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

Abstract

The present disclosure relates to a communication method, a terminal, a network device, a system and a storage medium. The method comprises: receiving first information sent by a network device, the first information comprising configuration parameter information for measuring channel state information (CSI) and / or configuration parameter information for predicting CSI; determining a first AI model according to the first information, the first AI model being used for CSI prediction. Thus, the consistency of inference models and training models is solved, and performance loss caused by the mismatch between inference models and training models is avoided.

Description

Communication method, terminal, network device, system and storage medium TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of communication, and particularly relates to a communication method, a terminal, a network device, a system and a storage medium. BACKGROUND

[0002] With the development of application of AI (Artificial Intelligence) technology, AI has been widely applied to a wireless communication physical layer. In the related art, CSI prediction based on a CSI prediction algorithm can also use an AI model to infer and predict channel information at a future time. Current simulation evaluation shows that the prediction performance based on AI is better than the prediction performance of a traditional non-AI algorithm.

[0003] SUMMARY

[0004] To overcome the technical problem of AI model performance degradation in the related art, the present disclosure provides a communication method, a terminal, a network device, a system and a storage medium.

[0005] According to a first aspect of an embodiment of the present disclosure, a communication method is provided, which is performed by a terminal, and the method comprises:

[0006] receiving first information sent by a network device, wherein the first information comprises configuration parameter information for measuring channel state information (CSI) and / or configuration parameter information for predicting CSI;

[0007] determining a first AI model according to the first information, wherein the first AI model is used for CSI prediction.

[0008] According to a second aspect of an embodiment of the present disclosure, a communication method is provided, which is performed by a network device, and the method comprises:

[0009] sending first information to a terminal, wherein the first information comprises configuration parameter information for measuring CSI and / or configuration parameter information for predicting CSI, and the first information is used for the terminal to determine a first AI model according to the first information, wherein the first AI model is used for CSI prediction.

[0010] According to a third aspect of an embodiment of the present disclosure, a terminal is provided, comprising:

[0011] a transceiver module, configured to receive first information sent by a network device, wherein the first information comprises configuration parameter information for measuring channel state information (CSI) and / or configuration parameter information for predicting CSI;

[0012] a processing module, configured to determine a first AI model according to the first information, wherein the first AI model is used for CSI prediction.

[0013] According to a fourth aspect of embodiments of the present disclosure, a network device is provided, comprising:

[0014] a transceiver configured to transmit first information to a terminal, the first information comprising configuration parameter information for measuring CSI and / or configuration parameter information for predicting CSI, the first information being used by the terminal to determine a first AI model for CSI prediction according to the first information.

[0015] According to a fifth aspect of embodiments of the present disclosure, a terminal is provided, comprising:

[0016] one or more processors;

[0017] The terminal is configured to perform the communication method according to any one of the first aspect of the present disclosure.

[0018] According to a sixth aspect of embodiments of the present disclosure, a network device is provided, comprising:

[0019] one or more processors;

[0020] The network device is configured to perform the communication method according to any one of the second aspect of the present disclosure.

[0021] According to a seventh aspect of embodiments of the present disclosure, a communication system is provided, comprising:

[0022] a terminal configured to perform the method according to any one of the first aspect of the present disclosure;

[0023] a network device configured to perform the method according to any one of the second aspect of the present disclosure.

[0024] According to an eighth aspect of embodiments of the present disclosure, a storage medium is provided, the storage medium storing instructions, when the instructions are executed on a communication device, causing the communication device to perform the communication method according to any one of the first aspect of the present disclosure, or causing the communication device to perform the communication method according to any one of the second aspect of the present disclosure.

[0025] According to a ninth aspect of embodiments of the present disclosure, a computer program product is provided, comprising a computer program and / or instructions, when the computer program and / or instructions are executed on a communication device, implementing the communication method according to any one of the first aspect of the present disclosure, or when the computer program and / or instructions are executed on a communication device, implementing the communication method according to any one of the second aspect of the present disclosure.

[0026] With the above technical solutions, at least the following beneficial technical effects can be achieved:

[0027] The network device sends first information, and the first information includes configuration parameter information of measuring channel state information (CSI) and / or configuration parameter information of predicting the CSI. According to the first information, a first AI model is determined, and the first AI model is used for CSI prediction. In this way, the consistency of the inference model and the training model is solved, and the performance loss caused by the mismatch between the inference model and the training model is avoided. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following describes the drawings required for the embodiments, and the following drawings are only some embodiments of the present disclosure, and do not specifically limit the protection scope of the present disclosure.

[0029] FIG. 1A is a schematic diagram of an architecture of a communication system according to an embodiment of the present disclosure.

[0030] FIG. 1B is a schematic diagram of an observation window and a prediction window according to an embodiment of the present disclosure.

[0031] FIG. 2A is a schematic diagram of an interaction of a communication method according to an embodiment of the present disclosure.

[0032] FIG. 2B is a schematic diagram of parameter information and CSI-RS resource transmission according to an embodiment of the present disclosure.

[0033] FIG. 2C is a schematic diagram of an interaction flow of a communication method according to an embodiment of the present disclosure.

[0034] FIG. 3 is a schematic diagram of a flow of a communication method according to an embodiment of the present disclosure.

[0035] FIG. 4 is a schematic diagram of a flow of a communication method according to an embodiment of the present disclosure.

[0036] FIG. 5A is a schematic diagram of an interaction of a communication method according to an embodiment of the present disclosure.

[0037] FIG. 5B is a schematic diagram of an interaction of a communication method according to an embodiment of the present disclosure.

[0038] FIG. 6 is a schematic diagram of a structure of a terminal according to an embodiment of the present disclosure.

[0039] FIG. 7 is a schematic diagram of a structure of a network device according to an embodiment of the present disclosure.

[0040] FIG. 8 is a schematic diagram of a structure of a communication device 8100 according to an embodiment of the present disclosure.

[0041] FIG. 9 is a schematic diagram of a structure of a chip 8200 according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0042] The embodiments of the present disclosure provide a communication method, a terminal, a network device, a system and a storage medium.

[0043] In a first aspect, the embodiments of the present disclosure provide a communication method performed by a terminal, comprising:

[0044] receiving first information sent by a network device, wherein the first information comprises configuration parameter information of measured channel state information (CSI) and / or configuration parameter information of predicted CSI;

[0045] determining a first AI model according to the first information, wherein the first AI model is used for CSI prediction.

[0046] In some embodiments of the first aspect, determining the first AI model according to the first information comprises:

[0047] obtaining a first training data set according to the first information;

[0048] performing model training on a second AI model according to the first training data set to generate the first AI model, wherein the second AI model is a preset AI model.

[0049] In the above embodiments, the terminal obtains a training data set according to the first information configured by the network device to perform model training, thereby ensuring consistency of the trained model and the inference model and service performance of the AI model.

[0050] In some embodiments of the first aspect, determining the first AI model according to the first information comprises:

[0051] determining that the first information is a subset of configuration parameters of a second training data set, wherein the second training data set is a training data set corresponding to a third AI model, and the third AI model is obtained by training according to the second training data set;

[0052] taking the third AI model as the first AI model.

[0053] In the above embodiments, when it is determined that the first information is a subset of configuration parameters of an AI model that has been trained, the AI model is taken as an AI model for inference, thereby ensuring service performance of the AI model.

[0054] In some embodiments of the first aspect, the first information comprises at least one of:

[0055] a number of reference signal resources for measuring the CSI;

[0056] a time domain behavior of each reference signal resource;

[0057] a time domain interval between adjacent measurement CSIs;

[0058] a number of predicted CSIs;

[0059] a time domain interval between adjacent predicted CSIs;

[0060] a time domain interval between a first predicted CSI and a reporting time of the predicted CSI;

[0061] time information of reporting the predicted CSI.

[0062] In the above embodiment, the first information can include multiple parameter types, and the network device can configure different types of parameter information based on different network environments. Therefore, the AI model used for inference has high robustness.

[0063] In some embodiments of the first aspect, the time domain behavior includes at least one of the following:

[0064] periodic time domain behavior;

[0065] semi-persistent periodic time domain behavior;

[0066] aperiodic time domain behavior.

[0067] In the above embodiment, the AI model-based CSI prediction in the communication system can be applied to multiple time domain behaviors of reference signal resources, so that the inference AI model can be applied to multiple time domain behavior scenarios, and the robustness of the AI model is improved.

[0068] In some embodiments of the first aspect, the first information includes an association ID, and the association ID is used to indicate at least one of the following:

[0069] first condition information of the terminal, the first condition information being used to indicate configuration information or a running state of the terminal;

[0070] a combination mode of the measurement CSI configuration parameter information and / or the predicted CSI configuration information;

[0071] second condition information of the network device, the second condition information being used to indicate a configuration state of the network device.

[0072] In the above embodiment, the network environment supported by the network device is indicated to the terminal by the association ID, the information overhead in the indication process is reduced, and the CSI prediction efficiency is improved.

[0073] In some embodiments of the first aspect, the determining the first AI model according to the first information includes:

[0074] determine, as the first AI model, an AI model matching the association ID from a plurality of third AI models, the third AI models being pre-trained AI models.

[0075] In the above embodiment, the configuration mode of the terminal supporting parameter information is adopted in the manner of association ID, the overhead in the signal transmission process is reduced, and the CSI prediction efficiency is improved.

[0076] In combination with some embodiments of the first aspect, in some embodiments, the method further includes:

[0077] sending the first condition information to the network device, the first condition information being used for the network device to determine the first information according to the first condition information.

[0078] In the above embodiment, the terminal reports the condition information to the network device, and the network device configures the first information according to the condition information, so as to ensure that the configuration parameters are consistent with the current conditions of the terminal, and further ensure the service performance of the AI model.

[0079] In combination with some embodiments of the first aspect, in some embodiments, the method further includes:

[0080] sending signaling information to the network device, the signaling information being used for requesting the network device to send a reference signal to the terminal based on a reference signal resource;

[0081] receiving the reference signal sent by the network device;

[0082] measuring the reference signal to determine the first condition information.

[0083] In the above embodiment, the terminal can trigger the measurement of the condition information currently supported by the terminal, and the network device can send the reference signal to the terminal, so that the terminal measures the reference signal to obtain the condition information. Therefore, the flexibility and accuracy of the AI model prediction are improved.

[0084] In combination with some embodiments of the first aspect, in some embodiments, the first information includes configuration parameter information of the measured CSI and configuration parameter information of the predicted CSI, and the determining the first AI model according to the first information includes:

[0085] performing performance detection on a plurality of third AI models according to the first information to generate a plurality of performance detection results corresponding to the plurality of third AI models, each third AI model being a pre-trained AI model;

[0086] determining the first AI model according to the plurality of performance detection results.

[0087] In the above embodiments, the terminal can perform performance detection on the trained multiple AI models according to the first information, and determine the AI model for CSI prediction based on the performance detection result. Thus, the prediction performance of the AI model is ensured, and the accuracy of the performance service is improved.

[0088] In some embodiments of the first aspect, determining the first AI model according to the multiple performance detection results comprises:

[0089] According to the multiple performance detection results, determining, from the multiple third AI models, an AI model that meets a set performance requirement as the first AI model.

[0090] In the above embodiments, the terminal measures the performance requirement of the AI model based on the set performance requirement, and selects the AI model that meets the requirement as the AI model for predicting the CSI. Thus, the performance detection is performed based on the set criterion, the accuracy of the performance detection result is improved, and the accuracy of the AI model is further improved.

[0091] In some embodiments of the first aspect, the method further comprises:

[0092] sending second information to the network device, the second information being used to indicate the first AI model.

[0093] In the above embodiments, after the terminal determines the AI model for predicting the CSI, the terminal reports second information to the network device, so as to ensure the consistency of the AI model in the network device and the terminal, and enable the network device to perform parameter configuration according to the AI model.

[0094] In some embodiments of the first aspect, determining the first AI model according to the multiple performance detection results comprises:

[0095] According to the multiple performance detection results, determining that multiple AI models in the multiple third AI models all meet a set performance requirement.

[0096] sending the multiple performance detection results to the network device, the multiple performance detection results being used by the network device to determine the first AI model from the multiple third AI models;

[0097] receiving third information sent by the network device, the third information being used to indicate the first AI model.

[0098] determining the first AI model according to the third information.

[0099] In the above embodiments, the terminal reports the performance detection result, and the network device determines the AI model according to the performance detection result, avoids random selection of the AI model, ensures that the AI models determined by the terminal side and the network device side are consistent, and ensures the prediction performance of the AI model.

[0100] In some embodiments of the first aspect, determining the first AI model according to the plurality of performance detection results comprises:

[0101] determining that none of the plurality of third AI models meets the set performance requirement according to the plurality of performance detection results;

[0102] sending the plurality of performance detection results to the network device, the plurality of performance detection results being used by the network device to determine the first AI model from the plurality of third AI models;

[0103] receiving third information sent by the network device, the third information being used to indicate the first AI model;

[0104] determining the first AI model according to the third information.

[0105] In the above embodiments, the network device determines the AI model for CSI prediction according to the performance detection result reported by the terminal, and avoids performance loss caused by mismatch between the inference model and the trained model.

[0106] In a second aspect, the embodiments of the present disclosure provide a communication method, performed by a network device, the method comprising:

[0107] sending first information to a terminal, the first information comprising configuration parameter information for measuring CSI and / or configuration parameter information for predicting CSI, the first information being used by the terminal to determine a first AI model according to the first information, the first AI model being used for CSI prediction.

[0108] In some embodiments of the second aspect, the first information comprises at least one of the following:

[0109] a number of reference signal resources for measuring CSI;

[0110] a time domain behavior of each reference signal resource;

[0111] a time domain interval between adjacent measured CSIs;

[0112] a number of predicted CSIs;

[0113] a time domain interval between adjacent predicted CSIs;

[0114] a time domain interval between the first predicted CSI and a reporting time of the predicted CSI;

[0115] reporting time information of the predicted CSI.

[0116] In some embodiments of the second aspect, the time domain behavior comprises at least one of:

[0117] a periodic time domain behavior;

[0118] a semi-persistent periodic time domain behavior;

[0119] an aperiodic time domain behavior.

[0120] In some embodiments of the second aspect, the first information comprises an association ID, the association ID being used to indicate at least one of:

[0121] first condition information of the terminal, the first condition information being used to indicate configuration information or a running state of the terminal;

[0122] a combination manner of the measurement CSI configuration parameter information and / or the predicted CSI configuration information;

[0123] second condition information of the network device, the second condition information being used to indicate a configuration state of the network device.

[0124] In some embodiments of the second aspect, the method further comprises:

[0125] receiving the first condition information sent by the terminal;

[0126] determining the first information according to the first condition information.

[0127] In some embodiments of the second aspect, the method further comprises:

[0128] receiving signaling information sent by the terminal;

[0129] sending, according to the signaling information, a reference signal to the terminal based on a reference signal resource, the reference signal being used for the terminal to measure the reference signal to determine the first condition information.

[0130] In some embodiments of the second aspect, the first information comprises the configuration parameter information of the measurement CSI and the configuration parameter information of the predicted CSI, and the method further comprises:

[0131] receive a plurality of performance detection results sent by the terminal, the plurality of performance detection results being generated by the terminal performing performance testing on a plurality of third AI models according to first information, the third AI models being pre-trained AI models;

[0132] determine the first AI model from the plurality of third AI models according to the plurality of performance detection results;

[0133] send third information to the terminal based on the first AI model, the third information being used to indicate the first AI model.

[0134] In a third aspect, an embodiment of the present disclosure provides a terminal, comprising:

[0135] a transceiver configured to receive first information sent by a network device, the first information comprising configuration parameter information for measuring channel state information (CSI) and / or configuration parameter information for predicting CSI

[0136] a processing module configured to determine a first AI model according to the first information, the first AI model being used for CSI prediction.

[0137] In a fourth aspect, an embodiment of the present disclosure provides a network device, comprising:

[0138] a transceiver configured to send first information to a terminal, the first information comprising configuration parameter information for measuring CSI and / or configuration parameter information for predicting CSI, the first information being used for the terminal to determine a first AI model according to the first information, the first AI model being used for CSI prediction.

[0139] In a fifth aspect, an embodiment of the present disclosure provides a terminal, comprising:

[0140] one or more processors;

[0141] The terminal is configured to perform the communication method of any one of the first aspect of the present disclosure.

[0142] In a sixth aspect, an embodiment of the present disclosure provides a network device, comprising:

[0143] one or more processors;

[0144] The network device is configured to perform the communication method of any one of the second aspect of the present disclosure.

[0145] In a seventh aspect, an embodiment of the present disclosure provides a communication system, comprising:

[0146] a terminal configured to perform the method of any one of the first aspect of the present disclosure;

[0147] A network device configured to perform the method of any of the second aspects of the disclosure.

[0148] In an eighth aspect, an embodiment of the disclosure provides a storage medium, which stores instructions, when the instructions are executed on a communication device, cause the communication device to perform the communication method of any of the first aspects of the disclosure, or cause the communication device to perform the communication method of any of the second aspects of the disclosure.

[0149] In a ninth aspect, an embodiment of the disclosure provides a computer program product, which includes a computer program and / or instructions, when the computer program and / or instructions are executed by a communication device, implement the communication method of any of the first aspects of the disclosure, or when the computer program and / or instructions are executed by a communication device, implement the communication method of any of the second aspects of the disclosure.

[0150] In a tenth aspect, an embodiment of the disclosure provides a computer program, when executed on a computer, causes the computer to perform the method described in the optional implementation of the first aspect and / or the second aspect.

[0151] In an eleventh aspect, an embodiment of the disclosure provides a chip or chip system. The chip or chip system includes processing circuitry configured to perform the method described in the above first aspect and / or the optional implementation of the second aspect.

[0152] It can be understood that the above terminal, access network device, first network element, second network element, core network device, communication system, storage medium, program product, computer program, chip or chip system are all used to perform the method proposed in the embodiments of the disclosure. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0153] Embodiments of the disclosure provide a communication method, a terminal, a network device, a system and a storage medium. In some embodiments, the terms of communication method and information processing method can be replaced with each other, the terms of information processing system and communication system can be replaced with each other.

[0154] The embodiments of the present disclosure are not exhaustive, but only illustrate some embodiments, and are not specific limitations on the protection scope of the present disclosure. In the case of no contradiction, each step in an embodiment can be implemented as an independent embodiment, and the steps can be combined arbitrarily, for example, the scheme after removing part of the steps in an embodiment can also be implemented as an independent embodiment, and the order of the steps in an embodiment can be exchanged arbitrarily, in addition, the optional implementation manners in an embodiment can be combined arbitrarily; in addition, the embodiments can be combined arbitrarily, for example, part or all steps of different embodiments can be combined arbitrarily, an embodiment can be combined with optional implementation manners of other embodiments arbitrarily.

[0155] In each embodiment of the present disclosure, the terms and / or descriptions between the embodiments are consistent if there is no special description and logical conflict, and can be referred to each other, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0156] The terms used in the embodiments of the present disclosure are only for the purpose of describing the specific embodiments, and not as a limitation on the present disclosure.

[0157] In the embodiments of the present disclosure, unless otherwise specified, the elements expressed in singular form, such as "one", "a", "the", "above", "said", "preceding", "this" and the like, can represent "one and only one", and can also represent "one or more", "at least one" and the like. For example, in the case of using articles such as "a", "an", "the" and the like in English, the noun after the article can be understood as singular expression, and can also be understood as plural expression.

[0158] In the embodiments of the present disclosure, "plurality" means two or more.

[0159] In some embodiments, the terms "at least one of", "one or more", "a plurality of", "multiple" and the like can be replaced with each other.

[0160] In some embodiments, "at least one of A, B", "A and / or B", "in one case A, in another case B", "responsive to case A, responsive to case B" and the like, can be interpreted to include both cases, A and B, in some embodiments, A (A is performed regardless of B), in some embodiments, B (B is performed regardless of A), in some embodiments, selected from the group consisting of A and B (the selection between A and B is an option), in some embodiments, A and B (both A and B are performed).

[0161] In some embodiments, "A or B" and the like, can be interpreted to include both cases, A and B, in some embodiments, A (A is performed regardless of B), in some embodiments, B (B is performed regardless of A), in some embodiments, selected from the group consisting of A and B (the selection between A and B is an option).

[0162] In some embodiments, the prefix words "first", "second" and the like in the disclosure do not limit the position, order, priority, number or content of the described objects, and the description of the described objects should be referred to the context of the claims or embodiments, and should not be construed as redundant limitations. For example, the described objects are "fields", and the ordinal words before "fields" in "first field" and "second field" do not limit the position or order between "fields", and "first" and "second" do not limit whether the "fields" modified by them are in the same message or not, nor limit the order of "first field" and "second field". For another example, the described objects are "levels", and the ordinal words before "levels" in "first level" and "second level" do not limit the priority between "levels". For another example, the number of the described objects is not limited by the ordinal words, and can be one or more. For example, "first device", where the number of "devices" can be one or more. In addition, the objects modified by different prefix words can be the same or different, for example, the described objects are "devices", and "first device" and "second device" can be the same device or different devices, and their types can be the same or different; for another example, the described objects are "information", and "first information" and "second information" can be the same information or different information, and their contents can be the same or different.

[0163] In some embodiments, "including A", "containing A", "for indicating A", "carrying A" can be interpreted as directly carrying A, or indirectly indicating A.

[0164] In some embodiments, the terms "in response to", "in response to determining", "in the case of", "when", "when", "if", "if" and the like can be replaced with each other.

[0165] 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", "above", and the like can be replaced with each other, and the terms "less than", "less than or equal to", "not greater than", "fewer than", "fewer than or equal to", "not more than", "lower than", "lower than or equal to", "not higher than", "below", and the like can be replaced with each other.

[0166] In some embodiments, the apparatuses and devices can be interpreted as physical or virtual, and their names are not limited to the names described in the embodiments, and in some cases can also be understood as "equipment", "device", "circuit", "network element", "node", "function", "unit", "section", "system", "network", "chip", "chip system", "entity", "subject", and the like.

[0167] In some embodiments, "network" can be interpreted as an apparatus included in the network, such as an access network device, a core network device, and the like.

[0168] In some embodiments, an “access network device (AN device)” can also be referred to as a “radio access network device (RAN device),” a “base station (BS),” a “radio base station,” a “fixed station,” and in some embodiments can also be understood as a “node,” an “access point,” a “transmission point (TP),” a “reception point (RP),” a “transmission / reception point (TRP),” a “panel,” an “antenna panel,” an “antenna array,” a “cell,” a “macro cell,” a “small cell,” a “femto cell,” a “pico cell,” a “sector,” a “cell group,” a “serving cell,” a “carrier,” a “component carrier,” a “bandwidth part (BWP),” and the like.

[0169] In some embodiments, a "terminal" or "terminal device" can be referred to as a "user equipment" (UE), a "user terminal," a "mobile station" (MS), a "mobile terminal" (MT), a subscriber station, a mobile unit, a subscriber unit, a wireless unit, a remote unit, a mobile device, a wireless device, a wireless communication device, a remote device, a mobile subscriber station, an access terminal, a mobile terminal, a wireless terminal, a remote terminal, a handset, a user agent, a mobile client, a client, and / or the like.

[0170] In some embodiments, data, information and / or the like can be obtained in compliance with laws and regulations of a country in which a location is situated.

[0171] In some embodiments, data, information and / or the like can be obtained after consent of a user is obtained.

[0172] In addition, each element, each row, or each column in a table of embodiments of the present disclosure can be implemented as an independent embodiment, and a combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0173] FIG. 1A is an architecture diagram of a communication system, according to an embodiment of the present disclosure. As shown in FIG. 1A, the communication system 100 includes a terminal 101 and a network device 102.

[0174] In some embodiments, the terminal 101 includes at least one of a mobile phone, a wearable device, an Internet of Things device, a communication-capable automobile, a smart automobile, a tablet (Pad), a wireless transceiver-equipped computer, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in a smart grid, a wireless terminal device in transportation safety, a wireless terminal device in a smart city, a wireless terminal device in a smart home, or the like, but is not limited thereto.

[0175] In some embodiments, the network device 102 is at least one of a node or a device that accesses a terminal to a wireless network, and can include an evolved NodeB (eNB) in a 5G communication system, a next generation eNB (ng-eNB), a next generation NodeB (gNB), a node B (NB), a home node B (HNB), a home evolved node B (HeNB), a wireless backhaul device, a radio network controller (RNC), a base station controller (BSC), a base transceiver station (BTS), a base band unit (BBU), a mobile switching center, a base station in a 6G communication system, an Open RAN, a Cloud RAN, a base station in other communication systems, an access node in a Wi-Fi system, or the like, but is not limited thereto.

[0176] In some embodiments, the technical solutions of the present disclosure can be applied to an Open RAN architecture, in which case, the interfaces between or within the access network devices involved in the embodiments of the present disclosure can become internal interfaces of the Open RAN, and the processes and information interactions between these internal interfaces can be implemented through software or programs.

[0177] In some embodiments, the access network device can be composed of a central unit (CU) and a distributed unit (DU), where the CU can also be referred to as a control unit. The CU-DU structure can split the protocol layers of the access network device, with some of the protocol layers being controlled by the CU and the rest of the protocol layers or all of the protocol layers being distributed in the DUs and controlled by the CU, but is not limited thereto.

[0178] In some embodiments, the network device 102 can be one device including the first network element 1031, the second network element 1032, and the like, or can be multiple devices or device groups, each including all or part of the first network element 1031, the second network element 1032, and the like. The network element can be virtual or physical. The core network includes at least one of an evolved packet core (EPC), a 5G core network (5GCN), and a next generation core (NGC), for example.

[0179] It can be understood that the communication system described in the embodiments of the present disclosure is for more clearly illustrating the technical solutions of the embodiments of the present disclosure, and does not constitute a limitation on the technical solutions proposed in the embodiments of the present disclosure. It can be known by those skilled in the art that, as the system architecture evolves and new business scenarios appear, the technical solutions proposed in the embodiments of the present disclosure are also applicable to similar technical problems.

[0180] The following embodiments of the present disclosure can be applied to the communication system 100 shown in FIG. 1A or part of the subject, but are not limited thereto. The subjects shown in FIG. 1A are examples, and the communication system can include all or part of the subjects in FIG. 1A, or include other subjects other than those in FIG. 1A. The number and form of each subject is arbitrary, each subject can be physical or virtual, the connection relationship between each subject is an example, each subject can not be connected or can be connected, the connection can be in any way, can be direct connection or indirect connection, can be wired connection or wireless connection.

[0181] Embodiments of the present disclosure 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), 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 (registered trademark)), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.20, Ultra-WideBand (UWB), Bluetooth (Bluetooth (registered trademark)), Public Land Mobile Network (PLMN) network, Device-to-Device (D2D) system, Machine to Machine (M2M) system, Internet of Things (IoT) system, Vehicle-to-Everything (V2X), system using other communication methods, next-generation system expanded based thereon, and the like. Further, a plurality of systems can be applied in combination (for example, combination of LTE or LTE-A and 5G, and the like).

[0182] In some embodiments, for terminals moving at medium or high speed, if the traditional Rel-15 / 16 / 17 Type II codebook is still used to feed back the CSI (Channel State Information), the rapid change of channel information in the time domain will lead to a decrease in system performance. To solve this problem, the Rel-18 Type II increased codebook is introduced in the 3GPP Rel-18 standardization. The Rel-18 Type II codebook is based on the downlink channel information estimated by the terminal side at the historical time, and uses the autoregressive or LMMSE (Linear Minimum Mean Square Error) algorithm to predict the downlink channel information at the future time, and then calculates the precoding information corresponding to the future time according to the predicted downlink channel information. In the medium or high speed moving scenario, compared with the Rel-16 / 17 Type II codebook, the Rel-18 Type II codebook can significantly improve the system performance.

[0183] In some embodiments, whether using AI or non-AI prediction algorithms, the CSI at multiple historical times needs to be used. The downlink channel information is estimated based on the CSI-RS (Channel State Information-Reference Signal) sent at multiple historical times, and the CSI at multiple future times is predicted. The time domain range corresponding to the CSI-RS sent at multiple historical times is the observation window. The time domain range corresponding to the CSI at multiple future times is called the prediction window.

[0184] FIG. 1B is a schematic diagram of an observation window and a prediction window according to an embodiment of the present disclosure. As shown in FIG. 1B, based on different prediction requirements, the CSI at multiple future times predicted based on the same CSI at multiple historical times is different. In the figure, the parameters are defined as follows: N represents the observation window of N historical time sending CSI-RS, the value of N can be N = 4 / 5 / 8 / 10; M represents the time domain interval between adjacent CSI-RS, the value of M can be M = D = 2.5 / 4 / 5 slots; K represents the prediction window with K time predicted CSI, the value of K can be K = 1 / 3 / 4 slots; D represents the time domain interval between adjacent time predicted by the prediction window, the value of D can be D = 1 / 2.5 / 4 / 5 / 8 slots. The length of the prediction window is w d = K·D.

[0185] In some embodiments, different AI models can be trained based on different data sets on the terminal side. In the inference stage, how to ensure that the AI model used for inference is consistent with the trained model is a problem to be solved. From the perspective of RAN1 (Radio Access Network Working Group 1), for the UE (User Equipment) side model developed (such as training, updating) on the UE side, the following procedures are an example (denoted as AI-Example 1) of MI-Option1 for further study (including feasibility / necessity):

[0186] A: For data collection, the NW (Network) signals the configuration related to data collection and its related ID; and the related ID of each sub-use case related to the additional conditions on the NW side;

[0187] B: The UE collects data corresponding to the related ID;

[0188] C: The UE develops (such as training, updating) AI / ML models according to the collected data corresponding to the related ID;

[0189] D: The UE reports AI / ML model information corresponding to the associated ID to the NW. The AI / ML model information can include at least one of the following:

[0190] Determine / assign a model ID for each AI / ML model;

[0191] The relationship between the model ID and the related ID;

[0192] How to determine / assign the model ID, for example: (1) the NW assigns the model ID; (2) the UE assigns / reports the model ID; (3) assume the related ID as the model ID; for example, in the above step D, it is not necessary to assign a model ID for each AI model; (4) the model ID is determined by the pre-defined rules in the specification;

[0193] How to report;

[0194] Note: Step D is to facilitate AI / ML model inference.

[0195] In some embodiments, according to the above steps A / B / C and the additional interaction between the UE and the NW related to the ID, it can be a solution to ensure the consistency of the terminal inference model and the trained model.

[0196] In some embodiments, for AI model based CSI prediction, which parameter information is configured by the network for training data collection is not given in the above embodiments. In addition, the model trained only by collecting training data by configuring parameter information cannot accurately describe the characteristics of the model, and the channel variation characteristics of the terminal side also need to be known, which belongs to the condition information of the terminal. Therefore, the above method cannot be directly applied to AI model based CSI prediction. Therefore, the present embodiment proposes a communication method for solving the consistency problem of the inference model and the training model used by the terminal, that is, ensuring that the condition information of the inference model and the training model used is consistent, and avoiding performance loss caused by mismatch between the inference model and the training model.

[0197] FIG. 2A is an interaction diagram of a communication method according to an embodiment of the present disclosure. As shown in FIG. 2A, the present embodiment relates to a communication method, and the method comprises:

[0198] In step S2101, the network device sends first information to the terminal.

[0199] In some embodiments, the terminal 101 receives the first information.

[0200] In some embodiments, the first information includes configuration parameter information for measuring CSI and / or configuration parameter information for predicting CSI.

[0201] For example, in the present embodiment, the CSI prediction based on the AI model. Based on the downlink channel information estimated by the CSI-RS sent at multiple historical time points, the AI model takes the CSI-RS sent at multiple historical time points as input information, and predicts the CSI at multiple future time points through the AI model. Wherein, the configuration parameter information for measuring CSI is used to indicate the configuration parameters in the observation window in the above embodiment, including configuration parameter type, configuration parameter number, etc., for example, the number of CSIs in the measurement window, the time interval between adjacent CSI-RSs in the measurement window, etc. The configuration parameter information for predicting CSI is used to indicate the configuration parameters in the prediction window in the above embodiment, for example: the number of CSIs in the prediction window, the time interval between adjacent predicted CSIs in the prediction window, etc.

[0202] In the process of CSI prediction based on the AI model, the AI model includes a model training stage and a model application stage. To ensure the accuracy of the CSI prediction result, in the model application stage of the AI model, the model input of the AI model needs to be matched with the model input of the AI model in the model training stage. For example, the data type of the input data in the model application stage is the same as the data type of the input data in the model training stage. For example, the AI model is an AI model for CSI prediction based on historical CSI. In the model training stage of the AI model, the parameter type of the input data is: N=4, M=2.5 slots of parameter information. Therefore, in the model application stage of the AI model, the AI model needs to input the configuration parameter of the parameter type: N=4, M=2.5 slots, so as to ensure the performance of the AI model. Therefore, to ensure that the training model is consistent with the inference model, the terminal trains the AI model based on the configuration parameter sent by the network device in the training process. To obtain the AI model matched with the configuration parameter, the inference is performed based on the trained AI model, so as to realize the consistency of the training model and the inference model.

[0203] For example, in the process of CSI prediction based on the AI model, different AI models are trained based on a plurality of different combinations of configuration parameters. For example, based on the configuration parameter N=5, M=4 slots of CSI-RS, the trained AI model performs AI prediction on the configuration parameter of the parameter type N=5, M=4 slots of CSI-RS in the model application stage; based on the configuration parameter N=5, M=5 slots of CSI-RS, the trained AI model performs AI prediction on the configuration parameter of the parameter type N=5, M=5 slots of CSI-RS in the model application stage; so as to ensure the model prediction performance of the AI model. The terminal may be configured with a plurality of trained AI models. According to the configuration parameter sent by the network device, the configuration parameter type is identified, and the corresponding AI model is selected to perform CSI prediction. Therefore, the training model is consistent with the inference model.

[0204] In some embodiments, the first information includes at least one of:

[0205] The number of reference signal resources for measuring the CSI;

[0206] The time domain behavior of each reference signal resource;

[0207] The time domain interval between adjacent measurement CSIs;

[0208] The number of predicted CSIs;

[0209] The time domain interval between adjacent predicted CSIs;

[0210] a time domain interval between the first predicted CSI and a reporting time of the predicted CSI;

[0211] a time information of reporting the predicted CSI.

[0212] In an example, the network device configures, for the terminal, first information, which can be used to indicate configuration parameter information of the measured CSI, and the configuration parameter information of the measured CSI can be a parameter type input by the AI model, and the parameter type includes at least one of the following: a quantity of reference signal resources for measuring the CSI, a time domain behavior of each reference signal resource, and a time domain interval between adjacent measured CSIs. The first information can also be used to indicate configuration parameter information of the predicted CSI, and the configuration parameter information of the predicted CSI can be a parameter type output by the AI model, and the parameter type includes at least one of the following: a quantity of predicted CSIs, a time domain interval between adjacent predicted CSIs, a time domain interval between the first predicted CSI and a reporting time of the predicted CSI, and a time information of reporting the predicted CSI.

[0213] Optionally, in some embodiments, the time domain behavior includes at least one of the following: a periodic time domain behavior, a semi-persistent periodic time domain behavior, and an aperiodic time domain behavior.

[0214] In an example, the first information in the embodiment includes the configuration parameter information of the measured CSI, which can be used to indicate a time domain behavior of a corresponding reference signal resource, and the time domain behavior can be any one of a periodic time domain behavior, a semi-persistent periodic time domain behavior, and an aperiodic time domain behavior. That is, the measured CSI configured by the network device can be a periodically transmitted CSI, a semi-persistently periodically transmitted CSI, and an aperiodically transmitted CSI.

[0215] In step S2102, the terminal determines a first AI model according to the first information.

[0216] In some embodiments, the first AI model is used for CSI prediction.

[0217] In an example, to ensure consistency between the trained model and the inference model, the terminal can obtain a data training set matched with the first information according to the first information, perform model training on a preset AI model according to the data training set, so as to obtain a first AI model matched with the first information, and then use the first information as a model input of the first AI model to perform CSI prediction.

[0218] The terminal can also be configured with multiple AI models that have been trained, select an AI model corresponding to the first information as the first AI model for CSI prediction. For example, obtain multiple training data sets corresponding to the multiple AI models, determine a target training data set most similar to the first information from the multiple training data sets, and select an AI model corresponding to the target training data set as the first AI model.

[0219] Optionally, in some embodiments, the step S2102 comprises:

[0220] The terminal obtains a first training data set according to the first information.

[0221] The terminal trains a second AI model according to the first training data set to generate the first AI model.

[0222] In some embodiments, the second AI model is a preset AI model.

[0223] For example, the terminal receives the first information sent by the network device, and obtains a first training data set according to the first information. The data type of the configuration parameter in the first training data set is the same as the data type of the configuration parameter indicated in the first information. For example, the first information includes the number of reference signal resources N = 10 for measuring CSI, and the time interval M = 4 slots between adjacent CSI-RSs. The terminal obtains multiple measured CSIs with reference signal resource number N = 10 and time interval M = 4 slots between adjacent CSI-RSs, and the predicted CSI corresponding to the multiple measured CSIs as the first training data set. The second AI model is trained according to the first training data set to generate the first AI model. The second AI model is a preset AI model in the terminal, and the first AI model is generated by training the preset AI model based on the first training data set. For example, the second AI model can also be an AI model that has been trained. The model parameters of the AI model are updated according to the first training data set to obtain the first AI model.

[0224] Optionally, in some embodiments, the step S2102 comprises:

[0225] The terminal determines that the first information is a subset of the configuration parameters of the second training data set.

[0226] The terminal selects a third AI model as the first AI model.

[0227] In some embodiments, the second training data set is a training data set corresponding to the third AI model, and the third AI model is trained according to the second training data set.

[0228] For example, the terminal is configured with multiple AI models that have completed training, and the multiple AI models correspond to multiple training data sets. The terminal performs model training based on the multiple training data sets to obtain the multiple AI models. The types of the training data sets corresponding to the respective AI models are different. After receiving the first information, the terminal compares the first information with the configuration parameters of the training data sets corresponding to the respective AI models, and determines that the first information is a subset of the configuration parameters of the second training data set, and determines that a third AI model trained based on the second training data set is the first AI model. For example, the configuration parameters of the second training data set are a configuration parameter range indicating that N=4-8 and M=2.5-5 slots are used to measure CSI, and the parameter information corresponding to the first information indicated by the network device indicates that N=4 and M=5 slots are used to measure CSI. Therefore, the first information is a subset of the configuration parameters of the second training data set, and the third AI model trained based on the second training data set is used as the first AI model.

[0229] For example, the AI model on the UE side is trained by the UE or a server on the UE side according to a collected training data set, and the UE collects the data set according to the configuration parameter information of the gNB and the corresponding reference signal resource. Assuming that the UE trains an AI model for CSI prediction, N=5 historical time instants of measured CSI are used as inputs of the AI model, and K=4 future time instants of predicted CSI are used as outputs of the AI model. In order to enable the UE to collect the corresponding data set, the gNB can configure a CSI-RS continuous signal including 5 aperiodic CSI-RS resources, and trigger the transmission of the aperiodic CSI-RS through multiple DCI (Downlink Control Information) signaling to measure the CSI at the historical time instants. The measurement of the CSI at the historical time instants is recorded as an observation window, and the interval between the adjacent two time instants of the transmitted CSI-RS configured by the gNB in the observation window is 2 slots. In order to collect the CSI at the future time instants as the training data set of the label, the gNB further configures 4 aperiodic CSI-RSs for measuring the CSI at the future time instants. The CSI at the future time instants is recorded as a prediction window, and the interval between the adjacent two time instants of the transmitted CSI-RS configured by the gNB in the prediction window is 4 slots. The time instant of the first transmitted CSI in the prediction window can be determined according to the reporting time instant n, for example, the interval between the time instant of the first transmitted CSI and the reporting time instant n is defined as d, and the value of d can be positive or negative. When d is negative, it means before the reporting time instant, and vice versa. Assuming that d=3, the first CSI time instant in the prediction window is n+3.

[0230] FIG. 2B is a schematic diagram of parameter information and CSI-RS resource transmission according to an embodiment of the present disclosure. As shown in FIG. 2B, the two CSI-RS bursts within the observation window and the prediction window contain CSI-RS resources belonging to one CSI-RS resource set or two CSI-RS resource sets. It should be noted that the above is only an example of aperiodic CSI-RS resources. Alternatively, the first CSI-RS burst corresponds to periodic CSI-RS resources, and the second CSI-RS burst is aperiodic CSI-RS resources.

[0231] In some embodiments, the UE measures the CSI within the observation window and the prediction window as a training data set according to the above parameter information and CSI-RS resource configuration information, and then completes the training of the prediction CSI model based on the training data set. The above only gives an example under a network parameter configuration. If the gNB configures different parameter information or CSI-RS resources, the UE can collect more different training data sets. The UE can train one or more models based on these different data sets.

[0232] In some embodiments, it is assumed that the UE has trained multiple AI models for CSI prediction based on different parameters and resource configurations. In the inference stage, the UE determines which AI model to use according to the parameters and resources configured by the network, ensuring that the parameters corresponding to the data set used by the inference model and the training model are the same, so that the inference model and the training model are consistent, avoiding performance loss due to inconsistent models. Alternatively,

[0233] In some embodiments, the UE trains an AI model with generalization based on different parameter configurations. For example, if the gNB configures parameters N∈{2,3,4}, K∈{1,2,3,4}, d∈{2,3}, M∈{1,2}, D∈{1,2}, the UE generates a data set to train an AI model based on these different parameter configurations. When the gNB configures a subset of parameters N∈{2,3,4}, K∈{1,2,3,4}, d∈{2,3}, M∈{1,2}, D∈{1,2}, the UE will use this model for inference, also ensuring the consistency of the inference model and the training model.

[0234] Optionally, in some embodiments, the configuration parameter or resource information is associated with an ID (associated ID). Optionally, the associated ID is further associated with condition information on the NW side. The condition information on the NW side can correspond to information such as cell ID, antenna downtilt angle, or analog beam used. Since these information belong to private data on the NW side, they are invisible to the UE. These condition information are informed to the UE in the form of associated ID. The UE will determine to use the AI model according to the associated ID.

[0235] In some embodiments, the first information includes an associated ID, and the associated ID is used to indicate at least one of the following:

[0236] First condition information of the terminal, the first condition information being used to indicate configuration information or running state of the terminal;

[0237] Combination mode of the measurement CSI configuration parameter information and / or the prediction CSI configuration information;

[0238] Second condition information of the network device, the second condition information being used to indicate configuration state of the network device.

[0239] In an example, the combination mode of the measurement CSI configuration parameter information and / or the prediction CSI configuration information can be indicated to the terminal in the form of the associated ID in the present embodiment, a mapping relationship between the associated ID and the combination mode is configured in the terminal and the network device, the network device indicates the associated ID to the terminal, and the terminal determines the combination mode corresponding to the associated ID based on the mapping relationship. For example, the following mapping relationship can be configured in the terminal and the network device:

[0240] When the network device configures the associated ID as 3 to the terminal, the terminal can determine, according to the above mapping relationship, that the configuration parameter information indicated by the network device is combined as: N=3, M=4 slots, K=3 slots, and D=4 slots.

[0241] It should be noted that the association ID can also be associated with the first condition information of the terminal and / or the second condition information of the network device, and the first condition information and / or the second condition information are indicated by the association ID. Among them, the first condition information is used to indicate the current configuration information or running state of the terminal, for example, the first condition information indicates the channel state information of the terminal, and the configuration information of the terminal when the terminal currently performs a communication process, including terminal remaining power information, terminal running state information, terminal storage space information, etc. For example, when performing CSI prediction, the terminal reports the first condition information to the network device, so that the network device configures the configuration parameters conforming to the current condition to the terminal according to the first condition information, thereby improving the AI model performance in the CSI prediction process. The second condition information is used to indicate the configuration state of the network device, and in the embodiment, the network device can indicate the second condition information to the terminal, and the terminal determines the corresponding type of AI model for CSI prediction according to the second condition information.

[0242] Optionally, in some embodiments, the step S2102 comprises:

[0243] The AI model matched with the association ID is determined from a plurality of third AI models as the first AI model, and the third AI model is a pre-trained AI model.

[0244] For example, in the embodiment, there is a mapping relationship between the association ID and the configuration parameter combination, the network device sends the association ID to the terminal, the terminal determines the configuration parameter combination corresponding to the association ID, and then determines the AI model similar to the configuration parameter combination from a plurality of third AI models that have been trained as the first AI model. The training data set. Thus, by indicating the association ID by the network device, the configuration parameters are indicated to the terminal, the signaling overhead is reduced, and the inference AI model and the training AI model are consistent.

[0245] Optionally, in some embodiments, the method further comprises:

[0246] sending the first condition information to the network device,

[0247] In some embodiments, the first condition information is used by the network device to determine the first information according to the first condition information.

[0248] For example, in the embodiment, the terminal reports the first condition information to the network device, indicating the communication condition currently supported by the terminal. The network device can determine the configuration parameter information suitable for the terminal at the time of CSI prediction according to the first condition information. After determining the configuration parameter information according to the first condition information, generate the first information, and send the first information to the terminal. Thus, the matching between the configuration parameters and the current environment of the terminal is realized, and the prediction performance of the AI model is ensured.

[0249] Optionally, in some embodiments, the method further comprises:

[0250] sending signaling information to the network device, the signaling information being used to request the network device to send a reference signal to the terminal based on the reference signal resource;

[0251] receiving the reference signal sent by the network device;

[0252] measuring the reference signal to determine the first condition information.

[0253] For example, the first condition information needs to be measured by the terminal based on the current environment and the current communication state of the terminal. The terminal can actively obtain the first condition information corresponding to the terminal. When the terminal has the demand for obtaining, the signaling information is sent to the network device to request the network device to configure the reference signal resource, and the terminal measures the reference signal resource to obtain the first condition information of the current communication environment of the terminal.

[0254] For example, when generating the training data set on the UE side, in addition to the corresponding NW side condition information, the condition information on the UE side also needs to be considered. Because the CSI corresponding to the UE in different mobile states shows great difference in the time domain. Therefore, to ensure the consistency of the inference model and the training model, the condition information of the UE needs to be considered.

[0255] In some embodiments, in order to obtain the condition information on the UE side, the gNB can configure one or more TRS (tracking reference signal) resource sets for measuring the time domain channel characteristics of the UE, and then configure the parameters and CSI-RS resources according to the time domain channel characteristics. The UE performs model training based on the collected data set. In the inference stage, the UE selects the AI model according to the parameter and resource information configured by the network and the condition information on the UE side.

[0256] In some embodiments, the condition information on the UE side can be associated with an associated ID, which can also be associated with the condition information of the gNB and the parameter and CSI-RS resource information configured by the gNB. That is, the training data set generated by the UE is associated with the associated ID, and in the inference stage, the UE can determine the AI model to be used according to the associated ID, wherein the associated ID to be used can be configured by the network to the UE.

[0257] In some embodiments, the names of information and the like are not limited to the names described in the embodiments, and the terms of "information", "message", "signal", "signaling", "report", "configuration", "indication", "instruction", "command", "channel", "parameter", "field", "symbol", "codebook", "codeword", "codepoint", "bit", "data", "program", "chip", and the like can be replaced with each other.

[0258] In some embodiments, the terms of "codebook", "codeword", "precoding matrix", and the like can be replaced with each other. For example, a codebook can be a collection of one or more codewords / precoding matrices.

[0259] In some embodiments, the terms of "uplink", "uplink", "physical uplink", and the like can be replaced with each other, the terms of "downlink", "downlink", "physical downlink", and the like can be replaced with each other, and the terms of "side", "sidelink", "sidelink communication", "sidelink communication", "direct connection", "direct connection link", "direct connection", "direct connection link communication", and the like can be replaced with each other.

[0260] In some embodiments, the terms of "downlink control information (DCI)", "downlink (DL) assignment", "DL DCI", "uplink (UL) grant", "UL DCI", and the like can be replaced with each other.

[0261] In some embodiments, the terms of "physical downlink shared channel (PDSCH)", "DL data", and the like can be replaced with each other, and the terms of "physical uplink shared channel (PUSCH)", "UL data", and the like can be replaced with each other.

[0262] In some embodiments, the terms “radio,” “wireless,” “radio access network” (RAN), “access network” (AN), “RAN-based,” and the like can be replaced with each other.

[0263] In some embodiments, the terms “search space,” “search space set,” “search space configuration,” “search space set configuration,” “control resource set” (CORESET), “CORESET configuration,” and the like can be replaced with each other.

[0264] In some embodiments, the terms “synchronization signal” (SS), “synchronization signal block” (SSB), “reference signal” (RS), “pilot,” “pilot signal,” and the like can be replaced with each other.

[0265] In some embodiments, the terms “time instant,” “time point,” “time,” “time location,” and the like can be replaced with each other, and the terms “time duration,” “time period,” “time window,” “window,” “time,” and the like can be replaced with each other.

[0266] In some embodiments, the terms “component carrier” (CC), “cell,” “frequency carrier,” “carrier frequency,” and the like can be replaced with each other.

[0267] In some embodiments, the terms “resource block” (RB), “physical resource block” (PRB), “sub-carrier group” (SCG), “resource element group” (REG), “PRB pair,” “RB pair,” “resource element” (RE), “sub-carrier,” and the like can be replaced with each other.

[0268] In some embodiments, the terms “wireless access scheme”, “waveform”, etc. can be replaced by each other.

[0269] In some embodiments, the terms “precoding”, “precoder”, “weight”, “precoding weight”, “quasi-co-location (QCL)”, “transmission configuration indication (TCI) state”, “spatial relation”, “spatial domain filter”, “transmission power”, “phase rotation”, “antenna port”, “antenna port group”, “layer”, “the number of layers”, “rank”, “resource”, “resource set”, “resource group”, “beam”, “beam width”, “beam angular degree”, “antenna”, “antenna element”, “panel”, etc. can be replaced by each other.

[0270] In some embodiments, the terms “frame”, “radio frame”, “subframe”, “slot”, “sub-slot”, “mini-slot”, “symbol”, “symbol”, “transmission time interval (TTI)”, etc. can be replaced by each other.

[0271] In some embodiments, the terms “acquire”, “obtain”, “get”, “receive”, “transmit”, “bidirectional transmission”, “transmit and / or receive” can be replaced by each other, which can be interpreted as receiving from other subjects, obtaining from protocols, obtaining from higher layers, obtaining by oneself, implementing autonomously, etc.

[0272] In some embodiments, the terms “sending”, “transmitting”, “reporting”, “issuing”, “transferring”, “bidirectional transferring”, “sending and / or receiving” and the like can be replaced by each other.

[0273] In some embodiments, the terms “certain”, “preseted”, “preset”, “set”, “indicated”, “certain”, “arbitrary”, “first” and the like can be replaced by each other, and “certain A”, “preset A”, “preset A”, “set A”, “indicated A”, “certain A”, “arbitrary A”, “first A” can be interpreted as A specified in advance in a protocol or the like, or A obtained by setting, configuration, or indication, or A specific, certain, arbitrary, or first A, but not limited thereto.

[0274] In some embodiments, the determination or judgment can be made by a value represented by 1 bit (0 or 1), or by a true or false value (Boolean value) represented by true or false, or by comparison of numerical values (for example, comparison with a predetermined value), but not limited thereto.

[0275] In some embodiments, “not expecting to receive” can be interpreted as not receiving on the time domain resource and / or frequency domain resource, or as not performing subsequent processing on the data or the like after receiving the data or the like; “not expecting to send” can be interpreted as not sending, or as sending but not expecting the receiving party to respond to the content of the sending.

[0276] The communication method related to the embodiments of the present disclosure can include at least one of steps S2101-S2102. For example, step S2101 can be implemented as an independent embodiment, step S2102 can be implemented as an independent embodiment, step 1+3 can be implemented as an independent embodiment, and step S2101+S2102 can be implemented as an independent embodiment (arrangement and combination of important steps related to the invention points are exemplified), but not limited thereto.

[0277] In some embodiments, the order of steps S2101 and S2102 can be exchanged or performed simultaneously.

[0278] In some embodiments, step S2101 is optional, and one or more of the steps can be omitted or replaced in different embodiments.

[0279] In some embodiments, other optional implementations described before or after the corresponding description of FIG. 2A can be referred to.

[0280] FIG. 2C is an interaction flow diagram of a communication method, according to an embodiment of the present disclosure. As shown in FIG. 2C, the embodiment of the present disclosure relates to a communication method, and the method comprises:

[0281] In step S2201, the network device sends first information to the terminal.

[0282] The optional implementation of step S2201 can refer to the optional implementation of step S2101 in FIG. 2A and other associated parts in the embodiments related to FIG. 2A, which will not be described here.

[0283] In step S2202, the terminal performs performance detection on the plurality of third AI models according to the first information, and generates a plurality of performance detection results corresponding to the plurality of third AI models one by one.

[0284] In some embodiments, the first information comprises configuration parameter information for measuring CSI and configuration parameter information for predicting CSI.

[0285] In some embodiments, the third AI models are all pre-trained AI models.

[0286] For example, a plurality of third AI models that have been trained are configured in the terminal, the third AI models are pre-trained AI models, performance detection is performed on the plurality of third AI models according to the first information sent by the network device, and a plurality of performance detection results corresponding to the plurality of third AI models one by one are generated. In this embodiment, the first information comprises configuration parameter information for measuring CSI and configuration parameter information for predicting CSI, wherein the configuration parameter information for measuring CSI is the model input of the AI model, and the configuration parameter information for predicting CSI is the standard model output corresponding to the model input. The terminal is configured with a performance detection index, and performance detection is performed on the plurality of third AI models in the terminal according to the first information to obtain the performance detection result of each third AI model.

[0287] In step S2203, the terminal determines the first AI model according to the plurality of performance detection results.

[0288] For example, the terminal determines the AI model that best meets the first information from the plurality of third AI models as the first AI model according to the plurality of performance detection results. For example, the terminal determines that only one AI model from the plurality of third AI models meets the corresponding performance index according to the plurality of performance detection results, and then determines that the AI model is the first AI model.

[0289] Optionally, in some embodiments, step S2203 comprises:

[0290] The terminal determines the AI model that meets the set performance requirement from the plurality of third AI models as the first AI model according to the plurality of performance detection results.

[0291] For example, the terminal screens the AI model according to the plurality of performance detection results with reference to the set performance requirement. If it is determined according to the plurality of performance detection results that only one AI model in the plurality of third AI models meets the set performance requirement, the AI model meeting the set performance requirement is taken as the first AI model. If it is determined according to the plurality of performance detection results that a plurality of AI models in the plurality of third AI models all meet the set performance requirement, the terminal can select any one of the AI models meeting the set performance requirement as the first AI model, or the terminal selects one AI model from the plurality of AI models meeting the set performance requirement as the first AI model based on a preset rule.

[0292] Optionally, in some embodiments, the method further includes:

[0293] sending the second information to the network device.

[0294] In some embodiments, the second information is used to indicate the first AI model.

[0295] For example, after the terminal determines the first AI model based on the plurality of performance detection results, the terminal needs to report second information to the network device to ensure consistency of the AI model on the terminal side and the network side. The second information is used to indicate the first AI model selected by the terminal.

[0296] Optionally, in some embodiments, the step S2203 includes:

[0297] The terminal determines, according to the plurality of performance detection results, that a plurality of AI models in the plurality of third AI models all meet the set performance requirement.

[0298] The terminal sends the plurality of performance detection results to the network device, and the plurality of performance detection results are used by the network device to determine the first AI model from the plurality of third AI models.

[0299] The terminal receives third information sent by the network device, and the third information is used to indicate the first AI model.

[0300] The terminal determines the first AI model according to the third information.

[0301] In an example, the terminal determines, according to the plurality of performance detection results, that there is no AI model in the plurality of third AI models that meets the set performance requirement, and the terminal cannot determine which AI model that meets the set performance requirement is used as the first AI model. The terminal reports the plurality of performance detection results to the network device, and each performance detection result is associated with model ID information of a corresponding AI model. The network device selects an AI model from the plurality of third AI models as the first AI model based on the plurality of performance detection results, and generates third information based on model ID information of the first AI model and sends the third information to the terminal, so as to instruct the terminal to perform CSI prediction based on the first AI model.

[0302] Optionally, in some embodiments, the step S2203 comprises:

[0303] The terminal determines, according to the plurality of performance detection results, that none of the plurality of third AI models meets the set performance requirement.

[0304] The terminal sends the plurality of performance detection results to the network device, and the plurality of performance detection results are used by the network device to determine the first AI model from the plurality of third AI models.

[0305] The terminal receives third information sent by the network device, and the third information is used to indicate the first AI model.

[0306] The terminal determines the first AI model according to the third information.

[0307] In an example, the terminal determines, according to the plurality of performance detection results, that there is no AI model in the plurality of third AI models that meets the set performance requirement, and the terminal cannot determine which AI model that meets the set performance requirement is used as the first AI model. The terminal reports the plurality of performance detection results to the network device, and each performance detection result is associated with model ID information of a corresponding AI model. The network device selects an AI model from the plurality of third AI models as the first AI model based on the plurality of performance detection results, and generates third information based on model ID information of the first AI model and sends the third information to the terminal, so as to instruct the terminal to perform CSI prediction based on the first AI model.

[0308] For example, assuming that the UE side trains multiple AI models based on different parameter configurations, the multiple AI models can also include an AI model that has generalization ability under a parameter configuration. If the gNB configures a parameter and a CSI-RS resource configuration, the gNB can also send an index requirement of corresponding performance under the configuration to the UE through one of an RRC (Radio Resource Control) protocol, a MAC-CE (MAC-Control Element), or a DCI signaling. The UE measures the historical time CSI based on the configured parameter and resource. If multiple AI models are deployed on an OTT (Over-The-Top server) server at the UE side, the UE can send the measured historical time CSI and / or the performance index requirement to the OTT server, and the OTT server can complete performance measurement of each model based on the CSI prediction. The OTT server selects an AI model that meets the performance requirement and has the lowest model complexity, for example, and then indicates the selected AI model to the UE. The UE performs inference by using the indicated model. If the performance of each AI model tested by the OTT server does not meet the performance index requirement, the UE can send the measurement result to the gNB through uplink signaling, and the gNB can configure corresponding parameter information and resources to enable the UE to generate a corresponding training data set to train a new AI model. Alternatively, the model can be updated or adjusted based on the newly generated data set on the basis of the original AI model. Thus, an AI model that meets the performance index requirement is obtained for inference.

[0309] In some embodiments, the terminal can send the test result to the gNB. The gNB determines which model meets the performance requirement based on the performance of each model, and indicates the AI model that meets the performance requirement to the UE. If none of the models meets the requirement, the gNB can perform the processing as described above, which is not repeated here.

[0310] FIG. 3 is a flow diagram of a communication method according to an embodiment of the present disclosure. As shown in FIG. 3, the embodiment of the present disclosure relates to a communication method performed by a terminal, and the method includes the following steps.

[0311] In step S3101, first information sent by a network device is received.

[0312] In some embodiments, the first information includes configuration parameter information for measuring channel state information (CSI) and / or configuration parameter information for predicting the CSI.

[0313] The optional implementation of step S3101 can refer to the optional implementation of step S2101 in FIG. 2A and other associated parts in the embodiments related to FIG. 2A, which are not repeated here.

[0314] In step S3102, a first AI model is determined according to the first information, and the first AI model is used for CSI prediction.

[0315] In some embodiments, the step S3102 includes:

[0316] obtaining a first training data set according to the first information;

[0317] performing model training on a second AI model according to the first training data set to generate the first AI model, and the second AI model is a preset AI model.

[0318] In some embodiments, the step S3102 includes:

[0319] the first information is determined as a subset of configuration parameters of a second training data set, the second training data set is a training data set corresponding to a third AI model, and the third AI model is trained according to the second training data set;

[0320] the third AI model is used as the first AI model.

[0321] In some embodiments, the first information includes at least one of the following:

[0322] a number of reference signal resources for measuring CSI;

[0323] a time domain behavior of each reference signal resource;

[0324] a time domain interval between adjacent CSI measurements;

[0325] a number of predicted CSIs;

[0326] a time domain interval between adjacent predicted CSIs;

[0327] a time domain interval between a first predicted CSI and a predicted CSI reporting time point;

[0328] time point information of reporting the predicted CSI.

[0329] In some embodiments, the time domain behavior includes at least one of the following:

[0330] a periodic time domain behavior;

[0331] a semi-persistent periodic time domain behavior;

[0332] an aperiodic time domain behavior.

[0333] In some embodiments, the first information includes an association ID, and the association ID is used to indicate at least one of the following:

[0334] The first condition information of the terminal, the first condition information being used to indicate configuration information or a running state of the terminal.

[0335] A combination manner of measuring the CSI configuration parameter information and / or predicting the CSI configuration information.

[0336] The second condition information of the network device, the second condition information being used to indicate a configuration state of the network device.

[0337] In some embodiments, the step S3102 comprises:

[0338] Determining, as the first AI model, an AI model matching the association ID from a plurality of third AI models, the third AI models being pre-trained AI models.

[0339] In some embodiments, the method further comprises:

[0340] Sending the first condition information to the network device, the first condition information being used for the network device to determine the first information according to the first condition information.

[0341] In some embodiments, the method further comprises:

[0342] Sending the signaling information to the network device, the signaling information being used to request the network device to send the reference signal to the terminal based on the reference signal resource.

[0343] Receiving the reference signal sent by the network device.

[0344] Measuring the reference signal to determine the first condition information.

[0345] In some embodiments, the first information comprises configuration parameter information of measured CSI and configuration parameter information of predicted CSI, and the step S3102 comprises:

[0346] According to the first information, performing performance detection on the plurality of third AI models to generate a plurality of performance detection results corresponding to the plurality of third AI models, each third AI model being a pre-trained AI model.

[0347] Determining the first AI model according to the plurality of performance detection results.

[0348] In some embodiments, the step of “determining the first AI model according to the plurality of performance detection results” comprises:

[0349] According to the plurality of performance detection results, determining, as the first AI model, an AI model meeting a set performance requirement from the plurality of third AI models.

[0350] In some embodiments, the method further comprises:

[0351] The second information is used to indicate the first AI model.

[0352] In some embodiments, the step of determining the first AI model according to the plurality of performance detection results comprises:

[0353] According to the plurality of performance detection results, it is determined that the plurality of AI models in the plurality of third AI models all meet the set performance requirement.

[0354] The plurality of performance detection results are sent to the network device, and the plurality of performance detection results are used by the network device to determine the first AI model from the plurality of third AI models.

[0355] The third information sent by the network device is received, and the third information is used to indicate the first AI model.

[0356] The first AI model is determined according to the third information.

[0357] In some embodiments, the step of determining the first AI model according to the plurality of performance detection results comprises:

[0358] According to the plurality of performance detection results, it is determined that the plurality of third AI models all do not meet the set performance requirement.

[0359] The plurality of performance detection results are sent to the network device, and the plurality of performance detection results are used by the network device to determine the first AI model from the plurality of third AI models.

[0360] The third information sent by the network device is received, and the third information is used to indicate the first AI model.

[0361] The first AI model is determined according to the third information.

[0362] The communication method related to the embodiments of the present disclosure can include at least one of steps S3101-S3102. For example, step S3101 can be implemented as an independent embodiment, step S3102 can be implemented as an independent embodiment, step S3101+step S3102 can be implemented as an independent embodiment, but not limited thereto.

[0363] In some embodiments, the steps S3101, S3102 can exchange order or be executed simultaneously.

[0364] In some embodiments, steps S3101, S3102 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0365] In some embodiments, other optional implementations described before or after the corresponding description of FIG. 3 can be referred to.

[0366] FIG. 4 is a flow diagram of a communication method according to an embodiment of the present disclosure. As shown in FIG. 4, the embodiment of the present disclosure relates to a communication method, which is performed by a network device, and the method comprises the following steps:

[0367] In step S4101, the first information is sent to the terminal.

[0368] In some embodiments, the first information comprises configuration parameter information of the measured CSI and / or configuration parameter information of the predicted CSI, and the first information is used by the terminal to determine a first AI model according to the first information, and the first AI model is used to perform CSI prediction.

[0369] The optional implementation of step S4101 can refer to the optional implementation of step S2101 in FIG. 2A and other associated parts in the embodiments involved by FIG. 2A, which will not be repeated here.

[0370] In some embodiments, the first information comprises at least one of the following:

[0371] The number of reference signal resources for measuring the CSI;

[0372] The time domain behavior of each reference signal resource;

[0373] The time domain interval between adjacent measured CSIs;

[0374] The number of predicted CSIs;

[0375] The time domain interval between adjacent predicted CSIs;

[0376] The time domain interval between the first predicted CSI and the reporting time of the predicted CSI;

[0377] The time information of reporting the predicted CSI.

[0378] In some embodiments, the time domain behavior comprises at least one of the following:

[0379] Periodic time domain behavior;

[0380] Semi-persistent periodic time domain behavior;

[0381] Aperiodic time domain behavior.

[0382] In some embodiments, the first information comprises an association ID, and the association ID is used to indicate at least one of the following:

[0383] First condition information of the terminal, the first condition information being used to indicate configuration information or a running state of the terminal;

[0384] The combination mode of the measured CSI configuration parameter information and / or the predicted CSI configuration information;

[0385] Second condition information of the network device, the second condition information being used to indicate a configuration state of the network device.

[0386] In some embodiments, the method further includes:

[0387] receiving first condition information sent by the terminal;

[0388] determining first information according to the first condition information.

[0389] In some embodiments, the method further includes:

[0390] receiving signaling information sent by the terminal;

[0391] sending, according to the signaling information, a reference signal to the terminal based on a reference signal resource, the reference signal being used for the terminal to determine the first condition information by measuring the reference signal.

[0392] In some embodiments, the first information includes configuration parameter information of measured CSI and configuration parameter information of predicted CSI, and the method further includes:

[0393] receiving a plurality of performance detection results sent by the terminal, the plurality of performance detection results being generated by the terminal performing performance testing on a plurality of third AI models according to the first information, the third AI models being pre-trained AI models;

[0394] determining, according to the plurality of performance detection results, a first AI model from the plurality of third AI models;

[0395] sending, based on the first AI model, third information to the terminal, the third information being used to indicate the first AI model.

[0396] FIG. 5A is an interaction schematic diagram of a communication method according to an embodiment of the present disclosure. As shown in FIG. 5A, the embodiment of the present disclosure relates to a communication method, and the above method includes:

[0397] Step S5101, the network device configures parameter information to the terminal.

[0398] In some embodiments, the network configuration parameter information includes at least one of the following:

[0399] a number of reference signal resources for measuring CSI; a time domain behavior of each reference signal resource, such as a reference signal resource period, semi-persistent or aperiodic; an interval between adjacent reference signal resources;

[0400] a number of predicted CSIs; an interval between two adjacent predicted CSIs; an interval between a first predicted CSI and a CSI reporting time point.

[0401] indicating time point information for reporting the predicted CSI.

[0402] Optionally, in some embodiments, the network side first obtains the condition information of the terminal side before configuring the parameter information, and the model consistency can be determined based on the following procedures:

[0403] (1) The network configures the reference signal resource for obtaining the condition information of the terminal side, and then configures the parameter information based on the condition information of the UE side. Optionally, the condition information of the UE side is associated with an ID, which is not only associated with the channel condition of the terminal side, but also associated with the parameter information configured by the network side and / or the condition information of the network side. In the inference stage, the terminal determines which model in the training model to use for inference according to the associated ID, so as to ensure the consistency of the inference model and the training model. The associated ID can be configured by the network, reported by the terminal, or predefined by the terminal and the network.

[0404] (2) The terminal collects the training data set based on the configured parameter and reference signal resource information, and then trains the model based on the data set. In the inference stage, the terminal determines the AI model to be used for inference according to the condition information of the terminal side and the configuration parameter information of the network. For example, if the terminal actively obtains the terminal condition information, the terminal can actively send a signaling request to the network to configure the reference signal resource for measuring and obtaining the terminal condition information.

[0405] In step S5102, the terminal collects the training data set according to the parameter information, and then trains a corresponding AI model based on the training data set.

[0406] For example, in the inference stage, the terminal determines to use the corresponding training model for inference according to the configuration parameter information of the network, so as to ensure the consistency of the inference model and the training model.

[0407] Optionally, if the UE trains an AI model with good generalization based on different parameter information, and the network configuration parameter information is a subset of the parameter information corresponding to the trained model, then the UE will use the generalization AI model according to the configured parameter information.

[0408] In some embodiments, when collecting the training data set, the network can configure the time domain behavior of the reference signal resource for measuring the CSI to be one or a combination of periodic, semi-persistent, or aperiodic time domain behaviors.

[0409] Optionally, in some embodiments, each of the above-mentioned configuration parameter information combinations can be associated with an ID, which is not only associated with the network configuration parameter information, but also associated with the condition information of the network side. The associated ID can be configured by the network to the terminal, and the terminal determines which model in the training model to use for inference according to the configured associated ID, so as to ensure the consistency of the inference model and the training model.

[0410] Through the above manner, the terminal trains a model according to the parameter information configured by the network, obtains an AI model corresponding to the parameter information, and performs model inference based on the AI model to predict the CSI at future time points. Thus, the inference model and the training model are consistent, and the service performance of the AI model is ensured.

[0411] FIG. 5B is an interaction diagram of a communication method according to an embodiment of the present disclosure. As shown in FIG. 5A, the embodiment of the present disclosure relates to a communication method, and the method includes:

[0412] In step S5201, the network device sends, to the terminal, reference signal resources corresponding to the measured CSI and the predicted CSI.

[0413] In step S5202, the terminal tests the performance of each AI model according to the reference signal resources corresponding to the measured CSI and the predicted CSI, respectively, and generates performance detection results of the AI models.

[0414] For example, the terminal tests the performance of each AI model based on channel information estimated based on reference signals corresponding to the time points of the measured CSI and the predicted CSI. The terminal determines the consistency of the inference model and the training model according to an AI model performance index. The performance index can be predefined by the network or negotiated by the terminal and the network or determined by the terminal.

[0415] Optionally, in some embodiments, if the performance of each AI model tested does not meet the performance requirement, the UE sends the test results to the network, and the network determines whether to fall back to a traditional algorithm for CSI prediction or to configure parameter and resource information to enable the UE to generate a data set for training of the AI model to obtain a corresponding AI model for inference under the current parameter configuration.

[0416] In step S5203, the terminal sends the performance detection results to the network device.

[0417] If the consistency of the training model and the inference model is determined by the network side, the terminal needs to report the performance detection results of each AI model tested to the network device. The network device determines the consistency of the training model and the inference model according to the performance of each model received. Finally, the network device sends the determination result to the terminal. For example, the network device selects an AI model that meets the performance index from a plurality of AI models configured by the terminal according to the performance detection results, and sends indication information of the AI model to the terminal. The terminal determines to use the corresponding AI model for inference according to the indication information.

[0418] Through the above manner, the consistency of the inference model and the training model used by the terminal is solved, that is, the condition information of the inference model and the training model used is consistent, and the performance loss caused by the mismatch between the inference model and the training model is avoided.

[0419] The embodiments of the present disclosure further provide a device for implementing any of the above methods, for example, a device comprising units or modules for implementing the steps performed by the terminal in any of the above methods. For another example, another device is provided, comprising units or modules for implementing the steps performed by the network equipment (such as an access network device, a core network function node, a core network device, etc.) in any of the above methods.

[0420] It should be understood that the division of each unit or module in the above device is only a logical function division, and all or part of them can be integrated into one physical entity or physically separated in actual implementation. In addition, the units or modules in the device can be implemented in the form of processor invoking software: for example, the device comprises a processor connected with a memory, the memory stores instructions, and the processor invokes the instructions stored in the memory to implement any of the above methods or to implement the functions of each unit or module of the device, wherein the processor is a general processor such as a central processing unit (CPU) or a microprocessor, and the memory is a memory in the device or a memory outside the device. Alternatively, the units or modules in the device can be implemented in the form of hardware circuit, and the functions of part or all of the units or modules can be implemented by the design of the hardware circuit, and the hardware circuit can be understood as one or more processors; for example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC), and the functions of part or all of the units or modules are implemented by the design of the logical relationship of the elements in the circuit; for another example, in another implementation, the hardware circuit is a programmable logic device (PLD), and taking a field programmable gate array (FPGA) as an example, it can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to implement the functions of part or all of the units or modules. All units or modules of the above device can be implemented in the form of processor invoking software, or all units or modules can be implemented in the form of hardware circuit, or part of the units or modules are implemented in the form of processor invoking software, and the remaining part is implemented in the form of hardware circuit.

[0421] In the embodiments of the present disclosure, the processor is a circuit with signal processing capability. In one implementation, the processor can be a circuit with instruction reading and running capability, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), a digital signal processor (DSP), or the like. In another implementation, the processor can implement certain functions through a logical relationship of hardware circuits, and the logical relationship of the hardware circuits is fixed or can be reconfigured. For example, the processor is a hardware circuit implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In the reconfigurable hardware circuit, the processor loads a configuration document to implement the configuration of the hardware circuit. It can be understood that the processor loads instructions to implement the functions of the above part or all units or modules. In addition, the hardware circuit can also be designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), or the like.

[0422] FIG. 6 is a structural schematic diagram of a terminal according to the embodiments of the present disclosure. As shown in FIG. 6, the terminal 6100 can include a transceiver module 6101 and a processing module 6102. In some embodiments, the transceiver module 6101 is configured to receive first information sent by a network device, the first information including configuration parameter information for measuring channel state information (CSI) and / or configuration parameter information for predicting the CSI, and the processing module 6102 is configured to determine a first AI model according to the first information, the first AI model being used for CSI prediction. Optionally, the transceiver module 6101 and the processing module 6102 are configured to perform at least one of the determining and / or obtaining communication steps performed by the terminal 101 in any of the above methods, details are not described herein.

[0423] In some embodiments, the transceiver module can include a receiving module and a sending module, and the receiving module and the sending module can be separate or integrated together. Optionally, the sending module can be replaced by the transmitter. The receiving module can be replaced by the receiver.

[0424] In some embodiments, the processing module can be one module, or can include multiple sub-modules. Optionally, the multiple sub-modules perform all or part of the steps required to be performed by the processing module, respectively. Optionally, the processing module can be mutually replaced with the processor.

[0425] FIG. 7 is a structural schematic diagram of a network device according to an embodiment of the present disclosure. As shown in FIG. 7, the network device 7100 can include a transceiver module 7101. In some embodiments, the transceiver module 7101 is configured to send first information to a terminal, the first information including configuration parameter information for measuring CSI and / or configuration parameter information for predicting CSI, and the first information is used by the terminal to determine a first AI model according to the first information, and the first AI model is used for CSI prediction. Optionally, the transceiver module 7101 is configured to perform at least one of the determining and / or obtaining and / or the like communication steps performed by the network device 102 in any of the above methods, which will not be described herein.

[0426] In some embodiments, the transceiver module can include a receiving module and a sending module, which can be separate or integrated together. Optionally, the sending module can be mutually replaced with the transmitter. The receiving module can be mutually replaced with the receiver.

[0427] FIG. 8 is a structural schematic diagram of a communication device 8100 according to an embodiment of the present disclosure. The communication device 8100 can be a network device (such as an access network device, a core network device, etc.), a terminal (such as a user equipment, etc.), a chip, a chip system, or a processor supporting the network device to implement any of the above methods, or a chip, a chip system, or a processor supporting the terminal to implement any of the above methods. The communication device 8100 can be used to implement the methods described in the above method embodiments, and specific implementation can be referred to the descriptions in the above method embodiments.

[0428] As shown in FIG. 8, the communication device 8100 includes one or more third processors 8101. The third processor 8101 can be a general-purpose processor or a special-purpose processor, for example, a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the communication device (such as a base station, a baseband chip, a terminal device, a terminal device chip, a DU or a CU, etc.), execute programs, and process data of the programs. Optionally, the communication device 8100 is configured to perform any of the above methods. Optionally, the one or more third processors 8101 are configured to invoke instructions to cause the communication device 8100 to perform any of the above methods.

[0429] In some embodiments, the communication device 8100 further includes one or more third transceivers 8102. When the communication device 8100 includes one or more third transceivers 8102, the third transceiver 8102 performs at least one of the communication steps of sending and / or receiving in the above-described methods, and the third processor 8101 performs at least one of the other steps. In optional embodiments, a transceiver can include a receiver and / or a transmitter, which can be separate or integrated together. Optionally, the terms transceiver, transceiving unit, transceiver, transceiving circuit, interface circuit, interface, and the like can be replaced with each other, the terms transmitter, transmitting unit, transmitter, transmitting circuit, and the like can be replaced with each other, and the terms receiver, receiving unit, receiver, receiving circuit, and the like can be replaced with each other.

[0430] In some embodiments, the communication device 8100 further includes one or more third memories 8103 for storing data. Optionally, all or part of the third memory 8103 can also be outside the communication device 8100. In optional embodiments, the communication device 8100 can include one or more first interface circuits 8104. Optionally, the first interface circuit 8104 is connected to the third memory 8103, and the first interface circuit 8104 can be used to receive data from the third memory 8103 or other devices, and can be used to send data to the third processor 8101 or other devices. For example, the first interface circuit 8104 can read the data stored in the third memory 8103 and send the data to the third processor 8101.

[0431] The communication device 8100 described in the above embodiments can be a network device or a terminal, but the scope of the communication device 8100 described in the present disclosure is not limited thereto, and the structure of the communication device 8100 can not be limited by Figure 8. The communication device can be a standalone device or can be part of a larger device. For example, the communication device can be: 1) a standalone integrated circuit (IC), or a chip, or a chip system or subsystem; (2) a set of one or more ICs, which can optionally include storage components for storing data, programs; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, a terminal device, a smart terminal device, a cellular phone, a wireless device, a handset, a mobile unit, a vehicle-mounted device, a network device, a cloud device, an artificial intelligence device, and the like; (6) others, and the like.

[0432] Figure 9 is a structural schematic diagram of a chip 8200 according to an embodiment of the present disclosure. For the case where the communication device 8100 can be a chip or a chip system, reference can be made to the structural schematic diagram of the chip 8200 shown in Figure 9, but not limited thereto.

[0433] The chip 8200 includes one or more fourth processors 8201. The chip 8200 is configured to perform any of the above methods.

[0434] In some embodiments, the chip 8200 further includes one or more second interface circuits 8202. Optionally, the terms interface circuit, interface, transceiver pin, etc. can replace each other. In some embodiments, the chip 8200 further includes one or more fourth memories 8203 configured to store data. Optionally, all or part of the fourth memory 8203 can be outside the chip 8200. Optionally, the second interface circuit 8202 is connected with the fourth memory 8203, the second interface circuit 8202 can be configured to receive data from the fourth memory 8203 or other devices, and the second interface circuit 8202 can be configured to send data to the fourth memory 8203 or other devices. For example, the second interface circuit 8202 can read the data stored in the fourth memory 8203 and send the data to the fourth processor 8201.

[0435] In some embodiments, the second interface circuit 8202 performs at least one of the communication steps such as sending and / or receiving in the above methods. The second interface circuit 8202 performs the communication steps such as sending and / or receiving in the above methods, for example, means that the second interface circuit 8202 performs data interaction between the fourth processor 8201, the chip 8200, the fourth memory 8203 or the transceiver device. In some embodiments, the fourth processor 8201 performs at least one of the other steps.

[0436] The modules and / or devices described in each of the embodiments of the virtual device, the physical device, the chip, etc. can be combined or separated according to the situation. Optionally, part or all of the steps can also be performed by multiple modules and / or devices in cooperation, which is not limited here.

[0437] The disclosure also proposes a storage medium, and the above storage medium stores instructions, when the instructions run on the communication device 8100, the communication device 8100 performs any of the above methods. Optionally, the above storage medium is an electronic storage medium. Optionally, the above storage medium is a computer readable storage medium, but is not limited to this, it can also be a storage medium readable by other devices. Optionally, the above storage medium can be a non-transitory storage medium, but is not limited to this, it can also be a transitory storage medium.

[0438] The disclosure also proposes a program product, and the above program product is executed by the communication device 8100, so that the communication device 8100 performs any of the above methods. Optionally, the above program product is a computer program product.

[0439] The disclosure also proposes a computer program, when it runs on a computer, it makes the computer perform any of the above methods.

Claims

A communication method characterized by comprising: The method is performed by a terminal, and the method comprises: receiving first information sent by a network device, the first information comprising configuration parameter information of measured channel state information (CSI) and / or configuration parameter information of predicted CSI; determining a first AI model according to the first information, the first AI model being used for CSI prediction. The method of claim 1, wherein The determining of the first AI model according to the first information comprises: obtaining a first training data set according to the first information; performing model training on a second AI model according to the first training data set to generate the first AI model, the second AI model being a preset AI model. The method of claim 1, wherein The determining of the first AI model according to the first information comprises: determining that the first information is a subset of configuration parameters of a second training data set, the second training data set being a training data set corresponding to a third AI model, the third AI model being obtained by training according to the second training data set; taking the third AI model as the first AI model. The method according to any one of claims 1-3, characterized in that The first information comprises at least one of: a number of reference signal resources for measuring CSI; time domain behavior of each reference signal resource; a time domain interval between adjacent measured CSIs; a number of predicted CSIs; a time domain interval between adjacent predicted CSIs; a time domain interval between a first predicted CSI and a predicted CSI reporting time; time information of reporting a predicted CSI. The method according to claim 4, characterized in that The time domain behavior comprises at least one of: periodic time domain behavior; semi-persistent periodic time domain behavior; aperiodic time domain behavior. The method according to claim 4 or 5, characterized in that The first information comprises an association ID, the association ID being used for indicating at least one of: first condition information of the terminal, the first condition information being used for indicating configuration information or a running state of the terminal; a combination mode of the measured CSI configuration parameter information and / or the predicted CSI configuration information; second condition information of the network device, the second condition information being used for indicating a configuration state of the network device. The method according to claim 6, characterized in that The determining of the first AI model according to the first information comprises: determining an AI model matched with the association ID from a plurality of third AI models as the first AI model, the third AI model being a pre-trained AI model. The method according to any one of claims 6 or 7, characterized in that The method further comprises: sending the first condition information to the network device, the first condition information being used for the network device to determine the first information according to the first condition information. The method of claim 8, wherein The method further comprises: sending signaling information to the network device, the signaling information being used for requesting the network device to send a reference signal to the terminal based on a reference signal resource; receiving the reference signal sent by the network device; measuring the reference signal to determine the first condition information. The method of claim 1, wherein The first information comprises the configuration parameter information of the measured CSI and the configuration parameter information of the predicted CSI, and the determining of the first AI model according to the first information comprises: performing performance detection on a plurality of third AI models according to the first information to generate a plurality of performance detection results corresponding to the plurality of third AI models, each third AI model being a pre-trained AI model; According to the multiple performance detection results, the first AI model is determined. The method of claim 10, wherein The determining of the first AI model according to the multiple performance detection results comprises: According to the multiple performance detection results, an AI model meeting the set performance requirement is determined from the multiple third AI models as the first AI model. The method of claim 11, wherein The method further comprises: sending second information to the network device, the second information being used to indicate the first AI model. The method of claim 10, wherein The determining of the first AI model according to the multiple performance detection results comprises: According to the multiple performance detection results, it is determined that multiple AI models in the multiple third AI models all meet the set performance requirement. sending the multiple performance detection results to the network device, the multiple performance detection results being used for the network device to determine the first AI model from the multiple third AI models; receiving third information sent by the network device, the third information being used to indicate the first AI model; determining the first AI model according to the third information. The method of claim 10, wherein The determining of the first AI model according to the multiple performance detection results comprises: According to the multiple performance detection results, it is determined that the multiple third AI models all do not meet the set performance requirement. sending the multiple performance detection results to the network device, the multiple performance detection results being used for the network device to determine the first AI model from the multiple third AI models; receiving third information sent by the network device, the third information being used to indicate the first AI model; determining the first AI model according to the third information. A communication method characterized by comprising: The method comprises: sending first information to a terminal, the first information comprising configuration parameter information of measured CSI and / or configuration parameter information of predicted CSI, the first information being used for the terminal to determine a first AI model according to the first information, the first AI model being used for CSI prediction. The method of claim 15, wherein The first information comprises at least one of the following: a number of reference signal resources for measuring CSI; a time domain behavior of each reference signal resource; a time domain interval between adjacent measured CSIs; a number of predicted CSIs; a time domain interval between adjacent predicted CSIs; a time domain interval between a first predicted CSI and a predicted CSI reporting time; time information of reporting a predicted CSI. The method of claim 16, wherein The time domain behavior comprises at least one of the following: periodic time domain behavior; semi-persistent periodic time domain behavior; aperiodic time domain behavior. The method according to claim 16 or 17, characterized in that The first information comprises an association ID, the association ID being used to indicate at least one of the following: first condition information of the terminal, the first condition information being used to indicate configuration information or a running state of the terminal; a combination mode of the measured CSI configuration parameter information and / or the predicted CSI configuration information; second condition information of the network device, the second condition information being used to indicate a configuration state of the network device. The method of any one of claim 18, wherein The method further comprises: receiving the first condition information sent by the terminal; determining the first information according to the first condition information. The method of claim 19, wherein The method further comprises: receiving signaling information sent by the terminal; According to the signaling information, a reference signal is sent to the terminal based on a reference signal resource, and the reference signal is used for the terminal to measure the reference signal and determine the first condition information. The method of claim 15, wherein The first information includes configuration parameter information of the measured CSI and configuration parameter information of the predicted CSI, and the method further includes: Receiving a plurality of performance detection results sent by the terminal, the plurality of performance detection results being generated by the terminal performing performance testing on a plurality of third AI models according to first information, the third AI models being pre-trained AI models; According to the plurality of performance detection results, determining the first AI model from the plurality of third AI models; Based on the first AI model, sending third information to the terminal, the third information being used to indicate the first AI model. A terminal, characterized by comprising: Including: The transceiver module is configured to receive first information sent by a network device, the first information including configuration parameter information of measured channel state information (CSI) and / or configuration parameter information of predicted CSI. The processing module is configured to determine a first AI model according to the first information, the first AI model being used for CSI prediction. A network device, characterized in that Including: The transceiver module is configured to send first information to a terminal, the first information including configuration parameter information of measured CSI and / or configuration parameter information of predicted CSI, the first information being used by the terminal to determine a first AI model according to the first information, the first AI model being used for CSI prediction. A terminal, characterized by comprising: Including: One or more processors; The terminal is configured to perform the communication method of any one of claims 1-14. A network device, characterized in that Including: One or more processors; The network device is configured to perform the communication method of any one of claims 15-21. A communication system characterized by Including: A terminal is configured to perform the method of any one of claims 1-14. A network device is configured to perform the method of any one of claims 15-21. A storage medium storing instructions, the storage medium storing instructions, characterized in that, When the instructions run on a communication device, the communication device is caused to perform the communication method of any one of claims 1-14, or the communication device is caused to perform the communication method of any one of claims 15-21. A computer program product comprising computer programs and / or instructions, characterized in that, The computer program and / or instructions, when executed by a communication device, implement the communication method of any one of claims 1-14, or the computer program and / or instructions, when executed by a communication device, implement the communication method of any one of claims 15-21.

Images