by demodulating a first code block group, the data to be demodulated includes the first code block group, the first code block group being a code block group with demodulation errors.

11. The method as described in claim 10, characterized in that, The method further includes: Receive second indication information, which is used to indicate the first code block or the first code block group corresponding to the partial data.

12. A communication processing method, characterized in that, The method includes: Based on the fourth artificial intelligence model, the data to be demodulated corresponding to the fourth data is demodulated to obtain the demodulation result; the modulation of the fourth data is based on the second proxy model, which is used to simulate the third artificial intelligence model, and the third artificial intelligence model is used for modulation. Based on the demodulation results, the performance monitoring results of the third artificial intelligence model and / or the fourth artificial intelligence model are obtained.

13. A communication processing method, characterized in that, The method includes: The fourth data is received, and the modulation of the fourth data is based on the third artificial intelligence model; the performance monitoring results of the third artificial intelligence model and / or the fourth artificial intelligence model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the fourth data; the demodulation performance information is obtained based on the demodulation result; the demodulation result is obtained by demodulating the data to be demodulated based on the fourth artificial intelligence model.

14. The method as described in claim 13, characterized in that, The demodulation performance information includes at least one of the following: The signal-to-noise ratio of the received signal, the distribution characteristics of the received data, the bit error rate, the block error rate, the probability of a negative response to a hybrid automatic repeat request, and the confirmation response information for a hybrid automatic repeat request.

15. A communication processing method, characterized in that, The method includes: Send a third downlink control message, which is used to instruct at least two repeated transmissions, including at least one first transmission and at least one second transmission. The modulation of the first transmission is based on an artificial intelligence model, and the modulation of the second transmission is different from that of the first transmission. The performance monitoring result of the artificial intelligence model is obtained based on the demodulation result of the at least two repeated transmissions.

16. A communication processing method, characterized in that, The method includes: Receive first data, the modulation of which is based on a first artificial intelligence model; Send the demodulation result or the demodulation performance information of the data to be demodulated corresponding to the first data; the demodulation result is obtained by demodulating the data to be demodulated based on the second artificial intelligence model; the demodulation performance information is obtained based on the demodulation result, and the performance monitoring results of the first artificial intelligence model and / or the second artificial intelligence model are obtained based on the demodulation performance information.

17. The method as described in claim 16, characterized in that, The demodulation performance information includes at least one of the following: Bit error rate, block error rate, probability of negative response to hybrid automatic repeat request (HARQ) and acknowledgment information for hybrid automatic repeat request (HARQ).

18. A communication processing method, characterized in that, The method includes: Receive first data, the modulation of which is based on a first artificial intelligence model; Send second data, which includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the first data based on the second artificial intelligence model; the performance monitoring results of the first artificial intelligence model and / or the second artificial intelligence model are obtained based on the second data.

19. The method as described in claim 18, characterized in that, The method further includes: The system receives first downlink control information, which is used to schedule the transmission of the first data and the transmission of the second data; wherein the transmission of the second data includes transmitting the second data through the transmission resources scheduled by the first downlink control information.

20. The method as described in claim 19, characterized in that, The first data and the second data are transmitted based on the same frequency domain resources; and / or, The first data and the second data are transmitted based on symbol resources at the same location in different time slots; and / or, The time slot for transmitting the second data has an offset of k time slots from the time slot for transmitting the first data, where k is an integer greater than or equal to 1, and k is predefined, obtained based on network configuration information, or indicated based on the first downlink control information.

21. The method as described in claim 19, characterized in that, The first downlink control information includes the scheduling resources of the first data and the scheduling resources of the second data.

22. The method as described in claim 18, characterized in that, The method further includes: Receive second downlink control information; the second downlink control information includes first indication information, the first indication information being used to indicate the second data transmitted through the transmission resources scheduled by the second downlink control information.

23. The method as described in claim 22, characterized in that, The first indication information includes a first downlink hybrid automatic repeat request process identifier, and the second data includes partial or all data obtained by demodulating the data to be demodulated corresponding to the first downlink hybrid automatic repeat request process identifier; and / or, The first indication information includes an uplink hybrid automatic repeat request process identifier, and the second data includes partial or complete data obtained by demodulating the data to be demodulated corresponding to the second downlink hybrid automatic repeat request process identifier, wherein the second downlink hybrid automatic repeat request process identifier has the same value as the uplink hybrid automatic repeat request process identifier; and / or, The second data includes part or all of the data obtained by demodulating the data to be demodulated corresponding to the third data. The third data is the data in the time domain unit in the first time domain unit that is closest to the time domain unit of the second downlink control information. The first time domain unit is the time domain unit used to receive the first data.

24. The method according to any one of claims 18-23, characterized in that, The partial data includes data obtained by demodulating a first code block, the data to be demodulated includes the first code block, the first code block being a code block with demodulation errors; and / or, the partial data includes data obtained by demodulating a first code block group, the data to be demodulated includes the first code block group, the first code block group being a code block group with demodulation errors.

25. The method as described in claim 24, characterized in that, The method further includes: Send a second indication message, which is used to indicate the first code block or the first code block group corresponding to the partial data.

26. A communication processing method, characterized in that, The method includes: The fourth data is transmitted; the modulation of the fourth data is based on the third artificial intelligence model; the performance monitoring results of the third artificial intelligence model and / or the fourth artificial intelligence model are obtained based on the demodulation performance information of the data to be demodulated corresponding to the fourth data; the demodulation performance information is obtained based on the demodulation result; the demodulation result is obtained by demodulating the data to be demodulated based on the fourth artificial intelligence model.

27. The method as described in claim 26, characterized in that, The demodulation performance information includes at least one of the following: The signal-to-noise ratio of the received signal, the distribution characteristics of the received data, the bit error rate, the block error rate, the probability of a negative response to a hybrid automatic repeat request, and the confirmation response information for a hybrid automatic repeat request.

28. A communication processing method, characterized in that, The method includes: Receive third downlink control information, the third downlink control information being used to instruct at least two repeated transmissions, the at least two repeated transmissions including at least one first transmission and at least one second transmission, the modulation of the first transmission being based on an artificial intelligence model, and the modulation of the second transmission being different from that of the first transmission; The at least two repeated transmissions are performed; the performance monitoring results of the artificial intelligence model are obtained based on the demodulation results of the at least two repeated transmissions.

29. A communication device, characterized in that, It includes units for implementing the method of any one of claims 1-15, or includes units for implementing the method of any one of claims 16-28.

30. A communication device, characterized in that, It includes 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 of the method according to any one of claims 1-15; or, to implement the steps of the method according to any one of claims 16-28.

31. A chip, comprising a processor, characterized in that, The processor performs the steps of the method according to any one of claims 1-15, or performs the steps of the method according to any one of claims 16-28.

32. A chip module, comprising a communication interface and a chip, characterized in that, The chip includes a processor that performs the steps of the method according to any one of claims 1-15, or the steps of the method according to any one of claims 16-28.

33. A non-volatile computer-readable storage medium, characterized in that, It stores a computer program or instructions that, when executed, implement the steps of the method according to any one of claims 1-15, or the steps of the method according to any one of claims 16-28.

34. A computer program product, characterized in that, Includes a computer program or instructions, wherein when executed, the computer program or instructions implement the steps of the method according to any one of claims 1-15, or the steps of the method according to any one of claims 16-28.

Images