Communication method, terminal, network device, system, storage medium, and program product

Abstract

The present disclosure relates to a communication method, a terminal, a network device, a system, a storage medium, and a program product. The communication method comprises: when the overall performance of a first artificial intelligence (AI) model and a second AI model does not meet a requirement, a network device sending first information to a terminal, wherein the first AI model is deployed at the terminal, and the second AI model is deployed at the network device; the first AI model is used for compressing a first measured value to obtain a second measured value, and the second AI model is used for decompressing the second measured value to obtain a third measured value; a difference between the first measured value and the third measured value is used for representing the overall performance of the first AI model and the second AI model; and the first information is used for instructing to perform channel coding on the second measured value. The present disclosure can ameliorate problems caused by the poor model performance and improve the communication efficiency.

Description

Communication methods, terminals, network devices, systems, storage media and software products Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to communication methods, terminals, network devices, systems, storage media, and program products. Background Technology

[0002] In recent years, artificial intelligence (AI) and machine learning (ML) technologies have made continuous breakthroughs in many fields. Summary of the Invention

[0003] This disclosure provides communication methods, terminals, network devices, systems, storage media, and program products.

[0004] According to a first aspect of the present disclosure, a communication method is proposed, the method comprising: when the overall performance of a first artificial intelligence (AI) model and a second AI model does not meet requirements, a network device sends first information to a terminal; wherein the first AI model is deployed on the terminal, the second AI model is deployed on the network device, the first AI model is used to compress a first measurement value to obtain a second measurement value, the second AI model is used to decompress the second measurement value to obtain a third measurement value, the difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model; the first information is used to instruct channel coding of the second measurement value.

[0005] According to a second aspect of the present disclosure, a communication method is proposed, the method comprising: a terminal receiving first information sent by a network device, the first information being sent when the overall performance of a first artificial intelligence (AI) model and a second AI model fails to meet requirements; wherein the first AI model is deployed on the terminal, the second AI model is deployed on the network device, the first AI model is used to compress a first measurement value to obtain a second measurement value, the second AI model is used to decompress the second measurement value to obtain a third measurement value, the difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model; the first information is used to instruct channel coding of the second measurement value.

[0006] According to a third aspect of the present disclosure, a network device is provided, comprising: sending first information to a terminal when the overall performance of a first artificial intelligence (AI) model and a second AI model does not meet requirements; wherein the first AI model is deployed on the terminal, the second AI model is deployed on the network device, the first AI model is used to compress a first measurement value to obtain a second measurement value, the second AI model is used to decompress the second measurement value to obtain a third measurement value, the difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model; the first information is used to instruct channel coding of the second measurement value.

[0007] According to a fourth aspect of the present disclosure, a terminal device is provided, comprising: a transceiver module, configured to receive first information sent by a network device, the first information being sent when the overall performance of a first artificial intelligence (AI) model and a second AI model fails to meet requirements; wherein the first AI model is deployed on the terminal, the second AI model is deployed on the network device, the first AI model is configured to compress a first measurement value to obtain a second measurement value, the second AI model is configured to decompress the second measurement value to obtain a third measurement value, the difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model; the first information is used to instruct channel coding of the second measurement value.

[0008] According to a fifth aspect of the present disclosure, a network device is provided, comprising: one or more processors; wherein the network device is configured to perform the first aspect and any one of the communication methods in the first aspect.

[0009] According to a sixth aspect of the present disclosure, a terminal device is provided, comprising: one or more processors; wherein the terminal device is configured to execute the second aspect and any one of the communication methods in the second aspect.

[0010] According to a seventh aspect of the present disclosure, a communication system is provided, including a network device and a terminal device, wherein the network device is configured to implement the first aspect and any one of the communication methods in the first aspect, and the terminal device is configured to implement the second aspect and any one of the communication methods in the second aspect.

[0011] According to an eighth aspect of the present disclosure, a storage medium is provided that stores instructions which, when executed on a communication device, cause the communication device to perform a communication method as described in the first aspect and any one thereof, or the second aspect and any one thereof.

[0012] According to a ninth aspect of the present disclosure, a program product is provided, comprising: a computer program, which, when executed by a communication device, causes the communication device to perform a communication method as described in the first aspect and any one of the first aspects or the second aspect and the second aspect.

[0013] This disclosure involves a network device sending first information to a terminal when the overall performance of a first AI model and a second AI model fails to meet requirements. The first AI model is deployed on the terminal, and the second AI model is deployed on the network device. The first AI model compresses a first measurement value to obtain a second measurement value, and the second AI model decompresses the second measurement value to obtain a third measurement value. The difference between the first and third measurement values ​​characterizes the overall performance of the first and second AI models. By instructing the second measurement value to be channel-coded using the first information, it is possible to further rule out whether the failure to meet performance requirements is due to the transmission of the second measurement value, thus allowing for targeted subsequent operations to improve the problem caused by low model performance and increase communication efficiency. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this disclosure, the accompanying drawings required for the description of the embodiments are introduced below. The following drawings are only some embodiments of this disclosure and do not impose specific limitations on the protection scope of this disclosure.

[0015] Figure 1a is a schematic diagram of a communication system architecture according to an embodiment of the present disclosure.

[0016] Figure 1b is a schematic diagram of CSI compressed feedback and recovery based on a bilateral model.

[0017] Figure 1c is a schematic diagram of CSI feedback based on joint source and channel coding using a bilateral AI model.

[0018] Figure 2 is a schematic diagram of a communication method interaction according to an embodiment of the present disclosure.

[0019] Figure 3 is a flowchart illustrating a communication method according to an embodiment of the present disclosure.

[0020] Figure 4 is a flowchart illustrating a communication method according to an embodiment of the present disclosure.

[0021] Figure 5a is a schematic diagram of a communication method interaction according to an embodiment of the present disclosure.

[0022] Figure 5b is an interactive schematic diagram of a bilateral model performance monitoring method according to an embodiment of the present disclosure.

[0023] Figure 5c is an interactive schematic diagram of a bilateral model performance monitoring method according to an embodiment of the present disclosure.

[0024] Figure 6a is a schematic diagram of the structure of the terminal device proposed in an embodiment of this disclosure.

[0025] Figure 6b is a schematic diagram of the structure of the network device proposed in an embodiment of this disclosure.

[0026] Figure 7a is a schematic diagram of the structure of a communication device proposed in an embodiment of this disclosure.

[0027] Figure 7b is a schematic diagram of the chip structure proposed in an embodiment of this disclosure. Detailed Implementation

[0028] This disclosure provides communication methods, terminals, network devices, systems, storage media, and program products.

[0029] In a first aspect, embodiments of this disclosure propose a communication method, the method comprising: when the overall performance of a first artificial intelligence (AI) model and a second AI model fails to meet requirements, a network device sends first information to a terminal; wherein the first AI model is deployed on the terminal, the second AI model is deployed on the network device, the first AI model is used to compress a first measurement value to obtain a second measurement value, the second AI model is used to decompress the second measurement value to obtain a third measurement value, the difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model; the first information is used to instruct channel coding of the second measurement value.

[0030] In some alternative embodiments of the first aspect, the method further includes: the network device receiving a fourth measurement value sent by the terminal, the fourth measurement value being obtained by the terminal performing channel coding on the second measurement value; the network device performing channel decoding on the fourth measurement value to obtain a fifth measurement value; the network device determining the transmission status of the second measurement value based on the difference between the fifth measurement value and the second measurement value; and the network device determining, based on the transmission status, whether the reason for the overall performance not meeting the requirements includes the transmission of the second measurement value.

[0031] In some alternative embodiments of the first aspect, if the cause is determined to include the transmission of the second measurement value, the method further includes: the network device sending second information to the terminal, the second information being used for channel coding of the measurement value output by the first AI model.

[0032] In some alternative embodiments of the first aspect, if the cause is determined to include the transmission of the second measurement value, the method further includes: the network device sending third information to the terminal, the third information being used to indicate: deactivating the first AI model; or, switching the first AI model; or, reverting to a non-AI mode.

[0033] In some alternative embodiments of the first aspect, the method further includes: the network device receiving the first measurement value and the second measurement value sent by the terminal; or, the network device measuring the uplink signal to obtain the first measurement value and receiving the second measurement value sent by the terminal; the network device inputting the second measurement value into the second AI model to obtain the third measurement value; and the network device determining the overall performance of the first AI model and the second AI model based on the difference between the first measurement value and the third measurement value.

[0034] In some alternative embodiments of the first aspect, the method further includes: the network device receiving the second measurement value sent by the terminal and inputting the second measurement value into the second AI model to obtain the third measurement value; the network device sending a precoded downlink signal to the terminal, the precoding being determined based on the third measurement value, the precoded downlink signal being used to monitor the overall performance of the first AI model and the second AI model; and the network device receiving fourth information sent by the terminal, the fourth information being used to indicate whether the overall performance of the first AI model and the second AI model meets the requirements.

[0035] In some alternative embodiments of the first aspect, the method further includes: the network device determining that performance monitoring is complete when the overall performance of the first AI model and the second AI model meets the requirements.

[0036] In a second aspect, a communication method is provided, comprising: a terminal receiving first information sent by a network device, the first information being sent when the overall performance of a first artificial intelligence (AI) model and a second AI model fails to meet requirements; wherein the first AI model is deployed on the terminal, the second AI model is deployed on the network device, the first AI model is used to compress a first measurement value to obtain a second measurement value, the second AI model is used to decompress the second measurement value to obtain a third measurement value, the difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model; the first information is used to instruct channel coding of the second measurement value.

[0037] In some alternative embodiments of the second aspect, the method further includes: the terminal performing channel coding on the second measurement value to obtain a fourth measurement value; the terminal sending the fourth measurement value to the network device, the fourth measurement value being used for channel decoding to obtain a fifth measurement value, the difference between the fifth measurement value and the second measurement value being used to determine the transmission status of the second measurement value, the transmission status being used to determine whether the reason for the overall performance not meeting the requirements includes the transmission of the second measurement value.

[0038] In some alternative embodiments of the second aspect, the method further includes: the terminal receiving second information sent by the network device, the second information being sent when it is determined that the cause includes the transmission of the second measurement value, the second information being used to instruct channel coding of the measurement value output by the first AI model.

[0039] In some alternative embodiments of the second aspect, the method further includes: the terminal receiving third information sent by the network device, the third information being sent when it is determined that the cause includes the transmission of the second measurement value, the third information being used to indicate: deactivating the first AI model; or, switching the first AI model; or, reverting to a non-AI mode.

[0040] In some alternative embodiments of the second aspect, the method further includes: the terminal sending the first measurement value and the second measurement value to the network device, the second measurement value being used as input to the second AI model to obtain the third measurement value, and the difference between the first measurement value and the third measurement value being used to determine the overall performance of the first AI model and the second AI model.

[0041] In some alternative embodiments of the second aspect, the method further includes: the terminal sending the second measurement value to the network device, the second measurement value being used as input to the second AI model to obtain the third measurement value; the terminal receiving a precoded downlink signal sent by the network device, the precoding being determined based on the third measurement value; the terminal monitoring the overall performance of the first AI model and the second AI model based on the precoded downlink signal; and the terminal sending fourth information to the network device, the fourth information being used to indicate whether the overall performance of the first AI model and the second AI model meets the requirements.

[0042] In some alternative embodiments of the second aspect, the method further includes: when the overall performance of the first AI model and the second AI model meets the requirements, the terminal determines that performance monitoring is complete.

[0043] Thirdly, a network device is provided, comprising: a transceiver module, configured to send first information to a terminal when the overall performance of a first artificial intelligence (AI) model and a second AI model fails to meet requirements; wherein the first AI model is deployed on the terminal, the second AI model is deployed on the network device, the first AI model is used to compress a first measurement value to obtain a second measurement value, the second AI model is used to decompress the second measurement value to obtain a third measurement value, the difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model; the first information is used to instruct channel coding of the second measurement value.

[0044] Fourthly, a terminal device is provided, comprising: a transceiver module for receiving first information sent by a network device, the first information being sent when the overall performance of a first artificial intelligence (AI) model and a second AI model fails to meet requirements; wherein the first AI model is deployed on the terminal, the second AI model is deployed on the network device, the first AI model is used to compress a first measurement value to obtain a second measurement value, the second AI model is used to decompress the second measurement value to obtain a third measurement value, the difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model; the first information is used to instruct channel coding of the second measurement value.

[0045] Fifthly, a network device is provided, including one or more processors; wherein the network device is configured to perform the first aspect and any one of the communication methods in the first aspect.

[0046] A sixth aspect provides a terminal device, comprising: one or more processors; wherein the terminal device is configured to execute the second aspect and any one of the communication methods in the second aspect.

[0047] A seventh aspect provides a communication system, including a network device and a terminal device, wherein the network device is configured to implement the first aspect and any one of the communication methods in the first aspect, and the terminal device is configured to implement the second aspect and any one of the communication methods in the second aspect.

[0048] Eighthly, a storage medium is provided that stores instructions, which, when executed on a communication device, cause the communication device to perform a communication method as described in the first aspect and any one thereof, or the second aspect and any one thereof.

[0049] Ninthly, embodiments of this disclosure provide a program product that, when executed by a communication device, causes the communication device to perform the method as described in the optional implementations of the first or second aspect.

[0050] In a tenth aspect, embodiments of this disclosure provide a computer program that, when run on a computer, causes the computer to perform the methods described in an optional implementation of the first or second aspect.

[0051] Eleventhly, embodiments of this disclosure provide a chip or chip system. The chip or chip system includes processing circuitry configured to perform the methods described in the optional implementations of the first or second aspect above.

[0052] It is understood that the terminals, access network devices, first network elements, other network elements, core network devices, communication systems, storage media, program products, computer programs, chips, or chip systems involved in the embodiments of this disclosure are all used to execute the methods proposed in the embodiments of this disclosure. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0053] This disclosure provides communication methods, terminals, network devices, systems, storage media, and program products. In some embodiments, the terms "communication method" and "information processing method" can be used interchangeably, as can the terms "communication device" and "information processing device" and "communication device," and the terms "information processing system" and "communication system."

[0054] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0055] In each of the disclosed embodiments, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of the embodiments are consistent and can be referenced by each other. The technical environments of different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0056] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0057] In this embodiment of the disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular expression or a plural expression.

[0058] In the embodiments disclosed herein, "multiple" refers to two or more.

[0059] In some embodiments, the terms “at least one of”, “one or more”, “a plurality of”, “multiple”, etc., may be used interchangeably.

[0060] In some embodiments, the notation "at least one of A and B", "A and / or B", "A in one case, B in another", "in response to one case A, in response to another case B", etc., may include the following technical solutions depending on the situation: in some embodiments, A (execute A regardless of B); in some embodiments, B (execute B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed); in some embodiments, A and B (both A and B are executed). The same applies when there are more branches such as A, B, C, etc.

[0061] In some embodiments, the notation "A or B" may include the following technical solutions, depending on the situation: in some embodiments, A (execution of A regardless of B); in some embodiments, B (execution of B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed). The same applies when there are more branches such as A, B, C, etc.

[0062] The prefixes "first," "second," etc., used in the embodiments of this disclosure are merely for distinguishing different descriptive objects and do not impose restrictions on the position, order, priority, quantity, or content of the descriptive objects. The description of the descriptive objects is found in the claims or the context of the embodiments, and the use of prefixes should not constitute unnecessary restrictions. For example, if the descriptive object is a "field," the ordinal numbers preceding "field" in "first field" and "second field" do not restrict the position or order of the "fields." "First" and "second" do not restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the descriptive object is a "level," the ordinal numbers preceding "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the number of descriptive objects is not limited by ordinal numbers and can be one or more. For example, in "first device," the number of "devices" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device", then "first device" and "second device" can be the same device or different devices, and their types can be the same or different. Similarly, if the object being described is "information", then "first information" and "second information" can be the same information or different information, and their content can be the same or different.

[0063] In some embodiments, “including A,” “containing A,” “for indicating A,” and “carrying A” can be interpreted as directly carrying A or indirectly indicating A.

[0064] In some embodiments, the terms “in response to…”, “in response to determining…”, “in the case of…”, “when…”, “if…”, “if…”, etc., can be used interchangeably.

[0065] In some embodiments, the terms “greater than,” “greater than or equal to,” “not less than,” “more than,” “more than or equal to,” “not less than,” “higher than,” “higher than or equal to,” “not lower than,” and “above” can be used interchangeably, as can the terms “less than,” “less than or equal to,” “not greater than,” “less than,” “less than or equal to,” “not more than,” “lower than,” “lower than or equal to,” “not higher than,” and “below”.

[0066] In some embodiments, the apparatus and device may be interpreted as physical or virtual, and their names are not limited to the names recorded in the embodiments. In some cases, they may also be understood as "equipment", "device", "circuit", "network element", "node", "function", "unit", "section", "system", "network", "chip", "chip system", "entity", "body", etc.

[0067] In some embodiments, "network" can be interpreted as devices included in the network, such as access network devices, core network devices, etc.

[0068] In some embodiments, "access network device (AN device)" may also be referred to as "radio access network device (RAN device)," "base station (BS)," "radio base station," or "fixed station." In some embodiments, it may also be understood as "node," "access point," "transmission point (TP)," "reception point (RP)," "transmission / reception point (TRP)," "panel," "antenna panel," "antenna array," "cell," "macro cell," "small cell," "femto cell," "pico cell," "sector," "cell group," "serving cell," "carrier," "component carrier," or "bandwidth part (BWP)."

[0069] In some embodiments, "terminal" or "terminal device" may be referred to as "user equipment (UE)," "user terminal," "mobile station (MS)," "mobile terminal (MT)," "subscriber station," "mobile unit," "subscriber unit," "wireless unit," "remote unit," "mobile device," "wireless device," "wireless communication device," "remote device," "mobile subscriber station," "access terminal," "mobile terminal," "wireless terminal," "remote terminal," "handset," "user agent," "mobile client," "client," etc.

[0070] In some embodiments, the acquisition of data, information, etc., may comply with the laws and regulations of the country where the location is situated.

[0071] In some embodiments, data, information, etc., may be obtained with the user's consent.

[0072] Furthermore, each element, each row, or each column in the table of this disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0073] Figure 1a is a schematic diagram of a communication system architecture according to an embodiment of the present disclosure.

[0074] As shown in Figure 1a, the communication system 100 includes a terminal 101 and a network device 102.

[0075] In some embodiments, terminal 101 includes, but is not limited to, at least one of the following: mobile phone, wearable device, Internet of Things device, car with communication function, smart car, tablet computer, computer with wireless transceiver function, virtual reality (VR) terminal device, augmented reality (AR) terminal device, wireless terminal device in industrial control, wireless terminal device in self-driving, wireless terminal device in remote medical surgery, wireless terminal device in smart grid, wireless terminal device in transportation safety, wireless terminal device in smart city, and wireless terminal device in smart home.

[0076] In some embodiments, network device 102 may include at least one of access network device and core network device.

[0077] In some embodiments, the access network device is, for example, a node or device that connects a terminal to a wireless network. The access network device may include, but is not limited to, at least one of the following in a 5G communication system: evolved Node B (eNB), next-generation eNB (ng-eNB), next-generation Node B (gNB), node B (NB), home node B (HNB), home evolved node B (HeNB), radio backhaul device, radio network controller (RNC), base station controller (BSC), base transceiver station (BTS), base band unit (BBU), mobile switching center, base station in a 6G communication system, open RAN, cloud RAN, base station in other communication systems, and access node in a Wi-Fi system.

[0078] In some embodiments, the technical solutions of this disclosure can be applied to the Open RAN architecture. In this case, the interfaces between or within access network devices involved in the embodiments of this disclosure can be transformed into internal interfaces of Open RAN. The processes and information interactions between these internal interfaces can be implemented by software or programs.

[0079] In some embodiments, the access network device may be composed of a central unit (CU) and a distributed unit (DU). The CU may also be called a control unit. The CU-DU structure can separate the protocol layer of the access network device. Some of the protocol layer functions are centrally controlled by the CU, while the remaining part or all of the protocol layer functions are distributed in the DU and centrally controlled by the CU. However, this is not the only possibility.

[0080] In some embodiments, a core network device may be a single device comprising one or more network elements, or it may be multiple devices or a group of devices, each comprising all or part of the aforementioned one or more network elements. Network elements may be virtual or physical. The core network may include, for example, at least one of an Evolved Packet Core (EPC), a 5G Core Network (5GCN), or a Next Generation Core (NGC).

[0081] It is understood that the communication system described in this disclosure is for the purpose of more clearly illustrating the technical solutions of this disclosure, and does not constitute a limitation on the technical solutions proposed in this disclosure. As those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions proposed in this disclosure are also applicable to similar technical problems.

[0082] The following embodiments of this disclosure can be applied to the communication system 100 shown in FIG1a, or to some of the main bodies, but are not limited thereto. The main bodies shown in FIG1a are illustrative. The communication system may include all or some of the main bodies in FIG1a, or it may include other main bodies outside of FIG1a. The number and form of each main body are arbitrary. Each main body may be physical or virtual. The connection relationship between the main bodies is illustrative. The main bodies may not be connected or may be connected. The connection may be in any way, such as direct connection or indirect connection, wired connection or wireless connection.

[0083] The embodiments disclosed herein can be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 6th generation mobile communication system (6G), 5G New Radio (NR), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New Radio Access (NX), Future Generation Radio Access (FX), Global System for Mobile Communications (GSM), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), and IEEE 802.20, Ultra-Wideband (UWB), Bluetooth (a registered trademark), Public Land Mobile Network (PLMN) networks, Device-to-Device (D2D) systems, Machine-to-Machine (M2M) systems, Internet of Things (IoT) systems, Vehicle-to-Everything (V2X) systems, systems utilizing other communication methods, and next-generation systems built upon them, etc. Furthermore, multiple systems can be combined (e.g., a combination of LTE or LTE-A with 5G).

[0084] In massively multi-input multiple-output (MIMO) systems, accurate CSI acquisition is crucial for improving system performance. Current 5G communication systems have designed methods for reporting Channel State Information (CSI) based on different types of high-precision Type II codebooks. Compared to low-precision Type I codebooks, high-precision Type II codebooks can significantly improve system performance. However, high-precision Type II codebooks also greatly increase feedback overhead. In particular, as the antenna size and bandwidth increase in MIMO systems, the CSI feedback overhead also increases. To reduce CSI feedback overhead or improve CSI feedback accuracy, research is currently underway on CSI compressed feedback standardization based on AI / ML models, as shown in Figure 1b. Figure 1b is a schematic diagram of CSI compressed feedback and recovery based on a bilateral model. The AI / ML-based CSI generation model (hereinafter referred to as the encoder) located on the terminal side implements CSI compression feedback of the input data, while the AI / ML-based CSI recovery model (hereinafter referred to as the decoder) located on the network side implements CSI recovery. Research shows that AI / ML-based CSI compression feedback can further improve CSI feedback accuracy or reduce CSI feedback overhead.

[0085] Whether it's codebook-based CSI feedback or AI / ML model-based CSI compression feedback, the CSI acquired by the UE needs to be channel-coded before being sent to the base station. CSI feedback and channel coding are executed separately and independently. This separation leads to quantization loss in CSI compression, and the redundancy introduced by channel coding consumes more communication resources. Furthermore, a mismatch between the channel coding and channel state in CSI feedback can cause the "cliff effect" and latency issues. To address these problems, recent research in academia and industry has focused on CSI feedback with joint source-channel coding. Joint source-channel coding CSI feedback leverages the characteristics of semantic communication systems, treating CSI feedback as a source. Through end-to-end AI / ML model optimization, it implements CSI compression feedback and recovery by deploying an encoder and decoder model on the terminal and base station sides respectively. The encoder performs both CSI compression and channel coding functions, as shown in Figure 1c. Figure 1c is a schematic diagram of CSI feedback with joint source and channel coding based on a bilateral AI model. The compressed CSI output from the encoder is modulated and transmitted to the NW side via a wireless channel. The demodulated data is then input to the decoder at the NW side, and the decoder recovers the CSI. Research shows that this AI / ML model design can improve CSI feedback accuracy or reduce CSI feedback overhead.

[0086] In some embodiments, performance monitoring of the bilateral AI model for CSI compression includes the following methods. Different monitoring methods can be categorized into NW-side performance monitoring and UE-side performance monitoring.

[0087] Traditional CSI compressed feedback bilateral AI model monitoring method:

[0088] NW-side monitoring model performance methods:

[0089] Method 1-1: Based on the target CSI reported by the UE through the traditional evolved type (eType) II codebook or the enhanced eType II codebook.

[0090] Methods 1-2: Monitoring based on Sounding Reference Signal (SRS).

[0091] UE-side monitoring model performance methods:

[0092] Method 2-1: Recovery of CSI based on decoder model output on the UE side.

[0093] Method 2-2: Key Performance Indicators (KPIs) directly estimated by UE.

[0094] Methods 2-3: Final performance monitoring results directly estimated by the UE.

[0095] Methods 2-4: Based on the precoding reference signal (such as Channel State Information-Reference Signal (CSI-RS) and Demodulation Reference Signal (DMRS)) transmitted by the NW, where the precoding is determined based on the recovered CSI output by the decoder on the NW side.

[0096] Methods 2-5: Recovering CSI based on sending decoder output via traditional eType II codebook or enhanced eType II codebook on the NW side.

[0097] However, for UE-side performance monitoring methods 2-1 to 2-3, since the results output by the UE based on these three methods or the monitoring results do not pass through the uplink channel, these three methods cannot effectively monitor the performance of the CSI feedback bilateral AI model based on source and channel joint coding.

[0098] Other methods for UE and NW side performance monitoring can also be applied to monitor the performance of the CSI feedback bilateral AI model based on source and channel joint coding. However, unlike traditional Rel-18 / 19 AI-based CSI feedback, the CSI feedback based on source and channel joint coding is not channel-coded, but modulated and transmitted to the NW via the radio channel. When the uplink channel becomes poor, this will cause errors in the transmission of the CSI feedback based on source and channel joint coding, and it cannot correct the transmission errors by using a lower code rate. Traditional AI-based CSI compression does not have this problem, because when the channel deteriorates, the NW will configure the UE to encode the CSI at a lower code rate before transmission, which can ensure the correctness of the transmitted CSI.

[0099] In summary, the performance degradation of the CSI feedback bilateral AI model based on joint source and channel coding is not only due to the AI ​​model itself, but may also be caused by a higher error rate during the CSI process of wireless channel transmission compression, leading to a deterioration in the AI ​​model's inference performance. Therefore, current performance monitoring methods need further improvement.

[0100] Therefore, this disclosure proposes a communication method in which, when the overall performance of a first AI model and a second AI model fails to meet requirements, a network device sends first information to a terminal. The first AI model is deployed on the terminal, and the second AI model is deployed on the network device. The first AI model compresses a first measurement value to obtain a second measurement value, and the second AI model decompresses the second measurement value to obtain a third measurement value. The difference between the first and third measurement values ​​characterizes the overall performance of the first and second AI models. By instructing the second measurement value to be channel-coded using the first information, it is possible to further rule out whether the failure to meet performance requirements is due to the transmission of the second measurement value, thus enabling targeted subsequent operations to improve the problem caused by low performance and increase communication efficiency.

[0101] Figure 2 is a schematic diagram of a communication method interaction according to an embodiment of the present disclosure. As shown in Figure 2, this embodiment of the present disclosure relates to a communication method for a communication system 100, the method including:

[0102] In step S2101, network device 102 determines that the overall performance of the first AI model and the second AI model does not meet the requirements.

[0103] In some embodiments, the terminal or network device may initiate monitoring of the overall performance of the first AI model and the second AI model. Alternatively, monitoring of overall performance may begin after the first AI model and the second AI model have been deployed or activated.

[0104] In some embodiments, a first AI model is deployed on a terminal, and a second AI model is deployed on a network device. The first and second AI models can be referred to as bilateral models. The first AI model is used to compress a first measurement value to obtain a second measurement value, and the second AI model is used to compress the second measurement value to obtain a third measurement value. The difference between the first and third measurement values ​​is used to characterize the overall performance of the first and second AI models. For example, if the difference is large, it is determined that the overall performance of the first and second AI models does not meet the requirements; if the difference is small, it is determined that the overall performance of the first and second AI models meets the requirements.

[0105] It is understandable that the criteria for judging whether the difference is large or small can be set according to the actual situation. For example, as an optional approach, the difference between the first and third measurements can be compared with a threshold. If the difference is greater than or equal to the threshold, the difference is considered large; if the difference is less than the threshold, the difference is considered small. This disclosure does not limit this.

[0106] It is understandable that AI models (the first AI model and / or the second AI model) can also be ML models, and ML models can be understood as a type of AI model.

[0107] In some embodiments, the measurement value may be CSI, but is not limited thereto.

[0108] In some embodiments, the first AI model and the second AI model can be AI models with encoding capabilities. For example, the first AI model can compress and encode the first measurement value to obtain the second measurement value. As another example, the second AI model can decompress and decode the second measurement value.

[0109] In some embodiments, both the terminal and the network device can monitor the overall performance of the first and second models. If the network device monitors, it can directly determine that the overall performance of the first and second AI models does not meet the requirements. If the terminal monitors, it can send fourth information to the network device to indicate whether the overall performance of the first and second AI models meets the requirements, based on the fourth information.

[0110] Optionally, the network device may determine whether the overall performance of the AI ​​model and the second AI model meets the requirements by: the network device receiving a first measurement value and a second measurement value sent by the terminal; or, the network device measuring the uplink signal to obtain a first measurement value and receiving a second measurement value sent by the terminal; the network device inputting the second measurement value into the second AI model to obtain a third measurement value; and the network device determining the overall performance of the first AI model and the second AI model based on the difference between the first measurement value and the third measurement value.

[0111] For example, the terminal reports a first measurement value and a second measurement value. The first measurement value can be a value measured by the terminal, also known as the target measurement value. The second measurement value can be the measurement value output by the first AI model after the terminal inputs the first measurement value; this second measurement value can also be called a compressed measurement value. The network device receives the first and second measurement values, and can input the second measurement value into the second AI model. The second AI model outputs a third measurement value. This third measurement value can also be called a restored measurement value or a decompressed measurement value. The network device compares the differences between the first and third measurement values ​​to determine the overall performance of the first and second AI models.

[0112] For example, the terminal can report the first and second measurements via the eType II codebook or an enhanced eType II codebook.

[0113] For example, taking the measurement value as CSI, the terminal reports the target CSI and the CSI output by the first AI model through the eType II codebook or the enhanced eType II codebook. The network device recovers the CSI and the received target CSI based on the output of the second AI model to determine the performance of the bilateral AI model.

[0114] Optionally, the network device determining whether the overall performance of the AI ​​model and the second AI model meets the requirements may include: the network device receiving a second measurement value sent by the terminal and inputting the second measurement value into the second AI model to obtain a third measurement value; the network device sending a precoded downlink signal to the terminal, the precoding being determined based on the third measurement value, the precoded downlink signal being used to monitor the overall performance of the first AI model and the second AI model; and the network device receiving fourth information sent by the terminal, the fourth information being used to indicate whether the overall performance of the first AI model and the second AI model meets the requirements.

[0115] For example, the terminal reports a second measurement value. This second measurement value can be the value output by the first AI model after the terminal inputs the first measurement value; it can also be called a compressed measurement value. The network device can input the second measurement value into the second AI model to obtain a third measurement value, which can also be called a restored or decompressed measurement value. Based on the third measurement value, the network device can determine the precoding and send the precoded downlink signal to the terminal. The terminal monitors the precoded downlink signal to obtain the overall performance of the first and second AI models and uses fourth information to indicate whether the overall performance of the first and second AI models meets the requirements.

[0116] For example, the terminal obtains the overall performance of the first AI model and the second AI model based on monitoring the precoded downlink signal. This can be achieved by the terminal measuring the precoded downlink signal to obtain a measured value, referred to herein as the sixth measured value for ease of distinction. The sixth measured value is compared with a threshold value to determine the overall performance of the first AI model and the second AI model. For instance, if the sixth measured value is greater than or equal to the threshold value, the overall performance of the first AI model and the second AI model is determined to meet the requirements; if the sixth measured value is less than the threshold value, the overall performance of the first AI model and the second AI model is determined to not meet the requirements.

[0117] For example, the sixth measurement may be a Channel Quality Indicator (CQI), a Signal to Interference plus Noise Ratio (SINR), etc., but is not limited to these.

[0118] In step S2102, if the overall performance of the first AI model and the second AI model does not meet the requirements, the network device 102 sends the first information to the terminal 101.

[0119] In some embodiments, terminal 101 receives first information sent by network device 102, the first information being sent when the overall performance of the first AI model and the second AI model does not meet the requirements.

[0120] In some embodiments, the first information is used to instruct channel coding of the second measurement value.

[0121] Understandably, if the overall performance of the first and second AI models fails to meet requirements, it may be due to the transmission of the second measurement value. For example, since the first AI model has channel coding capabilities, when the terminal feeds back the measurement value output by the first AI model to the network device, it often only performs modulation and other operations on the measurement value without channel coding. However, in traditional feedback, when the uplink channel becomes poor, the network device can configure a lower code rate for the terminal. The terminal then performs channel coding on the measurement value based on the lower code rate before feeding it back, which can ensure the correctness of the measurement value transmission as much as possible. Without channel coding, when the uplink channel becomes poor, it may lead to measurement value transmission errors. Therefore, when the overall performance of the first and second AI models fails to meet requirements, considering that it may be due to a problem with the transmission of the second measurement value, the network device sends a first message to the terminal, instructing the terminal to perform channel coding on the second measurement value, thereby further determining whether the failure of the overall performance of the first and second AI models to meet requirements is due to the transmission of the second measurement value.

[0122] In step S2103, terminal 101 performs channel coding on the second measurement value to obtain the fourth measurement value.

[0123] In some embodiments, the first information may include a code rate, or parameters for determining the code rate. For example, parameters may include, but are not limited to, the adjustment level indicated by the network device, the code rate for Physical Uplink Shared Channel (PUSCH) data transmission, the PUSCH resource size, and the code rate compensation factor for the second measurement. The terminal may perform channel coding on the second measurement based on the code rate in the first information, or the code rate determined based on the parameters in the first information, or the code rate determined by other means (e.g., the code rate pre-configured by the network device), to obtain the fourth measurement.

[0124] In step S2104, terminal 101 sends the fourth measurement value to network device 102.

[0125] In some embodiments, network device 102 receives a fourth measurement value sent by terminal 101.

[0126] In some embodiments, the terminal may send a fourth measurement value to the network device so that the network device can decode the fifth measurement value and compare the difference between the fifth measurement value and the second measurement value to determine the transmission status of the second measurement value.

[0127] In step S2105, network device 102 decodes the fourth measurement value to obtain the fifth measurement value.

[0128] In some embodiments, the network device may decode the fourth measurement value to obtain a fifth measurement value, so that the network device can determine the transmission status of the second measurement value by comparing the difference between the fifth measurement value and the second measurement value.

[0129] In step S2106, network device 102 determines the transmission status of the second measurement value based on the difference between the fifth measurement value and the second measurement value.

[0130] In some embodiments, the difference between the fifth measurement and the second measurement can be used to determine the transmission status of the second measurement. For example, the transmission error rate of the second measurement can be determined.

[0131] In step S2107, network device 102 determines, based on the transmission status, whether the reason why the overall performance of the first AI model and the second AI model does not meet the requirements includes the transmission of the second measurement value.

[0132] In some embodiments, the transmission status can be characterized by the transmission error rate. If the difference between the fifth measurement and the second measurement is large, it indicates a high transmission error rate, and thus it can be determined that the transmission of the second measurement is included as a reason why the overall performance of the first AI model and the second AI model does not meet the requirements. Conversely, if the difference between the fifth measurement and the second measurement is small, it indicates a low transmission error rate, and thus it can be determined that the transmission of the second measurement is not included as a reason why the overall performance of the first AI model and the second AI model does not meet the requirements.

[0133] Understandably, the overall performance of the first and second AI models fails to meet requirements by comparing a first and a third measurement value. The third measurement value is obtained by inputting the second measurement value into the second AI model. If the second measurement value is transmitted incorrectly, the third measurement value will also be inaccurate, leading to the conclusion that the overall performance of the first and second AI models fails to meet requirements when comparing the first and third measurements. Therefore, determining whether the transmission of the second measurement value is incorrect can help determine the reason why the overall performance of the first and second AI models fails to meet requirements.

[0134] In step S2108, if it is determined that the reason why the overall performance of the first AI model and the second AI model does not meet the requirements includes the transmission of the second measurement value, the network device 102 sends the second information to the terminal 101.

[0135] In some embodiments, terminal 101 receives second information sent by network device 102. The second information is sent when it is determined that the reason for the overall performance not meeting the requirements includes the transmission of a second measurement value.

[0136] In some embodiments, the second information is used to instruct the channel coding of the measurement values ​​output by the first AI model.

[0137] In some embodiments, if the reason why the overall performance of the first AI model and the second AI model fails to meet the requirements includes the transmission of the second measurement value, the network device can send second information to the terminal to instruct the terminal to perform channel coding on the measurement value output by the first AI model. That is, the terminal no longer directly feeds back the measurement value output by the first AI model to the network device, but performs channel coding before feeding it back to the network device, thereby solving the performance failure caused by transmission problems.

[0138] It is understood that step S2108 is optional. For example, if the network device determines that the reason why the overall performance of the first AI model and the second AI model does not meet the requirements does not include the transmission of the second measurement value, then step S2108 can be omitted. Alternatively, if the network device determines that the reason why the overall performance of the first AI model and the second AI model does not meet the requirements includes the transmission of the second measurement value, it can send third information to the terminal, i.e., execute step S2109, omitting step S2108.

[0139] In step S2109, if it is determined that the reason why the overall performance of the first AI model and the second AI model does not meet the requirements includes the transmission of the second measurement value, the network device 102 sends the third information to the terminal 101.

[0140] In some embodiments, terminal 101 receives third information sent by network device 102. The third information is sent when it is determined that the reason for the overall performance not meeting the requirements includes the transmission of a second measurement value.

[0141] In some embodiments, the third information is used to indicate: deactivate the first AI model; or switch the first AI model; or revert to a non-AI mode.

[0142] Optionally, the third information can be used to instruct the activation of the first AI model.

[0143] Optionally, the third information can be used to indicate switching to the first AI model. For example, the terminal can switch to a third AI model, which can compress the measurement values ​​and also has channel coding capabilities, but differs from the first AI model. That is, it can switch to a different AI model with the same functionality. As another example, the terminal can switch to a fourth AI model, which can compress the measurement values ​​but does not have channel coding capabilities. That is, it can switch to an AI model with different capabilities. Since the fourth AI model does not have channel coding capabilities, the terminal can perform channel coding on the measurement values ​​output by the fourth AI model before feeding them back to the network device.

[0144] Optionally, the third information can be used to indicate a fallback to a non-AI approach. For example, the terminal can feed back measurement values ​​to the network device in a non-AI manner, thereby avoiding problems caused by performance failures.

[0145] In some embodiments, when a network device instructs a terminal to switch or deactivate the first AI model, it may also switch or deactivate the second AI model, but this is not limited to that. For example, the network device may also instruct the terminal to switch or deactivate the first AI model, but the network device may not switch or deactivate the second AI model.

[0146] It is understood that step S2109 is optional. For example, if the network device determines that the reason why the overall performance of the first AI model and the second AI model does not meet the requirements does not include the transmission of the second measurement value, then step S2109 can be omitted. Alternatively, if the network device determines that the reason why the overall performance of the first AI model and the second AI model does not meet the requirements includes the transmission of the second measurement value, it can send the second information to the terminal, i.e., omit step S2109 and execute step S2108.

[0147] In step S2110, if the overall performance of the first AI model and the second AI model meets the requirements, the terminal device 101 and the network device 102 determine that the performance monitoring is complete.

[0148] In some embodiments, if the overall performance of the first AI model and the second AI model meets the requirements, the terminal device 101 and the network device determine that performance monitoring is complete. For example, no other operations may be required; the first AI model and the second AI model can be used normally for inference.

[0149] The communication method involved in the embodiments of this disclosure may include at least one of steps S2101 to S2110. For example, step S2102 may be implemented as a separate embodiment, but is not limited thereto.

[0150] In some embodiments, steps S2101, S2103 to S2110 are optional and may be omitted or replaced in different embodiments.

[0151] In some embodiments, other optional implementations described before or after the specification corresponding to FIG2 may be referred to.

[0152] Figure 3 is a flowchart illustrating a communication method according to an embodiment of the present disclosure. As shown in Figure 3, this embodiment of the present disclosure relates to a communication method executed by terminal 101, the method including:

[0153] Step S3101: Obtain the first information.

[0154] The optional implementation of step S3101 can be found in the optional implementation of step S2102 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0155] In some embodiments, terminal 101 receives first information sent by network device 102, but is not limited thereto; it may also receive first information sent by other entities.

[0156] In some embodiments, terminal 101 obtains first information as defined by the protocol.

[0157] In some embodiments, terminal 101 obtains first information from upper layer(s).

[0158] In some embodiments, the terminal 101 processes the information to obtain the first information.

[0159] In some embodiments, step S3101 is omitted, and the terminal 101 autonomously implements the function indicated by the first information, or the above function is a default or default setting.

[0160] Step S3102: Channel coding is performed on the second measurement value to obtain the fourth measurement value.

[0161] The optional implementation of step S3102 can be found in the optional implementation of step S2103 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0162] Step S3103: Send the fourth measurement value.

[0163] The optional implementation of step S3103 can be found in the optional implementation of step S2104 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0164] In some embodiments, terminal 101 sends a fourth measurement value to network device 102, but is not limited thereto; it may also send the fourth measurement value to other entities.

[0165] Step S3104: Obtain the second information.

[0166] The optional implementation of step S3104 can be found in the optional implementation of step S2108 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0167] In some embodiments, terminal 101 receives second information sent by network device 102, but is not limited thereto; it may also receive second information sent by other entities.

[0168] In some embodiments, terminal 101 obtains second information as defined by the protocol.

[0169] In some embodiments, terminal 101 obtains second information from upper layer(s).

[0170] In some embodiments, the terminal 101 performs processing to obtain the second information.

[0171] In some embodiments, step S3104 is omitted, and the terminal 101 autonomously implements the function indicated by the second information, or the above function is defaulted or set to default.

[0172] Step S3105: Obtain third information.

[0173] The optional implementation of step S3105 can be found in the optional implementation of step S2109 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0174] In some embodiments, terminal 101 receives third information sent by network device 102, but is not limited thereto; it may also receive third information sent by other entities.

[0175] In some embodiments, terminal 101 obtains third information as defined by the protocol.

[0176] In some embodiments, terminal 101 obtains third information from upper layer(s).

[0177] In some embodiments, terminal 101 processes the information to obtain third information.

[0178] In some embodiments, step S3105 is omitted, and the terminal 101 autonomously implements the function indicated by the third information, or the above function is defaulted or set to default.

[0179] Step S3106: If the overall performance of the first AI model and the second AI model meets the requirements, determine that the performance monitoring is complete.

[0180] The optional implementation of step S3106 can be found in the optional implementation of step S2110 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0181] The communication method involved in the embodiments of this disclosure may include at least one of steps S3101 to S3106. For example, step S3101 may be implemented as a standalone embodiment, but is not limited thereto.

[0182] In some embodiments, steps S3102 to S3106 are optional and may be omitted or replaced in different embodiments.

[0183] In some embodiments, other optional implementations may be described before or after the specification corresponding to FIG3.

[0184] Figure 4 is a flowchart illustrating a communication method according to an embodiment of the present disclosure. As shown in Figure 4, this embodiment of the present disclosure relates to a communication method executed by a network device 102, the method comprising:

[0185] Step S4101: Determine that the overall performance of the first AI model and the second AI model does not meet the requirements.

[0186] The optional implementation of step S4101 can be found in the optional implementation of step S2101 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0187] In step S4102, if the overall performance of the first AI model and the second AI model does not meet the requirements, the first message is sent.

[0188] The optional implementation of step S4102 can be found in the optional implementation of step S2102 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0189] In some embodiments, network device 102 may send first information to terminal 101, but is not limited thereto; it may also send first information to other entities.

[0190] Step S4103: Obtain the fourth measurement value.

[0191] The optional implementation of step S4103 can be found in the optional implementation of step S2104 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0192] In some embodiments, network device 102 receives a fourth measurement value sent by terminal 101, but is not limited thereto; it may also receive a fourth measurement value sent by other entities.

[0193] In some embodiments, network device 102 acquires a fourth measurement value as defined by a protocol.

[0194] In some embodiments, network device 102 obtains a fourth measurement from upper layer(s).

[0195] In some embodiments, network device 102 processes the data to obtain a fourth measurement value.

[0196] In some embodiments, step S4103 is omitted, and the network device 102 autonomously implements the function indicated by the fourth measurement value, or the above function is default or default.

[0197] Step S4104: Decode the fourth measurement value to obtain the fifth measurement value.

[0198] The optional implementation of step S4104 can be found in the optional implementation of step S2105 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0199] Step S4105: Determine the transmission status of the second measurement value based on the difference between the fifth measurement value and the second measurement value.

[0200] The optional implementation of step S4105 can be found in the optional implementation of step S2106 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0201] Step S4106: Determine whether the transmission of the second measurement value is included as a reason why the overall performance of the first AI model and the second AI model does not meet the requirements based on the transmission status.

[0202] The optional implementation of step S4106 can be found in the optional implementation of step S2107 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0203] Step S4107: If it is determined that the reason why the overall performance of the first AI model and the second AI model does not meet the requirements includes the transmission of the second measurement value, send the second information.

[0204] The optional implementation of step S4107 can be found in the optional implementation of step S2108 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0205] In some embodiments, network device 102 may send second information to terminal 101, but is not limited thereto; it may also send second information to other entities.

[0206] Step S4108: If it is determined that the reason why the overall performance of the first AI model and the second AI model does not meet the requirements includes the transmission of the second measurement value, send the third information.

[0207] The optional implementation of step S4108 can be found in the optional implementation of step S2109 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0208] In some embodiments, network device 102 may send third information to terminal 101, but is not limited thereto; it may also send third information to other entities.

[0209] Step S4109: If the overall performance of the first AI model and the second AI model meets the requirements, determine that the performance monitoring is complete.

[0210] The optional implementation of step S4109 can be found in the optional implementation of step S2110 in Figure 2, as well as other related parts in the embodiments involved in Figure 2, which will not be repeated here.

[0211] The communication method involved in the embodiments of this disclosure may include at least one of steps S4101 to S4109. For example, step S4102 may be implemented as a separate embodiment, but is not limited thereto.

[0212] In some embodiments, steps S4101, S4103 to S4109 are optional and may be omitted or substituted in different embodiments.

[0213] In some embodiments, other optional implementations may be described before or after the specification corresponding to Figure 4.

[0214] Figure 5a is an interactive schematic diagram of a communication method according to an embodiment of the present disclosure. As shown in Figure 5a, the present disclosure relates to a communication method, which includes:

[0215] In step S5101, if the overall performance of the first AI model and the second AI model does not meet the requirements, the network device 102 sends the first information to the terminal 101.

[0216] In some embodiments, the above methods may include the methods of the embodiments related to the communication system 100, terminal 101, and network device 102, which will not be described again here.

[0217] This disclosure provides a method for monitoring the performance of a bilateral AI model with CSI compressed feedback based on joint source and channel coding.

[0218] Figure 5b is an interactive schematic diagram of a bilateral model performance monitoring method according to an embodiment of the present disclosure. As shown in Figure 5b, the bilateral model performance monitoring method is as follows:

[0219] Method 1: The NW side first monitors whether the AI ​​model meets the performance requirements:

[0220] Step S5201: The UE or NW initiates bilateral AI model performance monitoring.

[0221] In some embodiments, the UE may be a terminal 101, and the NW may be a network device 102.

[0222] Step S5201 can be omitted; for example, monitoring of AI model performance can begin after the AI ​​model has been deployed or activated.

[0223] In step S5202, the UE reports the target CSI and the first CSI through the eType II codebook or the enhanced eType II codebook.

[0224] In some embodiments, the first CSI is the CSI output by the encoder.

[0225] In some embodiments, the NW recovers the CSI from the decoder output and the received target CSI to determine the bilateral AI model performance. The decoder input is the encoder output CSI. The encoder output CSI is defined as the first CSI, and the first CSI is not channel-coded. If the NW detects that the AI ​​model meets the performance requirements, the performance monitoring is complete. If the NW detects that the AI ​​model does not meet the performance requirements, step S5203 is executed to further monitor whether the reason for the AI ​​model's failure is due to an error in the transmission of the encoder output CSI. The NW also sends channel coding indication information (e.g., first indication information) of the output CSI to the UE. Optionally, the target CSI is measured based on the uplink pilot signal.

[0226] In step S5203, the UE performs channel coding on the first CSI according to the received channel coding indication information, and reports the second CSI to the NW.

[0227] In some embodiments, the output CSI channel coding indication information may be first information, and its name is not limited.

[0228] In some embodiments, the channel-coded output CSI is used as the second CSI.

[0229] In step S5204, NW determines whether the problem is due to a transmission issue with the first CSI based on the third CSI and the first CSI.

[0230] In some embodiments, the third CSI is obtained by the NW after channel decoding the second CSI. If the NW detects that the error rate of the first CSI transmission is within an acceptable range, the NW can deactivate the bilateral model or switch to another AI model, or fall back to the traditional CSI feedback method. If the NW detects that the error rate of the first CSI transmission is not within an acceptable range, the NW sends an indication message (second indication message) to instruct the UE to report the CSI codebook according to the second CSI based on the indicated code rate, or the code rate determined by the UE according to the NW configuration parameters, allocated uplink transmission resource information, etc., or perform other operations such as model switching or falling back to the traditional CSI reporting method. This indication message instructs the UE to report the second CSI according to the code rate determined by the parameters configured by the NW. This indication message is different from the indication message in step S5202. Although both instruct the UE to report the second CSI, the indication message in step S5204 instructs reporting within the performance monitoring time range, while the indication message in step S5204 instructs the UE to also report the second CSI during the inference phase.

[0231] In step S5205, the UE determines whether to report the second CSI during the inference phase based on the received instruction information.

[0232] Figure 5c is an interactive schematic diagram of a bilateral model performance monitoring method according to an embodiment of the present disclosure. As shown in Figure 5c, the bilateral model performance monitoring method is as follows:

[0233] The UE side first monitors whether the AI ​​model meets the performance requirements.

[0234] Steps S5301 and S5201 are similar, that is, the NW side or UE side initiates bilateral AI model performance monitoring, or starts monitoring AI model performance after AI model deployment or activation, which will not be described in detail in this disclosure.

[0235] In step S5302, the NW sends a precoded reference signal to the UE.

[0236] In some embodiments, the pre-coded CSI is recovered by the NW-side decoder. The input to the decoder is the first CSI reported by the UE.

[0237] In step S5303, the UE measures the downlink CSI based on the received precoded reference signal and compares it with the threshold value to determine that the performance of the bilateral AI model meets the performance requirements.

[0238] In some embodiments, the downlink CSI can be CQI or SINR, etc. The threshold value is predefined or configured to the UE by the NW. If the performance monitored by the UE meets the requirements, the UE sends an indication to the NW indicating whether the current AI model meets the requirements. Optionally, the UE reports the measured downlink CSI to the NW, and then the NW determines whether the current bilateral model meets the performance requirements.

[0239] In step S5304, the NW determines the model performance based on the model performance indication information received from the UE.

[0240] In some embodiments, if the indication information indicates that the model does not meet the requirements, step S5305 is executed to further monitor whether the reason why the AI ​​model cannot meet the requirements is due to an error in the CSI output by the encoder. The NW sends channel coding indication information of the output CSI to the UE, which includes relevant uplink transmission resources, modulation scheme and other parameter information.

[0241] In step S5305, the UE performs channel coding on the first CSI according to the received output CSI channel coding indication information, and reports the second CSI obtained after channel coding to the NW.

[0242] In step S5306, NW determines whether the problem is due to a transmission issue with the first CSI based on the third CSI and the first CSI.

[0243] In some embodiments, the third CSI is obtained by channel decoding of the second CSI. If the NW detects that the error rate of the first CSI transmission is within an acceptable range, the NW can deactivate the bilateral model or switch to another AI model, or fall back to the traditional CSI feedback method. If the NW detects that the error rate of the first CSI transmission is not within an acceptable range, the NW instructs the UE to report the CSI according to the second CSI based on the indicated code rate, or the code rate determined by the UE according to the NW configuration parameters, allocated uplink transmission resource information, etc., or perform other operations such as model switching or falling back to the traditional CSI reporting method. This indication information (second indication information) indicates that the second CSI should be reported according to the indicated code rate. This indication information is different from the indication information in step S5304. Although both indicate that the UE should report the second CSI, the indication information in step S5304 indicates that the reporting should be done within the performance monitoring time range, while the indication information in step S5306 indicates that the UE should also report the second CSI during the inference phase.

[0244] In step S5307, the UE determines whether to report the second CSI during the inference phase based on the received instruction information.

[0245] Example 1 (Method 1):

[0246] In a wireless communication network environment, encoder and decoder models are deployed on the NW and UE sides, respectively. The encoder model implements CSI compression and channel coding functions; that is, the data output by the encoder is modulated and processed before being sent to the NW side. Finally, the decoder on the NW side infers a CSI similar to that before compression. Due to channel variations, it is necessary to ensure that the deployed AI models meet certain performance requirements to optimize wireless communication quality. Therefore, performance monitoring of these two models is required.

[0247] Step 1: Start performance monitoring.

[0248] After the encoder / decoder is deployed, or during the application process, the performance monitoring process is initiated through the NW's internal monitoring mechanism or a request from the UE side. For example, the NW sends a control signal to the UE to instruct it to start monitoring the AI ​​model performance.

[0249] Step 2: CSI reporting and performance determination.

[0250] After receiving an instruction from the NW side to monitor the bilateral model, the UE reports the target CSI via an eType II codebook or an enhanced eType II codebook. Additionally, the UE's encoder compresses the CSI and sends it to the NW. The NW's decoder receives the CSI recovered from the encoder output and compares it with the received target CSI to determine the AI ​​model's performance. If the performance meets the requirements, monitoring ends; otherwise, the NW instructs the UE to proceed to the next step. For example, it instructs the UE to input the CSI into another encoder model (the second encoder) that only has CSI compression functionality, and the NW sends a first instruction via signaling to instruct the UE to perform channel coding on the CSI output from the second encoder.

[0251] Step 3: Further monitoring.

[0252] The UE compresses the CSI using the second encoder based on the received instruction information to obtain the second CSI, and then reports the second CSI to the NW after channel coding.

[0253] Step 4: Determine if the CSI issue is a transmission problem.

[0254] The NW compares the first CSI and the third CSI to determine if the performance failure is due to CSI transmission issues. If the first CSI transmission error rate is not within acceptable limits, the NW can choose to activate the AI ​​model or switch to another model; if it is still not within acceptable limits, the NW will send a second indication message to the UE, requesting the UE to report the second CSI according to the code rate determined by the configured parameter information.

[0255] Step 5: CSI reporting during the inference phase.

[0256] The UE determines whether to continue reporting the second CSI during the inference phase based on the received second indication information.

[0257] Example 2 (Method 2):

[0258] The assumptions for monitoring model performance are similar to those in Example 1. Example 2 involves the User Equipment (UE) first monitoring the performance of the AI ​​model, and then engaging in a series of interactions with the Network (NW) to determine whether the AI ​​model meets performance requirements. The specific implementation steps and examples are as follows.

[0259] Step 1: Start performance monitoring.

[0260] After the encoder / decoder is deployed, or during the application process, the performance monitoring process is initiated through the NW's internal monitoring mechanism or a request from the UE side. For example, the NW sends a control signal to the UE to instruct it to start monitoring the AI ​​model performance.

[0261] Step 2: Transmission of precoded reference signal.

[0262] The UE reports the first CSI output from the encoder to the NW. The first CSI reported by the UE is used as the input to the decoder. The NW determines the precoding of the transmission CSI-RS based on the received first CSI and the output of the decoder. The NW then sends the precoding reference signal to the UE.

[0263] Step 3: Measurement and comparison of downlink CSI.

[0264] The UE measures the downlink CSI (such as CQI or SINR) based on the received precoded reference signal and compares it with a predefined or NW-configured threshold value to determine whether the AI ​​model meets the performance requirements. For example, if the downlink CSI measured by the UE is CQI, it compares it with the threshold value. If the CQI value is greater than the threshold, the AI ​​model is considered to meet the performance requirements. The UE sends a message to the NW indicating that the current model meets the performance requirements. If the model does not meet the performance requirements, the UE also sends an indication message to the NW indicating the model's performance status.

[0265] Step 4: Determining model performance.

[0266] If the indication information received by the NW indicates that the current model cannot meet the performance requirements and the performance of the AI ​​model cannot be monitored, the next step will be executed to further investigate the reason why the AI ​​model cannot meet the requirements. The NW indicates the uplink transmission MCS through DCI or the uplink transmission PUCCH or PUSCH resource size and other parameter information through RRC signaling.

[0267] Step 5: Report the channel coding CSI.

[0268] The UE determines the channel coding rate for transmitting the CSI based on the received instruction information, performs channel coding on the CSI output by the encoder to obtain the second CSI, and reports it to the NW.

[0269] Step 6: Identifying CSI transmission issues.

[0270] Similar to Step 4 in Implementation 1, the NW compares the first CSI and the third CSI to determine whether the performance is not meeting requirements due to CSI transmission issues. If the first CSI transmission error rate is not within an acceptable range, the NW can choose to activate the AI ​​model or switch to another model; if it is still not within an acceptable range, the NW will send a second indication message to the UE, requesting the UE to report the second CSI according to the code rate determined by the configured parameter information.

[0271] Step 7: CSI reporting during the inference phase.

[0272] The UE determines whether to continue reporting the second CSI during the inference phase based on the received second indication information.

[0273] This disclosure also provides an apparatus for implementing any of the above methods. For example, an apparatus is provided that includes units or modules for implementing the steps performed by the terminal in any of the above methods. Alternatively, another apparatus is provided that includes units or modules for implementing the steps performed by a network device (e.g., an access network device, a core network functional node, a core network device, etc.) in any of the above methods.

[0274] It should be understood that the division of units or modules in the above device is only a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, the units or modules in the device can be implemented by a processor calling software: for example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of the units or modules in the above device. The processor can be, for example, a general-purpose processor, such as a Central Processing Unit (CPU) or a microprocessor, and the memory can be internal or external to the device. Alternatively, the units or modules in the device can be implemented in the form of hardware circuits. The functionality of some or all of the units or modules can be achieved through the design of these hardware circuits, which can be understood as one or more processors. For example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC). The functionality of some or all of the units or modules is achieved through the design of the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a programmable logic device (PLD). Taking a field-programmable gate array (FPGA) as an example, it can include a large number of logic gates. The connection relationships between the logic gates are configured through configuration files, thereby achieving the functionality of some or all of the units or modules. All units or modules of the above device can be implemented entirely through processor-called software, entirely through hardware circuits, or partially through processor-called software with the remaining parts implemented through hardware circuits.

[0275] In this embodiment, the processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a Central Processing Unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), or a digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. The logical relationships of the aforementioned hardware circuits are fixed or reconfigurable. For example, the processor is a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units or modules. Furthermore, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a Neural Network Processing Unit (NPU), a Tensor Processing Unit (TPU), or a Deep Learning Processing Unit (DPU).

[0276] Figure 6a is a schematic diagram of the terminal structure proposed in an embodiment of this disclosure. As shown in Figure 6a, the terminal 6100 may include at least one of a transceiver module 6101 and a processing module 6102. The transceiver module 6101 is used to receive first information sent by a network device. The first information is sent when the overall performance of the first artificial intelligence (AI) model and the second AI model does not meet the requirements. The first AI model is deployed on the terminal, and the second AI model is deployed on the network device. The first AI model is used to compress a first measurement value to obtain a second measurement value, and the second AI model is used to decompress the second measurement value to obtain a third measurement value. The difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model. The first information is used to instruct channel coding of the second measurement value.

[0277] In some embodiments, the transceiver module 6101 is further configured to: perform channel coding on the second measurement value to obtain a fourth measurement value; send the fourth measurement value to the network device, the fourth measurement value is used for channel decoding to obtain a fifth measurement value, the difference between the fifth measurement value and the second measurement value is used to determine the transmission status of the second measurement value, and the transmission status is used to determine whether the reason for the overall performance not meeting the requirements includes the transmission of the second measurement value.

[0278] In some embodiments, the transceiver module 6101 is further configured to: receive second information sent by a network device, the second information being sent when the cause of transmission includes the second measurement value, the second information being used to instruct channel coding of the measurement value output by the first AI model.

[0279] In some embodiments, the transceiver module 6101 is further configured to: send third information to the terminal, the third information being sent when the cause of transmission includes the second measurement value, the third information being used to indicate: deactivating the first AI model; or, switching the first AI model; or, reverting to a non-AI mode.

[0280] In some embodiments, the transceiver module 6101 is further configured to: send a first measurement value and a second measurement value to a network device, wherein the second measurement value is used to input a second AI model to obtain a third measurement value, and the difference between the first measurement value and the third measurement value is used to determine the overall performance of the first AI model and the second AI model.

[0281] In some embodiments, the transceiver module 6101 is further configured to: send a second measurement value to the network device, the second measurement value being used to input the second AI model to obtain a third measurement value; receive a precoded downlink signal sent by the network device, the precoding being determined based on the third measurement value; monitor the overall performance of the first AI model and the second AI model based on the precoded downlink signal; and send fourth information to the network device, the fourth information being used to indicate whether the overall performance of the first AI model and the second AI model meets the requirements.

[0282] In some embodiments, the processing module 6102 is used to determine that performance monitoring is complete when the overall performance of the first AI model and the second AI model meets the requirements.

[0283] Figure 6b is a schematic diagram of the network device proposed in an embodiment of this disclosure. As shown in Figure 6b, the network device 6200 may include at least one of a transceiver module 6201 and a processing module 6202. The transceiver module 6201 is used to send first information to the terminal when the overall performance of the first artificial intelligence (AI) model and the second AI model does not meet the requirements. The first AI model is deployed on the terminal, and the second AI model is deployed on the network device. The first AI model is used to compress a first measurement value to obtain a second measurement value, and the second AI model is used to decompress the second measurement value to obtain a third measurement value. The difference between the first and third measurement values ​​is used to characterize the overall performance of the first and second AI models. The first information is used to instruct channel coding of the second measurement value.

[0284] In some embodiments, the transceiver module 6201 is further configured to: receive a fourth measurement value sent by the terminal, the fourth measurement value being obtained by the terminal performing channel coding on the second measurement value; the processing module 6202 is configured to: perform channel decoding on the fourth measurement value to obtain a fifth measurement value; determine the transmission status of the second measurement value based on the difference between the fifth measurement value and the second measurement value; and determine, based on the transmission status, whether the reason for the overall performance not meeting the requirements includes the transmission of the second measurement value.

[0285] In some embodiments, if the determined cause includes the transmission of a second measurement value, the transceiver module 6201 is further configured to: send second information to the terminal, the second information being used for channel coding of the measurement value output by the first AI model.

[0286] In some embodiments, if the cause is determined to include the transmission of the second measurement value, the transceiver module 6201 is further configured to: send third information to the terminal via the network device, the third information being used to indicate: deactivate the first AI model; or, switch the first AI model; or, revert to a non-AI mode.

[0287] In some embodiments, the transceiver module 6201 is further configured to: receive a first measurement value and a second measurement value sent by the terminal; or, the processing module 6202 is further configured to: measure the uplink signal to obtain a first measurement value and receive a second measurement value sent by the terminal; the network device inputs the second measurement value into the second AI model to obtain a third measurement value; the network device determines the overall performance of the first AI model and the second AI model based on the difference between the first measurement value and the third measurement value.

[0288] In some embodiments, the transceiver module 6201 is further configured to: receive a second measurement value sent by the terminal, and input the second measurement value into the second AI model to obtain a third measurement value; the network device sends a precoded downlink signal to the terminal, the precoding being determined based on the third measurement value, the precoded downlink signal being used to monitor the overall performance of the first AI model and the second AI model; the network device receives fourth information sent by the terminal, the fourth information being used to indicate whether the overall performance of the first AI model and the second AI model meets the requirements.

[0289] In some embodiments, the processing module 6202 is further configured to: determine that performance monitoring is complete when the overall performance of the first AI model and the second AI model meets the requirements.

[0290] Figure 7a is a schematic diagram of a communication device according to an embodiment of this disclosure. The communication device 7100 can be a network device, a terminal device, or a chip, chip system, or processor that supports the implementation of any of the above methods in a network device, or a chip, chip system, or processor that supports the implementation of any of the above methods in a terminal device. Optionally, the network device can be an access network device, a core network device, etc. Optionally, the terminal device can be a user equipment, etc. The communication device 7100 can be used to implement the methods described in the above method embodiments; for details, please refer to the descriptions in the above method embodiments.

[0291] As shown in Figure 7a, the communication device 7100 includes one or more processors 7101. The processor 7101 can be a general-purpose processor or a dedicated processor, such as a baseband processor or a central processing unit (CPU). The baseband processor can be used to process communication protocols and communication data, while the CPU can be used to control the communication device, execute programs, and process program data. The communication device 7100 is used to execute any of the above methods. Optionally, the communication device can be a base station, a baseband chip, a terminal device, a terminal device chip, a DU (Distributed Unit), or a CU (Computer Integrated Circuit), etc.

[0292] In some embodiments, the communication device 7100 further includes one or more memories 7102 for storing instructions. Optionally, all or part of the memories 7102 may also be located outside the communication device 7100.

[0293] In some embodiments, the communication device 7100 further includes one or more transceivers 7103. When the communication device 7100 includes one or more transceivers 7103, the transceivers 7103 perform communication steps such as sending and / or receiving in the above method, such as step S2102, but are not limited thereto. The processor 7201 performs other steps, such as step S2101, but is not limited thereto.

[0294] In some embodiments, a transceiver may include a receiver and / or a transmitter, which may be separate or integrated. Optionally, the terms transceiver, transceiver unit, transceiver, transceiver circuit, etc., may be used interchangeably; the terms transmitter, transmitting unit, transmitter, transmitting circuit, etc., may be used interchangeably; and the terms receiver, receiving unit, receiver, receiving circuit, etc., may be used interchangeably.

[0295] In some embodiments, the communication device 7100 may include one or more interface circuits 7104. Optionally, the interface circuit 7104 is connected to the memory 7102, and the interface circuit 7104 can be used to receive signals from the memory 7102 or other devices, and can be used to send signals to the memory 7102 or other devices. For example, the interface circuit 7104 can read instructions stored in the memory 7102 and send the instructions to the processor 7101.

[0296] The communication device 7100 described in the above embodiments may be a network device or a terminal device, but the scope of the communication device 7100 described in this disclosure is not limited thereto, and the structure of the communication device 7100 may not be limited by FIG. 7a. The communication device may be a standalone device or a part of a larger device. For example, the communication device may be: (1) a standalone integrated circuit IC, or chip, or chip system or subsystem; (2) a collection of one or more ICs, optionally, the IC collection may also include storage components for storing data and programs; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, terminal device, smart terminal device, cellular phone, wireless device, handheld device, mobile unit, vehicle device, network device, cloud device, artificial intelligence device, etc.; (6) others, etc.

[0297] Figure 7b is a schematic diagram of the chip structure proposed in an embodiment of this disclosure. For cases where the communication device 7100 can be a chip or a chip system, please refer to the schematic diagram of the chip 7200 shown in Figure 7b, but it is not limited thereto.

[0298] Chip 7200 includes one or more processors 7201, which are used to perform any of the above methods.

[0299] In some embodiments, chip 7200 further includes one or more interface circuits 7202. Optionally, the interface circuit 7202 is connected to memory 7203, and the interface circuit 7202 can be used to receive signals from memory 7203 or other devices, and the interface circuit 7202 can be used to send signals to memory 7203 or other devices. For example, the interface circuit 7202 can read instructions stored in memory 7203 and send the instructions to processor 7201.

[0300] In some embodiments, the interface circuit 7202 performs communication steps such as sending and / or receiving in the above method, such as step S2102, but is not limited thereto. The processor 7201 performs other steps, such as step S2101, but is not limited thereto.

[0301] In some embodiments, the terms interface circuit, interface, transceiver pin, transceiver, etc., can be used interchangeably.

[0302] In some embodiments, chip 7200 further includes one or more memories 7203 for storing instructions. Optionally, all or part of the memories 7203 may be located outside of chip 7200.

[0303] This disclosure also proposes a storage medium storing instructions that, when executed on the communication device 7100, cause the communication device 7100 to perform any of the above methods. Optionally, the storage medium is an electronic storage medium. Optionally, the storage medium is a computer-readable storage medium, but not limited thereto; it may also be a storage medium readable by other devices. Optionally, the storage medium may be a non-transitory storage medium, but not limited thereto; it may also be a temporary storage medium.

[0304] This disclosure also provides a program product that, when executed by the communication device 7100, causes the communication device 7100 to perform any of the above methods. Optionally, the program product is a computer program product.

[0305] This disclosure also proposes a computer program that, when run on a computer, causes the computer to perform any of the above methods.

Claims

1. A communication method, characterized in that, The method includes: If the overall performance of the first AI model and the second AI model does not meet the requirements, the network device sends the first information to the terminal; The first AI model is deployed on the terminal, and the second AI model is deployed on the network device. The first AI model is used to compress a first measurement value to obtain a second measurement value, and the second AI model is used to decompress the second measurement value to obtain a third measurement value. The difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model. The first information is used to instruct channel coding of the second measurement value.

2. The method according to claim 1, characterized in that, The method further includes: The network device receives a fourth measurement value sent by the terminal, the fourth measurement value being obtained by the terminal through channel coding of the second measurement value; The network device performs channel decoding on the fourth measurement value to obtain the fifth measurement value; The network device determines the transmission status of the second measurement value based on the difference between the fifth measurement value and the second measurement value; The network device determines, based on the transmission status, whether the reason for the overall performance not meeting the requirements includes the transmission of the second measurement value.

3. The method according to claim 2, characterized in that, If the cause is determined to include the transmission of the second measurement value, the method further includes: The network device sends a second message to the terminal, the second message being used to instruct channel coding of the measurement values ​​output by the first AI model.

4. The method according to claim 2, characterized in that, If the cause is determined to include the transmission of the second measurement value, the method further includes: the network device sending third information to the terminal, the third information indicating: Deactivate the first AI model; or, Switch to the first AI model; or, Revert to a non-AI mode.

5. The method according to any one of claims 1-4, characterized in that, The method further includes: The network device receives the first measurement value and the second measurement value sent by the terminal; or, the network device measures the uplink signal to obtain the first measurement value and receives the second measurement value sent by the terminal. The network device inputs the second measurement value into the second AI model to obtain the third measurement value; The network device determines the overall performance of the first AI model and the second AI model based on the difference between the first measurement value and the third measurement value.

6. The method according to any one of claims 1-4, characterized in that, The method further includes: The network device receives the second measurement value sent by the terminal and inputs the second measurement value into the second AI model to obtain the third measurement value; The network device sends a precoded downlink signal to the terminal. The precoding is determined based on the third measurement value. The precoded downlink signal is used to monitor the overall performance of the first AI model and the second AI model. The network device receives a fourth message sent by the terminal, the fourth message indicating whether the overall performance of the first AI model and the second AI model meets the requirements.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: If the overall performance of the first AI model and the second AI model meets the requirements, the network device determines that performance monitoring is complete.

8. A communication method, characterized in that, The method includes: The terminal receives first information sent by the network device, which is sent when the overall performance of the first artificial intelligence (AI) model and the second AI model does not meet the requirements. The first AI model is deployed on the terminal, and the second AI model is deployed on the network device. The first AI model is used to compress a first measurement value to obtain a second measurement value, and the second AI model is used to decompress the second measurement value to obtain a third measurement value. The difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model. The first information is used to instruct channel coding of the second measurement value.

9. The method according to claim 8, characterized in that, The method further includes: The terminal performs channel coding on the second measurement value to obtain the fourth measurement value; The terminal sends the fourth measurement value to the network device. The fourth measurement value is used for channel decoding to obtain the fifth measurement value. The difference between the fifth measurement value and the second measurement value is used to determine the transmission status of the second measurement value. The transmission status is used to determine whether the reason why the overall performance does not meet the requirements includes the transmission of the second measurement value.

10. The method according to claim 9, characterized in that, The method further includes: The terminal receives second information sent by the network device, the second information being sent when it is determined that the cause includes the transmission of the second measurement value, the second information being used to instruct channel coding of the measurement value output by the first AI model.

11. The method according to claim 9, characterized in that, The method further includes: The terminal receives third information sent by the network device, the third information being sent when it is determined that the cause includes the transmission of the second measurement value, and the third information is used to indicate: Deactivate the first AI model; or, Switch to the first AI model; or, Revert to a non-AI mode.

12. The method according to any one of claims 8-11, characterized in that, The method further includes: The terminal sends the first measurement value and the second measurement value to the network device. The second measurement value is used to input the second AI model to obtain the third measurement value. The difference between the first measurement value and the third measurement value is used to determine the overall performance of the first AI model and the second AI model.

13. The method according to any one of claims 8-11, characterized in that, The method further includes: The terminal sends the second measurement value to the network device, and the second measurement value is used to input the second AI model to obtain the third measurement value; The terminal receives a precoded downlink signal sent by the network device, the precoding being determined based on the third measurement value; The terminal monitors the overall performance of the first AI model and the second AI model based on the precoded downlink signal. The terminal sends a fourth message to the network device, the fourth message indicating whether the overall performance of the first AI model and the second AI model meets the requirements.

14. The method according to any one of claims 8-13, characterized in that, The method further includes: If the overall performance of the first AI model and the second AI model meets the requirements, the terminal determines that performance monitoring is complete.

15. A network device, characterized in that, include: The transceiver module is used to send the first information to the terminal when the overall performance of the first AI model and the second AI model does not meet the requirements. The first AI model is deployed on the terminal, and the second AI model is deployed on the network device. The first AI model is used to compress a first measurement value to obtain a second measurement value, and the second AI model is used to decompress the second measurement value to obtain a third measurement value. The difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model. The first information is used to instruct channel coding of the second measurement value.

16. A terminal, characterized in that, include: The transceiver module is used to receive first information sent by the network device. The first information is sent when the overall performance of the first artificial intelligence (AI) model and the second AI model does not meet the requirements. The first AI model is deployed on the terminal, and the second AI model is deployed on the network device. The first AI model is used to compress a first measurement value to obtain a second measurement value, and the second AI model is used to decompress the second measurement value to obtain a third measurement value. The difference between the first measurement value and the third measurement value is used to characterize the overall performance of the first AI model and the second AI model. The first information is used to instruct channel coding of the second measurement value.

17. A network device, characterized in that, include: One or more processors; The processor is used to execute the communication method according to any one of claims 1-7.

18. A terminal, characterized in that, include: One or more processors; The processor is used to execute the communication method according to any one of claims 8-14.

19. A communication system, characterized in that, include: A network device and a terminal, wherein the network device is configured to implement the communication method of any one of claims 1-7, and the terminal is configured to implement the communication method of any one of claims 8-14.

20. A storage medium, characterized in that, include: The storage medium stores instructions that, when executed on a communication device, cause the communication device to perform the communication method as described in any one of claims 1-7 or 8-14.

21. A program product, characterized in that, include: A computer program, when executed by a communication device, causes the communication device to perform the communication method as described in any one of claims 1-7 or 8-14.

Images